Flood Mitigation on the Raritan River Initial Progress Report Prepared for the Department of Homeland Security FEMA Region II by Rutgers the State University of New Jersey December 20 2011 Submitted by Judy Shaw PhD Principal Investigator On behalf of Joseph J Seneca PhD Fred Roberts PhD Paul Kantor PhD Qizhong Guo PhD David Robinson PhD Will Irving Stacy Perrine Kate Davidoff

2

SUMMARY FOR INITIAL REPORT

The following is the first report (December 1 2011) for the FEMA-Rutgers project on Flood Mitigation on the Raritan River We have completed the initial work as described in the contract and continue to collect additional data that we believe may be useful to the project Task One Community Assessment This includes profiles on each of the communities in the Raritan watershed including census reports (mayors years in office and municipal engineers) flooding data (insurance payouts) and social environmental and political lsquomedia snapshotsrsquo In addition we have created several maps to help us understand the regional impacts of flooding and possible areas to consider in future planning maps by watershed region with overall payout data and regional planning area maps based on local zoning

We have also identified the key regional stakeholders and are in the process of conducting interviews

Task Two Economic Modeling The economic modeling element of this project is designed to look at best ways to estimate benefits to municipalities (and larger governmental units) of flood mitigation projects for the Raritan watershed For this section two draft reports are provided a flood expense taxonomy (separate spreadsheet) and a literature review Task Three Ideas on Risk Modeling The third element of the project includes the first stages of the development of a risk analysis identification model to focus on all hazardsrisks The identification of data to be analyzed will be developed by experts in both mathematical modeling and human health and environmental modeling from engineering and public policy perspectives The team initially Fred Roberts and Paul Kantor now includes Qizhong Guo Associate Professor of Civil and Environmental Engineering and David Robinson NJ State Climatologist and Geography Professor

3

Contents Introduction 3

Task One Community Assessment 4

Task Two Economic Modeling 8

Task Three Ideas on Risk Assessment for Flooding on the Raritan 9

Appendix A News Snapshots 14

Appendix B-1 Safety UCR 20

Appendix B-2 Safety UCR Detail 23

Appendix C Maps 24

Appendix D Literature 30

Appendix E Guo CV 31

Appendix F Robinson CV 36

Introduction Traditionally national flood mitigation strategies sought to move flood waters away from vulnerable communities Today FEMA explores how reducing the volume of water moving during flood events can deliver cost-effective results Working with Rutgers University FEMA seeks to use public dialogue in the Raritan River watershed and the collective expertise of leaders across the university research community to evaluate flood mitigation from engineering social environmental political and economic perspectives

The project hopes to provide strategies that will improve mitigation through partnerships Those partnerships span public-private and authorities crossing all levels of government and include key stakeholders within the watershed Using the strategies developed here FEMA hopes to improve understanding information sharing governance and partnering in this critical area The successful model will be transferrable to other watersheds both smaller and larger

Rutgers University through the E J Bloustein School of Planning and Public Policy engaged partners from across the University to bring expertise in regional planning public engagement communication economics mathematics engineering and geography into this effort Together this team will develop strategies and models to help validate this approach

This initial report provides status on the three major elements of the project Community Economics and Risk This initial report is provided to illustrate the direction being taken within each project element and to give the FEMA partners and other advisors within the Collaboratorium an understanding of our progress and the challenges we have encountered

This report also includes draft spreadsheets attached as separate files

Spreadsheet 1 Demographic Data Spreadsheet 2 Flood Expense Taxonomy

4

Task One Community Assessment The deliverables for Community Assessment are to develop a background of census data a general report on each of the ninety-nine Raritan Basin municipalities (including watershed area maps for certain data) and the initial stages of a series of interviews with key municipal stakeholders The data presented here is preliminary and will continue to be updated throughout the project For all of this work the study area is the Raritan River Basin and is presented either by municipality or by watershed management area (WMA) WMA 8 ( the North and Upper Branches) WMA 9 (Lower Raritan) and WMA 10 (the Millstone River)

Demographics

The first step in the Community Assessment is a report on certain Census data for the ninety-nine towns in the watershed as described in Table 1 below The report on this work is presented in Spreadsheet 1 Demographic Data (See Separate Spreadsheet File)

Municipality County by watershed Population Upper Raritan (8) 1990 2000 2010 Lower Raritan (9) Change in Population 00-10 Millstone (10) Wealth Education (25 yo and up) Med household income high school graduate or higher Average household income Bachelorrsquos degree or higher Med Home Value (Owner Occupied) Employment (16 yo and up) Insurance of workforce employed Policies in force as of 83111 commuting 1hour+ to work Claims reported to FEMA 1178-83111 Workforce bull Total Losses Population over 16 yo bull Closed Losses Pop over 16 as of workforce bull Open Losses bull CWOP Losses Government bull Total Payments Mayor Term Engineer

Table 1 Demographic Data 1990 2000 2010

5

Map 1 Watershed Management Area 8 by FEMA Payout

Map 2 Watershed Management Area 9 by FEMA Payout

In addition we reviewed media coverage for these municipalities looking at social environmental and political issues That initial report is provided ldquoSocial Environmental and Political Snapshotsrdquo and is attached as Appendix A A summary of Safety produced from the NJ State Police Uniform Crime Reports is presented in Attachment B Data Maps The second component of the initial phase of the Community Assessment looks at the geography of the region to identify areas where enhanced mitigation strategies might be most productive in reducing impacts to the most affected communities This resulted in several series of maps for the region

NFIP Payout Maps The first set of maps display the National Flood Insurance Program (NFIP) payouts in the study region by watershed management area delineated by the NJDEP and NJ Geological Survey) The payout amounts were joined to the maps using the NFIP statistics labeled Loss Statistics New Jersey as of 83111 This information does not take into account damages sustained by Hurricane Irene but it does offer a 33 year history of the NFIP loss claims and payouts to policy holders from 1978 to immediately prior to that event The maps were color coded to easily display the breakdown in payouts in the individual municipalities in each watershed The data shows some municipalities suffer dramatically more loss than some of their neighbors particularly in WMA 9 where Bound Brook has a cumulative amount of NFIP payouts totaling over $53M while other municipalities in the watershed Metuchen Borough for example have under $10000 in total NFIP payouts in the same period of time The data are presented in Maps 1 2 and 3 below and in full size in Appendix C

Land Use Maps To further examine opportunities for enhanced prevention of damage from flood waters we created three maps to explore land use (Maps 4 5 and 6) These maps were created from the New Jersey State Development and Redevelopment Plan (SDRP) administered by the NJ Office for Planning Advocacy which delineates planning areas throughout the state The SDRP seeks to provide a

Map 3 Watershed Management Area 10 by FEMA Payout

6

balance between growth and conservation by designating planning areas that share common conditions with regard to development and environmental features These planning areas (PAs) are as follows

Growth Areas PA1 Metropolitan Planning Areas(up to 80-90 impervious coverage with development) PA2 Suburban Planning Areas (up to 30 impervious coverage with development) Designated Centers

Limited Growth Areas planning in these planning areas should promote a balance of conservation and limited growth PA3 Fringe Planning Areas PA4 Rural Planning Areas PA5 Environmentally Sensitive Planning Areas Conservation Areas (below 30 impervious coverage with development and limits or excludes sewer infrastructure) PA3 Fringe Planning Areas PA4 Rural Planning Areas PA5 Environmentally Sensitive Planning Areas

Map 6 WMA 10 Planning Areas

Map 5 WMA 9 Planning Areas

Map 4 WMA 8 Planning Areas

7

Interviews

The second phase of this work currently underway involves in-person interviews with 20-25 key regional municipal officials to establish the following

bull Importance of flood mitigation among other concerns schools police fire service traffic management trash collection parking street repair housing maintenance and repair aesthetics and other quality of life issues

bull What they know about flood mitigation flooding the degree of damage and the causes bull Their understanding of climate impacts on flooding and mitigation strategies bull Local stormwater management measures (retention basin management to rain barrels bull Preferred flood mitigation strategies both locally and regionally bull Level of local engagement and what causes citizens to become engaged in local issues

At this time the following municipalities are to be interviewed with either the mayor the administrator and or the director of the local Office of Emergency Management

Municipality CO Municipality CO Berkeley Heights Union Middlesex Boro Msx Bound Brook Mayor amp OEM Som Milltown Msx Branchburg Som Monroe (Engineer Ernie Feist) Msx Clinton Township Hun North Plainfield Msx Edison John Medina Msx Scotch Plains Union Franklin Som Raritan Township Hun Hightstown Mer Watchung Som Hopewell Mer Washington Twp Morris Manville Administrator amp OEM Som Woodbridge Msx Marlboro Mon

Next Steps

The next steps will involve closer analysis of the census data as we proceed with municipal interviews The goal is to talk to people at various levels not just those who are most directly affected but also those who may be upstream from more impacted communities The next mapping exercise will be to add contours in order to verify the drainage and to calculate current potential mitigation in the uppermost communities through major stormwater mitigation efforts The use of impervious surface coverages from the state as well as the application of independent models will lend more insight into the potential for stormwater management BMPs to aid in overall impact reduction

The findings of the interviews will form a discussion guide at a dinner meeting with municipal and county leaders at the Bloustein School on Tuesday January 31 2012 The discussion will focus on further action for a regional volume-focused flood mitigation strategy Focus Groups Several focus groups are also scheduled Milltown and Manville are scheduled for January Two others will be held in February and March These meetings will be held with groups of key citizen leaders in selected municipalities to determine their knowledge of the issue significance to them their desire for further information about the subject and desired mode of transmission of the information and willingness to become involved in the issue in their own town These sessions will also address the questions asked of the initial stakeholders to verify unity of thought and understanding of issues across the community

8

In addition a series of meetings with regional engineers are being held to engage them in the mitigation conversation from the perspective of stormwater management and overall adaptation of green infrastructure practices Their inputs will be gathered in January and February through a series of conference calls on the following topics

bull regulation changes bull policy directions bull professional education needs bull public education needs bull financing mitigation

They will meet together in early March to share their findings and engage with municipal officials and another cohort regional business leaders

Pilot Survey A draft survey will be created and a pilot conducted for a larger regional survey This would be tested with a limited number of residents of the basin to determine their knowledge of the issue significance to them their desire for further information about the subject and desired mode of transmission of the information and willingness to become involved in the issue in their own towns A larger application of the survey would not be fully implemented as part of this project

Task Two Economic Modeling The economic modeling element of this project is designed to look at best ways to estimate benefits to municipalities (and larger governmental units) of flood mitigation projects for the Raritan watershed For this section two draft reports are provided a flood expense taxonomy (separate spreadsheet) and a literature review

Flood Expense Taxonomy

The flood expense taxonomy (see Separate Spreadsheet File) provides a framework for identifying and categorizing the costs associated with flooding events We developed the taxonomy drawing from academic literature prior reports of flooding events and newspaper articles covering flooding events The taxonomy identifies potential flood-related expenses for four sectors of the economy households commercial businesses municipalities and agriculture The potential costs to each sector are cross-categorized with respect to the time sequence of the flooding event pre-flooding flooding and post flooding We will continue to identify additional potential costs and where possible populate those findings with estimated costs Additionally we will add to this taxonomy the potential economic benefits (if any) to the same four sectors as a result of removing structures located in the floodplain

Literature Review

The annotated literature review (Attachment D) includes two types of articles those that examine the relationship between floods and housing prices and those that estimate the costs of flooding on municipalities and their residents and businesses The articles that examine the relationship between flooding events and home prices generally use hedonic pricing models to explain the difference in home value (as measured by sales price) between homes that are on the floodplain and similar houses that are not on it The articles generally indicate that prices for homes located on floodplains are discounted relative to similar homes not on the floodplain One article offers a meta-analysis of the empirical literature on home prices and flooding events The articles in our taxonomy examine the economic impacts of flooding events from England the Red-River Basin in the Western US and the Delaware

9

River in New Jersey We also include a meta-analysis that re-estimates National Weather Service estimates of national economic damages We will build on this literature review to develop a better understanding of the existing estimates of the full economic costs of flooding events and the potential benefits of removing structures (private and public) from floodplains

Task Three Ideas on Risk Assessment for Flooding on the Raritan The third element of the project includes the first stages of the development of a risk analysis identification model to focus on all hazardsrisks The identification of data to be analyzed will be developed by experts in both mathematical modeling and human health and environmental modeling from engineering and public policy perspectives The team initially Fred Roberts and Paul Kantor now includes Qizhong Guo Associate Professor of Civil and Environmental Engineering and David Robinson and State Climatologist and Geography Professor Brief bio sketches are attached as Appendix E Qizhong Guo and Appendix F David Robinson)

Overview

We have spent the past few months researching alternative approaches to flood risk assessment and gathering as much background information as possible We have gotten information from numerous individuals in the US and abroad and numerous agencies including New Jersey Department of Environmental Protection Department of Homeland Security Risk Management and Analysis USGS We came to understand that it was important to expand our team and so we have added a climatologist Professor David Robinson of the Rutgers Department of Geography the New Jersey State Climatologist and a hydrologist Professor Qizhong Guo of the Rutgers Department of Civil and Environmental Engineering We present here a high level sketch of the approach we are examining and discuss the kinds of detailed information that its implementation will require

Problem Formulation

Our multidisciplinary discussions have led us to understand many of the subtleties involved in doing a risk assessment for flooding on the Raritan We have nearly completed outlining an approach based on what we have learned and the next steps will be to share that approach with those working on other tasks in the project revise as needed and then begin gathering the data needed to do the required modeling

To explain our approach and the issues we have to resolve let us consider first a diagram (Figure 1) which illustrates the kinds of factors and tools we need to take into account in understanding flood modeling and applying that modeling to a risk assessment

10

Figure 1 The interaction of nature models and controllable activities (shown in yellow)

Engineering models can combine information on rainfall specified as to time and space with information on soil moisture conditions seasonality and river levels prior to the rain event the fixed properties of the watershed (not indicated) and the land cover (natural and built environment) to produce flood inundation maps To obtain the precision needed to specify events that might occur every five or ten years requires LiDAR information on elevations We refer to the soil moisture conditions and river levels as ldquoantecedent eventsrdquo and note that seasonality is important because it affects the ability of the soil to absorb water (due to vegetation cover freezing etc) The land coverenvironment is viewed as modifiable (yellow) over the span of a few years That would in turn move the flood contours Flood inundation maps can be used for insurance andor regulatory purposes for risk assessments and to communicate complex geology and meteorology to stakeholder groups and thereby facilitate consideration of alternative policies

In the simplest case we might think of four types of soil moisture conditions dry ldquoaveragerdquo wet or frozen Probably it is sufficient to think of four different seasons We might also think of three basic river levels low average and high Thus there are in principle 4 x 4 x 3 = 48 different combinations of these conditions In reality not all of these combinations are feasible We wonrsquot have frozen ground in summer In our analysis we will probably demonstrate the methods choosing only several of the sensible alternatives for detailed investigation

Risk Modeling

The basic outline for how DHS does risk assessment is shown in Figure 2 below which is taken from a presentation by Isaac Maya of the CREATE Center It defines ldquoriskrdquo R as a function of threat T vulnerability V and consequence C R = f(TVC) Often this is represented as a product R = T x V x C Generally the factors T and V are some kinds of probabilities while C is some measure of loss In our case ldquoattackrdquo is interpreted to mean weather event and ldquosuccessrdquo is interpreted to mean flooding

Precipitation(xt) )tthe past

The models

Soil moisture conditions + seasonality + river levels seasonality

Land Cover (natural + built environment)

River gauge levels

LiDAR information

Flood Contour Maps

waterway

5 year contour

10 year contour

11

Figure 2 Sketch of the DHS computation of risk

There are two branches to the tree in this figure The top branch is the ldquostatus quordquo and the bottom branch involves some mitigation strategy or combination of such strategies The probability of attack P is the probability that there will be a weather event of the particular kind being considered and the probability of ldquosuccessrdquo Q is the conditional probability that such an event if it does occur will lead to flooding at a site being considered The consequences could be of various types In the case of floods they include loss of life economic damage to homes and businesses (direct or indirect) and psychological damage There is an attempt to put each of these types of consequences in terms of dollars The consequence (cost L) of an attack (flood) is usually calculated as a weighted sum of the different kinds of costs in our case of the cost in terms of lives lost economic damage and psychological damage What weights one uses and whether it makes sense to add up these weighted values are issues that will be viewed differently by different stakeholder groups

Complexities of the Watershed Situation

In fact we expect to be able to do more than simply treat ldquofloodrdquo as a yes or no matter as is done in this tree diagram Flood inundation maps will present levels of flood waters and the losses incurred change if there is a different level of flooding (or if there is a different duration for the flooding a factor we will disregard for now) The flood inundation levels are numerical values obtained from hydrological models (they can be later translated into major moderate minor near flood stage and none in making flood forecasts) A more complete model would replace the ldquosuccessfailurerdquo bifurcation in the tree diagram with a number of branches depending on the levelcategory of flood as measured for example by river gauges In our case it makes sense to define category 0 as no flooding category 1 as 0 to 1 feet category 2 as 1 to 2 feet etc

In practice obtaining the flood inundation levels may be complicated even before we introduce mitigating strategies The inundation maps are maps for theoretical storms not maps used to predict floods from an approaching storm These maps for the storms of a 100-year and 500-year ldquoreturn periodrdquo1 are already available and are being updated by FEMA using the recently available high resolution LiDAR data on the topography These maps are called flood insurance rate maps (FIRM) However they are developed from a kind of discrete-event modeling and may conceal hidden

1 This technical expression can be interpreted as representing a 1 of 15 probability per year

Success

Failure Attack

Vulnerability

No Attack

No Risk Management Plan

Threat of Attack

Success

Failure

Attack

No Attack

Risk Management Plan

Risk Management

PlansConsequences

L = Expected[$-Eqiuvalent Loss]

No Loss

No Loss

C + LC = Cost of Risk Management Plan

L = Expected[Reduced Loss] where L lt L

C

C

PQ

(1 - P)

(1 - Q)

P where P lt P

(1 - P)

Q where Q lt Q

(1 - Q)

12

uncertainties Moreover the inundation maps for more frequent storms are not readily available or non-existent It would be possible to build flood maps for 2-year 5-year 10-year 25-year and 50-year storms from available data This would imply gathering data and modeling and will depend on having access to the data Moreover the discrete-event models do not take into account antecedent conditions or seasonality Continuous modeling could take these factors into account but the cost of modeling would then be much higher

Assuming we can get flood inundation levels from hydrological models we would try to calculate the probability Qi that a flood of category i occurs Each category of flooding i would have a level of loss Li associated with it The level of flooding from a given amount of rainfall over a period of time depends not only on the precipitation but also on the soil moisture conditionsseasonalityriver levels and on the land cover (natural + built environment) Initially we may assume a fixed land cover (although this can be changed by mitigation) As noted above there are many different combinations of parameters for soil moisture conditions seasonality river levels We could in principle estimate the probability of each of these combinations (but would need to work out a method for doing so) For each combination we could use the hydrological models to calculate the flood level and thus the flood category Putting all of these calculations together will allow us to calculate the Qi parameters This is probably much too complicated and certainly much too expensive Instead we will pick out several more common combinations among the sensible ones and combine them with different precipitation patterns to define a set of weather events

In short our threats will be different kinds of weather events defined in terms of the profile of rainfall over a given period of time (say in number of inches in first hour number in second hour etc over a 24 hour period) plus antecedent precipitation and seasonal effects and pre-existing river levels We will aim to identify six to ten weather scenarios to analyze For each the probability P will have to be determined Information about precipitation totals in 24 hour periods is available and we could then use this to define a 2-year 5-year 10-year 25-year 50-year 100-year event in terms of precipitation profile per 24-hour period The probability of a 5-year event would be 15 of a 50 year event 150 etc except for the other factors (antecedent events plus seasonality) So we may want to modify P based on these other factors We will need to determine how to do that Note also that climate change will likely affect the value of P but we do not know how to calculate modified P with any degree of confidence

Now what is the interpretation of the Qi parameters If the weather event (precipitation profile) is defined and the antecedent conditions and seasonality are known in principle the hydrological models will produce just one category of flood But it could be that it makes more sense to build some uncertainty into the predictions of the models or it may make sense to assume that we have some uncertainty in knowing the soil moisture conditions or the river levels Thus there is surely still some value to studying the risk assessments using different flood categories

When there is a mitigation strategy presumably the probability of an attack would not be affected so Prsquo in the diagram is the same as P not less than P The probability of ldquosuccessrdquo or flood given a weather event is Qrsquo which is less than Q If we use different flood level categories then we would replace each Qi with a number Qirsquo but note that Qirsquo would be higher than Qi for some i Also the consequences of a flood could be less with expected consequences now Lrsquo presumably less than L (or Lirsquo presumably less than Li) But there is a cost C of the mitigation strategy and that cost applies even if there is no flood

Still using the notation in the diagram we see that if f(TVC) = TxVxC then with no mitigation strategy the risk or total expected cost is PxQxL but if there is a mitigation strategy then (since P = Prsquo) the risk or total expected cost is C + PxQrsquoxLrsquo The difference G = PxQxL ndash (C + PxQrsquoxLrsquo) in these expected costs would be the expected gain or expected risk reduction due to the strategy (If we have different flood level categories the calculation would be modified appropriately)

13

Mitigation Strategies and Elicitation of Some Key Parameters

We will need to decide if we look at flood damage (vulnerability) to a larger region or concentrate on a few towns Our preliminary plan is to do the analysis for a few sample towns for instance Manville and Millstone We will need to coordinate with the other parts of the project on this decision

Consequences will need to be measured by our partners in the project and we will need to coordinate with them

We will look at a variety of mitigation strategies Among them could be improved flood forecasting and warning systems retrofit of green infrastructure (eg impervious area disconnection rain gardens cisterns and rain barrels green roofs pervious concrete) to existing development areas to reduce upstream runoff volume regional (upstream) stormwater detention basins acquisition and demolition and relocation of properties flood-proofing and elevation in place disaster preparedness and response plans flood insurance cleanout of urban drainage systems construction of dams and reservoirs construction of dikes levees and floodwalls channel alterations and high flow diversions and spillways We will want to coordinate with other parts of the project on which mitigation strategies we look at

We will need to work with other parts of the project to get ballpark estimates of reduced costs Lrsquo (or Lirsquo) of floods

Under a given mitigation strategy the challenge is to produce inundation maps that will allow us to determine vulnerability (whether there is a flood) In principle there are hydrological models for doing this or such models could be developed but to do the actual calculations will be much more expensive than we can afford in this modest project Thus the best we can do is pick a few sample mitigation strategies for which we could make educated guesses as to vulnerabilities ndash just to illustrate the methodology Our thoughts are to start with three types of strategies retrofit of green infrastructure some sort of upstream mitigation that would have impact on downstream flooding and improved forecasting (eg improved monitoring from an optimum number of precipitation gauges merged station and radar precipitation estimates soil moisture observations evapotranspiration estimates and additional stream gauges) The last of these three would have no impact on the probabilities Q or Qi and also will not require changed inundation maps Only our calculation as to consequence would be affected since the whole idea of improved forecasting would be to allow preparation so as to reduce the potential flood damage

A full risk assessment would want to look not at individual mitigation strategies but at ldquoportfoliosrdquo of mitigation strategies Of course all of what was said above about ldquomitigation strategyrdquo could apply to a portfolio of strategies But the inundation maps will be much harder to obtain for portfolios of strategies than for individual ones As noted some of the mitigation strategies will not affect inundation maps Looking at mitigation strategies separately avoids the complication that several of these strategies may interact with each other in complex ways Some portfolios of mitigation strategies will be too expensive or infeasible for other reasons We may want to look at budgetary constraints and try to look at a few simple portfolios of strategies that have total investment cost (sum of Crsquos for each strategy in the portfolio) less than a given amount Then the question will become What is a measure of the value of a given portfolio Are we trying to maximize total expected gain (sum of G for all strategies in the portfolio) Or are we optimizing something else Or is this a ldquomulti-attributed utilityrdquo problem

14

Social Environmental and Political News Snapshots2 Hunterdon Mercer Middlesex Monmouth Morris Somerset Union

RARITAN RIVER TOWN SOCIAL ENVIRONMENTAL AND POLITICAL NEWS SNAPSHOTS 34

Alexandria 4-way stop to be set up near Delaware Valley High School

Bedminster

bull Clarence Dillon Library in Bedminster to host a series of Salons (92711) bull Bedminster bicyclist killed after crash with truck in Far Hills (12111)

Berkeley Heights Berkeley Heights childrenrsquos library bounces back Bernards Outraged NJ towns ask state to let them pull plug on JCPampL (12211)

Bernardsville Looking for updates on trail conditions since Hurricane Irene report in if you have them (9211)

Bethlehem Franklin Bethlehem school boards nearing alliance

Bound Brook

bull Flood prevention project in Bound Brook keeps Hurricane Irene from becoming another Floyd (83011)

bull Parts of Bound Brook South Bound Brook Manville under mandatory evacuation order (82811)

Branchburg Bald eagle sighting

Bridgewater

bull Pfizer anticipates EPA cleanup plan for American Cyanamid site in Bridgewater (121611)

bull French drug maker plans to move its US business to Bridgewater (121511)

bull Federal environmental officials close to decision on $205M clean-up proposal of Bridgewaters Cyanamid property (121111)

bull The bike path is clear and the going is easy at Duke Island Park in Bridgewater (13111)

Califon

bull Hunterdon complaints of power outages prompt review of utilities by state Board of Public Utilities (112911)

bull Snowstorm results in $15000 bill to Califon for tree branch clean-up (112611)

Chester Borough New Jersey residents and communities cope with effects of pre-Halloween snowstorm (103111)

Chester Township North Hunterdon High School sets metro record for donating blood Clinton Town Clinton seeking new bids for sewer line manhole work (111411)

Clinton Township Village Green development in Annandale gets final approval to build on former Warren Lumber site (111411)

Cranbury Horse owner forestalls sale to slaughterhouse (121011)

Delaware Eroded Delaware Township roads popular with cyclists pedestrians prompt speed limit worries (121311)

Dunellen Thunderstorms roll through NJ causing lightning strikes and roof collapse (61711)

East Amwell Hunterdon Freeholders delay changes for Solid Waste Advisory Council after SWAC members object (12711)

2 Gathered July-December 2011 3 With assistance from Melissa Basile MCRP 2012 4 Gathered from the Star Ledger New Jersey County and Town Pages httpwwwnjcomlocal under ldquoMorerdquo ndashFor example httptopicsnjcomtagbernardspostshtml substituting the name of the town of interest

Appendix A News Snapshots

2

SUMMARY FOR INITIAL REPORT

The following is the first report (December 1 2011) for the FEMA-Rutgers project on Flood Mitigation on the Raritan River We have completed the initial work as described in the contract and continue to collect additional data that we believe may be useful to the project Task One Community Assessment This includes profiles on each of the communities in the Raritan watershed including census reports (mayors years in office and municipal engineers) flooding data (insurance payouts) and social environmental and political lsquomedia snapshotsrsquo In addition we have created several maps to help us understand the regional impacts of flooding and possible areas to consider in future planning maps by watershed region with overall payout data and regional planning area maps based on local zoning

We have also identified the key regional stakeholders and are in the process of conducting interviews

Task Two Economic Modeling The economic modeling element of this project is designed to look at best ways to estimate benefits to municipalities (and larger governmental units) of flood mitigation projects for the Raritan watershed For this section two draft reports are provided a flood expense taxonomy (separate spreadsheet) and a literature review Task Three Ideas on Risk Modeling The third element of the project includes the first stages of the development of a risk analysis identification model to focus on all hazardsrisks The identification of data to be analyzed will be developed by experts in both mathematical modeling and human health and environmental modeling from engineering and public policy perspectives The team initially Fred Roberts and Paul Kantor now includes Qizhong Guo Associate Professor of Civil and Environmental Engineering and David Robinson NJ State Climatologist and Geography Professor

3

Contents Introduction 3

Task One Community Assessment 4

Task Two Economic Modeling 8

Task Three Ideas on Risk Assessment for Flooding on the Raritan 9

Appendix A News Snapshots 14

Appendix B-1 Safety UCR 20

Appendix B-2 Safety UCR Detail 23

Appendix C Maps 24

Appendix D Literature 30

Appendix E Guo CV 31

Appendix F Robinson CV 36

Introduction Traditionally national flood mitigation strategies sought to move flood waters away from vulnerable communities Today FEMA explores how reducing the volume of water moving during flood events can deliver cost-effective results Working with Rutgers University FEMA seeks to use public dialogue in the Raritan River watershed and the collective expertise of leaders across the university research community to evaluate flood mitigation from engineering social environmental political and economic perspectives

The project hopes to provide strategies that will improve mitigation through partnerships Those partnerships span public-private and authorities crossing all levels of government and include key stakeholders within the watershed Using the strategies developed here FEMA hopes to improve understanding information sharing governance and partnering in this critical area The successful model will be transferrable to other watersheds both smaller and larger

Rutgers University through the E J Bloustein School of Planning and Public Policy engaged partners from across the University to bring expertise in regional planning public engagement communication economics mathematics engineering and geography into this effort Together this team will develop strategies and models to help validate this approach

This initial report provides status on the three major elements of the project Community Economics and Risk This initial report is provided to illustrate the direction being taken within each project element and to give the FEMA partners and other advisors within the Collaboratorium an understanding of our progress and the challenges we have encountered

This report also includes draft spreadsheets attached as separate files

Spreadsheet 1 Demographic Data Spreadsheet 2 Flood Expense Taxonomy

4

Task One Community Assessment The deliverables for Community Assessment are to develop a background of census data a general report on each of the ninety-nine Raritan Basin municipalities (including watershed area maps for certain data) and the initial stages of a series of interviews with key municipal stakeholders The data presented here is preliminary and will continue to be updated throughout the project For all of this work the study area is the Raritan River Basin and is presented either by municipality or by watershed management area (WMA) WMA 8 ( the North and Upper Branches) WMA 9 (Lower Raritan) and WMA 10 (the Millstone River)

Demographics

The first step in the Community Assessment is a report on certain Census data for the ninety-nine towns in the watershed as described in Table 1 below The report on this work is presented in Spreadsheet 1 Demographic Data (See Separate Spreadsheet File)

Municipality County by watershed Population Upper Raritan (8) 1990 2000 2010 Lower Raritan (9) Change in Population 00-10 Millstone (10) Wealth Education (25 yo and up) Med household income high school graduate or higher Average household income Bachelorrsquos degree or higher Med Home Value (Owner Occupied) Employment (16 yo and up) Insurance of workforce employed Policies in force as of 83111 commuting 1hour+ to work Claims reported to FEMA 1178-83111 Workforce bull Total Losses Population over 16 yo bull Closed Losses Pop over 16 as of workforce bull Open Losses bull CWOP Losses Government bull Total Payments Mayor Term Engineer

Table 1 Demographic Data 1990 2000 2010

5

Map 1 Watershed Management Area 8 by FEMA Payout

Map 2 Watershed Management Area 9 by FEMA Payout

In addition we reviewed media coverage for these municipalities looking at social environmental and political issues That initial report is provided ldquoSocial Environmental and Political Snapshotsrdquo and is attached as Appendix A A summary of Safety produced from the NJ State Police Uniform Crime Reports is presented in Attachment B Data Maps The second component of the initial phase of the Community Assessment looks at the geography of the region to identify areas where enhanced mitigation strategies might be most productive in reducing impacts to the most affected communities This resulted in several series of maps for the region

NFIP Payout Maps The first set of maps display the National Flood Insurance Program (NFIP) payouts in the study region by watershed management area delineated by the NJDEP and NJ Geological Survey) The payout amounts were joined to the maps using the NFIP statistics labeled Loss Statistics New Jersey as of 83111 This information does not take into account damages sustained by Hurricane Irene but it does offer a 33 year history of the NFIP loss claims and payouts to policy holders from 1978 to immediately prior to that event The maps were color coded to easily display the breakdown in payouts in the individual municipalities in each watershed The data shows some municipalities suffer dramatically more loss than some of their neighbors particularly in WMA 9 where Bound Brook has a cumulative amount of NFIP payouts totaling over $53M while other municipalities in the watershed Metuchen Borough for example have under $10000 in total NFIP payouts in the same period of time The data are presented in Maps 1 2 and 3 below and in full size in Appendix C

Land Use Maps To further examine opportunities for enhanced prevention of damage from flood waters we created three maps to explore land use (Maps 4 5 and 6) These maps were created from the New Jersey State Development and Redevelopment Plan (SDRP) administered by the NJ Office for Planning Advocacy which delineates planning areas throughout the state The SDRP seeks to provide a

Map 3 Watershed Management Area 10 by FEMA Payout

6

balance between growth and conservation by designating planning areas that share common conditions with regard to development and environmental features These planning areas (PAs) are as follows

Growth Areas PA1 Metropolitan Planning Areas(up to 80-90 impervious coverage with development) PA2 Suburban Planning Areas (up to 30 impervious coverage with development) Designated Centers

Limited Growth Areas planning in these planning areas should promote a balance of conservation and limited growth PA3 Fringe Planning Areas PA4 Rural Planning Areas PA5 Environmentally Sensitive Planning Areas Conservation Areas (below 30 impervious coverage with development and limits or excludes sewer infrastructure) PA3 Fringe Planning Areas PA4 Rural Planning Areas PA5 Environmentally Sensitive Planning Areas

Map 6 WMA 10 Planning Areas

Map 5 WMA 9 Planning Areas

Map 4 WMA 8 Planning Areas

7

Interviews

The second phase of this work currently underway involves in-person interviews with 20-25 key regional municipal officials to establish the following

bull Importance of flood mitigation among other concerns schools police fire service traffic management trash collection parking street repair housing maintenance and repair aesthetics and other quality of life issues

bull What they know about flood mitigation flooding the degree of damage and the causes bull Their understanding of climate impacts on flooding and mitigation strategies bull Local stormwater management measures (retention basin management to rain barrels bull Preferred flood mitigation strategies both locally and regionally bull Level of local engagement and what causes citizens to become engaged in local issues

At this time the following municipalities are to be interviewed with either the mayor the administrator and or the director of the local Office of Emergency Management

Municipality CO Municipality CO Berkeley Heights Union Middlesex Boro Msx Bound Brook Mayor amp OEM Som Milltown Msx Branchburg Som Monroe (Engineer Ernie Feist) Msx Clinton Township Hun North Plainfield Msx Edison John Medina Msx Scotch Plains Union Franklin Som Raritan Township Hun Hightstown Mer Watchung Som Hopewell Mer Washington Twp Morris Manville Administrator amp OEM Som Woodbridge Msx Marlboro Mon

Next Steps

The next steps will involve closer analysis of the census data as we proceed with municipal interviews The goal is to talk to people at various levels not just those who are most directly affected but also those who may be upstream from more impacted communities The next mapping exercise will be to add contours in order to verify the drainage and to calculate current potential mitigation in the uppermost communities through major stormwater mitigation efforts The use of impervious surface coverages from the state as well as the application of independent models will lend more insight into the potential for stormwater management BMPs to aid in overall impact reduction

The findings of the interviews will form a discussion guide at a dinner meeting with municipal and county leaders at the Bloustein School on Tuesday January 31 2012 The discussion will focus on further action for a regional volume-focused flood mitigation strategy Focus Groups Several focus groups are also scheduled Milltown and Manville are scheduled for January Two others will be held in February and March These meetings will be held with groups of key citizen leaders in selected municipalities to determine their knowledge of the issue significance to them their desire for further information about the subject and desired mode of transmission of the information and willingness to become involved in the issue in their own town These sessions will also address the questions asked of the initial stakeholders to verify unity of thought and understanding of issues across the community

8

In addition a series of meetings with regional engineers are being held to engage them in the mitigation conversation from the perspective of stormwater management and overall adaptation of green infrastructure practices Their inputs will be gathered in January and February through a series of conference calls on the following topics

bull regulation changes bull policy directions bull professional education needs bull public education needs bull financing mitigation

They will meet together in early March to share their findings and engage with municipal officials and another cohort regional business leaders

Pilot Survey A draft survey will be created and a pilot conducted for a larger regional survey This would be tested with a limited number of residents of the basin to determine their knowledge of the issue significance to them their desire for further information about the subject and desired mode of transmission of the information and willingness to become involved in the issue in their own towns A larger application of the survey would not be fully implemented as part of this project

Task Two Economic Modeling The economic modeling element of this project is designed to look at best ways to estimate benefits to municipalities (and larger governmental units) of flood mitigation projects for the Raritan watershed For this section two draft reports are provided a flood expense taxonomy (separate spreadsheet) and a literature review

Flood Expense Taxonomy

The flood expense taxonomy (see Separate Spreadsheet File) provides a framework for identifying and categorizing the costs associated with flooding events We developed the taxonomy drawing from academic literature prior reports of flooding events and newspaper articles covering flooding events The taxonomy identifies potential flood-related expenses for four sectors of the economy households commercial businesses municipalities and agriculture The potential costs to each sector are cross-categorized with respect to the time sequence of the flooding event pre-flooding flooding and post flooding We will continue to identify additional potential costs and where possible populate those findings with estimated costs Additionally we will add to this taxonomy the potential economic benefits (if any) to the same four sectors as a result of removing structures located in the floodplain

Literature Review

The annotated literature review (Attachment D) includes two types of articles those that examine the relationship between floods and housing prices and those that estimate the costs of flooding on municipalities and their residents and businesses The articles that examine the relationship between flooding events and home prices generally use hedonic pricing models to explain the difference in home value (as measured by sales price) between homes that are on the floodplain and similar houses that are not on it The articles generally indicate that prices for homes located on floodplains are discounted relative to similar homes not on the floodplain One article offers a meta-analysis of the empirical literature on home prices and flooding events The articles in our taxonomy examine the economic impacts of flooding events from England the Red-River Basin in the Western US and the Delaware

9

River in New Jersey We also include a meta-analysis that re-estimates National Weather Service estimates of national economic damages We will build on this literature review to develop a better understanding of the existing estimates of the full economic costs of flooding events and the potential benefits of removing structures (private and public) from floodplains

Task Three Ideas on Risk Assessment for Flooding on the Raritan The third element of the project includes the first stages of the development of a risk analysis identification model to focus on all hazardsrisks The identification of data to be analyzed will be developed by experts in both mathematical modeling and human health and environmental modeling from engineering and public policy perspectives The team initially Fred Roberts and Paul Kantor now includes Qizhong Guo Associate Professor of Civil and Environmental Engineering and David Robinson and State Climatologist and Geography Professor Brief bio sketches are attached as Appendix E Qizhong Guo and Appendix F David Robinson)

Overview

We have spent the past few months researching alternative approaches to flood risk assessment and gathering as much background information as possible We have gotten information from numerous individuals in the US and abroad and numerous agencies including New Jersey Department of Environmental Protection Department of Homeland Security Risk Management and Analysis USGS We came to understand that it was important to expand our team and so we have added a climatologist Professor David Robinson of the Rutgers Department of Geography the New Jersey State Climatologist and a hydrologist Professor Qizhong Guo of the Rutgers Department of Civil and Environmental Engineering We present here a high level sketch of the approach we are examining and discuss the kinds of detailed information that its implementation will require

Problem Formulation

Our multidisciplinary discussions have led us to understand many of the subtleties involved in doing a risk assessment for flooding on the Raritan We have nearly completed outlining an approach based on what we have learned and the next steps will be to share that approach with those working on other tasks in the project revise as needed and then begin gathering the data needed to do the required modeling

To explain our approach and the issues we have to resolve let us consider first a diagram (Figure 1) which illustrates the kinds of factors and tools we need to take into account in understanding flood modeling and applying that modeling to a risk assessment

10

Figure 1 The interaction of nature models and controllable activities (shown in yellow)

Engineering models can combine information on rainfall specified as to time and space with information on soil moisture conditions seasonality and river levels prior to the rain event the fixed properties of the watershed (not indicated) and the land cover (natural and built environment) to produce flood inundation maps To obtain the precision needed to specify events that might occur every five or ten years requires LiDAR information on elevations We refer to the soil moisture conditions and river levels as ldquoantecedent eventsrdquo and note that seasonality is important because it affects the ability of the soil to absorb water (due to vegetation cover freezing etc) The land coverenvironment is viewed as modifiable (yellow) over the span of a few years That would in turn move the flood contours Flood inundation maps can be used for insurance andor regulatory purposes for risk assessments and to communicate complex geology and meteorology to stakeholder groups and thereby facilitate consideration of alternative policies

In the simplest case we might think of four types of soil moisture conditions dry ldquoaveragerdquo wet or frozen Probably it is sufficient to think of four different seasons We might also think of three basic river levels low average and high Thus there are in principle 4 x 4 x 3 = 48 different combinations of these conditions In reality not all of these combinations are feasible We wonrsquot have frozen ground in summer In our analysis we will probably demonstrate the methods choosing only several of the sensible alternatives for detailed investigation

Risk Modeling

The basic outline for how DHS does risk assessment is shown in Figure 2 below which is taken from a presentation by Isaac Maya of the CREATE Center It defines ldquoriskrdquo R as a function of threat T vulnerability V and consequence C R = f(TVC) Often this is represented as a product R = T x V x C Generally the factors T and V are some kinds of probabilities while C is some measure of loss In our case ldquoattackrdquo is interpreted to mean weather event and ldquosuccessrdquo is interpreted to mean flooding

Precipitation(xt) )tthe past

The models

Soil moisture conditions + seasonality + river levels seasonality

Land Cover (natural + built environment)

River gauge levels

LiDAR information

Flood Contour Maps

waterway

5 year contour

10 year contour

11

Figure 2 Sketch of the DHS computation of risk

There are two branches to the tree in this figure The top branch is the ldquostatus quordquo and the bottom branch involves some mitigation strategy or combination of such strategies The probability of attack P is the probability that there will be a weather event of the particular kind being considered and the probability of ldquosuccessrdquo Q is the conditional probability that such an event if it does occur will lead to flooding at a site being considered The consequences could be of various types In the case of floods they include loss of life economic damage to homes and businesses (direct or indirect) and psychological damage There is an attempt to put each of these types of consequences in terms of dollars The consequence (cost L) of an attack (flood) is usually calculated as a weighted sum of the different kinds of costs in our case of the cost in terms of lives lost economic damage and psychological damage What weights one uses and whether it makes sense to add up these weighted values are issues that will be viewed differently by different stakeholder groups

Complexities of the Watershed Situation

In fact we expect to be able to do more than simply treat ldquofloodrdquo as a yes or no matter as is done in this tree diagram Flood inundation maps will present levels of flood waters and the losses incurred change if there is a different level of flooding (or if there is a different duration for the flooding a factor we will disregard for now) The flood inundation levels are numerical values obtained from hydrological models (they can be later translated into major moderate minor near flood stage and none in making flood forecasts) A more complete model would replace the ldquosuccessfailurerdquo bifurcation in the tree diagram with a number of branches depending on the levelcategory of flood as measured for example by river gauges In our case it makes sense to define category 0 as no flooding category 1 as 0 to 1 feet category 2 as 1 to 2 feet etc

In practice obtaining the flood inundation levels may be complicated even before we introduce mitigating strategies The inundation maps are maps for theoretical storms not maps used to predict floods from an approaching storm These maps for the storms of a 100-year and 500-year ldquoreturn periodrdquo1 are already available and are being updated by FEMA using the recently available high resolution LiDAR data on the topography These maps are called flood insurance rate maps (FIRM) However they are developed from a kind of discrete-event modeling and may conceal hidden

1 This technical expression can be interpreted as representing a 1 of 15 probability per year

Success

Failure Attack

Vulnerability

No Attack

No Risk Management Plan

Threat of Attack

Success

Failure

Attack

No Attack

Risk Management Plan

Risk Management

PlansConsequences

L = Expected[$-Eqiuvalent Loss]

No Loss

No Loss

C + LC = Cost of Risk Management Plan

L = Expected[Reduced Loss] where L lt L

C

C

PQ

(1 - P)

(1 - Q)

P where P lt P

(1 - P)

Q where Q lt Q

(1 - Q)

12

uncertainties Moreover the inundation maps for more frequent storms are not readily available or non-existent It would be possible to build flood maps for 2-year 5-year 10-year 25-year and 50-year storms from available data This would imply gathering data and modeling and will depend on having access to the data Moreover the discrete-event models do not take into account antecedent conditions or seasonality Continuous modeling could take these factors into account but the cost of modeling would then be much higher

Assuming we can get flood inundation levels from hydrological models we would try to calculate the probability Qi that a flood of category i occurs Each category of flooding i would have a level of loss Li associated with it The level of flooding from a given amount of rainfall over a period of time depends not only on the precipitation but also on the soil moisture conditionsseasonalityriver levels and on the land cover (natural + built environment) Initially we may assume a fixed land cover (although this can be changed by mitigation) As noted above there are many different combinations of parameters for soil moisture conditions seasonality river levels We could in principle estimate the probability of each of these combinations (but would need to work out a method for doing so) For each combination we could use the hydrological models to calculate the flood level and thus the flood category Putting all of these calculations together will allow us to calculate the Qi parameters This is probably much too complicated and certainly much too expensive Instead we will pick out several more common combinations among the sensible ones and combine them with different precipitation patterns to define a set of weather events

In short our threats will be different kinds of weather events defined in terms of the profile of rainfall over a given period of time (say in number of inches in first hour number in second hour etc over a 24 hour period) plus antecedent precipitation and seasonal effects and pre-existing river levels We will aim to identify six to ten weather scenarios to analyze For each the probability P will have to be determined Information about precipitation totals in 24 hour periods is available and we could then use this to define a 2-year 5-year 10-year 25-year 50-year 100-year event in terms of precipitation profile per 24-hour period The probability of a 5-year event would be 15 of a 50 year event 150 etc except for the other factors (antecedent events plus seasonality) So we may want to modify P based on these other factors We will need to determine how to do that Note also that climate change will likely affect the value of P but we do not know how to calculate modified P with any degree of confidence

Now what is the interpretation of the Qi parameters If the weather event (precipitation profile) is defined and the antecedent conditions and seasonality are known in principle the hydrological models will produce just one category of flood But it could be that it makes more sense to build some uncertainty into the predictions of the models or it may make sense to assume that we have some uncertainty in knowing the soil moisture conditions or the river levels Thus there is surely still some value to studying the risk assessments using different flood categories

When there is a mitigation strategy presumably the probability of an attack would not be affected so Prsquo in the diagram is the same as P not less than P The probability of ldquosuccessrdquo or flood given a weather event is Qrsquo which is less than Q If we use different flood level categories then we would replace each Qi with a number Qirsquo but note that Qirsquo would be higher than Qi for some i Also the consequences of a flood could be less with expected consequences now Lrsquo presumably less than L (or Lirsquo presumably less than Li) But there is a cost C of the mitigation strategy and that cost applies even if there is no flood

Still using the notation in the diagram we see that if f(TVC) = TxVxC then with no mitigation strategy the risk or total expected cost is PxQxL but if there is a mitigation strategy then (since P = Prsquo) the risk or total expected cost is C + PxQrsquoxLrsquo The difference G = PxQxL ndash (C + PxQrsquoxLrsquo) in these expected costs would be the expected gain or expected risk reduction due to the strategy (If we have different flood level categories the calculation would be modified appropriately)

13

Mitigation Strategies and Elicitation of Some Key Parameters

We will need to decide if we look at flood damage (vulnerability) to a larger region or concentrate on a few towns Our preliminary plan is to do the analysis for a few sample towns for instance Manville and Millstone We will need to coordinate with the other parts of the project on this decision

Consequences will need to be measured by our partners in the project and we will need to coordinate with them

We will look at a variety of mitigation strategies Among them could be improved flood forecasting and warning systems retrofit of green infrastructure (eg impervious area disconnection rain gardens cisterns and rain barrels green roofs pervious concrete) to existing development areas to reduce upstream runoff volume regional (upstream) stormwater detention basins acquisition and demolition and relocation of properties flood-proofing and elevation in place disaster preparedness and response plans flood insurance cleanout of urban drainage systems construction of dams and reservoirs construction of dikes levees and floodwalls channel alterations and high flow diversions and spillways We will want to coordinate with other parts of the project on which mitigation strategies we look at

We will need to work with other parts of the project to get ballpark estimates of reduced costs Lrsquo (or Lirsquo) of floods

Under a given mitigation strategy the challenge is to produce inundation maps that will allow us to determine vulnerability (whether there is a flood) In principle there are hydrological models for doing this or such models could be developed but to do the actual calculations will be much more expensive than we can afford in this modest project Thus the best we can do is pick a few sample mitigation strategies for which we could make educated guesses as to vulnerabilities ndash just to illustrate the methodology Our thoughts are to start with three types of strategies retrofit of green infrastructure some sort of upstream mitigation that would have impact on downstream flooding and improved forecasting (eg improved monitoring from an optimum number of precipitation gauges merged station and radar precipitation estimates soil moisture observations evapotranspiration estimates and additional stream gauges) The last of these three would have no impact on the probabilities Q or Qi and also will not require changed inundation maps Only our calculation as to consequence would be affected since the whole idea of improved forecasting would be to allow preparation so as to reduce the potential flood damage

A full risk assessment would want to look not at individual mitigation strategies but at ldquoportfoliosrdquo of mitigation strategies Of course all of what was said above about ldquomitigation strategyrdquo could apply to a portfolio of strategies But the inundation maps will be much harder to obtain for portfolios of strategies than for individual ones As noted some of the mitigation strategies will not affect inundation maps Looking at mitigation strategies separately avoids the complication that several of these strategies may interact with each other in complex ways Some portfolios of mitigation strategies will be too expensive or infeasible for other reasons We may want to look at budgetary constraints and try to look at a few simple portfolios of strategies that have total investment cost (sum of Crsquos for each strategy in the portfolio) less than a given amount Then the question will become What is a measure of the value of a given portfolio Are we trying to maximize total expected gain (sum of G for all strategies in the portfolio) Or are we optimizing something else Or is this a ldquomulti-attributed utilityrdquo problem

14

Social Environmental and Political News Snapshots2 Hunterdon Mercer Middlesex Monmouth Morris Somerset Union

RARITAN RIVER TOWN SOCIAL ENVIRONMENTAL AND POLITICAL NEWS SNAPSHOTS 34

Alexandria 4-way stop to be set up near Delaware Valley High School

Bedminster

bull Clarence Dillon Library in Bedminster to host a series of Salons (92711) bull Bedminster bicyclist killed after crash with truck in Far Hills (12111)

Berkeley Heights Berkeley Heights childrenrsquos library bounces back Bernards Outraged NJ towns ask state to let them pull plug on JCPampL (12211)

Bernardsville Looking for updates on trail conditions since Hurricane Irene report in if you have them (9211)

Bethlehem Franklin Bethlehem school boards nearing alliance

Bound Brook

bull Flood prevention project in Bound Brook keeps Hurricane Irene from becoming another Floyd (83011)

bull Parts of Bound Brook South Bound Brook Manville under mandatory evacuation order (82811)

Branchburg Bald eagle sighting

Bridgewater

bull Pfizer anticipates EPA cleanup plan for American Cyanamid site in Bridgewater (121611)

bull French drug maker plans to move its US business to Bridgewater (121511)

bull Federal environmental officials close to decision on $205M clean-up proposal of Bridgewaters Cyanamid property (121111)

bull The bike path is clear and the going is easy at Duke Island Park in Bridgewater (13111)

Califon

bull Hunterdon complaints of power outages prompt review of utilities by state Board of Public Utilities (112911)

bull Snowstorm results in $15000 bill to Califon for tree branch clean-up (112611)

Chester Borough New Jersey residents and communities cope with effects of pre-Halloween snowstorm (103111)

Chester Township North Hunterdon High School sets metro record for donating blood Clinton Town Clinton seeking new bids for sewer line manhole work (111411)

Clinton Township Village Green development in Annandale gets final approval to build on former Warren Lumber site (111411)

Cranbury Horse owner forestalls sale to slaughterhouse (121011)

Delaware Eroded Delaware Township roads popular with cyclists pedestrians prompt speed limit worries (121311)

Dunellen Thunderstorms roll through NJ causing lightning strikes and roof collapse (61711)

East Amwell Hunterdon Freeholders delay changes for Solid Waste Advisory Council after SWAC members object (12711)

2 Gathered July-December 2011 3 With assistance from Melissa Basile MCRP 2012 4 Gathered from the Star Ledger New Jersey County and Town Pages httpwwwnjcomlocal under ldquoMorerdquo ndashFor example httptopicsnjcomtagbernardspostshtml substituting the name of the town of interest

Appendix A News Snapshots

3

Contents Introduction 3

Task One Community Assessment 4

Task Two Economic Modeling 8

Task Three Ideas on Risk Assessment for Flooding on the Raritan 9

Appendix A News Snapshots 14

Appendix B-1 Safety UCR 20

Appendix B-2 Safety UCR Detail 23

Appendix C Maps 24

Appendix D Literature 30

Appendix E Guo CV 31

Appendix F Robinson CV 36

Introduction Traditionally national flood mitigation strategies sought to move flood waters away from vulnerable communities Today FEMA explores how reducing the volume of water moving during flood events can deliver cost-effective results Working with Rutgers University FEMA seeks to use public dialogue in the Raritan River watershed and the collective expertise of leaders across the university research community to evaluate flood mitigation from engineering social environmental political and economic perspectives

The project hopes to provide strategies that will improve mitigation through partnerships Those partnerships span public-private and authorities crossing all levels of government and include key stakeholders within the watershed Using the strategies developed here FEMA hopes to improve understanding information sharing governance and partnering in this critical area The successful model will be transferrable to other watersheds both smaller and larger

Rutgers University through the E J Bloustein School of Planning and Public Policy engaged partners from across the University to bring expertise in regional planning public engagement communication economics mathematics engineering and geography into this effort Together this team will develop strategies and models to help validate this approach

This initial report provides status on the three major elements of the project Community Economics and Risk This initial report is provided to illustrate the direction being taken within each project element and to give the FEMA partners and other advisors within the Collaboratorium an understanding of our progress and the challenges we have encountered

This report also includes draft spreadsheets attached as separate files

Spreadsheet 1 Demographic Data Spreadsheet 2 Flood Expense Taxonomy

4

Task One Community Assessment The deliverables for Community Assessment are to develop a background of census data a general report on each of the ninety-nine Raritan Basin municipalities (including watershed area maps for certain data) and the initial stages of a series of interviews with key municipal stakeholders The data presented here is preliminary and will continue to be updated throughout the project For all of this work the study area is the Raritan River Basin and is presented either by municipality or by watershed management area (WMA) WMA 8 ( the North and Upper Branches) WMA 9 (Lower Raritan) and WMA 10 (the Millstone River)

Demographics

The first step in the Community Assessment is a report on certain Census data for the ninety-nine towns in the watershed as described in Table 1 below The report on this work is presented in Spreadsheet 1 Demographic Data (See Separate Spreadsheet File)

Municipality County by watershed Population Upper Raritan (8) 1990 2000 2010 Lower Raritan (9) Change in Population 00-10 Millstone (10) Wealth Education (25 yo and up) Med household income high school graduate or higher Average household income Bachelorrsquos degree or higher Med Home Value (Owner Occupied) Employment (16 yo and up) Insurance of workforce employed Policies in force as of 83111 commuting 1hour+ to work Claims reported to FEMA 1178-83111 Workforce bull Total Losses Population over 16 yo bull Closed Losses Pop over 16 as of workforce bull Open Losses bull CWOP Losses Government bull Total Payments Mayor Term Engineer

Table 1 Demographic Data 1990 2000 2010

5

Map 1 Watershed Management Area 8 by FEMA Payout

Map 2 Watershed Management Area 9 by FEMA Payout

In addition we reviewed media coverage for these municipalities looking at social environmental and political issues That initial report is provided ldquoSocial Environmental and Political Snapshotsrdquo and is attached as Appendix A A summary of Safety produced from the NJ State Police Uniform Crime Reports is presented in Attachment B Data Maps The second component of the initial phase of the Community Assessment looks at the geography of the region to identify areas where enhanced mitigation strategies might be most productive in reducing impacts to the most affected communities This resulted in several series of maps for the region

NFIP Payout Maps The first set of maps display the National Flood Insurance Program (NFIP) payouts in the study region by watershed management area delineated by the NJDEP and NJ Geological Survey) The payout amounts were joined to the maps using the NFIP statistics labeled Loss Statistics New Jersey as of 83111 This information does not take into account damages sustained by Hurricane Irene but it does offer a 33 year history of the NFIP loss claims and payouts to policy holders from 1978 to immediately prior to that event The maps were color coded to easily display the breakdown in payouts in the individual municipalities in each watershed The data shows some municipalities suffer dramatically more loss than some of their neighbors particularly in WMA 9 where Bound Brook has a cumulative amount of NFIP payouts totaling over $53M while other municipalities in the watershed Metuchen Borough for example have under $10000 in total NFIP payouts in the same period of time The data are presented in Maps 1 2 and 3 below and in full size in Appendix C

Land Use Maps To further examine opportunities for enhanced prevention of damage from flood waters we created three maps to explore land use (Maps 4 5 and 6) These maps were created from the New Jersey State Development and Redevelopment Plan (SDRP) administered by the NJ Office for Planning Advocacy which delineates planning areas throughout the state The SDRP seeks to provide a

Map 3 Watershed Management Area 10 by FEMA Payout

6

balance between growth and conservation by designating planning areas that share common conditions with regard to development and environmental features These planning areas (PAs) are as follows

Growth Areas PA1 Metropolitan Planning Areas(up to 80-90 impervious coverage with development) PA2 Suburban Planning Areas (up to 30 impervious coverage with development) Designated Centers

Limited Growth Areas planning in these planning areas should promote a balance of conservation and limited growth PA3 Fringe Planning Areas PA4 Rural Planning Areas PA5 Environmentally Sensitive Planning Areas Conservation Areas (below 30 impervious coverage with development and limits or excludes sewer infrastructure) PA3 Fringe Planning Areas PA4 Rural Planning Areas PA5 Environmentally Sensitive Planning Areas

Map 6 WMA 10 Planning Areas

Map 5 WMA 9 Planning Areas

Map 4 WMA 8 Planning Areas

7

Interviews

The second phase of this work currently underway involves in-person interviews with 20-25 key regional municipal officials to establish the following

bull Importance of flood mitigation among other concerns schools police fire service traffic management trash collection parking street repair housing maintenance and repair aesthetics and other quality of life issues

bull What they know about flood mitigation flooding the degree of damage and the causes bull Their understanding of climate impacts on flooding and mitigation strategies bull Local stormwater management measures (retention basin management to rain barrels bull Preferred flood mitigation strategies both locally and regionally bull Level of local engagement and what causes citizens to become engaged in local issues

At this time the following municipalities are to be interviewed with either the mayor the administrator and or the director of the local Office of Emergency Management

Municipality CO Municipality CO Berkeley Heights Union Middlesex Boro Msx Bound Brook Mayor amp OEM Som Milltown Msx Branchburg Som Monroe (Engineer Ernie Feist) Msx Clinton Township Hun North Plainfield Msx Edison John Medina Msx Scotch Plains Union Franklin Som Raritan Township Hun Hightstown Mer Watchung Som Hopewell Mer Washington Twp Morris Manville Administrator amp OEM Som Woodbridge Msx Marlboro Mon

Next Steps

The next steps will involve closer analysis of the census data as we proceed with municipal interviews The goal is to talk to people at various levels not just those who are most directly affected but also those who may be upstream from more impacted communities The next mapping exercise will be to add contours in order to verify the drainage and to calculate current potential mitigation in the uppermost communities through major stormwater mitigation efforts The use of impervious surface coverages from the state as well as the application of independent models will lend more insight into the potential for stormwater management BMPs to aid in overall impact reduction

The findings of the interviews will form a discussion guide at a dinner meeting with municipal and county leaders at the Bloustein School on Tuesday January 31 2012 The discussion will focus on further action for a regional volume-focused flood mitigation strategy Focus Groups Several focus groups are also scheduled Milltown and Manville are scheduled for January Two others will be held in February and March These meetings will be held with groups of key citizen leaders in selected municipalities to determine their knowledge of the issue significance to them their desire for further information about the subject and desired mode of transmission of the information and willingness to become involved in the issue in their own town These sessions will also address the questions asked of the initial stakeholders to verify unity of thought and understanding of issues across the community

8

In addition a series of meetings with regional engineers are being held to engage them in the mitigation conversation from the perspective of stormwater management and overall adaptation of green infrastructure practices Their inputs will be gathered in January and February through a series of conference calls on the following topics

bull regulation changes bull policy directions bull professional education needs bull public education needs bull financing mitigation

They will meet together in early March to share their findings and engage with municipal officials and another cohort regional business leaders

Pilot Survey A draft survey will be created and a pilot conducted for a larger regional survey This would be tested with a limited number of residents of the basin to determine their knowledge of the issue significance to them their desire for further information about the subject and desired mode of transmission of the information and willingness to become involved in the issue in their own towns A larger application of the survey would not be fully implemented as part of this project

Task Two Economic Modeling The economic modeling element of this project is designed to look at best ways to estimate benefits to municipalities (and larger governmental units) of flood mitigation projects for the Raritan watershed For this section two draft reports are provided a flood expense taxonomy (separate spreadsheet) and a literature review

Flood Expense Taxonomy

The flood expense taxonomy (see Separate Spreadsheet File) provides a framework for identifying and categorizing the costs associated with flooding events We developed the taxonomy drawing from academic literature prior reports of flooding events and newspaper articles covering flooding events The taxonomy identifies potential flood-related expenses for four sectors of the economy households commercial businesses municipalities and agriculture The potential costs to each sector are cross-categorized with respect to the time sequence of the flooding event pre-flooding flooding and post flooding We will continue to identify additional potential costs and where possible populate those findings with estimated costs Additionally we will add to this taxonomy the potential economic benefits (if any) to the same four sectors as a result of removing structures located in the floodplain

Literature Review

The annotated literature review (Attachment D) includes two types of articles those that examine the relationship between floods and housing prices and those that estimate the costs of flooding on municipalities and their residents and businesses The articles that examine the relationship between flooding events and home prices generally use hedonic pricing models to explain the difference in home value (as measured by sales price) between homes that are on the floodplain and similar houses that are not on it The articles generally indicate that prices for homes located on floodplains are discounted relative to similar homes not on the floodplain One article offers a meta-analysis of the empirical literature on home prices and flooding events The articles in our taxonomy examine the economic impacts of flooding events from England the Red-River Basin in the Western US and the Delaware

9

River in New Jersey We also include a meta-analysis that re-estimates National Weather Service estimates of national economic damages We will build on this literature review to develop a better understanding of the existing estimates of the full economic costs of flooding events and the potential benefits of removing structures (private and public) from floodplains

Task Three Ideas on Risk Assessment for Flooding on the Raritan The third element of the project includes the first stages of the development of a risk analysis identification model to focus on all hazardsrisks The identification of data to be analyzed will be developed by experts in both mathematical modeling and human health and environmental modeling from engineering and public policy perspectives The team initially Fred Roberts and Paul Kantor now includes Qizhong Guo Associate Professor of Civil and Environmental Engineering and David Robinson and State Climatologist and Geography Professor Brief bio sketches are attached as Appendix E Qizhong Guo and Appendix F David Robinson)

Overview

We have spent the past few months researching alternative approaches to flood risk assessment and gathering as much background information as possible We have gotten information from numerous individuals in the US and abroad and numerous agencies including New Jersey Department of Environmental Protection Department of Homeland Security Risk Management and Analysis USGS We came to understand that it was important to expand our team and so we have added a climatologist Professor David Robinson of the Rutgers Department of Geography the New Jersey State Climatologist and a hydrologist Professor Qizhong Guo of the Rutgers Department of Civil and Environmental Engineering We present here a high level sketch of the approach we are examining and discuss the kinds of detailed information that its implementation will require

Problem Formulation

Our multidisciplinary discussions have led us to understand many of the subtleties involved in doing a risk assessment for flooding on the Raritan We have nearly completed outlining an approach based on what we have learned and the next steps will be to share that approach with those working on other tasks in the project revise as needed and then begin gathering the data needed to do the required modeling

To explain our approach and the issues we have to resolve let us consider first a diagram (Figure 1) which illustrates the kinds of factors and tools we need to take into account in understanding flood modeling and applying that modeling to a risk assessment

10

Figure 1 The interaction of nature models and controllable activities (shown in yellow)

Engineering models can combine information on rainfall specified as to time and space with information on soil moisture conditions seasonality and river levels prior to the rain event the fixed properties of the watershed (not indicated) and the land cover (natural and built environment) to produce flood inundation maps To obtain the precision needed to specify events that might occur every five or ten years requires LiDAR information on elevations We refer to the soil moisture conditions and river levels as ldquoantecedent eventsrdquo and note that seasonality is important because it affects the ability of the soil to absorb water (due to vegetation cover freezing etc) The land coverenvironment is viewed as modifiable (yellow) over the span of a few years That would in turn move the flood contours Flood inundation maps can be used for insurance andor regulatory purposes for risk assessments and to communicate complex geology and meteorology to stakeholder groups and thereby facilitate consideration of alternative policies

In the simplest case we might think of four types of soil moisture conditions dry ldquoaveragerdquo wet or frozen Probably it is sufficient to think of four different seasons We might also think of three basic river levels low average and high Thus there are in principle 4 x 4 x 3 = 48 different combinations of these conditions In reality not all of these combinations are feasible We wonrsquot have frozen ground in summer In our analysis we will probably demonstrate the methods choosing only several of the sensible alternatives for detailed investigation

Risk Modeling

The basic outline for how DHS does risk assessment is shown in Figure 2 below which is taken from a presentation by Isaac Maya of the CREATE Center It defines ldquoriskrdquo R as a function of threat T vulnerability V and consequence C R = f(TVC) Often this is represented as a product R = T x V x C Generally the factors T and V are some kinds of probabilities while C is some measure of loss In our case ldquoattackrdquo is interpreted to mean weather event and ldquosuccessrdquo is interpreted to mean flooding

Precipitation(xt) )tthe past

The models

Soil moisture conditions + seasonality + river levels seasonality

Land Cover (natural + built environment)

River gauge levels

LiDAR information

Flood Contour Maps

waterway

5 year contour

10 year contour

11

Figure 2 Sketch of the DHS computation of risk

There are two branches to the tree in this figure The top branch is the ldquostatus quordquo and the bottom branch involves some mitigation strategy or combination of such strategies The probability of attack P is the probability that there will be a weather event of the particular kind being considered and the probability of ldquosuccessrdquo Q is the conditional probability that such an event if it does occur will lead to flooding at a site being considered The consequences could be of various types In the case of floods they include loss of life economic damage to homes and businesses (direct or indirect) and psychological damage There is an attempt to put each of these types of consequences in terms of dollars The consequence (cost L) of an attack (flood) is usually calculated as a weighted sum of the different kinds of costs in our case of the cost in terms of lives lost economic damage and psychological damage What weights one uses and whether it makes sense to add up these weighted values are issues that will be viewed differently by different stakeholder groups

Complexities of the Watershed Situation

In fact we expect to be able to do more than simply treat ldquofloodrdquo as a yes or no matter as is done in this tree diagram Flood inundation maps will present levels of flood waters and the losses incurred change if there is a different level of flooding (or if there is a different duration for the flooding a factor we will disregard for now) The flood inundation levels are numerical values obtained from hydrological models (they can be later translated into major moderate minor near flood stage and none in making flood forecasts) A more complete model would replace the ldquosuccessfailurerdquo bifurcation in the tree diagram with a number of branches depending on the levelcategory of flood as measured for example by river gauges In our case it makes sense to define category 0 as no flooding category 1 as 0 to 1 feet category 2 as 1 to 2 feet etc

In practice obtaining the flood inundation levels may be complicated even before we introduce mitigating strategies The inundation maps are maps for theoretical storms not maps used to predict floods from an approaching storm These maps for the storms of a 100-year and 500-year ldquoreturn periodrdquo1 are already available and are being updated by FEMA using the recently available high resolution LiDAR data on the topography These maps are called flood insurance rate maps (FIRM) However they are developed from a kind of discrete-event modeling and may conceal hidden

1 This technical expression can be interpreted as representing a 1 of 15 probability per year

Success

Failure Attack

Vulnerability

No Attack

No Risk Management Plan

Threat of Attack

Success

Failure

Attack

No Attack

Risk Management Plan

Risk Management

PlansConsequences

L = Expected[$-Eqiuvalent Loss]

No Loss

No Loss

C + LC = Cost of Risk Management Plan

L = Expected[Reduced Loss] where L lt L

C

C

PQ

(1 - P)

(1 - Q)

P where P lt P

(1 - P)

Q where Q lt Q

(1 - Q)

12

uncertainties Moreover the inundation maps for more frequent storms are not readily available or non-existent It would be possible to build flood maps for 2-year 5-year 10-year 25-year and 50-year storms from available data This would imply gathering data and modeling and will depend on having access to the data Moreover the discrete-event models do not take into account antecedent conditions or seasonality Continuous modeling could take these factors into account but the cost of modeling would then be much higher

Assuming we can get flood inundation levels from hydrological models we would try to calculate the probability Qi that a flood of category i occurs Each category of flooding i would have a level of loss Li associated with it The level of flooding from a given amount of rainfall over a period of time depends not only on the precipitation but also on the soil moisture conditionsseasonalityriver levels and on the land cover (natural + built environment) Initially we may assume a fixed land cover (although this can be changed by mitigation) As noted above there are many different combinations of parameters for soil moisture conditions seasonality river levels We could in principle estimate the probability of each of these combinations (but would need to work out a method for doing so) For each combination we could use the hydrological models to calculate the flood level and thus the flood category Putting all of these calculations together will allow us to calculate the Qi parameters This is probably much too complicated and certainly much too expensive Instead we will pick out several more common combinations among the sensible ones and combine them with different precipitation patterns to define a set of weather events

In short our threats will be different kinds of weather events defined in terms of the profile of rainfall over a given period of time (say in number of inches in first hour number in second hour etc over a 24 hour period) plus antecedent precipitation and seasonal effects and pre-existing river levels We will aim to identify six to ten weather scenarios to analyze For each the probability P will have to be determined Information about precipitation totals in 24 hour periods is available and we could then use this to define a 2-year 5-year 10-year 25-year 50-year 100-year event in terms of precipitation profile per 24-hour period The probability of a 5-year event would be 15 of a 50 year event 150 etc except for the other factors (antecedent events plus seasonality) So we may want to modify P based on these other factors We will need to determine how to do that Note also that climate change will likely affect the value of P but we do not know how to calculate modified P with any degree of confidence

Now what is the interpretation of the Qi parameters If the weather event (precipitation profile) is defined and the antecedent conditions and seasonality are known in principle the hydrological models will produce just one category of flood But it could be that it makes more sense to build some uncertainty into the predictions of the models or it may make sense to assume that we have some uncertainty in knowing the soil moisture conditions or the river levels Thus there is surely still some value to studying the risk assessments using different flood categories

When there is a mitigation strategy presumably the probability of an attack would not be affected so Prsquo in the diagram is the same as P not less than P The probability of ldquosuccessrdquo or flood given a weather event is Qrsquo which is less than Q If we use different flood level categories then we would replace each Qi with a number Qirsquo but note that Qirsquo would be higher than Qi for some i Also the consequences of a flood could be less with expected consequences now Lrsquo presumably less than L (or Lirsquo presumably less than Li) But there is a cost C of the mitigation strategy and that cost applies even if there is no flood

Still using the notation in the diagram we see that if f(TVC) = TxVxC then with no mitigation strategy the risk or total expected cost is PxQxL but if there is a mitigation strategy then (since P = Prsquo) the risk or total expected cost is C + PxQrsquoxLrsquo The difference G = PxQxL ndash (C + PxQrsquoxLrsquo) in these expected costs would be the expected gain or expected risk reduction due to the strategy (If we have different flood level categories the calculation would be modified appropriately)

13

Mitigation Strategies and Elicitation of Some Key Parameters

We will need to decide if we look at flood damage (vulnerability) to a larger region or concentrate on a few towns Our preliminary plan is to do the analysis for a few sample towns for instance Manville and Millstone We will need to coordinate with the other parts of the project on this decision

Consequences will need to be measured by our partners in the project and we will need to coordinate with them

We will look at a variety of mitigation strategies Among them could be improved flood forecasting and warning systems retrofit of green infrastructure (eg impervious area disconnection rain gardens cisterns and rain barrels green roofs pervious concrete) to existing development areas to reduce upstream runoff volume regional (upstream) stormwater detention basins acquisition and demolition and relocation of properties flood-proofing and elevation in place disaster preparedness and response plans flood insurance cleanout of urban drainage systems construction of dams and reservoirs construction of dikes levees and floodwalls channel alterations and high flow diversions and spillways We will want to coordinate with other parts of the project on which mitigation strategies we look at

We will need to work with other parts of the project to get ballpark estimates of reduced costs Lrsquo (or Lirsquo) of floods

Under a given mitigation strategy the challenge is to produce inundation maps that will allow us to determine vulnerability (whether there is a flood) In principle there are hydrological models for doing this or such models could be developed but to do the actual calculations will be much more expensive than we can afford in this modest project Thus the best we can do is pick a few sample mitigation strategies for which we could make educated guesses as to vulnerabilities ndash just to illustrate the methodology Our thoughts are to start with three types of strategies retrofit of green infrastructure some sort of upstream mitigation that would have impact on downstream flooding and improved forecasting (eg improved monitoring from an optimum number of precipitation gauges merged station and radar precipitation estimates soil moisture observations evapotranspiration estimates and additional stream gauges) The last of these three would have no impact on the probabilities Q or Qi and also will not require changed inundation maps Only our calculation as to consequence would be affected since the whole idea of improved forecasting would be to allow preparation so as to reduce the potential flood damage

A full risk assessment would want to look not at individual mitigation strategies but at ldquoportfoliosrdquo of mitigation strategies Of course all of what was said above about ldquomitigation strategyrdquo could apply to a portfolio of strategies But the inundation maps will be much harder to obtain for portfolios of strategies than for individual ones As noted some of the mitigation strategies will not affect inundation maps Looking at mitigation strategies separately avoids the complication that several of these strategies may interact with each other in complex ways Some portfolios of mitigation strategies will be too expensive or infeasible for other reasons We may want to look at budgetary constraints and try to look at a few simple portfolios of strategies that have total investment cost (sum of Crsquos for each strategy in the portfolio) less than a given amount Then the question will become What is a measure of the value of a given portfolio Are we trying to maximize total expected gain (sum of G for all strategies in the portfolio) Or are we optimizing something else Or is this a ldquomulti-attributed utilityrdquo problem

14

Social Environmental and Political News Snapshots2 Hunterdon Mercer Middlesex Monmouth Morris Somerset Union

RARITAN RIVER TOWN SOCIAL ENVIRONMENTAL AND POLITICAL NEWS SNAPSHOTS 34

Alexandria 4-way stop to be set up near Delaware Valley High School

Bedminster

bull Clarence Dillon Library in Bedminster to host a series of Salons (92711) bull Bedminster bicyclist killed after crash with truck in Far Hills (12111)

Berkeley Heights Berkeley Heights childrenrsquos library bounces back Bernards Outraged NJ towns ask state to let them pull plug on JCPampL (12211)

Bernardsville Looking for updates on trail conditions since Hurricane Irene report in if you have them (9211)

Bethlehem Franklin Bethlehem school boards nearing alliance

Bound Brook

bull Flood prevention project in Bound Brook keeps Hurricane Irene from becoming another Floyd (83011)

bull Parts of Bound Brook South Bound Brook Manville under mandatory evacuation order (82811)

Branchburg Bald eagle sighting

Bridgewater

bull Pfizer anticipates EPA cleanup plan for American Cyanamid site in Bridgewater (121611)

bull French drug maker plans to move its US business to Bridgewater (121511)

bull Federal environmental officials close to decision on $205M clean-up proposal of Bridgewaters Cyanamid property (121111)

bull The bike path is clear and the going is easy at Duke Island Park in Bridgewater (13111)

Califon

bull Hunterdon complaints of power outages prompt review of utilities by state Board of Public Utilities (112911)

bull Snowstorm results in $15000 bill to Califon for tree branch clean-up (112611)

Chester Borough New Jersey residents and communities cope with effects of pre-Halloween snowstorm (103111)

Chester Township North Hunterdon High School sets metro record for donating blood Clinton Town Clinton seeking new bids for sewer line manhole work (111411)

Clinton Township Village Green development in Annandale gets final approval to build on former Warren Lumber site (111411)

Cranbury Horse owner forestalls sale to slaughterhouse (121011)

Delaware Eroded Delaware Township roads popular with cyclists pedestrians prompt speed limit worries (121311)

Dunellen Thunderstorms roll through NJ causing lightning strikes and roof collapse (61711)

East Amwell Hunterdon Freeholders delay changes for Solid Waste Advisory Council after SWAC members object (12711)

2 Gathered July-December 2011 3 With assistance from Melissa Basile MCRP 2012 4 Gathered from the Star Ledger New Jersey County and Town Pages httpwwwnjcomlocal under ldquoMorerdquo ndashFor example httptopicsnjcomtagbernardspostshtml substituting the name of the town of interest

Appendix A News Snapshots

4

Task One Community Assessment The deliverables for Community Assessment are to develop a background of census data a general report on each of the ninety-nine Raritan Basin municipalities (including watershed area maps for certain data) and the initial stages of a series of interviews with key municipal stakeholders The data presented here is preliminary and will continue to be updated throughout the project For all of this work the study area is the Raritan River Basin and is presented either by municipality or by watershed management area (WMA) WMA 8 ( the North and Upper Branches) WMA 9 (Lower Raritan) and WMA 10 (the Millstone River)

Demographics

The first step in the Community Assessment is a report on certain Census data for the ninety-nine towns in the watershed as described in Table 1 below The report on this work is presented in Spreadsheet 1 Demographic Data (See Separate Spreadsheet File)

Municipality County by watershed Population Upper Raritan (8) 1990 2000 2010 Lower Raritan (9) Change in Population 00-10 Millstone (10) Wealth Education (25 yo and up) Med household income high school graduate or higher Average household income Bachelorrsquos degree or higher Med Home Value (Owner Occupied) Employment (16 yo and up) Insurance of workforce employed Policies in force as of 83111 commuting 1hour+ to work Claims reported to FEMA 1178-83111 Workforce bull Total Losses Population over 16 yo bull Closed Losses Pop over 16 as of workforce bull Open Losses bull CWOP Losses Government bull Total Payments Mayor Term Engineer

Table 1 Demographic Data 1990 2000 2010

5

Map 1 Watershed Management Area 8 by FEMA Payout

Map 2 Watershed Management Area 9 by FEMA Payout

In addition we reviewed media coverage for these municipalities looking at social environmental and political issues That initial report is provided ldquoSocial Environmental and Political Snapshotsrdquo and is attached as Appendix A A summary of Safety produced from the NJ State Police Uniform Crime Reports is presented in Attachment B Data Maps The second component of the initial phase of the Community Assessment looks at the geography of the region to identify areas where enhanced mitigation strategies might be most productive in reducing impacts to the most affected communities This resulted in several series of maps for the region

NFIP Payout Maps The first set of maps display the National Flood Insurance Program (NFIP) payouts in the study region by watershed management area delineated by the NJDEP and NJ Geological Survey) The payout amounts were joined to the maps using the NFIP statistics labeled Loss Statistics New Jersey as of 83111 This information does not take into account damages sustained by Hurricane Irene but it does offer a 33 year history of the NFIP loss claims and payouts to policy holders from 1978 to immediately prior to that event The maps were color coded to easily display the breakdown in payouts in the individual municipalities in each watershed The data shows some municipalities suffer dramatically more loss than some of their neighbors particularly in WMA 9 where Bound Brook has a cumulative amount of NFIP payouts totaling over $53M while other municipalities in the watershed Metuchen Borough for example have under $10000 in total NFIP payouts in the same period of time The data are presented in Maps 1 2 and 3 below and in full size in Appendix C

Land Use Maps To further examine opportunities for enhanced prevention of damage from flood waters we created three maps to explore land use (Maps 4 5 and 6) These maps were created from the New Jersey State Development and Redevelopment Plan (SDRP) administered by the NJ Office for Planning Advocacy which delineates planning areas throughout the state The SDRP seeks to provide a

Map 3 Watershed Management Area 10 by FEMA Payout

6

balance between growth and conservation by designating planning areas that share common conditions with regard to development and environmental features These planning areas (PAs) are as follows

Growth Areas PA1 Metropolitan Planning Areas(up to 80-90 impervious coverage with development) PA2 Suburban Planning Areas (up to 30 impervious coverage with development) Designated Centers

Limited Growth Areas planning in these planning areas should promote a balance of conservation and limited growth PA3 Fringe Planning Areas PA4 Rural Planning Areas PA5 Environmentally Sensitive Planning Areas Conservation Areas (below 30 impervious coverage with development and limits or excludes sewer infrastructure) PA3 Fringe Planning Areas PA4 Rural Planning Areas PA5 Environmentally Sensitive Planning Areas

Map 6 WMA 10 Planning Areas

Map 5 WMA 9 Planning Areas

Map 4 WMA 8 Planning Areas

7

Interviews

The second phase of this work currently underway involves in-person interviews with 20-25 key regional municipal officials to establish the following

bull Importance of flood mitigation among other concerns schools police fire service traffic management trash collection parking street repair housing maintenance and repair aesthetics and other quality of life issues

bull What they know about flood mitigation flooding the degree of damage and the causes bull Their understanding of climate impacts on flooding and mitigation strategies bull Local stormwater management measures (retention basin management to rain barrels bull Preferred flood mitigation strategies both locally and regionally bull Level of local engagement and what causes citizens to become engaged in local issues

At this time the following municipalities are to be interviewed with either the mayor the administrator and or the director of the local Office of Emergency Management

Municipality CO Municipality CO Berkeley Heights Union Middlesex Boro Msx Bound Brook Mayor amp OEM Som Milltown Msx Branchburg Som Monroe (Engineer Ernie Feist) Msx Clinton Township Hun North Plainfield Msx Edison John Medina Msx Scotch Plains Union Franklin Som Raritan Township Hun Hightstown Mer Watchung Som Hopewell Mer Washington Twp Morris Manville Administrator amp OEM Som Woodbridge Msx Marlboro Mon

Next Steps

The next steps will involve closer analysis of the census data as we proceed with municipal interviews The goal is to talk to people at various levels not just those who are most directly affected but also those who may be upstream from more impacted communities The next mapping exercise will be to add contours in order to verify the drainage and to calculate current potential mitigation in the uppermost communities through major stormwater mitigation efforts The use of impervious surface coverages from the state as well as the application of independent models will lend more insight into the potential for stormwater management BMPs to aid in overall impact reduction

The findings of the interviews will form a discussion guide at a dinner meeting with municipal and county leaders at the Bloustein School on Tuesday January 31 2012 The discussion will focus on further action for a regional volume-focused flood mitigation strategy Focus Groups Several focus groups are also scheduled Milltown and Manville are scheduled for January Two others will be held in February and March These meetings will be held with groups of key citizen leaders in selected municipalities to determine their knowledge of the issue significance to them their desire for further information about the subject and desired mode of transmission of the information and willingness to become involved in the issue in their own town These sessions will also address the questions asked of the initial stakeholders to verify unity of thought and understanding of issues across the community

8

In addition a series of meetings with regional engineers are being held to engage them in the mitigation conversation from the perspective of stormwater management and overall adaptation of green infrastructure practices Their inputs will be gathered in January and February through a series of conference calls on the following topics

bull regulation changes bull policy directions bull professional education needs bull public education needs bull financing mitigation

They will meet together in early March to share their findings and engage with municipal officials and another cohort regional business leaders

Pilot Survey A draft survey will be created and a pilot conducted for a larger regional survey This would be tested with a limited number of residents of the basin to determine their knowledge of the issue significance to them their desire for further information about the subject and desired mode of transmission of the information and willingness to become involved in the issue in their own towns A larger application of the survey would not be fully implemented as part of this project

Task Two Economic Modeling The economic modeling element of this project is designed to look at best ways to estimate benefits to municipalities (and larger governmental units) of flood mitigation projects for the Raritan watershed For this section two draft reports are provided a flood expense taxonomy (separate spreadsheet) and a literature review

Flood Expense Taxonomy

The flood expense taxonomy (see Separate Spreadsheet File) provides a framework for identifying and categorizing the costs associated with flooding events We developed the taxonomy drawing from academic literature prior reports of flooding events and newspaper articles covering flooding events The taxonomy identifies potential flood-related expenses for four sectors of the economy households commercial businesses municipalities and agriculture The potential costs to each sector are cross-categorized with respect to the time sequence of the flooding event pre-flooding flooding and post flooding We will continue to identify additional potential costs and where possible populate those findings with estimated costs Additionally we will add to this taxonomy the potential economic benefits (if any) to the same four sectors as a result of removing structures located in the floodplain

Literature Review

The annotated literature review (Attachment D) includes two types of articles those that examine the relationship between floods and housing prices and those that estimate the costs of flooding on municipalities and their residents and businesses The articles that examine the relationship between flooding events and home prices generally use hedonic pricing models to explain the difference in home value (as measured by sales price) between homes that are on the floodplain and similar houses that are not on it The articles generally indicate that prices for homes located on floodplains are discounted relative to similar homes not on the floodplain One article offers a meta-analysis of the empirical literature on home prices and flooding events The articles in our taxonomy examine the economic impacts of flooding events from England the Red-River Basin in the Western US and the Delaware

9

River in New Jersey We also include a meta-analysis that re-estimates National Weather Service estimates of national economic damages We will build on this literature review to develop a better understanding of the existing estimates of the full economic costs of flooding events and the potential benefits of removing structures (private and public) from floodplains

Task Three Ideas on Risk Assessment for Flooding on the Raritan The third element of the project includes the first stages of the development of a risk analysis identification model to focus on all hazardsrisks The identification of data to be analyzed will be developed by experts in both mathematical modeling and human health and environmental modeling from engineering and public policy perspectives The team initially Fred Roberts and Paul Kantor now includes Qizhong Guo Associate Professor of Civil and Environmental Engineering and David Robinson and State Climatologist and Geography Professor Brief bio sketches are attached as Appendix E Qizhong Guo and Appendix F David Robinson)

Overview

We have spent the past few months researching alternative approaches to flood risk assessment and gathering as much background information as possible We have gotten information from numerous individuals in the US and abroad and numerous agencies including New Jersey Department of Environmental Protection Department of Homeland Security Risk Management and Analysis USGS We came to understand that it was important to expand our team and so we have added a climatologist Professor David Robinson of the Rutgers Department of Geography the New Jersey State Climatologist and a hydrologist Professor Qizhong Guo of the Rutgers Department of Civil and Environmental Engineering We present here a high level sketch of the approach we are examining and discuss the kinds of detailed information that its implementation will require

Problem Formulation

Our multidisciplinary discussions have led us to understand many of the subtleties involved in doing a risk assessment for flooding on the Raritan We have nearly completed outlining an approach based on what we have learned and the next steps will be to share that approach with those working on other tasks in the project revise as needed and then begin gathering the data needed to do the required modeling

To explain our approach and the issues we have to resolve let us consider first a diagram (Figure 1) which illustrates the kinds of factors and tools we need to take into account in understanding flood modeling and applying that modeling to a risk assessment

10

Figure 1 The interaction of nature models and controllable activities (shown in yellow)

Engineering models can combine information on rainfall specified as to time and space with information on soil moisture conditions seasonality and river levels prior to the rain event the fixed properties of the watershed (not indicated) and the land cover (natural and built environment) to produce flood inundation maps To obtain the precision needed to specify events that might occur every five or ten years requires LiDAR information on elevations We refer to the soil moisture conditions and river levels as ldquoantecedent eventsrdquo and note that seasonality is important because it affects the ability of the soil to absorb water (due to vegetation cover freezing etc) The land coverenvironment is viewed as modifiable (yellow) over the span of a few years That would in turn move the flood contours Flood inundation maps can be used for insurance andor regulatory purposes for risk assessments and to communicate complex geology and meteorology to stakeholder groups and thereby facilitate consideration of alternative policies

In the simplest case we might think of four types of soil moisture conditions dry ldquoaveragerdquo wet or frozen Probably it is sufficient to think of four different seasons We might also think of three basic river levels low average and high Thus there are in principle 4 x 4 x 3 = 48 different combinations of these conditions In reality not all of these combinations are feasible We wonrsquot have frozen ground in summer In our analysis we will probably demonstrate the methods choosing only several of the sensible alternatives for detailed investigation

Risk Modeling

The basic outline for how DHS does risk assessment is shown in Figure 2 below which is taken from a presentation by Isaac Maya of the CREATE Center It defines ldquoriskrdquo R as a function of threat T vulnerability V and consequence C R = f(TVC) Often this is represented as a product R = T x V x C Generally the factors T and V are some kinds of probabilities while C is some measure of loss In our case ldquoattackrdquo is interpreted to mean weather event and ldquosuccessrdquo is interpreted to mean flooding

Precipitation(xt) )tthe past

The models

Soil moisture conditions + seasonality + river levels seasonality

Land Cover (natural + built environment)

River gauge levels

LiDAR information

Flood Contour Maps

waterway

5 year contour

10 year contour

11

Figure 2 Sketch of the DHS computation of risk

There are two branches to the tree in this figure The top branch is the ldquostatus quordquo and the bottom branch involves some mitigation strategy or combination of such strategies The probability of attack P is the probability that there will be a weather event of the particular kind being considered and the probability of ldquosuccessrdquo Q is the conditional probability that such an event if it does occur will lead to flooding at a site being considered The consequences could be of various types In the case of floods they include loss of life economic damage to homes and businesses (direct or indirect) and psychological damage There is an attempt to put each of these types of consequences in terms of dollars The consequence (cost L) of an attack (flood) is usually calculated as a weighted sum of the different kinds of costs in our case of the cost in terms of lives lost economic damage and psychological damage What weights one uses and whether it makes sense to add up these weighted values are issues that will be viewed differently by different stakeholder groups

Complexities of the Watershed Situation

In fact we expect to be able to do more than simply treat ldquofloodrdquo as a yes or no matter as is done in this tree diagram Flood inundation maps will present levels of flood waters and the losses incurred change if there is a different level of flooding (or if there is a different duration for the flooding a factor we will disregard for now) The flood inundation levels are numerical values obtained from hydrological models (they can be later translated into major moderate minor near flood stage and none in making flood forecasts) A more complete model would replace the ldquosuccessfailurerdquo bifurcation in the tree diagram with a number of branches depending on the levelcategory of flood as measured for example by river gauges In our case it makes sense to define category 0 as no flooding category 1 as 0 to 1 feet category 2 as 1 to 2 feet etc

In practice obtaining the flood inundation levels may be complicated even before we introduce mitigating strategies The inundation maps are maps for theoretical storms not maps used to predict floods from an approaching storm These maps for the storms of a 100-year and 500-year ldquoreturn periodrdquo1 are already available and are being updated by FEMA using the recently available high resolution LiDAR data on the topography These maps are called flood insurance rate maps (FIRM) However they are developed from a kind of discrete-event modeling and may conceal hidden

1 This technical expression can be interpreted as representing a 1 of 15 probability per year

Success

Failure Attack

Vulnerability

No Attack

No Risk Management Plan

Threat of Attack

Success

Failure

Attack

No Attack

Risk Management Plan

Risk Management

PlansConsequences

L = Expected[$-Eqiuvalent Loss]

No Loss

No Loss

C + LC = Cost of Risk Management Plan

L = Expected[Reduced Loss] where L lt L

C

C

PQ

(1 - P)

(1 - Q)

P where P lt P

(1 - P)

Q where Q lt Q

(1 - Q)

12

uncertainties Moreover the inundation maps for more frequent storms are not readily available or non-existent It would be possible to build flood maps for 2-year 5-year 10-year 25-year and 50-year storms from available data This would imply gathering data and modeling and will depend on having access to the data Moreover the discrete-event models do not take into account antecedent conditions or seasonality Continuous modeling could take these factors into account but the cost of modeling would then be much higher

Assuming we can get flood inundation levels from hydrological models we would try to calculate the probability Qi that a flood of category i occurs Each category of flooding i would have a level of loss Li associated with it The level of flooding from a given amount of rainfall over a period of time depends not only on the precipitation but also on the soil moisture conditionsseasonalityriver levels and on the land cover (natural + built environment) Initially we may assume a fixed land cover (although this can be changed by mitigation) As noted above there are many different combinations of parameters for soil moisture conditions seasonality river levels We could in principle estimate the probability of each of these combinations (but would need to work out a method for doing so) For each combination we could use the hydrological models to calculate the flood level and thus the flood category Putting all of these calculations together will allow us to calculate the Qi parameters This is probably much too complicated and certainly much too expensive Instead we will pick out several more common combinations among the sensible ones and combine them with different precipitation patterns to define a set of weather events

In short our threats will be different kinds of weather events defined in terms of the profile of rainfall over a given period of time (say in number of inches in first hour number in second hour etc over a 24 hour period) plus antecedent precipitation and seasonal effects and pre-existing river levels We will aim to identify six to ten weather scenarios to analyze For each the probability P will have to be determined Information about precipitation totals in 24 hour periods is available and we could then use this to define a 2-year 5-year 10-year 25-year 50-year 100-year event in terms of precipitation profile per 24-hour period The probability of a 5-year event would be 15 of a 50 year event 150 etc except for the other factors (antecedent events plus seasonality) So we may want to modify P based on these other factors We will need to determine how to do that Note also that climate change will likely affect the value of P but we do not know how to calculate modified P with any degree of confidence

Now what is the interpretation of the Qi parameters If the weather event (precipitation profile) is defined and the antecedent conditions and seasonality are known in principle the hydrological models will produce just one category of flood But it could be that it makes more sense to build some uncertainty into the predictions of the models or it may make sense to assume that we have some uncertainty in knowing the soil moisture conditions or the river levels Thus there is surely still some value to studying the risk assessments using different flood categories

When there is a mitigation strategy presumably the probability of an attack would not be affected so Prsquo in the diagram is the same as P not less than P The probability of ldquosuccessrdquo or flood given a weather event is Qrsquo which is less than Q If we use different flood level categories then we would replace each Qi with a number Qirsquo but note that Qirsquo would be higher than Qi for some i Also the consequences of a flood could be less with expected consequences now Lrsquo presumably less than L (or Lirsquo presumably less than Li) But there is a cost C of the mitigation strategy and that cost applies even if there is no flood

Still using the notation in the diagram we see that if f(TVC) = TxVxC then with no mitigation strategy the risk or total expected cost is PxQxL but if there is a mitigation strategy then (since P = Prsquo) the risk or total expected cost is C + PxQrsquoxLrsquo The difference G = PxQxL ndash (C + PxQrsquoxLrsquo) in these expected costs would be the expected gain or expected risk reduction due to the strategy (If we have different flood level categories the calculation would be modified appropriately)

13

Mitigation Strategies and Elicitation of Some Key Parameters

We will need to decide if we look at flood damage (vulnerability) to a larger region or concentrate on a few towns Our preliminary plan is to do the analysis for a few sample towns for instance Manville and Millstone We will need to coordinate with the other parts of the project on this decision

Consequences will need to be measured by our partners in the project and we will need to coordinate with them

We will look at a variety of mitigation strategies Among them could be improved flood forecasting and warning systems retrofit of green infrastructure (eg impervious area disconnection rain gardens cisterns and rain barrels green roofs pervious concrete) to existing development areas to reduce upstream runoff volume regional (upstream) stormwater detention basins acquisition and demolition and relocation of properties flood-proofing and elevation in place disaster preparedness and response plans flood insurance cleanout of urban drainage systems construction of dams and reservoirs construction of dikes levees and floodwalls channel alterations and high flow diversions and spillways We will want to coordinate with other parts of the project on which mitigation strategies we look at

We will need to work with other parts of the project to get ballpark estimates of reduced costs Lrsquo (or Lirsquo) of floods

Under a given mitigation strategy the challenge is to produce inundation maps that will allow us to determine vulnerability (whether there is a flood) In principle there are hydrological models for doing this or such models could be developed but to do the actual calculations will be much more expensive than we can afford in this modest project Thus the best we can do is pick a few sample mitigation strategies for which we could make educated guesses as to vulnerabilities ndash just to illustrate the methodology Our thoughts are to start with three types of strategies retrofit of green infrastructure some sort of upstream mitigation that would have impact on downstream flooding and improved forecasting (eg improved monitoring from an optimum number of precipitation gauges merged station and radar precipitation estimates soil moisture observations evapotranspiration estimates and additional stream gauges) The last of these three would have no impact on the probabilities Q or Qi and also will not require changed inundation maps Only our calculation as to consequence would be affected since the whole idea of improved forecasting would be to allow preparation so as to reduce the potential flood damage

A full risk assessment would want to look not at individual mitigation strategies but at ldquoportfoliosrdquo of mitigation strategies Of course all of what was said above about ldquomitigation strategyrdquo could apply to a portfolio of strategies But the inundation maps will be much harder to obtain for portfolios of strategies than for individual ones As noted some of the mitigation strategies will not affect inundation maps Looking at mitigation strategies separately avoids the complication that several of these strategies may interact with each other in complex ways Some portfolios of mitigation strategies will be too expensive or infeasible for other reasons We may want to look at budgetary constraints and try to look at a few simple portfolios of strategies that have total investment cost (sum of Crsquos for each strategy in the portfolio) less than a given amount Then the question will become What is a measure of the value of a given portfolio Are we trying to maximize total expected gain (sum of G for all strategies in the portfolio) Or are we optimizing something else Or is this a ldquomulti-attributed utilityrdquo problem

14

Social Environmental and Political News Snapshots2 Hunterdon Mercer Middlesex Monmouth Morris Somerset Union

RARITAN RIVER TOWN SOCIAL ENVIRONMENTAL AND POLITICAL NEWS SNAPSHOTS 34

Alexandria 4-way stop to be set up near Delaware Valley High School

Bedminster

bull Clarence Dillon Library in Bedminster to host a series of Salons (92711) bull Bedminster bicyclist killed after crash with truck in Far Hills (12111)

Berkeley Heights Berkeley Heights childrenrsquos library bounces back Bernards Outraged NJ towns ask state to let them pull plug on JCPampL (12211)

Bernardsville Looking for updates on trail conditions since Hurricane Irene report in if you have them (9211)

Bethlehem Franklin Bethlehem school boards nearing alliance

Bound Brook

bull Flood prevention project in Bound Brook keeps Hurricane Irene from becoming another Floyd (83011)

bull Parts of Bound Brook South Bound Brook Manville under mandatory evacuation order (82811)

Branchburg Bald eagle sighting

Bridgewater

bull Pfizer anticipates EPA cleanup plan for American Cyanamid site in Bridgewater (121611)

bull French drug maker plans to move its US business to Bridgewater (121511)

bull Federal environmental officials close to decision on $205M clean-up proposal of Bridgewaters Cyanamid property (121111)

bull The bike path is clear and the going is easy at Duke Island Park in Bridgewater (13111)

Califon

bull Hunterdon complaints of power outages prompt review of utilities by state Board of Public Utilities (112911)

bull Snowstorm results in $15000 bill to Califon for tree branch clean-up (112611)

Chester Borough New Jersey residents and communities cope with effects of pre-Halloween snowstorm (103111)

Chester Township North Hunterdon High School sets metro record for donating blood Clinton Town Clinton seeking new bids for sewer line manhole work (111411)

Clinton Township Village Green development in Annandale gets final approval to build on former Warren Lumber site (111411)

Cranbury Horse owner forestalls sale to slaughterhouse (121011)

Delaware Eroded Delaware Township roads popular with cyclists pedestrians prompt speed limit worries (121311)

Dunellen Thunderstorms roll through NJ causing lightning strikes and roof collapse (61711)

East Amwell Hunterdon Freeholders delay changes for Solid Waste Advisory Council after SWAC members object (12711)

2 Gathered July-December 2011 3 With assistance from Melissa Basile MCRP 2012 4 Gathered from the Star Ledger New Jersey County and Town Pages httpwwwnjcomlocal under ldquoMorerdquo ndashFor example httptopicsnjcomtagbernardspostshtml substituting the name of the town of interest

Appendix A News Snapshots

5

Map 1 Watershed Management Area 8 by FEMA Payout

Map 2 Watershed Management Area 9 by FEMA Payout

In addition we reviewed media coverage for these municipalities looking at social environmental and political issues That initial report is provided ldquoSocial Environmental and Political Snapshotsrdquo and is attached as Appendix A A summary of Safety produced from the NJ State Police Uniform Crime Reports is presented in Attachment B Data Maps The second component of the initial phase of the Community Assessment looks at the geography of the region to identify areas where enhanced mitigation strategies might be most productive in reducing impacts to the most affected communities This resulted in several series of maps for the region

NFIP Payout Maps The first set of maps display the National Flood Insurance Program (NFIP) payouts in the study region by watershed management area delineated by the NJDEP and NJ Geological Survey) The payout amounts were joined to the maps using the NFIP statistics labeled Loss Statistics New Jersey as of 83111 This information does not take into account damages sustained by Hurricane Irene but it does offer a 33 year history of the NFIP loss claims and payouts to policy holders from 1978 to immediately prior to that event The maps were color coded to easily display the breakdown in payouts in the individual municipalities in each watershed The data shows some municipalities suffer dramatically more loss than some of their neighbors particularly in WMA 9 where Bound Brook has a cumulative amount of NFIP payouts totaling over $53M while other municipalities in the watershed Metuchen Borough for example have under $10000 in total NFIP payouts in the same period of time The data are presented in Maps 1 2 and 3 below and in full size in Appendix C

Land Use Maps To further examine opportunities for enhanced prevention of damage from flood waters we created three maps to explore land use (Maps 4 5 and 6) These maps were created from the New Jersey State Development and Redevelopment Plan (SDRP) administered by the NJ Office for Planning Advocacy which delineates planning areas throughout the state The SDRP seeks to provide a

Map 3 Watershed Management Area 10 by FEMA Payout

6

balance between growth and conservation by designating planning areas that share common conditions with regard to development and environmental features These planning areas (PAs) are as follows

Growth Areas PA1 Metropolitan Planning Areas(up to 80-90 impervious coverage with development) PA2 Suburban Planning Areas (up to 30 impervious coverage with development) Designated Centers

Limited Growth Areas planning in these planning areas should promote a balance of conservation and limited growth PA3 Fringe Planning Areas PA4 Rural Planning Areas PA5 Environmentally Sensitive Planning Areas Conservation Areas (below 30 impervious coverage with development and limits or excludes sewer infrastructure) PA3 Fringe Planning Areas PA4 Rural Planning Areas PA5 Environmentally Sensitive Planning Areas

Map 6 WMA 10 Planning Areas

Map 5 WMA 9 Planning Areas

Map 4 WMA 8 Planning Areas

7

Interviews

The second phase of this work currently underway involves in-person interviews with 20-25 key regional municipal officials to establish the following

bull Importance of flood mitigation among other concerns schools police fire service traffic management trash collection parking street repair housing maintenance and repair aesthetics and other quality of life issues

bull What they know about flood mitigation flooding the degree of damage and the causes bull Their understanding of climate impacts on flooding and mitigation strategies bull Local stormwater management measures (retention basin management to rain barrels bull Preferred flood mitigation strategies both locally and regionally bull Level of local engagement and what causes citizens to become engaged in local issues

At this time the following municipalities are to be interviewed with either the mayor the administrator and or the director of the local Office of Emergency Management

Municipality CO Municipality CO Berkeley Heights Union Middlesex Boro Msx Bound Brook Mayor amp OEM Som Milltown Msx Branchburg Som Monroe (Engineer Ernie Feist) Msx Clinton Township Hun North Plainfield Msx Edison John Medina Msx Scotch Plains Union Franklin Som Raritan Township Hun Hightstown Mer Watchung Som Hopewell Mer Washington Twp Morris Manville Administrator amp OEM Som Woodbridge Msx Marlboro Mon

Next Steps

The next steps will involve closer analysis of the census data as we proceed with municipal interviews The goal is to talk to people at various levels not just those who are most directly affected but also those who may be upstream from more impacted communities The next mapping exercise will be to add contours in order to verify the drainage and to calculate current potential mitigation in the uppermost communities through major stormwater mitigation efforts The use of impervious surface coverages from the state as well as the application of independent models will lend more insight into the potential for stormwater management BMPs to aid in overall impact reduction

The findings of the interviews will form a discussion guide at a dinner meeting with municipal and county leaders at the Bloustein School on Tuesday January 31 2012 The discussion will focus on further action for a regional volume-focused flood mitigation strategy Focus Groups Several focus groups are also scheduled Milltown and Manville are scheduled for January Two others will be held in February and March These meetings will be held with groups of key citizen leaders in selected municipalities to determine their knowledge of the issue significance to them their desire for further information about the subject and desired mode of transmission of the information and willingness to become involved in the issue in their own town These sessions will also address the questions asked of the initial stakeholders to verify unity of thought and understanding of issues across the community

8

In addition a series of meetings with regional engineers are being held to engage them in the mitigation conversation from the perspective of stormwater management and overall adaptation of green infrastructure practices Their inputs will be gathered in January and February through a series of conference calls on the following topics

bull regulation changes bull policy directions bull professional education needs bull public education needs bull financing mitigation

They will meet together in early March to share their findings and engage with municipal officials and another cohort regional business leaders

Pilot Survey A draft survey will be created and a pilot conducted for a larger regional survey This would be tested with a limited number of residents of the basin to determine their knowledge of the issue significance to them their desire for further information about the subject and desired mode of transmission of the information and willingness to become involved in the issue in their own towns A larger application of the survey would not be fully implemented as part of this project

Task Two Economic Modeling The economic modeling element of this project is designed to look at best ways to estimate benefits to municipalities (and larger governmental units) of flood mitigation projects for the Raritan watershed For this section two draft reports are provided a flood expense taxonomy (separate spreadsheet) and a literature review

Flood Expense Taxonomy

The flood expense taxonomy (see Separate Spreadsheet File) provides a framework for identifying and categorizing the costs associated with flooding events We developed the taxonomy drawing from academic literature prior reports of flooding events and newspaper articles covering flooding events The taxonomy identifies potential flood-related expenses for four sectors of the economy households commercial businesses municipalities and agriculture The potential costs to each sector are cross-categorized with respect to the time sequence of the flooding event pre-flooding flooding and post flooding We will continue to identify additional potential costs and where possible populate those findings with estimated costs Additionally we will add to this taxonomy the potential economic benefits (if any) to the same four sectors as a result of removing structures located in the floodplain

Literature Review

The annotated literature review (Attachment D) includes two types of articles those that examine the relationship between floods and housing prices and those that estimate the costs of flooding on municipalities and their residents and businesses The articles that examine the relationship between flooding events and home prices generally use hedonic pricing models to explain the difference in home value (as measured by sales price) between homes that are on the floodplain and similar houses that are not on it The articles generally indicate that prices for homes located on floodplains are discounted relative to similar homes not on the floodplain One article offers a meta-analysis of the empirical literature on home prices and flooding events The articles in our taxonomy examine the economic impacts of flooding events from England the Red-River Basin in the Western US and the Delaware

9

River in New Jersey We also include a meta-analysis that re-estimates National Weather Service estimates of national economic damages We will build on this literature review to develop a better understanding of the existing estimates of the full economic costs of flooding events and the potential benefits of removing structures (private and public) from floodplains

Task Three Ideas on Risk Assessment for Flooding on the Raritan The third element of the project includes the first stages of the development of a risk analysis identification model to focus on all hazardsrisks The identification of data to be analyzed will be developed by experts in both mathematical modeling and human health and environmental modeling from engineering and public policy perspectives The team initially Fred Roberts and Paul Kantor now includes Qizhong Guo Associate Professor of Civil and Environmental Engineering and David Robinson and State Climatologist and Geography Professor Brief bio sketches are attached as Appendix E Qizhong Guo and Appendix F David Robinson)

Overview

We have spent the past few months researching alternative approaches to flood risk assessment and gathering as much background information as possible We have gotten information from numerous individuals in the US and abroad and numerous agencies including New Jersey Department of Environmental Protection Department of Homeland Security Risk Management and Analysis USGS We came to understand that it was important to expand our team and so we have added a climatologist Professor David Robinson of the Rutgers Department of Geography the New Jersey State Climatologist and a hydrologist Professor Qizhong Guo of the Rutgers Department of Civil and Environmental Engineering We present here a high level sketch of the approach we are examining and discuss the kinds of detailed information that its implementation will require

Problem Formulation

Our multidisciplinary discussions have led us to understand many of the subtleties involved in doing a risk assessment for flooding on the Raritan We have nearly completed outlining an approach based on what we have learned and the next steps will be to share that approach with those working on other tasks in the project revise as needed and then begin gathering the data needed to do the required modeling

To explain our approach and the issues we have to resolve let us consider first a diagram (Figure 1) which illustrates the kinds of factors and tools we need to take into account in understanding flood modeling and applying that modeling to a risk assessment

10

Figure 1 The interaction of nature models and controllable activities (shown in yellow)

Engineering models can combine information on rainfall specified as to time and space with information on soil moisture conditions seasonality and river levels prior to the rain event the fixed properties of the watershed (not indicated) and the land cover (natural and built environment) to produce flood inundation maps To obtain the precision needed to specify events that might occur every five or ten years requires LiDAR information on elevations We refer to the soil moisture conditions and river levels as ldquoantecedent eventsrdquo and note that seasonality is important because it affects the ability of the soil to absorb water (due to vegetation cover freezing etc) The land coverenvironment is viewed as modifiable (yellow) over the span of a few years That would in turn move the flood contours Flood inundation maps can be used for insurance andor regulatory purposes for risk assessments and to communicate complex geology and meteorology to stakeholder groups and thereby facilitate consideration of alternative policies

In the simplest case we might think of four types of soil moisture conditions dry ldquoaveragerdquo wet or frozen Probably it is sufficient to think of four different seasons We might also think of three basic river levels low average and high Thus there are in principle 4 x 4 x 3 = 48 different combinations of these conditions In reality not all of these combinations are feasible We wonrsquot have frozen ground in summer In our analysis we will probably demonstrate the methods choosing only several of the sensible alternatives for detailed investigation

Risk Modeling

The basic outline for how DHS does risk assessment is shown in Figure 2 below which is taken from a presentation by Isaac Maya of the CREATE Center It defines ldquoriskrdquo R as a function of threat T vulnerability V and consequence C R = f(TVC) Often this is represented as a product R = T x V x C Generally the factors T and V are some kinds of probabilities while C is some measure of loss In our case ldquoattackrdquo is interpreted to mean weather event and ldquosuccessrdquo is interpreted to mean flooding

Precipitation(xt) )tthe past

The models

Soil moisture conditions + seasonality + river levels seasonality

Land Cover (natural + built environment)

River gauge levels

LiDAR information

Flood Contour Maps

waterway

5 year contour

10 year contour

11

Figure 2 Sketch of the DHS computation of risk

There are two branches to the tree in this figure The top branch is the ldquostatus quordquo and the bottom branch involves some mitigation strategy or combination of such strategies The probability of attack P is the probability that there will be a weather event of the particular kind being considered and the probability of ldquosuccessrdquo Q is the conditional probability that such an event if it does occur will lead to flooding at a site being considered The consequences could be of various types In the case of floods they include loss of life economic damage to homes and businesses (direct or indirect) and psychological damage There is an attempt to put each of these types of consequences in terms of dollars The consequence (cost L) of an attack (flood) is usually calculated as a weighted sum of the different kinds of costs in our case of the cost in terms of lives lost economic damage and psychological damage What weights one uses and whether it makes sense to add up these weighted values are issues that will be viewed differently by different stakeholder groups

Complexities of the Watershed Situation

In fact we expect to be able to do more than simply treat ldquofloodrdquo as a yes or no matter as is done in this tree diagram Flood inundation maps will present levels of flood waters and the losses incurred change if there is a different level of flooding (or if there is a different duration for the flooding a factor we will disregard for now) The flood inundation levels are numerical values obtained from hydrological models (they can be later translated into major moderate minor near flood stage and none in making flood forecasts) A more complete model would replace the ldquosuccessfailurerdquo bifurcation in the tree diagram with a number of branches depending on the levelcategory of flood as measured for example by river gauges In our case it makes sense to define category 0 as no flooding category 1 as 0 to 1 feet category 2 as 1 to 2 feet etc

In practice obtaining the flood inundation levels may be complicated even before we introduce mitigating strategies The inundation maps are maps for theoretical storms not maps used to predict floods from an approaching storm These maps for the storms of a 100-year and 500-year ldquoreturn periodrdquo1 are already available and are being updated by FEMA using the recently available high resolution LiDAR data on the topography These maps are called flood insurance rate maps (FIRM) However they are developed from a kind of discrete-event modeling and may conceal hidden

1 This technical expression can be interpreted as representing a 1 of 15 probability per year

Success

Failure Attack

Vulnerability

No Attack

No Risk Management Plan

Threat of Attack

Success

Failure

Attack

No Attack

Risk Management Plan

Risk Management

PlansConsequences

L = Expected[$-Eqiuvalent Loss]

No Loss

No Loss

C + LC = Cost of Risk Management Plan

L = Expected[Reduced Loss] where L lt L

C

C

PQ

(1 - P)

(1 - Q)

P where P lt P

(1 - P)

Q where Q lt Q

(1 - Q)

12

uncertainties Moreover the inundation maps for more frequent storms are not readily available or non-existent It would be possible to build flood maps for 2-year 5-year 10-year 25-year and 50-year storms from available data This would imply gathering data and modeling and will depend on having access to the data Moreover the discrete-event models do not take into account antecedent conditions or seasonality Continuous modeling could take these factors into account but the cost of modeling would then be much higher

Assuming we can get flood inundation levels from hydrological models we would try to calculate the probability Qi that a flood of category i occurs Each category of flooding i would have a level of loss Li associated with it The level of flooding from a given amount of rainfall over a period of time depends not only on the precipitation but also on the soil moisture conditionsseasonalityriver levels and on the land cover (natural + built environment) Initially we may assume a fixed land cover (although this can be changed by mitigation) As noted above there are many different combinations of parameters for soil moisture conditions seasonality river levels We could in principle estimate the probability of each of these combinations (but would need to work out a method for doing so) For each combination we could use the hydrological models to calculate the flood level and thus the flood category Putting all of these calculations together will allow us to calculate the Qi parameters This is probably much too complicated and certainly much too expensive Instead we will pick out several more common combinations among the sensible ones and combine them with different precipitation patterns to define a set of weather events

In short our threats will be different kinds of weather events defined in terms of the profile of rainfall over a given period of time (say in number of inches in first hour number in second hour etc over a 24 hour period) plus antecedent precipitation and seasonal effects and pre-existing river levels We will aim to identify six to ten weather scenarios to analyze For each the probability P will have to be determined Information about precipitation totals in 24 hour periods is available and we could then use this to define a 2-year 5-year 10-year 25-year 50-year 100-year event in terms of precipitation profile per 24-hour period The probability of a 5-year event would be 15 of a 50 year event 150 etc except for the other factors (antecedent events plus seasonality) So we may want to modify P based on these other factors We will need to determine how to do that Note also that climate change will likely affect the value of P but we do not know how to calculate modified P with any degree of confidence

Now what is the interpretation of the Qi parameters If the weather event (precipitation profile) is defined and the antecedent conditions and seasonality are known in principle the hydrological models will produce just one category of flood But it could be that it makes more sense to build some uncertainty into the predictions of the models or it may make sense to assume that we have some uncertainty in knowing the soil moisture conditions or the river levels Thus there is surely still some value to studying the risk assessments using different flood categories

When there is a mitigation strategy presumably the probability of an attack would not be affected so Prsquo in the diagram is the same as P not less than P The probability of ldquosuccessrdquo or flood given a weather event is Qrsquo which is less than Q If we use different flood level categories then we would replace each Qi with a number Qirsquo but note that Qirsquo would be higher than Qi for some i Also the consequences of a flood could be less with expected consequences now Lrsquo presumably less than L (or Lirsquo presumably less than Li) But there is a cost C of the mitigation strategy and that cost applies even if there is no flood

Still using the notation in the diagram we see that if f(TVC) = TxVxC then with no mitigation strategy the risk or total expected cost is PxQxL but if there is a mitigation strategy then (since P = Prsquo) the risk or total expected cost is C + PxQrsquoxLrsquo The difference G = PxQxL ndash (C + PxQrsquoxLrsquo) in these expected costs would be the expected gain or expected risk reduction due to the strategy (If we have different flood level categories the calculation would be modified appropriately)

13

Mitigation Strategies and Elicitation of Some Key Parameters

We will need to decide if we look at flood damage (vulnerability) to a larger region or concentrate on a few towns Our preliminary plan is to do the analysis for a few sample towns for instance Manville and Millstone We will need to coordinate with the other parts of the project on this decision

Consequences will need to be measured by our partners in the project and we will need to coordinate with them

We will look at a variety of mitigation strategies Among them could be improved flood forecasting and warning systems retrofit of green infrastructure (eg impervious area disconnection rain gardens cisterns and rain barrels green roofs pervious concrete) to existing development areas to reduce upstream runoff volume regional (upstream) stormwater detention basins acquisition and demolition and relocation of properties flood-proofing and elevation in place disaster preparedness and response plans flood insurance cleanout of urban drainage systems construction of dams and reservoirs construction of dikes levees and floodwalls channel alterations and high flow diversions and spillways We will want to coordinate with other parts of the project on which mitigation strategies we look at

We will need to work with other parts of the project to get ballpark estimates of reduced costs Lrsquo (or Lirsquo) of floods

Under a given mitigation strategy the challenge is to produce inundation maps that will allow us to determine vulnerability (whether there is a flood) In principle there are hydrological models for doing this or such models could be developed but to do the actual calculations will be much more expensive than we can afford in this modest project Thus the best we can do is pick a few sample mitigation strategies for which we could make educated guesses as to vulnerabilities ndash just to illustrate the methodology Our thoughts are to start with three types of strategies retrofit of green infrastructure some sort of upstream mitigation that would have impact on downstream flooding and improved forecasting (eg improved monitoring from an optimum number of precipitation gauges merged station and radar precipitation estimates soil moisture observations evapotranspiration estimates and additional stream gauges) The last of these three would have no impact on the probabilities Q or Qi and also will not require changed inundation maps Only our calculation as to consequence would be affected since the whole idea of improved forecasting would be to allow preparation so as to reduce the potential flood damage

A full risk assessment would want to look not at individual mitigation strategies but at ldquoportfoliosrdquo of mitigation strategies Of course all of what was said above about ldquomitigation strategyrdquo could apply to a portfolio of strategies But the inundation maps will be much harder to obtain for portfolios of strategies than for individual ones As noted some of the mitigation strategies will not affect inundation maps Looking at mitigation strategies separately avoids the complication that several of these strategies may interact with each other in complex ways Some portfolios of mitigation strategies will be too expensive or infeasible for other reasons We may want to look at budgetary constraints and try to look at a few simple portfolios of strategies that have total investment cost (sum of Crsquos for each strategy in the portfolio) less than a given amount Then the question will become What is a measure of the value of a given portfolio Are we trying to maximize total expected gain (sum of G for all strategies in the portfolio) Or are we optimizing something else Or is this a ldquomulti-attributed utilityrdquo problem

14

Social Environmental and Political News Snapshots2 Hunterdon Mercer Middlesex Monmouth Morris Somerset Union

RARITAN RIVER TOWN SOCIAL ENVIRONMENTAL AND POLITICAL NEWS SNAPSHOTS 34

Alexandria 4-way stop to be set up near Delaware Valley High School

Bedminster

bull Clarence Dillon Library in Bedminster to host a series of Salons (92711) bull Bedminster bicyclist killed after crash with truck in Far Hills (12111)

Berkeley Heights Berkeley Heights childrenrsquos library bounces back Bernards Outraged NJ towns ask state to let them pull plug on JCPampL (12211)

Bernardsville Looking for updates on trail conditions since Hurricane Irene report in if you have them (9211)

Bethlehem Franklin Bethlehem school boards nearing alliance

Bound Brook

bull Flood prevention project in Bound Brook keeps Hurricane Irene from becoming another Floyd (83011)

bull Parts of Bound Brook South Bound Brook Manville under mandatory evacuation order (82811)

Branchburg Bald eagle sighting

Bridgewater

bull Pfizer anticipates EPA cleanup plan for American Cyanamid site in Bridgewater (121611)

bull French drug maker plans to move its US business to Bridgewater (121511)

bull Federal environmental officials close to decision on $205M clean-up proposal of Bridgewaters Cyanamid property (121111)

bull The bike path is clear and the going is easy at Duke Island Park in Bridgewater (13111)

Califon

bull Hunterdon complaints of power outages prompt review of utilities by state Board of Public Utilities (112911)

bull Snowstorm results in $15000 bill to Califon for tree branch clean-up (112611)

Chester Borough New Jersey residents and communities cope with effects of pre-Halloween snowstorm (103111)

Chester Township North Hunterdon High School sets metro record for donating blood Clinton Town Clinton seeking new bids for sewer line manhole work (111411)

Clinton Township Village Green development in Annandale gets final approval to build on former Warren Lumber site (111411)

Cranbury Horse owner forestalls sale to slaughterhouse (121011)

Delaware Eroded Delaware Township roads popular with cyclists pedestrians prompt speed limit worries (121311)

Dunellen Thunderstorms roll through NJ causing lightning strikes and roof collapse (61711)

East Amwell Hunterdon Freeholders delay changes for Solid Waste Advisory Council after SWAC members object (12711)

2 Gathered July-December 2011 3 With assistance from Melissa Basile MCRP 2012 4 Gathered from the Star Ledger New Jersey County and Town Pages httpwwwnjcomlocal under ldquoMorerdquo ndashFor example httptopicsnjcomtagbernardspostshtml substituting the name of the town of interest

Appendix A News Snapshots

6

balance between growth and conservation by designating planning areas that share common conditions with regard to development and environmental features These planning areas (PAs) are as follows

Growth Areas PA1 Metropolitan Planning Areas(up to 80-90 impervious coverage with development) PA2 Suburban Planning Areas (up to 30 impervious coverage with development) Designated Centers

Limited Growth Areas planning in these planning areas should promote a balance of conservation and limited growth PA3 Fringe Planning Areas PA4 Rural Planning Areas PA5 Environmentally Sensitive Planning Areas Conservation Areas (below 30 impervious coverage with development and limits or excludes sewer infrastructure) PA3 Fringe Planning Areas PA4 Rural Planning Areas PA5 Environmentally Sensitive Planning Areas

Map 6 WMA 10 Planning Areas

Map 5 WMA 9 Planning Areas

Map 4 WMA 8 Planning Areas

7

Interviews

The second phase of this work currently underway involves in-person interviews with 20-25 key regional municipal officials to establish the following

bull Importance of flood mitigation among other concerns schools police fire service traffic management trash collection parking street repair housing maintenance and repair aesthetics and other quality of life issues

bull What they know about flood mitigation flooding the degree of damage and the causes bull Their understanding of climate impacts on flooding and mitigation strategies bull Local stormwater management measures (retention basin management to rain barrels bull Preferred flood mitigation strategies both locally and regionally bull Level of local engagement and what causes citizens to become engaged in local issues

At this time the following municipalities are to be interviewed with either the mayor the administrator and or the director of the local Office of Emergency Management

Municipality CO Municipality CO Berkeley Heights Union Middlesex Boro Msx Bound Brook Mayor amp OEM Som Milltown Msx Branchburg Som Monroe (Engineer Ernie Feist) Msx Clinton Township Hun North Plainfield Msx Edison John Medina Msx Scotch Plains Union Franklin Som Raritan Township Hun Hightstown Mer Watchung Som Hopewell Mer Washington Twp Morris Manville Administrator amp OEM Som Woodbridge Msx Marlboro Mon

Next Steps

The next steps will involve closer analysis of the census data as we proceed with municipal interviews The goal is to talk to people at various levels not just those who are most directly affected but also those who may be upstream from more impacted communities The next mapping exercise will be to add contours in order to verify the drainage and to calculate current potential mitigation in the uppermost communities through major stormwater mitigation efforts The use of impervious surface coverages from the state as well as the application of independent models will lend more insight into the potential for stormwater management BMPs to aid in overall impact reduction

The findings of the interviews will form a discussion guide at a dinner meeting with municipal and county leaders at the Bloustein School on Tuesday January 31 2012 The discussion will focus on further action for a regional volume-focused flood mitigation strategy Focus Groups Several focus groups are also scheduled Milltown and Manville are scheduled for January Two others will be held in February and March These meetings will be held with groups of key citizen leaders in selected municipalities to determine their knowledge of the issue significance to them their desire for further information about the subject and desired mode of transmission of the information and willingness to become involved in the issue in their own town These sessions will also address the questions asked of the initial stakeholders to verify unity of thought and understanding of issues across the community

8

In addition a series of meetings with regional engineers are being held to engage them in the mitigation conversation from the perspective of stormwater management and overall adaptation of green infrastructure practices Their inputs will be gathered in January and February through a series of conference calls on the following topics

bull regulation changes bull policy directions bull professional education needs bull public education needs bull financing mitigation

They will meet together in early March to share their findings and engage with municipal officials and another cohort regional business leaders

Pilot Survey A draft survey will be created and a pilot conducted for a larger regional survey This would be tested with a limited number of residents of the basin to determine their knowledge of the issue significance to them their desire for further information about the subject and desired mode of transmission of the information and willingness to become involved in the issue in their own towns A larger application of the survey would not be fully implemented as part of this project

Task Two Economic Modeling The economic modeling element of this project is designed to look at best ways to estimate benefits to municipalities (and larger governmental units) of flood mitigation projects for the Raritan watershed For this section two draft reports are provided a flood expense taxonomy (separate spreadsheet) and a literature review

Flood Expense Taxonomy

The flood expense taxonomy (see Separate Spreadsheet File) provides a framework for identifying and categorizing the costs associated with flooding events We developed the taxonomy drawing from academic literature prior reports of flooding events and newspaper articles covering flooding events The taxonomy identifies potential flood-related expenses for four sectors of the economy households commercial businesses municipalities and agriculture The potential costs to each sector are cross-categorized with respect to the time sequence of the flooding event pre-flooding flooding and post flooding We will continue to identify additional potential costs and where possible populate those findings with estimated costs Additionally we will add to this taxonomy the potential economic benefits (if any) to the same four sectors as a result of removing structures located in the floodplain

Literature Review

The annotated literature review (Attachment D) includes two types of articles those that examine the relationship between floods and housing prices and those that estimate the costs of flooding on municipalities and their residents and businesses The articles that examine the relationship between flooding events and home prices generally use hedonic pricing models to explain the difference in home value (as measured by sales price) between homes that are on the floodplain and similar houses that are not on it The articles generally indicate that prices for homes located on floodplains are discounted relative to similar homes not on the floodplain One article offers a meta-analysis of the empirical literature on home prices and flooding events The articles in our taxonomy examine the economic impacts of flooding events from England the Red-River Basin in the Western US and the Delaware

9

River in New Jersey We also include a meta-analysis that re-estimates National Weather Service estimates of national economic damages We will build on this literature review to develop a better understanding of the existing estimates of the full economic costs of flooding events and the potential benefits of removing structures (private and public) from floodplains

Task Three Ideas on Risk Assessment for Flooding on the Raritan The third element of the project includes the first stages of the development of a risk analysis identification model to focus on all hazardsrisks The identification of data to be analyzed will be developed by experts in both mathematical modeling and human health and environmental modeling from engineering and public policy perspectives The team initially Fred Roberts and Paul Kantor now includes Qizhong Guo Associate Professor of Civil and Environmental Engineering and David Robinson and State Climatologist and Geography Professor Brief bio sketches are attached as Appendix E Qizhong Guo and Appendix F David Robinson)

Overview

We have spent the past few months researching alternative approaches to flood risk assessment and gathering as much background information as possible We have gotten information from numerous individuals in the US and abroad and numerous agencies including New Jersey Department of Environmental Protection Department of Homeland Security Risk Management and Analysis USGS We came to understand that it was important to expand our team and so we have added a climatologist Professor David Robinson of the Rutgers Department of Geography the New Jersey State Climatologist and a hydrologist Professor Qizhong Guo of the Rutgers Department of Civil and Environmental Engineering We present here a high level sketch of the approach we are examining and discuss the kinds of detailed information that its implementation will require

Problem Formulation

Our multidisciplinary discussions have led us to understand many of the subtleties involved in doing a risk assessment for flooding on the Raritan We have nearly completed outlining an approach based on what we have learned and the next steps will be to share that approach with those working on other tasks in the project revise as needed and then begin gathering the data needed to do the required modeling

To explain our approach and the issues we have to resolve let us consider first a diagram (Figure 1) which illustrates the kinds of factors and tools we need to take into account in understanding flood modeling and applying that modeling to a risk assessment

10

Figure 1 The interaction of nature models and controllable activities (shown in yellow)

Engineering models can combine information on rainfall specified as to time and space with information on soil moisture conditions seasonality and river levels prior to the rain event the fixed properties of the watershed (not indicated) and the land cover (natural and built environment) to produce flood inundation maps To obtain the precision needed to specify events that might occur every five or ten years requires LiDAR information on elevations We refer to the soil moisture conditions and river levels as ldquoantecedent eventsrdquo and note that seasonality is important because it affects the ability of the soil to absorb water (due to vegetation cover freezing etc) The land coverenvironment is viewed as modifiable (yellow) over the span of a few years That would in turn move the flood contours Flood inundation maps can be used for insurance andor regulatory purposes for risk assessments and to communicate complex geology and meteorology to stakeholder groups and thereby facilitate consideration of alternative policies

In the simplest case we might think of four types of soil moisture conditions dry ldquoaveragerdquo wet or frozen Probably it is sufficient to think of four different seasons We might also think of three basic river levels low average and high Thus there are in principle 4 x 4 x 3 = 48 different combinations of these conditions In reality not all of these combinations are feasible We wonrsquot have frozen ground in summer In our analysis we will probably demonstrate the methods choosing only several of the sensible alternatives for detailed investigation

Risk Modeling

The basic outline for how DHS does risk assessment is shown in Figure 2 below which is taken from a presentation by Isaac Maya of the CREATE Center It defines ldquoriskrdquo R as a function of threat T vulnerability V and consequence C R = f(TVC) Often this is represented as a product R = T x V x C Generally the factors T and V are some kinds of probabilities while C is some measure of loss In our case ldquoattackrdquo is interpreted to mean weather event and ldquosuccessrdquo is interpreted to mean flooding

Precipitation(xt) )tthe past

The models

Soil moisture conditions + seasonality + river levels seasonality

Land Cover (natural + built environment)

River gauge levels

LiDAR information

Flood Contour Maps

waterway

5 year contour

10 year contour

11

Figure 2 Sketch of the DHS computation of risk

There are two branches to the tree in this figure The top branch is the ldquostatus quordquo and the bottom branch involves some mitigation strategy or combination of such strategies The probability of attack P is the probability that there will be a weather event of the particular kind being considered and the probability of ldquosuccessrdquo Q is the conditional probability that such an event if it does occur will lead to flooding at a site being considered The consequences could be of various types In the case of floods they include loss of life economic damage to homes and businesses (direct or indirect) and psychological damage There is an attempt to put each of these types of consequences in terms of dollars The consequence (cost L) of an attack (flood) is usually calculated as a weighted sum of the different kinds of costs in our case of the cost in terms of lives lost economic damage and psychological damage What weights one uses and whether it makes sense to add up these weighted values are issues that will be viewed differently by different stakeholder groups

Complexities of the Watershed Situation

In fact we expect to be able to do more than simply treat ldquofloodrdquo as a yes or no matter as is done in this tree diagram Flood inundation maps will present levels of flood waters and the losses incurred change if there is a different level of flooding (or if there is a different duration for the flooding a factor we will disregard for now) The flood inundation levels are numerical values obtained from hydrological models (they can be later translated into major moderate minor near flood stage and none in making flood forecasts) A more complete model would replace the ldquosuccessfailurerdquo bifurcation in the tree diagram with a number of branches depending on the levelcategory of flood as measured for example by river gauges In our case it makes sense to define category 0 as no flooding category 1 as 0 to 1 feet category 2 as 1 to 2 feet etc

In practice obtaining the flood inundation levels may be complicated even before we introduce mitigating strategies The inundation maps are maps for theoretical storms not maps used to predict floods from an approaching storm These maps for the storms of a 100-year and 500-year ldquoreturn periodrdquo1 are already available and are being updated by FEMA using the recently available high resolution LiDAR data on the topography These maps are called flood insurance rate maps (FIRM) However they are developed from a kind of discrete-event modeling and may conceal hidden

1 This technical expression can be interpreted as representing a 1 of 15 probability per year

Success

Failure Attack

Vulnerability

No Attack

No Risk Management Plan

Threat of Attack

Success

Failure

Attack

No Attack

Risk Management Plan

Risk Management

PlansConsequences

L = Expected[$-Eqiuvalent Loss]

No Loss

No Loss

C + LC = Cost of Risk Management Plan

L = Expected[Reduced Loss] where L lt L

C

C

PQ

(1 - P)

(1 - Q)

P where P lt P

(1 - P)

Q where Q lt Q

(1 - Q)

12

uncertainties Moreover the inundation maps for more frequent storms are not readily available or non-existent It would be possible to build flood maps for 2-year 5-year 10-year 25-year and 50-year storms from available data This would imply gathering data and modeling and will depend on having access to the data Moreover the discrete-event models do not take into account antecedent conditions or seasonality Continuous modeling could take these factors into account but the cost of modeling would then be much higher

Assuming we can get flood inundation levels from hydrological models we would try to calculate the probability Qi that a flood of category i occurs Each category of flooding i would have a level of loss Li associated with it The level of flooding from a given amount of rainfall over a period of time depends not only on the precipitation but also on the soil moisture conditionsseasonalityriver levels and on the land cover (natural + built environment) Initially we may assume a fixed land cover (although this can be changed by mitigation) As noted above there are many different combinations of parameters for soil moisture conditions seasonality river levels We could in principle estimate the probability of each of these combinations (but would need to work out a method for doing so) For each combination we could use the hydrological models to calculate the flood level and thus the flood category Putting all of these calculations together will allow us to calculate the Qi parameters This is probably much too complicated and certainly much too expensive Instead we will pick out several more common combinations among the sensible ones and combine them with different precipitation patterns to define a set of weather events

In short our threats will be different kinds of weather events defined in terms of the profile of rainfall over a given period of time (say in number of inches in first hour number in second hour etc over a 24 hour period) plus antecedent precipitation and seasonal effects and pre-existing river levels We will aim to identify six to ten weather scenarios to analyze For each the probability P will have to be determined Information about precipitation totals in 24 hour periods is available and we could then use this to define a 2-year 5-year 10-year 25-year 50-year 100-year event in terms of precipitation profile per 24-hour period The probability of a 5-year event would be 15 of a 50 year event 150 etc except for the other factors (antecedent events plus seasonality) So we may want to modify P based on these other factors We will need to determine how to do that Note also that climate change will likely affect the value of P but we do not know how to calculate modified P with any degree of confidence

Now what is the interpretation of the Qi parameters If the weather event (precipitation profile) is defined and the antecedent conditions and seasonality are known in principle the hydrological models will produce just one category of flood But it could be that it makes more sense to build some uncertainty into the predictions of the models or it may make sense to assume that we have some uncertainty in knowing the soil moisture conditions or the river levels Thus there is surely still some value to studying the risk assessments using different flood categories

When there is a mitigation strategy presumably the probability of an attack would not be affected so Prsquo in the diagram is the same as P not less than P The probability of ldquosuccessrdquo or flood given a weather event is Qrsquo which is less than Q If we use different flood level categories then we would replace each Qi with a number Qirsquo but note that Qirsquo would be higher than Qi for some i Also the consequences of a flood could be less with expected consequences now Lrsquo presumably less than L (or Lirsquo presumably less than Li) But there is a cost C of the mitigation strategy and that cost applies even if there is no flood

Still using the notation in the diagram we see that if f(TVC) = TxVxC then with no mitigation strategy the risk or total expected cost is PxQxL but if there is a mitigation strategy then (since P = Prsquo) the risk or total expected cost is C + PxQrsquoxLrsquo The difference G = PxQxL ndash (C + PxQrsquoxLrsquo) in these expected costs would be the expected gain or expected risk reduction due to the strategy (If we have different flood level categories the calculation would be modified appropriately)

13

Mitigation Strategies and Elicitation of Some Key Parameters

We will need to decide if we look at flood damage (vulnerability) to a larger region or concentrate on a few towns Our preliminary plan is to do the analysis for a few sample towns for instance Manville and Millstone We will need to coordinate with the other parts of the project on this decision

Consequences will need to be measured by our partners in the project and we will need to coordinate with them

We will look at a variety of mitigation strategies Among them could be improved flood forecasting and warning systems retrofit of green infrastructure (eg impervious area disconnection rain gardens cisterns and rain barrels green roofs pervious concrete) to existing development areas to reduce upstream runoff volume regional (upstream) stormwater detention basins acquisition and demolition and relocation of properties flood-proofing and elevation in place disaster preparedness and response plans flood insurance cleanout of urban drainage systems construction of dams and reservoirs construction of dikes levees and floodwalls channel alterations and high flow diversions and spillways We will want to coordinate with other parts of the project on which mitigation strategies we look at

We will need to work with other parts of the project to get ballpark estimates of reduced costs Lrsquo (or Lirsquo) of floods

Under a given mitigation strategy the challenge is to produce inundation maps that will allow us to determine vulnerability (whether there is a flood) In principle there are hydrological models for doing this or such models could be developed but to do the actual calculations will be much more expensive than we can afford in this modest project Thus the best we can do is pick a few sample mitigation strategies for which we could make educated guesses as to vulnerabilities ndash just to illustrate the methodology Our thoughts are to start with three types of strategies retrofit of green infrastructure some sort of upstream mitigation that would have impact on downstream flooding and improved forecasting (eg improved monitoring from an optimum number of precipitation gauges merged station and radar precipitation estimates soil moisture observations evapotranspiration estimates and additional stream gauges) The last of these three would have no impact on the probabilities Q or Qi and also will not require changed inundation maps Only our calculation as to consequence would be affected since the whole idea of improved forecasting would be to allow preparation so as to reduce the potential flood damage

A full risk assessment would want to look not at individual mitigation strategies but at ldquoportfoliosrdquo of mitigation strategies Of course all of what was said above about ldquomitigation strategyrdquo could apply to a portfolio of strategies But the inundation maps will be much harder to obtain for portfolios of strategies than for individual ones As noted some of the mitigation strategies will not affect inundation maps Looking at mitigation strategies separately avoids the complication that several of these strategies may interact with each other in complex ways Some portfolios of mitigation strategies will be too expensive or infeasible for other reasons We may want to look at budgetary constraints and try to look at a few simple portfolios of strategies that have total investment cost (sum of Crsquos for each strategy in the portfolio) less than a given amount Then the question will become What is a measure of the value of a given portfolio Are we trying to maximize total expected gain (sum of G for all strategies in the portfolio) Or are we optimizing something else Or is this a ldquomulti-attributed utilityrdquo problem

14

Social Environmental and Political News Snapshots2 Hunterdon Mercer Middlesex Monmouth Morris Somerset Union

RARITAN RIVER TOWN SOCIAL ENVIRONMENTAL AND POLITICAL NEWS SNAPSHOTS 34

Alexandria 4-way stop to be set up near Delaware Valley High School

Bedminster

bull Clarence Dillon Library in Bedminster to host a series of Salons (92711) bull Bedminster bicyclist killed after crash with truck in Far Hills (12111)

Berkeley Heights Berkeley Heights childrenrsquos library bounces back Bernards Outraged NJ towns ask state to let them pull plug on JCPampL (12211)

Bernardsville Looking for updates on trail conditions since Hurricane Irene report in if you have them (9211)

Bethlehem Franklin Bethlehem school boards nearing alliance

Bound Brook

bull Flood prevention project in Bound Brook keeps Hurricane Irene from becoming another Floyd (83011)

bull Parts of Bound Brook South Bound Brook Manville under mandatory evacuation order (82811)

Branchburg Bald eagle sighting

Bridgewater

bull Pfizer anticipates EPA cleanup plan for American Cyanamid site in Bridgewater (121611)

bull French drug maker plans to move its US business to Bridgewater (121511)

bull Federal environmental officials close to decision on $205M clean-up proposal of Bridgewaters Cyanamid property (121111)

bull The bike path is clear and the going is easy at Duke Island Park in Bridgewater (13111)

Califon

bull Hunterdon complaints of power outages prompt review of utilities by state Board of Public Utilities (112911)

bull Snowstorm results in $15000 bill to Califon for tree branch clean-up (112611)

Chester Borough New Jersey residents and communities cope with effects of pre-Halloween snowstorm (103111)

Chester Township North Hunterdon High School sets metro record for donating blood Clinton Town Clinton seeking new bids for sewer line manhole work (111411)

Clinton Township Village Green development in Annandale gets final approval to build on former Warren Lumber site (111411)

Cranbury Horse owner forestalls sale to slaughterhouse (121011)

Delaware Eroded Delaware Township roads popular with cyclists pedestrians prompt speed limit worries (121311)

Dunellen Thunderstorms roll through NJ causing lightning strikes and roof collapse (61711)

East Amwell Hunterdon Freeholders delay changes for Solid Waste Advisory Council after SWAC members object (12711)

2 Gathered July-December 2011 3 With assistance from Melissa Basile MCRP 2012 4 Gathered from the Star Ledger New Jersey County and Town Pages httpwwwnjcomlocal under ldquoMorerdquo ndashFor example httptopicsnjcomtagbernardspostshtml substituting the name of the town of interest

Appendix A News Snapshots

7

Interviews

The second phase of this work currently underway involves in-person interviews with 20-25 key regional municipal officials to establish the following

bull Importance of flood mitigation among other concerns schools police fire service traffic management trash collection parking street repair housing maintenance and repair aesthetics and other quality of life issues

bull What they know about flood mitigation flooding the degree of damage and the causes bull Their understanding of climate impacts on flooding and mitigation strategies bull Local stormwater management measures (retention basin management to rain barrels bull Preferred flood mitigation strategies both locally and regionally bull Level of local engagement and what causes citizens to become engaged in local issues

At this time the following municipalities are to be interviewed with either the mayor the administrator and or the director of the local Office of Emergency Management

Municipality CO Municipality CO Berkeley Heights Union Middlesex Boro Msx Bound Brook Mayor amp OEM Som Milltown Msx Branchburg Som Monroe (Engineer Ernie Feist) Msx Clinton Township Hun North Plainfield Msx Edison John Medina Msx Scotch Plains Union Franklin Som Raritan Township Hun Hightstown Mer Watchung Som Hopewell Mer Washington Twp Morris Manville Administrator amp OEM Som Woodbridge Msx Marlboro Mon

Next Steps

The next steps will involve closer analysis of the census data as we proceed with municipal interviews The goal is to talk to people at various levels not just those who are most directly affected but also those who may be upstream from more impacted communities The next mapping exercise will be to add contours in order to verify the drainage and to calculate current potential mitigation in the uppermost communities through major stormwater mitigation efforts The use of impervious surface coverages from the state as well as the application of independent models will lend more insight into the potential for stormwater management BMPs to aid in overall impact reduction

The findings of the interviews will form a discussion guide at a dinner meeting with municipal and county leaders at the Bloustein School on Tuesday January 31 2012 The discussion will focus on further action for a regional volume-focused flood mitigation strategy Focus Groups Several focus groups are also scheduled Milltown and Manville are scheduled for January Two others will be held in February and March These meetings will be held with groups of key citizen leaders in selected municipalities to determine their knowledge of the issue significance to them their desire for further information about the subject and desired mode of transmission of the information and willingness to become involved in the issue in their own town These sessions will also address the questions asked of the initial stakeholders to verify unity of thought and understanding of issues across the community

8

In addition a series of meetings with regional engineers are being held to engage them in the mitigation conversation from the perspective of stormwater management and overall adaptation of green infrastructure practices Their inputs will be gathered in January and February through a series of conference calls on the following topics

bull regulation changes bull policy directions bull professional education needs bull public education needs bull financing mitigation

They will meet together in early March to share their findings and engage with municipal officials and another cohort regional business leaders

Pilot Survey A draft survey will be created and a pilot conducted for a larger regional survey This would be tested with a limited number of residents of the basin to determine their knowledge of the issue significance to them their desire for further information about the subject and desired mode of transmission of the information and willingness to become involved in the issue in their own towns A larger application of the survey would not be fully implemented as part of this project

Task Two Economic Modeling The economic modeling element of this project is designed to look at best ways to estimate benefits to municipalities (and larger governmental units) of flood mitigation projects for the Raritan watershed For this section two draft reports are provided a flood expense taxonomy (separate spreadsheet) and a literature review

Flood Expense Taxonomy

The flood expense taxonomy (see Separate Spreadsheet File) provides a framework for identifying and categorizing the costs associated with flooding events We developed the taxonomy drawing from academic literature prior reports of flooding events and newspaper articles covering flooding events The taxonomy identifies potential flood-related expenses for four sectors of the economy households commercial businesses municipalities and agriculture The potential costs to each sector are cross-categorized with respect to the time sequence of the flooding event pre-flooding flooding and post flooding We will continue to identify additional potential costs and where possible populate those findings with estimated costs Additionally we will add to this taxonomy the potential economic benefits (if any) to the same four sectors as a result of removing structures located in the floodplain

Literature Review

The annotated literature review (Attachment D) includes two types of articles those that examine the relationship between floods and housing prices and those that estimate the costs of flooding on municipalities and their residents and businesses The articles that examine the relationship between flooding events and home prices generally use hedonic pricing models to explain the difference in home value (as measured by sales price) between homes that are on the floodplain and similar houses that are not on it The articles generally indicate that prices for homes located on floodplains are discounted relative to similar homes not on the floodplain One article offers a meta-analysis of the empirical literature on home prices and flooding events The articles in our taxonomy examine the economic impacts of flooding events from England the Red-River Basin in the Western US and the Delaware

9

River in New Jersey We also include a meta-analysis that re-estimates National Weather Service estimates of national economic damages We will build on this literature review to develop a better understanding of the existing estimates of the full economic costs of flooding events and the potential benefits of removing structures (private and public) from floodplains

Task Three Ideas on Risk Assessment for Flooding on the Raritan The third element of the project includes the first stages of the development of a risk analysis identification model to focus on all hazardsrisks The identification of data to be analyzed will be developed by experts in both mathematical modeling and human health and environmental modeling from engineering and public policy perspectives The team initially Fred Roberts and Paul Kantor now includes Qizhong Guo Associate Professor of Civil and Environmental Engineering and David Robinson and State Climatologist and Geography Professor Brief bio sketches are attached as Appendix E Qizhong Guo and Appendix F David Robinson)

Overview

We have spent the past few months researching alternative approaches to flood risk assessment and gathering as much background information as possible We have gotten information from numerous individuals in the US and abroad and numerous agencies including New Jersey Department of Environmental Protection Department of Homeland Security Risk Management and Analysis USGS We came to understand that it was important to expand our team and so we have added a climatologist Professor David Robinson of the Rutgers Department of Geography the New Jersey State Climatologist and a hydrologist Professor Qizhong Guo of the Rutgers Department of Civil and Environmental Engineering We present here a high level sketch of the approach we are examining and discuss the kinds of detailed information that its implementation will require

Problem Formulation

Our multidisciplinary discussions have led us to understand many of the subtleties involved in doing a risk assessment for flooding on the Raritan We have nearly completed outlining an approach based on what we have learned and the next steps will be to share that approach with those working on other tasks in the project revise as needed and then begin gathering the data needed to do the required modeling

To explain our approach and the issues we have to resolve let us consider first a diagram (Figure 1) which illustrates the kinds of factors and tools we need to take into account in understanding flood modeling and applying that modeling to a risk assessment

10

Figure 1 The interaction of nature models and controllable activities (shown in yellow)

Engineering models can combine information on rainfall specified as to time and space with information on soil moisture conditions seasonality and river levels prior to the rain event the fixed properties of the watershed (not indicated) and the land cover (natural and built environment) to produce flood inundation maps To obtain the precision needed to specify events that might occur every five or ten years requires LiDAR information on elevations We refer to the soil moisture conditions and river levels as ldquoantecedent eventsrdquo and note that seasonality is important because it affects the ability of the soil to absorb water (due to vegetation cover freezing etc) The land coverenvironment is viewed as modifiable (yellow) over the span of a few years That would in turn move the flood contours Flood inundation maps can be used for insurance andor regulatory purposes for risk assessments and to communicate complex geology and meteorology to stakeholder groups and thereby facilitate consideration of alternative policies

In the simplest case we might think of four types of soil moisture conditions dry ldquoaveragerdquo wet or frozen Probably it is sufficient to think of four different seasons We might also think of three basic river levels low average and high Thus there are in principle 4 x 4 x 3 = 48 different combinations of these conditions In reality not all of these combinations are feasible We wonrsquot have frozen ground in summer In our analysis we will probably demonstrate the methods choosing only several of the sensible alternatives for detailed investigation

Risk Modeling

The basic outline for how DHS does risk assessment is shown in Figure 2 below which is taken from a presentation by Isaac Maya of the CREATE Center It defines ldquoriskrdquo R as a function of threat T vulnerability V and consequence C R = f(TVC) Often this is represented as a product R = T x V x C Generally the factors T and V are some kinds of probabilities while C is some measure of loss In our case ldquoattackrdquo is interpreted to mean weather event and ldquosuccessrdquo is interpreted to mean flooding

Precipitation(xt) )tthe past

The models

Soil moisture conditions + seasonality + river levels seasonality

Land Cover (natural + built environment)

River gauge levels

LiDAR information

Flood Contour Maps

waterway

5 year contour

10 year contour

11

Figure 2 Sketch of the DHS computation of risk

There are two branches to the tree in this figure The top branch is the ldquostatus quordquo and the bottom branch involves some mitigation strategy or combination of such strategies The probability of attack P is the probability that there will be a weather event of the particular kind being considered and the probability of ldquosuccessrdquo Q is the conditional probability that such an event if it does occur will lead to flooding at a site being considered The consequences could be of various types In the case of floods they include loss of life economic damage to homes and businesses (direct or indirect) and psychological damage There is an attempt to put each of these types of consequences in terms of dollars The consequence (cost L) of an attack (flood) is usually calculated as a weighted sum of the different kinds of costs in our case of the cost in terms of lives lost economic damage and psychological damage What weights one uses and whether it makes sense to add up these weighted values are issues that will be viewed differently by different stakeholder groups

Complexities of the Watershed Situation

In fact we expect to be able to do more than simply treat ldquofloodrdquo as a yes or no matter as is done in this tree diagram Flood inundation maps will present levels of flood waters and the losses incurred change if there is a different level of flooding (or if there is a different duration for the flooding a factor we will disregard for now) The flood inundation levels are numerical values obtained from hydrological models (they can be later translated into major moderate minor near flood stage and none in making flood forecasts) A more complete model would replace the ldquosuccessfailurerdquo bifurcation in the tree diagram with a number of branches depending on the levelcategory of flood as measured for example by river gauges In our case it makes sense to define category 0 as no flooding category 1 as 0 to 1 feet category 2 as 1 to 2 feet etc

In practice obtaining the flood inundation levels may be complicated even before we introduce mitigating strategies The inundation maps are maps for theoretical storms not maps used to predict floods from an approaching storm These maps for the storms of a 100-year and 500-year ldquoreturn periodrdquo1 are already available and are being updated by FEMA using the recently available high resolution LiDAR data on the topography These maps are called flood insurance rate maps (FIRM) However they are developed from a kind of discrete-event modeling and may conceal hidden

1 This technical expression can be interpreted as representing a 1 of 15 probability per year

Success

Failure Attack

Vulnerability

No Attack

No Risk Management Plan

Threat of Attack

Success

Failure

Attack

No Attack

Risk Management Plan

Risk Management

PlansConsequences

L = Expected[$-Eqiuvalent Loss]

No Loss

No Loss

C + LC = Cost of Risk Management Plan

L = Expected[Reduced Loss] where L lt L

C

C

PQ

(1 - P)

(1 - Q)

P where P lt P

(1 - P)

Q where Q lt Q

(1 - Q)

12

uncertainties Moreover the inundation maps for more frequent storms are not readily available or non-existent It would be possible to build flood maps for 2-year 5-year 10-year 25-year and 50-year storms from available data This would imply gathering data and modeling and will depend on having access to the data Moreover the discrete-event models do not take into account antecedent conditions or seasonality Continuous modeling could take these factors into account but the cost of modeling would then be much higher

Assuming we can get flood inundation levels from hydrological models we would try to calculate the probability Qi that a flood of category i occurs Each category of flooding i would have a level of loss Li associated with it The level of flooding from a given amount of rainfall over a period of time depends not only on the precipitation but also on the soil moisture conditionsseasonalityriver levels and on the land cover (natural + built environment) Initially we may assume a fixed land cover (although this can be changed by mitigation) As noted above there are many different combinations of parameters for soil moisture conditions seasonality river levels We could in principle estimate the probability of each of these combinations (but would need to work out a method for doing so) For each combination we could use the hydrological models to calculate the flood level and thus the flood category Putting all of these calculations together will allow us to calculate the Qi parameters This is probably much too complicated and certainly much too expensive Instead we will pick out several more common combinations among the sensible ones and combine them with different precipitation patterns to define a set of weather events

In short our threats will be different kinds of weather events defined in terms of the profile of rainfall over a given period of time (say in number of inches in first hour number in second hour etc over a 24 hour period) plus antecedent precipitation and seasonal effects and pre-existing river levels We will aim to identify six to ten weather scenarios to analyze For each the probability P will have to be determined Information about precipitation totals in 24 hour periods is available and we could then use this to define a 2-year 5-year 10-year 25-year 50-year 100-year event in terms of precipitation profile per 24-hour period The probability of a 5-year event would be 15 of a 50 year event 150 etc except for the other factors (antecedent events plus seasonality) So we may want to modify P based on these other factors We will need to determine how to do that Note also that climate change will likely affect the value of P but we do not know how to calculate modified P with any degree of confidence

Now what is the interpretation of the Qi parameters If the weather event (precipitation profile) is defined and the antecedent conditions and seasonality are known in principle the hydrological models will produce just one category of flood But it could be that it makes more sense to build some uncertainty into the predictions of the models or it may make sense to assume that we have some uncertainty in knowing the soil moisture conditions or the river levels Thus there is surely still some value to studying the risk assessments using different flood categories

When there is a mitigation strategy presumably the probability of an attack would not be affected so Prsquo in the diagram is the same as P not less than P The probability of ldquosuccessrdquo or flood given a weather event is Qrsquo which is less than Q If we use different flood level categories then we would replace each Qi with a number Qirsquo but note that Qirsquo would be higher than Qi for some i Also the consequences of a flood could be less with expected consequences now Lrsquo presumably less than L (or Lirsquo presumably less than Li) But there is a cost C of the mitigation strategy and that cost applies even if there is no flood

Still using the notation in the diagram we see that if f(TVC) = TxVxC then with no mitigation strategy the risk or total expected cost is PxQxL but if there is a mitigation strategy then (since P = Prsquo) the risk or total expected cost is C + PxQrsquoxLrsquo The difference G = PxQxL ndash (C + PxQrsquoxLrsquo) in these expected costs would be the expected gain or expected risk reduction due to the strategy (If we have different flood level categories the calculation would be modified appropriately)

13

Mitigation Strategies and Elicitation of Some Key Parameters

We will need to decide if we look at flood damage (vulnerability) to a larger region or concentrate on a few towns Our preliminary plan is to do the analysis for a few sample towns for instance Manville and Millstone We will need to coordinate with the other parts of the project on this decision

Consequences will need to be measured by our partners in the project and we will need to coordinate with them

We will look at a variety of mitigation strategies Among them could be improved flood forecasting and warning systems retrofit of green infrastructure (eg impervious area disconnection rain gardens cisterns and rain barrels green roofs pervious concrete) to existing development areas to reduce upstream runoff volume regional (upstream) stormwater detention basins acquisition and demolition and relocation of properties flood-proofing and elevation in place disaster preparedness and response plans flood insurance cleanout of urban drainage systems construction of dams and reservoirs construction of dikes levees and floodwalls channel alterations and high flow diversions and spillways We will want to coordinate with other parts of the project on which mitigation strategies we look at

We will need to work with other parts of the project to get ballpark estimates of reduced costs Lrsquo (or Lirsquo) of floods

Under a given mitigation strategy the challenge is to produce inundation maps that will allow us to determine vulnerability (whether there is a flood) In principle there are hydrological models for doing this or such models could be developed but to do the actual calculations will be much more expensive than we can afford in this modest project Thus the best we can do is pick a few sample mitigation strategies for which we could make educated guesses as to vulnerabilities ndash just to illustrate the methodology Our thoughts are to start with three types of strategies retrofit of green infrastructure some sort of upstream mitigation that would have impact on downstream flooding and improved forecasting (eg improved monitoring from an optimum number of precipitation gauges merged station and radar precipitation estimates soil moisture observations evapotranspiration estimates and additional stream gauges) The last of these three would have no impact on the probabilities Q or Qi and also will not require changed inundation maps Only our calculation as to consequence would be affected since the whole idea of improved forecasting would be to allow preparation so as to reduce the potential flood damage

A full risk assessment would want to look not at individual mitigation strategies but at ldquoportfoliosrdquo of mitigation strategies Of course all of what was said above about ldquomitigation strategyrdquo could apply to a portfolio of strategies But the inundation maps will be much harder to obtain for portfolios of strategies than for individual ones As noted some of the mitigation strategies will not affect inundation maps Looking at mitigation strategies separately avoids the complication that several of these strategies may interact with each other in complex ways Some portfolios of mitigation strategies will be too expensive or infeasible for other reasons We may want to look at budgetary constraints and try to look at a few simple portfolios of strategies that have total investment cost (sum of Crsquos for each strategy in the portfolio) less than a given amount Then the question will become What is a measure of the value of a given portfolio Are we trying to maximize total expected gain (sum of G for all strategies in the portfolio) Or are we optimizing something else Or is this a ldquomulti-attributed utilityrdquo problem

14

Social Environmental and Political News Snapshots2 Hunterdon Mercer Middlesex Monmouth Morris Somerset Union

RARITAN RIVER TOWN SOCIAL ENVIRONMENTAL AND POLITICAL NEWS SNAPSHOTS 34

Alexandria 4-way stop to be set up near Delaware Valley High School

Bedminster

bull Clarence Dillon Library in Bedminster to host a series of Salons (92711) bull Bedminster bicyclist killed after crash with truck in Far Hills (12111)

Berkeley Heights Berkeley Heights childrenrsquos library bounces back Bernards Outraged NJ towns ask state to let them pull plug on JCPampL (12211)

Bernardsville Looking for updates on trail conditions since Hurricane Irene report in if you have them (9211)

Bethlehem Franklin Bethlehem school boards nearing alliance

Bound Brook

bull Flood prevention project in Bound Brook keeps Hurricane Irene from becoming another Floyd (83011)

bull Parts of Bound Brook South Bound Brook Manville under mandatory evacuation order (82811)

Branchburg Bald eagle sighting

Bridgewater

bull Pfizer anticipates EPA cleanup plan for American Cyanamid site in Bridgewater (121611)

bull French drug maker plans to move its US business to Bridgewater (121511)

bull Federal environmental officials close to decision on $205M clean-up proposal of Bridgewaters Cyanamid property (121111)

bull The bike path is clear and the going is easy at Duke Island Park in Bridgewater (13111)

Califon

bull Hunterdon complaints of power outages prompt review of utilities by state Board of Public Utilities (112911)

bull Snowstorm results in $15000 bill to Califon for tree branch clean-up (112611)

Chester Borough New Jersey residents and communities cope with effects of pre-Halloween snowstorm (103111)

Chester Township North Hunterdon High School sets metro record for donating blood Clinton Town Clinton seeking new bids for sewer line manhole work (111411)

Clinton Township Village Green development in Annandale gets final approval to build on former Warren Lumber site (111411)

Cranbury Horse owner forestalls sale to slaughterhouse (121011)

Delaware Eroded Delaware Township roads popular with cyclists pedestrians prompt speed limit worries (121311)

Dunellen Thunderstorms roll through NJ causing lightning strikes and roof collapse (61711)

East Amwell Hunterdon Freeholders delay changes for Solid Waste Advisory Council after SWAC members object (12711)

2 Gathered July-December 2011 3 With assistance from Melissa Basile MCRP 2012 4 Gathered from the Star Ledger New Jersey County and Town Pages httpwwwnjcomlocal under ldquoMorerdquo ndashFor example httptopicsnjcomtagbernardspostshtml substituting the name of the town of interest

Appendix A News Snapshots

8

In addition a series of meetings with regional engineers are being held to engage them in the mitigation conversation from the perspective of stormwater management and overall adaptation of green infrastructure practices Their inputs will be gathered in January and February through a series of conference calls on the following topics

bull regulation changes bull policy directions bull professional education needs bull public education needs bull financing mitigation

They will meet together in early March to share their findings and engage with municipal officials and another cohort regional business leaders

Pilot Survey A draft survey will be created and a pilot conducted for a larger regional survey This would be tested with a limited number of residents of the basin to determine their knowledge of the issue significance to them their desire for further information about the subject and desired mode of transmission of the information and willingness to become involved in the issue in their own towns A larger application of the survey would not be fully implemented as part of this project

Task Two Economic Modeling The economic modeling element of this project is designed to look at best ways to estimate benefits to municipalities (and larger governmental units) of flood mitigation projects for the Raritan watershed For this section two draft reports are provided a flood expense taxonomy (separate spreadsheet) and a literature review

Flood Expense Taxonomy

The flood expense taxonomy (see Separate Spreadsheet File) provides a framework for identifying and categorizing the costs associated with flooding events We developed the taxonomy drawing from academic literature prior reports of flooding events and newspaper articles covering flooding events The taxonomy identifies potential flood-related expenses for four sectors of the economy households commercial businesses municipalities and agriculture The potential costs to each sector are cross-categorized with respect to the time sequence of the flooding event pre-flooding flooding and post flooding We will continue to identify additional potential costs and where possible populate those findings with estimated costs Additionally we will add to this taxonomy the potential economic benefits (if any) to the same four sectors as a result of removing structures located in the floodplain

Literature Review

The annotated literature review (Attachment D) includes two types of articles those that examine the relationship between floods and housing prices and those that estimate the costs of flooding on municipalities and their residents and businesses The articles that examine the relationship between flooding events and home prices generally use hedonic pricing models to explain the difference in home value (as measured by sales price) between homes that are on the floodplain and similar houses that are not on it The articles generally indicate that prices for homes located on floodplains are discounted relative to similar homes not on the floodplain One article offers a meta-analysis of the empirical literature on home prices and flooding events The articles in our taxonomy examine the economic impacts of flooding events from England the Red-River Basin in the Western US and the Delaware

9

River in New Jersey We also include a meta-analysis that re-estimates National Weather Service estimates of national economic damages We will build on this literature review to develop a better understanding of the existing estimates of the full economic costs of flooding events and the potential benefits of removing structures (private and public) from floodplains

Task Three Ideas on Risk Assessment for Flooding on the Raritan The third element of the project includes the first stages of the development of a risk analysis identification model to focus on all hazardsrisks The identification of data to be analyzed will be developed by experts in both mathematical modeling and human health and environmental modeling from engineering and public policy perspectives The team initially Fred Roberts and Paul Kantor now includes Qizhong Guo Associate Professor of Civil and Environmental Engineering and David Robinson and State Climatologist and Geography Professor Brief bio sketches are attached as Appendix E Qizhong Guo and Appendix F David Robinson)

Overview

We have spent the past few months researching alternative approaches to flood risk assessment and gathering as much background information as possible We have gotten information from numerous individuals in the US and abroad and numerous agencies including New Jersey Department of Environmental Protection Department of Homeland Security Risk Management and Analysis USGS We came to understand that it was important to expand our team and so we have added a climatologist Professor David Robinson of the Rutgers Department of Geography the New Jersey State Climatologist and a hydrologist Professor Qizhong Guo of the Rutgers Department of Civil and Environmental Engineering We present here a high level sketch of the approach we are examining and discuss the kinds of detailed information that its implementation will require

Problem Formulation

Our multidisciplinary discussions have led us to understand many of the subtleties involved in doing a risk assessment for flooding on the Raritan We have nearly completed outlining an approach based on what we have learned and the next steps will be to share that approach with those working on other tasks in the project revise as needed and then begin gathering the data needed to do the required modeling

To explain our approach and the issues we have to resolve let us consider first a diagram (Figure 1) which illustrates the kinds of factors and tools we need to take into account in understanding flood modeling and applying that modeling to a risk assessment

10

Figure 1 The interaction of nature models and controllable activities (shown in yellow)

Engineering models can combine information on rainfall specified as to time and space with information on soil moisture conditions seasonality and river levels prior to the rain event the fixed properties of the watershed (not indicated) and the land cover (natural and built environment) to produce flood inundation maps To obtain the precision needed to specify events that might occur every five or ten years requires LiDAR information on elevations We refer to the soil moisture conditions and river levels as ldquoantecedent eventsrdquo and note that seasonality is important because it affects the ability of the soil to absorb water (due to vegetation cover freezing etc) The land coverenvironment is viewed as modifiable (yellow) over the span of a few years That would in turn move the flood contours Flood inundation maps can be used for insurance andor regulatory purposes for risk assessments and to communicate complex geology and meteorology to stakeholder groups and thereby facilitate consideration of alternative policies

In the simplest case we might think of four types of soil moisture conditions dry ldquoaveragerdquo wet or frozen Probably it is sufficient to think of four different seasons We might also think of three basic river levels low average and high Thus there are in principle 4 x 4 x 3 = 48 different combinations of these conditions In reality not all of these combinations are feasible We wonrsquot have frozen ground in summer In our analysis we will probably demonstrate the methods choosing only several of the sensible alternatives for detailed investigation

Risk Modeling

The basic outline for how DHS does risk assessment is shown in Figure 2 below which is taken from a presentation by Isaac Maya of the CREATE Center It defines ldquoriskrdquo R as a function of threat T vulnerability V and consequence C R = f(TVC) Often this is represented as a product R = T x V x C Generally the factors T and V are some kinds of probabilities while C is some measure of loss In our case ldquoattackrdquo is interpreted to mean weather event and ldquosuccessrdquo is interpreted to mean flooding

Precipitation(xt) )tthe past

The models

Soil moisture conditions + seasonality + river levels seasonality

Land Cover (natural + built environment)

River gauge levels

LiDAR information

Flood Contour Maps

waterway

5 year contour

10 year contour

11

Figure 2 Sketch of the DHS computation of risk

There are two branches to the tree in this figure The top branch is the ldquostatus quordquo and the bottom branch involves some mitigation strategy or combination of such strategies The probability of attack P is the probability that there will be a weather event of the particular kind being considered and the probability of ldquosuccessrdquo Q is the conditional probability that such an event if it does occur will lead to flooding at a site being considered The consequences could be of various types In the case of floods they include loss of life economic damage to homes and businesses (direct or indirect) and psychological damage There is an attempt to put each of these types of consequences in terms of dollars The consequence (cost L) of an attack (flood) is usually calculated as a weighted sum of the different kinds of costs in our case of the cost in terms of lives lost economic damage and psychological damage What weights one uses and whether it makes sense to add up these weighted values are issues that will be viewed differently by different stakeholder groups

Complexities of the Watershed Situation

In fact we expect to be able to do more than simply treat ldquofloodrdquo as a yes or no matter as is done in this tree diagram Flood inundation maps will present levels of flood waters and the losses incurred change if there is a different level of flooding (or if there is a different duration for the flooding a factor we will disregard for now) The flood inundation levels are numerical values obtained from hydrological models (they can be later translated into major moderate minor near flood stage and none in making flood forecasts) A more complete model would replace the ldquosuccessfailurerdquo bifurcation in the tree diagram with a number of branches depending on the levelcategory of flood as measured for example by river gauges In our case it makes sense to define category 0 as no flooding category 1 as 0 to 1 feet category 2 as 1 to 2 feet etc

In practice obtaining the flood inundation levels may be complicated even before we introduce mitigating strategies The inundation maps are maps for theoretical storms not maps used to predict floods from an approaching storm These maps for the storms of a 100-year and 500-year ldquoreturn periodrdquo1 are already available and are being updated by FEMA using the recently available high resolution LiDAR data on the topography These maps are called flood insurance rate maps (FIRM) However they are developed from a kind of discrete-event modeling and may conceal hidden

1 This technical expression can be interpreted as representing a 1 of 15 probability per year

Success

Failure Attack

Vulnerability

No Attack

No Risk Management Plan

Threat of Attack

Success

Failure

Attack

No Attack

Risk Management Plan

Risk Management

PlansConsequences

L = Expected[$-Eqiuvalent Loss]

No Loss

No Loss

C + LC = Cost of Risk Management Plan

L = Expected[Reduced Loss] where L lt L

C

C

PQ

(1 - P)

(1 - Q)

P where P lt P

(1 - P)

Q where Q lt Q

(1 - Q)

12

uncertainties Moreover the inundation maps for more frequent storms are not readily available or non-existent It would be possible to build flood maps for 2-year 5-year 10-year 25-year and 50-year storms from available data This would imply gathering data and modeling and will depend on having access to the data Moreover the discrete-event models do not take into account antecedent conditions or seasonality Continuous modeling could take these factors into account but the cost of modeling would then be much higher

Assuming we can get flood inundation levels from hydrological models we would try to calculate the probability Qi that a flood of category i occurs Each category of flooding i would have a level of loss Li associated with it The level of flooding from a given amount of rainfall over a period of time depends not only on the precipitation but also on the soil moisture conditionsseasonalityriver levels and on the land cover (natural + built environment) Initially we may assume a fixed land cover (although this can be changed by mitigation) As noted above there are many different combinations of parameters for soil moisture conditions seasonality river levels We could in principle estimate the probability of each of these combinations (but would need to work out a method for doing so) For each combination we could use the hydrological models to calculate the flood level and thus the flood category Putting all of these calculations together will allow us to calculate the Qi parameters This is probably much too complicated and certainly much too expensive Instead we will pick out several more common combinations among the sensible ones and combine them with different precipitation patterns to define a set of weather events

In short our threats will be different kinds of weather events defined in terms of the profile of rainfall over a given period of time (say in number of inches in first hour number in second hour etc over a 24 hour period) plus antecedent precipitation and seasonal effects and pre-existing river levels We will aim to identify six to ten weather scenarios to analyze For each the probability P will have to be determined Information about precipitation totals in 24 hour periods is available and we could then use this to define a 2-year 5-year 10-year 25-year 50-year 100-year event in terms of precipitation profile per 24-hour period The probability of a 5-year event would be 15 of a 50 year event 150 etc except for the other factors (antecedent events plus seasonality) So we may want to modify P based on these other factors We will need to determine how to do that Note also that climate change will likely affect the value of P but we do not know how to calculate modified P with any degree of confidence

Now what is the interpretation of the Qi parameters If the weather event (precipitation profile) is defined and the antecedent conditions and seasonality are known in principle the hydrological models will produce just one category of flood But it could be that it makes more sense to build some uncertainty into the predictions of the models or it may make sense to assume that we have some uncertainty in knowing the soil moisture conditions or the river levels Thus there is surely still some value to studying the risk assessments using different flood categories

When there is a mitigation strategy presumably the probability of an attack would not be affected so Prsquo in the diagram is the same as P not less than P The probability of ldquosuccessrdquo or flood given a weather event is Qrsquo which is less than Q If we use different flood level categories then we would replace each Qi with a number Qirsquo but note that Qirsquo would be higher than Qi for some i Also the consequences of a flood could be less with expected consequences now Lrsquo presumably less than L (or Lirsquo presumably less than Li) But there is a cost C of the mitigation strategy and that cost applies even if there is no flood

Still using the notation in the diagram we see that if f(TVC) = TxVxC then with no mitigation strategy the risk or total expected cost is PxQxL but if there is a mitigation strategy then (since P = Prsquo) the risk or total expected cost is C + PxQrsquoxLrsquo The difference G = PxQxL ndash (C + PxQrsquoxLrsquo) in these expected costs would be the expected gain or expected risk reduction due to the strategy (If we have different flood level categories the calculation would be modified appropriately)

13

Mitigation Strategies and Elicitation of Some Key Parameters

We will need to decide if we look at flood damage (vulnerability) to a larger region or concentrate on a few towns Our preliminary plan is to do the analysis for a few sample towns for instance Manville and Millstone We will need to coordinate with the other parts of the project on this decision

Consequences will need to be measured by our partners in the project and we will need to coordinate with them

We will look at a variety of mitigation strategies Among them could be improved flood forecasting and warning systems retrofit of green infrastructure (eg impervious area disconnection rain gardens cisterns and rain barrels green roofs pervious concrete) to existing development areas to reduce upstream runoff volume regional (upstream) stormwater detention basins acquisition and demolition and relocation of properties flood-proofing and elevation in place disaster preparedness and response plans flood insurance cleanout of urban drainage systems construction of dams and reservoirs construction of dikes levees and floodwalls channel alterations and high flow diversions and spillways We will want to coordinate with other parts of the project on which mitigation strategies we look at

We will need to work with other parts of the project to get ballpark estimates of reduced costs Lrsquo (or Lirsquo) of floods

Under a given mitigation strategy the challenge is to produce inundation maps that will allow us to determine vulnerability (whether there is a flood) In principle there are hydrological models for doing this or such models could be developed but to do the actual calculations will be much more expensive than we can afford in this modest project Thus the best we can do is pick a few sample mitigation strategies for which we could make educated guesses as to vulnerabilities ndash just to illustrate the methodology Our thoughts are to start with three types of strategies retrofit of green infrastructure some sort of upstream mitigation that would have impact on downstream flooding and improved forecasting (eg improved monitoring from an optimum number of precipitation gauges merged station and radar precipitation estimates soil moisture observations evapotranspiration estimates and additional stream gauges) The last of these three would have no impact on the probabilities Q or Qi and also will not require changed inundation maps Only our calculation as to consequence would be affected since the whole idea of improved forecasting would be to allow preparation so as to reduce the potential flood damage

A full risk assessment would want to look not at individual mitigation strategies but at ldquoportfoliosrdquo of mitigation strategies Of course all of what was said above about ldquomitigation strategyrdquo could apply to a portfolio of strategies But the inundation maps will be much harder to obtain for portfolios of strategies than for individual ones As noted some of the mitigation strategies will not affect inundation maps Looking at mitigation strategies separately avoids the complication that several of these strategies may interact with each other in complex ways Some portfolios of mitigation strategies will be too expensive or infeasible for other reasons We may want to look at budgetary constraints and try to look at a few simple portfolios of strategies that have total investment cost (sum of Crsquos for each strategy in the portfolio) less than a given amount Then the question will become What is a measure of the value of a given portfolio Are we trying to maximize total expected gain (sum of G for all strategies in the portfolio) Or are we optimizing something else Or is this a ldquomulti-attributed utilityrdquo problem

14

Social Environmental and Political News Snapshots2 Hunterdon Mercer Middlesex Monmouth Morris Somerset Union

RARITAN RIVER TOWN SOCIAL ENVIRONMENTAL AND POLITICAL NEWS SNAPSHOTS 34

Alexandria 4-way stop to be set up near Delaware Valley High School

Bedminster

bull Clarence Dillon Library in Bedminster to host a series of Salons (92711) bull Bedminster bicyclist killed after crash with truck in Far Hills (12111)

Berkeley Heights Berkeley Heights childrenrsquos library bounces back Bernards Outraged NJ towns ask state to let them pull plug on JCPampL (12211)

Bernardsville Looking for updates on trail conditions since Hurricane Irene report in if you have them (9211)

Bethlehem Franklin Bethlehem school boards nearing alliance

Bound Brook

bull Flood prevention project in Bound Brook keeps Hurricane Irene from becoming another Floyd (83011)

bull Parts of Bound Brook South Bound Brook Manville under mandatory evacuation order (82811)

Branchburg Bald eagle sighting

Bridgewater

bull Pfizer anticipates EPA cleanup plan for American Cyanamid site in Bridgewater (121611)

bull French drug maker plans to move its US business to Bridgewater (121511)

bull Federal environmental officials close to decision on $205M clean-up proposal of Bridgewaters Cyanamid property (121111)

bull The bike path is clear and the going is easy at Duke Island Park in Bridgewater (13111)

Califon

bull Hunterdon complaints of power outages prompt review of utilities by state Board of Public Utilities (112911)

bull Snowstorm results in $15000 bill to Califon for tree branch clean-up (112611)

Chester Borough New Jersey residents and communities cope with effects of pre-Halloween snowstorm (103111)

Chester Township North Hunterdon High School sets metro record for donating blood Clinton Town Clinton seeking new bids for sewer line manhole work (111411)

Clinton Township Village Green development in Annandale gets final approval to build on former Warren Lumber site (111411)

Cranbury Horse owner forestalls sale to slaughterhouse (121011)

Delaware Eroded Delaware Township roads popular with cyclists pedestrians prompt speed limit worries (121311)

Dunellen Thunderstorms roll through NJ causing lightning strikes and roof collapse (61711)

East Amwell Hunterdon Freeholders delay changes for Solid Waste Advisory Council after SWAC members object (12711)

2 Gathered July-December 2011 3 With assistance from Melissa Basile MCRP 2012 4 Gathered from the Star Ledger New Jersey County and Town Pages httpwwwnjcomlocal under ldquoMorerdquo ndashFor example httptopicsnjcomtagbernardspostshtml substituting the name of the town of interest

Appendix A News Snapshots

9

River in New Jersey We also include a meta-analysis that re-estimates National Weather Service estimates of national economic damages We will build on this literature review to develop a better understanding of the existing estimates of the full economic costs of flooding events and the potential benefits of removing structures (private and public) from floodplains

Task Three Ideas on Risk Assessment for Flooding on the Raritan The third element of the project includes the first stages of the development of a risk analysis identification model to focus on all hazardsrisks The identification of data to be analyzed will be developed by experts in both mathematical modeling and human health and environmental modeling from engineering and public policy perspectives The team initially Fred Roberts and Paul Kantor now includes Qizhong Guo Associate Professor of Civil and Environmental Engineering and David Robinson and State Climatologist and Geography Professor Brief bio sketches are attached as Appendix E Qizhong Guo and Appendix F David Robinson)

Overview

We have spent the past few months researching alternative approaches to flood risk assessment and gathering as much background information as possible We have gotten information from numerous individuals in the US and abroad and numerous agencies including New Jersey Department of Environmental Protection Department of Homeland Security Risk Management and Analysis USGS We came to understand that it was important to expand our team and so we have added a climatologist Professor David Robinson of the Rutgers Department of Geography the New Jersey State Climatologist and a hydrologist Professor Qizhong Guo of the Rutgers Department of Civil and Environmental Engineering We present here a high level sketch of the approach we are examining and discuss the kinds of detailed information that its implementation will require

Problem Formulation

Our multidisciplinary discussions have led us to understand many of the subtleties involved in doing a risk assessment for flooding on the Raritan We have nearly completed outlining an approach based on what we have learned and the next steps will be to share that approach with those working on other tasks in the project revise as needed and then begin gathering the data needed to do the required modeling

To explain our approach and the issues we have to resolve let us consider first a diagram (Figure 1) which illustrates the kinds of factors and tools we need to take into account in understanding flood modeling and applying that modeling to a risk assessment

10

Figure 1 The interaction of nature models and controllable activities (shown in yellow)

Engineering models can combine information on rainfall specified as to time and space with information on soil moisture conditions seasonality and river levels prior to the rain event the fixed properties of the watershed (not indicated) and the land cover (natural and built environment) to produce flood inundation maps To obtain the precision needed to specify events that might occur every five or ten years requires LiDAR information on elevations We refer to the soil moisture conditions and river levels as ldquoantecedent eventsrdquo and note that seasonality is important because it affects the ability of the soil to absorb water (due to vegetation cover freezing etc) The land coverenvironment is viewed as modifiable (yellow) over the span of a few years That would in turn move the flood contours Flood inundation maps can be used for insurance andor regulatory purposes for risk assessments and to communicate complex geology and meteorology to stakeholder groups and thereby facilitate consideration of alternative policies

In the simplest case we might think of four types of soil moisture conditions dry ldquoaveragerdquo wet or frozen Probably it is sufficient to think of four different seasons We might also think of three basic river levels low average and high Thus there are in principle 4 x 4 x 3 = 48 different combinations of these conditions In reality not all of these combinations are feasible We wonrsquot have frozen ground in summer In our analysis we will probably demonstrate the methods choosing only several of the sensible alternatives for detailed investigation

Risk Modeling

The basic outline for how DHS does risk assessment is shown in Figure 2 below which is taken from a presentation by Isaac Maya of the CREATE Center It defines ldquoriskrdquo R as a function of threat T vulnerability V and consequence C R = f(TVC) Often this is represented as a product R = T x V x C Generally the factors T and V are some kinds of probabilities while C is some measure of loss In our case ldquoattackrdquo is interpreted to mean weather event and ldquosuccessrdquo is interpreted to mean flooding

Precipitation(xt) )tthe past

The models

Soil moisture conditions + seasonality + river levels seasonality

Land Cover (natural + built environment)

River gauge levels

LiDAR information

Flood Contour Maps

waterway

5 year contour

10 year contour

11

Figure 2 Sketch of the DHS computation of risk

There are two branches to the tree in this figure The top branch is the ldquostatus quordquo and the bottom branch involves some mitigation strategy or combination of such strategies The probability of attack P is the probability that there will be a weather event of the particular kind being considered and the probability of ldquosuccessrdquo Q is the conditional probability that such an event if it does occur will lead to flooding at a site being considered The consequences could be of various types In the case of floods they include loss of life economic damage to homes and businesses (direct or indirect) and psychological damage There is an attempt to put each of these types of consequences in terms of dollars The consequence (cost L) of an attack (flood) is usually calculated as a weighted sum of the different kinds of costs in our case of the cost in terms of lives lost economic damage and psychological damage What weights one uses and whether it makes sense to add up these weighted values are issues that will be viewed differently by different stakeholder groups

Complexities of the Watershed Situation

In fact we expect to be able to do more than simply treat ldquofloodrdquo as a yes or no matter as is done in this tree diagram Flood inundation maps will present levels of flood waters and the losses incurred change if there is a different level of flooding (or if there is a different duration for the flooding a factor we will disregard for now) The flood inundation levels are numerical values obtained from hydrological models (they can be later translated into major moderate minor near flood stage and none in making flood forecasts) A more complete model would replace the ldquosuccessfailurerdquo bifurcation in the tree diagram with a number of branches depending on the levelcategory of flood as measured for example by river gauges In our case it makes sense to define category 0 as no flooding category 1 as 0 to 1 feet category 2 as 1 to 2 feet etc

In practice obtaining the flood inundation levels may be complicated even before we introduce mitigating strategies The inundation maps are maps for theoretical storms not maps used to predict floods from an approaching storm These maps for the storms of a 100-year and 500-year ldquoreturn periodrdquo1 are already available and are being updated by FEMA using the recently available high resolution LiDAR data on the topography These maps are called flood insurance rate maps (FIRM) However they are developed from a kind of discrete-event modeling and may conceal hidden

1 This technical expression can be interpreted as representing a 1 of 15 probability per year

Success

Failure Attack

Vulnerability

No Attack

No Risk Management Plan

Threat of Attack

Success

Failure

Attack

No Attack

Risk Management Plan

Risk Management

PlansConsequences

L = Expected[$-Eqiuvalent Loss]

No Loss

No Loss

C + LC = Cost of Risk Management Plan

L = Expected[Reduced Loss] where L lt L

C

C

PQ

(1 - P)

(1 - Q)

P where P lt P

(1 - P)

Q where Q lt Q

(1 - Q)

12

uncertainties Moreover the inundation maps for more frequent storms are not readily available or non-existent It would be possible to build flood maps for 2-year 5-year 10-year 25-year and 50-year storms from available data This would imply gathering data and modeling and will depend on having access to the data Moreover the discrete-event models do not take into account antecedent conditions or seasonality Continuous modeling could take these factors into account but the cost of modeling would then be much higher

Assuming we can get flood inundation levels from hydrological models we would try to calculate the probability Qi that a flood of category i occurs Each category of flooding i would have a level of loss Li associated with it The level of flooding from a given amount of rainfall over a period of time depends not only on the precipitation but also on the soil moisture conditionsseasonalityriver levels and on the land cover (natural + built environment) Initially we may assume a fixed land cover (although this can be changed by mitigation) As noted above there are many different combinations of parameters for soil moisture conditions seasonality river levels We could in principle estimate the probability of each of these combinations (but would need to work out a method for doing so) For each combination we could use the hydrological models to calculate the flood level and thus the flood category Putting all of these calculations together will allow us to calculate the Qi parameters This is probably much too complicated and certainly much too expensive Instead we will pick out several more common combinations among the sensible ones and combine them with different precipitation patterns to define a set of weather events

In short our threats will be different kinds of weather events defined in terms of the profile of rainfall over a given period of time (say in number of inches in first hour number in second hour etc over a 24 hour period) plus antecedent precipitation and seasonal effects and pre-existing river levels We will aim to identify six to ten weather scenarios to analyze For each the probability P will have to be determined Information about precipitation totals in 24 hour periods is available and we could then use this to define a 2-year 5-year 10-year 25-year 50-year 100-year event in terms of precipitation profile per 24-hour period The probability of a 5-year event would be 15 of a 50 year event 150 etc except for the other factors (antecedent events plus seasonality) So we may want to modify P based on these other factors We will need to determine how to do that Note also that climate change will likely affect the value of P but we do not know how to calculate modified P with any degree of confidence

Now what is the interpretation of the Qi parameters If the weather event (precipitation profile) is defined and the antecedent conditions and seasonality are known in principle the hydrological models will produce just one category of flood But it could be that it makes more sense to build some uncertainty into the predictions of the models or it may make sense to assume that we have some uncertainty in knowing the soil moisture conditions or the river levels Thus there is surely still some value to studying the risk assessments using different flood categories

When there is a mitigation strategy presumably the probability of an attack would not be affected so Prsquo in the diagram is the same as P not less than P The probability of ldquosuccessrdquo or flood given a weather event is Qrsquo which is less than Q If we use different flood level categories then we would replace each Qi with a number Qirsquo but note that Qirsquo would be higher than Qi for some i Also the consequences of a flood could be less with expected consequences now Lrsquo presumably less than L (or Lirsquo presumably less than Li) But there is a cost C of the mitigation strategy and that cost applies even if there is no flood

Still using the notation in the diagram we see that if f(TVC) = TxVxC then with no mitigation strategy the risk or total expected cost is PxQxL but if there is a mitigation strategy then (since P = Prsquo) the risk or total expected cost is C + PxQrsquoxLrsquo The difference G = PxQxL ndash (C + PxQrsquoxLrsquo) in these expected costs would be the expected gain or expected risk reduction due to the strategy (If we have different flood level categories the calculation would be modified appropriately)

13

Mitigation Strategies and Elicitation of Some Key Parameters

We will need to decide if we look at flood damage (vulnerability) to a larger region or concentrate on a few towns Our preliminary plan is to do the analysis for a few sample towns for instance Manville and Millstone We will need to coordinate with the other parts of the project on this decision

Consequences will need to be measured by our partners in the project and we will need to coordinate with them

We will look at a variety of mitigation strategies Among them could be improved flood forecasting and warning systems retrofit of green infrastructure (eg impervious area disconnection rain gardens cisterns and rain barrels green roofs pervious concrete) to existing development areas to reduce upstream runoff volume regional (upstream) stormwater detention basins acquisition and demolition and relocation of properties flood-proofing and elevation in place disaster preparedness and response plans flood insurance cleanout of urban drainage systems construction of dams and reservoirs construction of dikes levees and floodwalls channel alterations and high flow diversions and spillways We will want to coordinate with other parts of the project on which mitigation strategies we look at

We will need to work with other parts of the project to get ballpark estimates of reduced costs Lrsquo (or Lirsquo) of floods

Under a given mitigation strategy the challenge is to produce inundation maps that will allow us to determine vulnerability (whether there is a flood) In principle there are hydrological models for doing this or such models could be developed but to do the actual calculations will be much more expensive than we can afford in this modest project Thus the best we can do is pick a few sample mitigation strategies for which we could make educated guesses as to vulnerabilities ndash just to illustrate the methodology Our thoughts are to start with three types of strategies retrofit of green infrastructure some sort of upstream mitigation that would have impact on downstream flooding and improved forecasting (eg improved monitoring from an optimum number of precipitation gauges merged station and radar precipitation estimates soil moisture observations evapotranspiration estimates and additional stream gauges) The last of these three would have no impact on the probabilities Q or Qi and also will not require changed inundation maps Only our calculation as to consequence would be affected since the whole idea of improved forecasting would be to allow preparation so as to reduce the potential flood damage

A full risk assessment would want to look not at individual mitigation strategies but at ldquoportfoliosrdquo of mitigation strategies Of course all of what was said above about ldquomitigation strategyrdquo could apply to a portfolio of strategies But the inundation maps will be much harder to obtain for portfolios of strategies than for individual ones As noted some of the mitigation strategies will not affect inundation maps Looking at mitigation strategies separately avoids the complication that several of these strategies may interact with each other in complex ways Some portfolios of mitigation strategies will be too expensive or infeasible for other reasons We may want to look at budgetary constraints and try to look at a few simple portfolios of strategies that have total investment cost (sum of Crsquos for each strategy in the portfolio) less than a given amount Then the question will become What is a measure of the value of a given portfolio Are we trying to maximize total expected gain (sum of G for all strategies in the portfolio) Or are we optimizing something else Or is this a ldquomulti-attributed utilityrdquo problem

14

Social Environmental and Political News Snapshots2 Hunterdon Mercer Middlesex Monmouth Morris Somerset Union

RARITAN RIVER TOWN SOCIAL ENVIRONMENTAL AND POLITICAL NEWS SNAPSHOTS 34

Alexandria 4-way stop to be set up near Delaware Valley High School

Bedminster

bull Clarence Dillon Library in Bedminster to host a series of Salons (92711) bull Bedminster bicyclist killed after crash with truck in Far Hills (12111)

Berkeley Heights Berkeley Heights childrenrsquos library bounces back Bernards Outraged NJ towns ask state to let them pull plug on JCPampL (12211)

Bernardsville Looking for updates on trail conditions since Hurricane Irene report in if you have them (9211)

Bethlehem Franklin Bethlehem school boards nearing alliance

Bound Brook

bull Flood prevention project in Bound Brook keeps Hurricane Irene from becoming another Floyd (83011)

bull Parts of Bound Brook South Bound Brook Manville under mandatory evacuation order (82811)

Branchburg Bald eagle sighting

Bridgewater

bull Pfizer anticipates EPA cleanup plan for American Cyanamid site in Bridgewater (121611)

bull French drug maker plans to move its US business to Bridgewater (121511)

bull Federal environmental officials close to decision on $205M clean-up proposal of Bridgewaters Cyanamid property (121111)

bull The bike path is clear and the going is easy at Duke Island Park in Bridgewater (13111)

Califon

bull Hunterdon complaints of power outages prompt review of utilities by state Board of Public Utilities (112911)

bull Snowstorm results in $15000 bill to Califon for tree branch clean-up (112611)

Chester Borough New Jersey residents and communities cope with effects of pre-Halloween snowstorm (103111)

Chester Township North Hunterdon High School sets metro record for donating blood Clinton Town Clinton seeking new bids for sewer line manhole work (111411)

Clinton Township Village Green development in Annandale gets final approval to build on former Warren Lumber site (111411)

Cranbury Horse owner forestalls sale to slaughterhouse (121011)

Delaware Eroded Delaware Township roads popular with cyclists pedestrians prompt speed limit worries (121311)

Dunellen Thunderstorms roll through NJ causing lightning strikes and roof collapse (61711)

East Amwell Hunterdon Freeholders delay changes for Solid Waste Advisory Council after SWAC members object (12711)

2 Gathered July-December 2011 3 With assistance from Melissa Basile MCRP 2012 4 Gathered from the Star Ledger New Jersey County and Town Pages httpwwwnjcomlocal under ldquoMorerdquo ndashFor example httptopicsnjcomtagbernardspostshtml substituting the name of the town of interest

Appendix A News Snapshots

10

Figure 1 The interaction of nature models and controllable activities (shown in yellow)

Engineering models can combine information on rainfall specified as to time and space with information on soil moisture conditions seasonality and river levels prior to the rain event the fixed properties of the watershed (not indicated) and the land cover (natural and built environment) to produce flood inundation maps To obtain the precision needed to specify events that might occur every five or ten years requires LiDAR information on elevations We refer to the soil moisture conditions and river levels as ldquoantecedent eventsrdquo and note that seasonality is important because it affects the ability of the soil to absorb water (due to vegetation cover freezing etc) The land coverenvironment is viewed as modifiable (yellow) over the span of a few years That would in turn move the flood contours Flood inundation maps can be used for insurance andor regulatory purposes for risk assessments and to communicate complex geology and meteorology to stakeholder groups and thereby facilitate consideration of alternative policies

In the simplest case we might think of four types of soil moisture conditions dry ldquoaveragerdquo wet or frozen Probably it is sufficient to think of four different seasons We might also think of three basic river levels low average and high Thus there are in principle 4 x 4 x 3 = 48 different combinations of these conditions In reality not all of these combinations are feasible We wonrsquot have frozen ground in summer In our analysis we will probably demonstrate the methods choosing only several of the sensible alternatives for detailed investigation

Risk Modeling

The basic outline for how DHS does risk assessment is shown in Figure 2 below which is taken from a presentation by Isaac Maya of the CREATE Center It defines ldquoriskrdquo R as a function of threat T vulnerability V and consequence C R = f(TVC) Often this is represented as a product R = T x V x C Generally the factors T and V are some kinds of probabilities while C is some measure of loss In our case ldquoattackrdquo is interpreted to mean weather event and ldquosuccessrdquo is interpreted to mean flooding

Precipitation(xt) )tthe past

The models

Soil moisture conditions + seasonality + river levels seasonality

Land Cover (natural + built environment)

River gauge levels

LiDAR information

Flood Contour Maps

waterway

5 year contour

10 year contour

11

Figure 2 Sketch of the DHS computation of risk

There are two branches to the tree in this figure The top branch is the ldquostatus quordquo and the bottom branch involves some mitigation strategy or combination of such strategies The probability of attack P is the probability that there will be a weather event of the particular kind being considered and the probability of ldquosuccessrdquo Q is the conditional probability that such an event if it does occur will lead to flooding at a site being considered The consequences could be of various types In the case of floods they include loss of life economic damage to homes and businesses (direct or indirect) and psychological damage There is an attempt to put each of these types of consequences in terms of dollars The consequence (cost L) of an attack (flood) is usually calculated as a weighted sum of the different kinds of costs in our case of the cost in terms of lives lost economic damage and psychological damage What weights one uses and whether it makes sense to add up these weighted values are issues that will be viewed differently by different stakeholder groups

Complexities of the Watershed Situation

In fact we expect to be able to do more than simply treat ldquofloodrdquo as a yes or no matter as is done in this tree diagram Flood inundation maps will present levels of flood waters and the losses incurred change if there is a different level of flooding (or if there is a different duration for the flooding a factor we will disregard for now) The flood inundation levels are numerical values obtained from hydrological models (they can be later translated into major moderate minor near flood stage and none in making flood forecasts) A more complete model would replace the ldquosuccessfailurerdquo bifurcation in the tree diagram with a number of branches depending on the levelcategory of flood as measured for example by river gauges In our case it makes sense to define category 0 as no flooding category 1 as 0 to 1 feet category 2 as 1 to 2 feet etc

In practice obtaining the flood inundation levels may be complicated even before we introduce mitigating strategies The inundation maps are maps for theoretical storms not maps used to predict floods from an approaching storm These maps for the storms of a 100-year and 500-year ldquoreturn periodrdquo1 are already available and are being updated by FEMA using the recently available high resolution LiDAR data on the topography These maps are called flood insurance rate maps (FIRM) However they are developed from a kind of discrete-event modeling and may conceal hidden

1 This technical expression can be interpreted as representing a 1 of 15 probability per year

Success

Failure Attack

Vulnerability

No Attack

No Risk Management Plan

Threat of Attack

Success

Failure

Attack

No Attack

Risk Management Plan

Risk Management

PlansConsequences

L = Expected[$-Eqiuvalent Loss]

No Loss

No Loss

C + LC = Cost of Risk Management Plan

L = Expected[Reduced Loss] where L lt L

C

C

PQ

(1 - P)

(1 - Q)

P where P lt P

(1 - P)

Q where Q lt Q

(1 - Q)

12

uncertainties Moreover the inundation maps for more frequent storms are not readily available or non-existent It would be possible to build flood maps for 2-year 5-year 10-year 25-year and 50-year storms from available data This would imply gathering data and modeling and will depend on having access to the data Moreover the discrete-event models do not take into account antecedent conditions or seasonality Continuous modeling could take these factors into account but the cost of modeling would then be much higher

Assuming we can get flood inundation levels from hydrological models we would try to calculate the probability Qi that a flood of category i occurs Each category of flooding i would have a level of loss Li associated with it The level of flooding from a given amount of rainfall over a period of time depends not only on the precipitation but also on the soil moisture conditionsseasonalityriver levels and on the land cover (natural + built environment) Initially we may assume a fixed land cover (although this can be changed by mitigation) As noted above there are many different combinations of parameters for soil moisture conditions seasonality river levels We could in principle estimate the probability of each of these combinations (but would need to work out a method for doing so) For each combination we could use the hydrological models to calculate the flood level and thus the flood category Putting all of these calculations together will allow us to calculate the Qi parameters This is probably much too complicated and certainly much too expensive Instead we will pick out several more common combinations among the sensible ones and combine them with different precipitation patterns to define a set of weather events

In short our threats will be different kinds of weather events defined in terms of the profile of rainfall over a given period of time (say in number of inches in first hour number in second hour etc over a 24 hour period) plus antecedent precipitation and seasonal effects and pre-existing river levels We will aim to identify six to ten weather scenarios to analyze For each the probability P will have to be determined Information about precipitation totals in 24 hour periods is available and we could then use this to define a 2-year 5-year 10-year 25-year 50-year 100-year event in terms of precipitation profile per 24-hour period The probability of a 5-year event would be 15 of a 50 year event 150 etc except for the other factors (antecedent events plus seasonality) So we may want to modify P based on these other factors We will need to determine how to do that Note also that climate change will likely affect the value of P but we do not know how to calculate modified P with any degree of confidence

Now what is the interpretation of the Qi parameters If the weather event (precipitation profile) is defined and the antecedent conditions and seasonality are known in principle the hydrological models will produce just one category of flood But it could be that it makes more sense to build some uncertainty into the predictions of the models or it may make sense to assume that we have some uncertainty in knowing the soil moisture conditions or the river levels Thus there is surely still some value to studying the risk assessments using different flood categories

When there is a mitigation strategy presumably the probability of an attack would not be affected so Prsquo in the diagram is the same as P not less than P The probability of ldquosuccessrdquo or flood given a weather event is Qrsquo which is less than Q If we use different flood level categories then we would replace each Qi with a number Qirsquo but note that Qirsquo would be higher than Qi for some i Also the consequences of a flood could be less with expected consequences now Lrsquo presumably less than L (or Lirsquo presumably less than Li) But there is a cost C of the mitigation strategy and that cost applies even if there is no flood

Still using the notation in the diagram we see that if f(TVC) = TxVxC then with no mitigation strategy the risk or total expected cost is PxQxL but if there is a mitigation strategy then (since P = Prsquo) the risk or total expected cost is C + PxQrsquoxLrsquo The difference G = PxQxL ndash (C + PxQrsquoxLrsquo) in these expected costs would be the expected gain or expected risk reduction due to the strategy (If we have different flood level categories the calculation would be modified appropriately)

13

Mitigation Strategies and Elicitation of Some Key Parameters

We will need to decide if we look at flood damage (vulnerability) to a larger region or concentrate on a few towns Our preliminary plan is to do the analysis for a few sample towns for instance Manville and Millstone We will need to coordinate with the other parts of the project on this decision

Consequences will need to be measured by our partners in the project and we will need to coordinate with them

We will look at a variety of mitigation strategies Among them could be improved flood forecasting and warning systems retrofit of green infrastructure (eg impervious area disconnection rain gardens cisterns and rain barrels green roofs pervious concrete) to existing development areas to reduce upstream runoff volume regional (upstream) stormwater detention basins acquisition and demolition and relocation of properties flood-proofing and elevation in place disaster preparedness and response plans flood insurance cleanout of urban drainage systems construction of dams and reservoirs construction of dikes levees and floodwalls channel alterations and high flow diversions and spillways We will want to coordinate with other parts of the project on which mitigation strategies we look at

We will need to work with other parts of the project to get ballpark estimates of reduced costs Lrsquo (or Lirsquo) of floods

Under a given mitigation strategy the challenge is to produce inundation maps that will allow us to determine vulnerability (whether there is a flood) In principle there are hydrological models for doing this or such models could be developed but to do the actual calculations will be much more expensive than we can afford in this modest project Thus the best we can do is pick a few sample mitigation strategies for which we could make educated guesses as to vulnerabilities ndash just to illustrate the methodology Our thoughts are to start with three types of strategies retrofit of green infrastructure some sort of upstream mitigation that would have impact on downstream flooding and improved forecasting (eg improved monitoring from an optimum number of precipitation gauges merged station and radar precipitation estimates soil moisture observations evapotranspiration estimates and additional stream gauges) The last of these three would have no impact on the probabilities Q or Qi and also will not require changed inundation maps Only our calculation as to consequence would be affected since the whole idea of improved forecasting would be to allow preparation so as to reduce the potential flood damage

A full risk assessment would want to look not at individual mitigation strategies but at ldquoportfoliosrdquo of mitigation strategies Of course all of what was said above about ldquomitigation strategyrdquo could apply to a portfolio of strategies But the inundation maps will be much harder to obtain for portfolios of strategies than for individual ones As noted some of the mitigation strategies will not affect inundation maps Looking at mitigation strategies separately avoids the complication that several of these strategies may interact with each other in complex ways Some portfolios of mitigation strategies will be too expensive or infeasible for other reasons We may want to look at budgetary constraints and try to look at a few simple portfolios of strategies that have total investment cost (sum of Crsquos for each strategy in the portfolio) less than a given amount Then the question will become What is a measure of the value of a given portfolio Are we trying to maximize total expected gain (sum of G for all strategies in the portfolio) Or are we optimizing something else Or is this a ldquomulti-attributed utilityrdquo problem

14

Social Environmental and Political News Snapshots2 Hunterdon Mercer Middlesex Monmouth Morris Somerset Union

RARITAN RIVER TOWN SOCIAL ENVIRONMENTAL AND POLITICAL NEWS SNAPSHOTS 34

Alexandria 4-way stop to be set up near Delaware Valley High School

Bedminster

bull Clarence Dillon Library in Bedminster to host a series of Salons (92711) bull Bedminster bicyclist killed after crash with truck in Far Hills (12111)

Berkeley Heights Berkeley Heights childrenrsquos library bounces back Bernards Outraged NJ towns ask state to let them pull plug on JCPampL (12211)

Bernardsville Looking for updates on trail conditions since Hurricane Irene report in if you have them (9211)

Bethlehem Franklin Bethlehem school boards nearing alliance

Bound Brook

bull Flood prevention project in Bound Brook keeps Hurricane Irene from becoming another Floyd (83011)

bull Parts of Bound Brook South Bound Brook Manville under mandatory evacuation order (82811)

Branchburg Bald eagle sighting

Bridgewater

bull Pfizer anticipates EPA cleanup plan for American Cyanamid site in Bridgewater (121611)

bull French drug maker plans to move its US business to Bridgewater (121511)

bull Federal environmental officials close to decision on $205M clean-up proposal of Bridgewaters Cyanamid property (121111)

bull The bike path is clear and the going is easy at Duke Island Park in Bridgewater (13111)

Califon

bull Hunterdon complaints of power outages prompt review of utilities by state Board of Public Utilities (112911)

bull Snowstorm results in $15000 bill to Califon for tree branch clean-up (112611)

Chester Borough New Jersey residents and communities cope with effects of pre-Halloween snowstorm (103111)

Chester Township North Hunterdon High School sets metro record for donating blood Clinton Town Clinton seeking new bids for sewer line manhole work (111411)

Clinton Township Village Green development in Annandale gets final approval to build on former Warren Lumber site (111411)

Cranbury Horse owner forestalls sale to slaughterhouse (121011)

Delaware Eroded Delaware Township roads popular with cyclists pedestrians prompt speed limit worries (121311)

Dunellen Thunderstorms roll through NJ causing lightning strikes and roof collapse (61711)

East Amwell Hunterdon Freeholders delay changes for Solid Waste Advisory Council after SWAC members object (12711)

2 Gathered July-December 2011 3 With assistance from Melissa Basile MCRP 2012 4 Gathered from the Star Ledger New Jersey County and Town Pages httpwwwnjcomlocal under ldquoMorerdquo ndashFor example httptopicsnjcomtagbernardspostshtml substituting the name of the town of interest

Appendix A News Snapshots

11

Figure 2 Sketch of the DHS computation of risk

There are two branches to the tree in this figure The top branch is the ldquostatus quordquo and the bottom branch involves some mitigation strategy or combination of such strategies The probability of attack P is the probability that there will be a weather event of the particular kind being considered and the probability of ldquosuccessrdquo Q is the conditional probability that such an event if it does occur will lead to flooding at a site being considered The consequences could be of various types In the case of floods they include loss of life economic damage to homes and businesses (direct or indirect) and psychological damage There is an attempt to put each of these types of consequences in terms of dollars The consequence (cost L) of an attack (flood) is usually calculated as a weighted sum of the different kinds of costs in our case of the cost in terms of lives lost economic damage and psychological damage What weights one uses and whether it makes sense to add up these weighted values are issues that will be viewed differently by different stakeholder groups

Complexities of the Watershed Situation

In fact we expect to be able to do more than simply treat ldquofloodrdquo as a yes or no matter as is done in this tree diagram Flood inundation maps will present levels of flood waters and the losses incurred change if there is a different level of flooding (or if there is a different duration for the flooding a factor we will disregard for now) The flood inundation levels are numerical values obtained from hydrological models (they can be later translated into major moderate minor near flood stage and none in making flood forecasts) A more complete model would replace the ldquosuccessfailurerdquo bifurcation in the tree diagram with a number of branches depending on the levelcategory of flood as measured for example by river gauges In our case it makes sense to define category 0 as no flooding category 1 as 0 to 1 feet category 2 as 1 to 2 feet etc

In practice obtaining the flood inundation levels may be complicated even before we introduce mitigating strategies The inundation maps are maps for theoretical storms not maps used to predict floods from an approaching storm These maps for the storms of a 100-year and 500-year ldquoreturn periodrdquo1 are already available and are being updated by FEMA using the recently available high resolution LiDAR data on the topography These maps are called flood insurance rate maps (FIRM) However they are developed from a kind of discrete-event modeling and may conceal hidden

1 This technical expression can be interpreted as representing a 1 of 15 probability per year

Success

Failure Attack

Vulnerability

No Attack

No Risk Management Plan

Threat of Attack

Success

Failure

Attack

No Attack

Risk Management Plan

Risk Management

PlansConsequences

L = Expected[$-Eqiuvalent Loss]

No Loss

No Loss

C + LC = Cost of Risk Management Plan

L = Expected[Reduced Loss] where L lt L

C

C

PQ

(1 - P)

(1 - Q)

P where P lt P

(1 - P)

Q where Q lt Q

(1 - Q)

12

uncertainties Moreover the inundation maps for more frequent storms are not readily available or non-existent It would be possible to build flood maps for 2-year 5-year 10-year 25-year and 50-year storms from available data This would imply gathering data and modeling and will depend on having access to the data Moreover the discrete-event models do not take into account antecedent conditions or seasonality Continuous modeling could take these factors into account but the cost of modeling would then be much higher

Assuming we can get flood inundation levels from hydrological models we would try to calculate the probability Qi that a flood of category i occurs Each category of flooding i would have a level of loss Li associated with it The level of flooding from a given amount of rainfall over a period of time depends not only on the precipitation but also on the soil moisture conditionsseasonalityriver levels and on the land cover (natural + built environment) Initially we may assume a fixed land cover (although this can be changed by mitigation) As noted above there are many different combinations of parameters for soil moisture conditions seasonality river levels We could in principle estimate the probability of each of these combinations (but would need to work out a method for doing so) For each combination we could use the hydrological models to calculate the flood level and thus the flood category Putting all of these calculations together will allow us to calculate the Qi parameters This is probably much too complicated and certainly much too expensive Instead we will pick out several more common combinations among the sensible ones and combine them with different precipitation patterns to define a set of weather events

In short our threats will be different kinds of weather events defined in terms of the profile of rainfall over a given period of time (say in number of inches in first hour number in second hour etc over a 24 hour period) plus antecedent precipitation and seasonal effects and pre-existing river levels We will aim to identify six to ten weather scenarios to analyze For each the probability P will have to be determined Information about precipitation totals in 24 hour periods is available and we could then use this to define a 2-year 5-year 10-year 25-year 50-year 100-year event in terms of precipitation profile per 24-hour period The probability of a 5-year event would be 15 of a 50 year event 150 etc except for the other factors (antecedent events plus seasonality) So we may want to modify P based on these other factors We will need to determine how to do that Note also that climate change will likely affect the value of P but we do not know how to calculate modified P with any degree of confidence

Now what is the interpretation of the Qi parameters If the weather event (precipitation profile) is defined and the antecedent conditions and seasonality are known in principle the hydrological models will produce just one category of flood But it could be that it makes more sense to build some uncertainty into the predictions of the models or it may make sense to assume that we have some uncertainty in knowing the soil moisture conditions or the river levels Thus there is surely still some value to studying the risk assessments using different flood categories

When there is a mitigation strategy presumably the probability of an attack would not be affected so Prsquo in the diagram is the same as P not less than P The probability of ldquosuccessrdquo or flood given a weather event is Qrsquo which is less than Q If we use different flood level categories then we would replace each Qi with a number Qirsquo but note that Qirsquo would be higher than Qi for some i Also the consequences of a flood could be less with expected consequences now Lrsquo presumably less than L (or Lirsquo presumably less than Li) But there is a cost C of the mitigation strategy and that cost applies even if there is no flood

Still using the notation in the diagram we see that if f(TVC) = TxVxC then with no mitigation strategy the risk or total expected cost is PxQxL but if there is a mitigation strategy then (since P = Prsquo) the risk or total expected cost is C + PxQrsquoxLrsquo The difference G = PxQxL ndash (C + PxQrsquoxLrsquo) in these expected costs would be the expected gain or expected risk reduction due to the strategy (If we have different flood level categories the calculation would be modified appropriately)

13

Mitigation Strategies and Elicitation of Some Key Parameters

We will need to decide if we look at flood damage (vulnerability) to a larger region or concentrate on a few towns Our preliminary plan is to do the analysis for a few sample towns for instance Manville and Millstone We will need to coordinate with the other parts of the project on this decision

Consequences will need to be measured by our partners in the project and we will need to coordinate with them

We will look at a variety of mitigation strategies Among them could be improved flood forecasting and warning systems retrofit of green infrastructure (eg impervious area disconnection rain gardens cisterns and rain barrels green roofs pervious concrete) to existing development areas to reduce upstream runoff volume regional (upstream) stormwater detention basins acquisition and demolition and relocation of properties flood-proofing and elevation in place disaster preparedness and response plans flood insurance cleanout of urban drainage systems construction of dams and reservoirs construction of dikes levees and floodwalls channel alterations and high flow diversions and spillways We will want to coordinate with other parts of the project on which mitigation strategies we look at

We will need to work with other parts of the project to get ballpark estimates of reduced costs Lrsquo (or Lirsquo) of floods

Under a given mitigation strategy the challenge is to produce inundation maps that will allow us to determine vulnerability (whether there is a flood) In principle there are hydrological models for doing this or such models could be developed but to do the actual calculations will be much more expensive than we can afford in this modest project Thus the best we can do is pick a few sample mitigation strategies for which we could make educated guesses as to vulnerabilities ndash just to illustrate the methodology Our thoughts are to start with three types of strategies retrofit of green infrastructure some sort of upstream mitigation that would have impact on downstream flooding and improved forecasting (eg improved monitoring from an optimum number of precipitation gauges merged station and radar precipitation estimates soil moisture observations evapotranspiration estimates and additional stream gauges) The last of these three would have no impact on the probabilities Q or Qi and also will not require changed inundation maps Only our calculation as to consequence would be affected since the whole idea of improved forecasting would be to allow preparation so as to reduce the potential flood damage

A full risk assessment would want to look not at individual mitigation strategies but at ldquoportfoliosrdquo of mitigation strategies Of course all of what was said above about ldquomitigation strategyrdquo could apply to a portfolio of strategies But the inundation maps will be much harder to obtain for portfolios of strategies than for individual ones As noted some of the mitigation strategies will not affect inundation maps Looking at mitigation strategies separately avoids the complication that several of these strategies may interact with each other in complex ways Some portfolios of mitigation strategies will be too expensive or infeasible for other reasons We may want to look at budgetary constraints and try to look at a few simple portfolios of strategies that have total investment cost (sum of Crsquos for each strategy in the portfolio) less than a given amount Then the question will become What is a measure of the value of a given portfolio Are we trying to maximize total expected gain (sum of G for all strategies in the portfolio) Or are we optimizing something else Or is this a ldquomulti-attributed utilityrdquo problem

14

Social Environmental and Political News Snapshots2 Hunterdon Mercer Middlesex Monmouth Morris Somerset Union

RARITAN RIVER TOWN SOCIAL ENVIRONMENTAL AND POLITICAL NEWS SNAPSHOTS 34

Alexandria 4-way stop to be set up near Delaware Valley High School

Bedminster

bull Clarence Dillon Library in Bedminster to host a series of Salons (92711) bull Bedminster bicyclist killed after crash with truck in Far Hills (12111)

Berkeley Heights Berkeley Heights childrenrsquos library bounces back Bernards Outraged NJ towns ask state to let them pull plug on JCPampL (12211)

Bernardsville Looking for updates on trail conditions since Hurricane Irene report in if you have them (9211)

Bethlehem Franklin Bethlehem school boards nearing alliance

Bound Brook

bull Flood prevention project in Bound Brook keeps Hurricane Irene from becoming another Floyd (83011)

bull Parts of Bound Brook South Bound Brook Manville under mandatory evacuation order (82811)

Branchburg Bald eagle sighting

Bridgewater

bull Pfizer anticipates EPA cleanup plan for American Cyanamid site in Bridgewater (121611)

bull French drug maker plans to move its US business to Bridgewater (121511)

bull Federal environmental officials close to decision on $205M clean-up proposal of Bridgewaters Cyanamid property (121111)

bull The bike path is clear and the going is easy at Duke Island Park in Bridgewater (13111)

Califon

bull Hunterdon complaints of power outages prompt review of utilities by state Board of Public Utilities (112911)

bull Snowstorm results in $15000 bill to Califon for tree branch clean-up (112611)

Chester Borough New Jersey residents and communities cope with effects of pre-Halloween snowstorm (103111)

Chester Township North Hunterdon High School sets metro record for donating blood Clinton Town Clinton seeking new bids for sewer line manhole work (111411)

Clinton Township Village Green development in Annandale gets final approval to build on former Warren Lumber site (111411)

Cranbury Horse owner forestalls sale to slaughterhouse (121011)

Delaware Eroded Delaware Township roads popular with cyclists pedestrians prompt speed limit worries (121311)

Dunellen Thunderstorms roll through NJ causing lightning strikes and roof collapse (61711)

East Amwell Hunterdon Freeholders delay changes for Solid Waste Advisory Council after SWAC members object (12711)

2 Gathered July-December 2011 3 With assistance from Melissa Basile MCRP 2012 4 Gathered from the Star Ledger New Jersey County and Town Pages httpwwwnjcomlocal under ldquoMorerdquo ndashFor example httptopicsnjcomtagbernardspostshtml substituting the name of the town of interest

Appendix A News Snapshots

12

uncertainties Moreover the inundation maps for more frequent storms are not readily available or non-existent It would be possible to build flood maps for 2-year 5-year 10-year 25-year and 50-year storms from available data This would imply gathering data and modeling and will depend on having access to the data Moreover the discrete-event models do not take into account antecedent conditions or seasonality Continuous modeling could take these factors into account but the cost of modeling would then be much higher

Assuming we can get flood inundation levels from hydrological models we would try to calculate the probability Qi that a flood of category i occurs Each category of flooding i would have a level of loss Li associated with it The level of flooding from a given amount of rainfall over a period of time depends not only on the precipitation but also on the soil moisture conditionsseasonalityriver levels and on the land cover (natural + built environment) Initially we may assume a fixed land cover (although this can be changed by mitigation) As noted above there are many different combinations of parameters for soil moisture conditions seasonality river levels We could in principle estimate the probability of each of these combinations (but would need to work out a method for doing so) For each combination we could use the hydrological models to calculate the flood level and thus the flood category Putting all of these calculations together will allow us to calculate the Qi parameters This is probably much too complicated and certainly much too expensive Instead we will pick out several more common combinations among the sensible ones and combine them with different precipitation patterns to define a set of weather events

In short our threats will be different kinds of weather events defined in terms of the profile of rainfall over a given period of time (say in number of inches in first hour number in second hour etc over a 24 hour period) plus antecedent precipitation and seasonal effects and pre-existing river levels We will aim to identify six to ten weather scenarios to analyze For each the probability P will have to be determined Information about precipitation totals in 24 hour periods is available and we could then use this to define a 2-year 5-year 10-year 25-year 50-year 100-year event in terms of precipitation profile per 24-hour period The probability of a 5-year event would be 15 of a 50 year event 150 etc except for the other factors (antecedent events plus seasonality) So we may want to modify P based on these other factors We will need to determine how to do that Note also that climate change will likely affect the value of P but we do not know how to calculate modified P with any degree of confidence

Now what is the interpretation of the Qi parameters If the weather event (precipitation profile) is defined and the antecedent conditions and seasonality are known in principle the hydrological models will produce just one category of flood But it could be that it makes more sense to build some uncertainty into the predictions of the models or it may make sense to assume that we have some uncertainty in knowing the soil moisture conditions or the river levels Thus there is surely still some value to studying the risk assessments using different flood categories

When there is a mitigation strategy presumably the probability of an attack would not be affected so Prsquo in the diagram is the same as P not less than P The probability of ldquosuccessrdquo or flood given a weather event is Qrsquo which is less than Q If we use different flood level categories then we would replace each Qi with a number Qirsquo but note that Qirsquo would be higher than Qi for some i Also the consequences of a flood could be less with expected consequences now Lrsquo presumably less than L (or Lirsquo presumably less than Li) But there is a cost C of the mitigation strategy and that cost applies even if there is no flood

Still using the notation in the diagram we see that if f(TVC) = TxVxC then with no mitigation strategy the risk or total expected cost is PxQxL but if there is a mitigation strategy then (since P = Prsquo) the risk or total expected cost is C + PxQrsquoxLrsquo The difference G = PxQxL ndash (C + PxQrsquoxLrsquo) in these expected costs would be the expected gain or expected risk reduction due to the strategy (If we have different flood level categories the calculation would be modified appropriately)

13

Mitigation Strategies and Elicitation of Some Key Parameters

We will need to decide if we look at flood damage (vulnerability) to a larger region or concentrate on a few towns Our preliminary plan is to do the analysis for a few sample towns for instance Manville and Millstone We will need to coordinate with the other parts of the project on this decision

Consequences will need to be measured by our partners in the project and we will need to coordinate with them

We will look at a variety of mitigation strategies Among them could be improved flood forecasting and warning systems retrofit of green infrastructure (eg impervious area disconnection rain gardens cisterns and rain barrels green roofs pervious concrete) to existing development areas to reduce upstream runoff volume regional (upstream) stormwater detention basins acquisition and demolition and relocation of properties flood-proofing and elevation in place disaster preparedness and response plans flood insurance cleanout of urban drainage systems construction of dams and reservoirs construction of dikes levees and floodwalls channel alterations and high flow diversions and spillways We will want to coordinate with other parts of the project on which mitigation strategies we look at

We will need to work with other parts of the project to get ballpark estimates of reduced costs Lrsquo (or Lirsquo) of floods

Under a given mitigation strategy the challenge is to produce inundation maps that will allow us to determine vulnerability (whether there is a flood) In principle there are hydrological models for doing this or such models could be developed but to do the actual calculations will be much more expensive than we can afford in this modest project Thus the best we can do is pick a few sample mitigation strategies for which we could make educated guesses as to vulnerabilities ndash just to illustrate the methodology Our thoughts are to start with three types of strategies retrofit of green infrastructure some sort of upstream mitigation that would have impact on downstream flooding and improved forecasting (eg improved monitoring from an optimum number of precipitation gauges merged station and radar precipitation estimates soil moisture observations evapotranspiration estimates and additional stream gauges) The last of these three would have no impact on the probabilities Q or Qi and also will not require changed inundation maps Only our calculation as to consequence would be affected since the whole idea of improved forecasting would be to allow preparation so as to reduce the potential flood damage

A full risk assessment would want to look not at individual mitigation strategies but at ldquoportfoliosrdquo of mitigation strategies Of course all of what was said above about ldquomitigation strategyrdquo could apply to a portfolio of strategies But the inundation maps will be much harder to obtain for portfolios of strategies than for individual ones As noted some of the mitigation strategies will not affect inundation maps Looking at mitigation strategies separately avoids the complication that several of these strategies may interact with each other in complex ways Some portfolios of mitigation strategies will be too expensive or infeasible for other reasons We may want to look at budgetary constraints and try to look at a few simple portfolios of strategies that have total investment cost (sum of Crsquos for each strategy in the portfolio) less than a given amount Then the question will become What is a measure of the value of a given portfolio Are we trying to maximize total expected gain (sum of G for all strategies in the portfolio) Or are we optimizing something else Or is this a ldquomulti-attributed utilityrdquo problem

14

Social Environmental and Political News Snapshots2 Hunterdon Mercer Middlesex Monmouth Morris Somerset Union

RARITAN RIVER TOWN SOCIAL ENVIRONMENTAL AND POLITICAL NEWS SNAPSHOTS 34

Alexandria 4-way stop to be set up near Delaware Valley High School

Bedminster

bull Clarence Dillon Library in Bedminster to host a series of Salons (92711) bull Bedminster bicyclist killed after crash with truck in Far Hills (12111)

Berkeley Heights Berkeley Heights childrenrsquos library bounces back Bernards Outraged NJ towns ask state to let them pull plug on JCPampL (12211)

Bernardsville Looking for updates on trail conditions since Hurricane Irene report in if you have them (9211)

Bethlehem Franklin Bethlehem school boards nearing alliance

Bound Brook

bull Flood prevention project in Bound Brook keeps Hurricane Irene from becoming another Floyd (83011)

bull Parts of Bound Brook South Bound Brook Manville under mandatory evacuation order (82811)

Branchburg Bald eagle sighting

Bridgewater

bull Pfizer anticipates EPA cleanup plan for American Cyanamid site in Bridgewater (121611)

bull French drug maker plans to move its US business to Bridgewater (121511)

bull Federal environmental officials close to decision on $205M clean-up proposal of Bridgewaters Cyanamid property (121111)

bull The bike path is clear and the going is easy at Duke Island Park in Bridgewater (13111)

Califon

bull Hunterdon complaints of power outages prompt review of utilities by state Board of Public Utilities (112911)

bull Snowstorm results in $15000 bill to Califon for tree branch clean-up (112611)

Chester Borough New Jersey residents and communities cope with effects of pre-Halloween snowstorm (103111)

Chester Township North Hunterdon High School sets metro record for donating blood Clinton Town Clinton seeking new bids for sewer line manhole work (111411)

Clinton Township Village Green development in Annandale gets final approval to build on former Warren Lumber site (111411)

Cranbury Horse owner forestalls sale to slaughterhouse (121011)

Delaware Eroded Delaware Township roads popular with cyclists pedestrians prompt speed limit worries (121311)

Dunellen Thunderstorms roll through NJ causing lightning strikes and roof collapse (61711)

East Amwell Hunterdon Freeholders delay changes for Solid Waste Advisory Council after SWAC members object (12711)

2 Gathered July-December 2011 3 With assistance from Melissa Basile MCRP 2012 4 Gathered from the Star Ledger New Jersey County and Town Pages httpwwwnjcomlocal under ldquoMorerdquo ndashFor example httptopicsnjcomtagbernardspostshtml substituting the name of the town of interest

Appendix A News Snapshots

13

Mitigation Strategies and Elicitation of Some Key Parameters

We will need to decide if we look at flood damage (vulnerability) to a larger region or concentrate on a few towns Our preliminary plan is to do the analysis for a few sample towns for instance Manville and Millstone We will need to coordinate with the other parts of the project on this decision

Consequences will need to be measured by our partners in the project and we will need to coordinate with them

We will look at a variety of mitigation strategies Among them could be improved flood forecasting and warning systems retrofit of green infrastructure (eg impervious area disconnection rain gardens cisterns and rain barrels green roofs pervious concrete) to existing development areas to reduce upstream runoff volume regional (upstream) stormwater detention basins acquisition and demolition and relocation of properties flood-proofing and elevation in place disaster preparedness and response plans flood insurance cleanout of urban drainage systems construction of dams and reservoirs construction of dikes levees and floodwalls channel alterations and high flow diversions and spillways We will want to coordinate with other parts of the project on which mitigation strategies we look at

We will need to work with other parts of the project to get ballpark estimates of reduced costs Lrsquo (or Lirsquo) of floods

Under a given mitigation strategy the challenge is to produce inundation maps that will allow us to determine vulnerability (whether there is a flood) In principle there are hydrological models for doing this or such models could be developed but to do the actual calculations will be much more expensive than we can afford in this modest project Thus the best we can do is pick a few sample mitigation strategies for which we could make educated guesses as to vulnerabilities ndash just to illustrate the methodology Our thoughts are to start with three types of strategies retrofit of green infrastructure some sort of upstream mitigation that would have impact on downstream flooding and improved forecasting (eg improved monitoring from an optimum number of precipitation gauges merged station and radar precipitation estimates soil moisture observations evapotranspiration estimates and additional stream gauges) The last of these three would have no impact on the probabilities Q or Qi and also will not require changed inundation maps Only our calculation as to consequence would be affected since the whole idea of improved forecasting would be to allow preparation so as to reduce the potential flood damage

A full risk assessment would want to look not at individual mitigation strategies but at ldquoportfoliosrdquo of mitigation strategies Of course all of what was said above about ldquomitigation strategyrdquo could apply to a portfolio of strategies But the inundation maps will be much harder to obtain for portfolios of strategies than for individual ones As noted some of the mitigation strategies will not affect inundation maps Looking at mitigation strategies separately avoids the complication that several of these strategies may interact with each other in complex ways Some portfolios of mitigation strategies will be too expensive or infeasible for other reasons We may want to look at budgetary constraints and try to look at a few simple portfolios of strategies that have total investment cost (sum of Crsquos for each strategy in the portfolio) less than a given amount Then the question will become What is a measure of the value of a given portfolio Are we trying to maximize total expected gain (sum of G for all strategies in the portfolio) Or are we optimizing something else Or is this a ldquomulti-attributed utilityrdquo problem

14

Social Environmental and Political News Snapshots2 Hunterdon Mercer Middlesex Monmouth Morris Somerset Union

RARITAN RIVER TOWN SOCIAL ENVIRONMENTAL AND POLITICAL NEWS SNAPSHOTS 34

Alexandria 4-way stop to be set up near Delaware Valley High School

Bedminster

bull Clarence Dillon Library in Bedminster to host a series of Salons (92711) bull Bedminster bicyclist killed after crash with truck in Far Hills (12111)

Berkeley Heights Berkeley Heights childrenrsquos library bounces back Bernards Outraged NJ towns ask state to let them pull plug on JCPampL (12211)

Bernardsville Looking for updates on trail conditions since Hurricane Irene report in if you have them (9211)

Bethlehem Franklin Bethlehem school boards nearing alliance

Bound Brook

bull Flood prevention project in Bound Brook keeps Hurricane Irene from becoming another Floyd (83011)

bull Parts of Bound Brook South Bound Brook Manville under mandatory evacuation order (82811)

Branchburg Bald eagle sighting

Bridgewater

bull Pfizer anticipates EPA cleanup plan for American Cyanamid site in Bridgewater (121611)

bull French drug maker plans to move its US business to Bridgewater (121511)

bull Federal environmental officials close to decision on $205M clean-up proposal of Bridgewaters Cyanamid property (121111)

bull The bike path is clear and the going is easy at Duke Island Park in Bridgewater (13111)

Califon

bull Hunterdon complaints of power outages prompt review of utilities by state Board of Public Utilities (112911)

bull Snowstorm results in $15000 bill to Califon for tree branch clean-up (112611)

Chester Borough New Jersey residents and communities cope with effects of pre-Halloween snowstorm (103111)

Chester Township North Hunterdon High School sets metro record for donating blood Clinton Town Clinton seeking new bids for sewer line manhole work (111411)

Clinton Township Village Green development in Annandale gets final approval to build on former Warren Lumber site (111411)

Cranbury Horse owner forestalls sale to slaughterhouse (121011)

Delaware Eroded Delaware Township roads popular with cyclists pedestrians prompt speed limit worries (121311)

Dunellen Thunderstorms roll through NJ causing lightning strikes and roof collapse (61711)

East Amwell Hunterdon Freeholders delay changes for Solid Waste Advisory Council after SWAC members object (12711)

2 Gathered July-December 2011 3 With assistance from Melissa Basile MCRP 2012 4 Gathered from the Star Ledger New Jersey County and Town Pages httpwwwnjcomlocal under ldquoMorerdquo ndashFor example httptopicsnjcomtagbernardspostshtml substituting the name of the town of interest

Appendix A News Snapshots

14

Social Environmental and Political News Snapshots2 Hunterdon Mercer Middlesex Monmouth Morris Somerset Union

RARITAN RIVER TOWN SOCIAL ENVIRONMENTAL AND POLITICAL NEWS SNAPSHOTS 34

Alexandria 4-way stop to be set up near Delaware Valley High School

Bedminster

bull Clarence Dillon Library in Bedminster to host a series of Salons (92711) bull Bedminster bicyclist killed after crash with truck in Far Hills (12111)

Berkeley Heights Berkeley Heights childrenrsquos library bounces back Bernards Outraged NJ towns ask state to let them pull plug on JCPampL (12211)

Bernardsville Looking for updates on trail conditions since Hurricane Irene report in if you have them (9211)

Bethlehem Franklin Bethlehem school boards nearing alliance

Bound Brook

bull Flood prevention project in Bound Brook keeps Hurricane Irene from becoming another Floyd (83011)

bull Parts of Bound Brook South Bound Brook Manville under mandatory evacuation order (82811)

Branchburg Bald eagle sighting

Bridgewater

bull Pfizer anticipates EPA cleanup plan for American Cyanamid site in Bridgewater (121611)

bull French drug maker plans to move its US business to Bridgewater (121511)

bull Federal environmental officials close to decision on $205M clean-up proposal of Bridgewaters Cyanamid property (121111)

bull The bike path is clear and the going is easy at Duke Island Park in Bridgewater (13111)

Califon

bull Hunterdon complaints of power outages prompt review of utilities by state Board of Public Utilities (112911)

bull Snowstorm results in $15000 bill to Califon for tree branch clean-up (112611)

Chester Borough New Jersey residents and communities cope with effects of pre-Halloween snowstorm (103111)

Chester Township North Hunterdon High School sets metro record for donating blood Clinton Town Clinton seeking new bids for sewer line manhole work (111411)

Clinton Township Village Green development in Annandale gets final approval to build on former Warren Lumber site (111411)

Cranbury Horse owner forestalls sale to slaughterhouse (121011)

Delaware Eroded Delaware Township roads popular with cyclists pedestrians prompt speed limit worries (121311)

Dunellen Thunderstorms roll through NJ causing lightning strikes and roof collapse (61711)

East Amwell Hunterdon Freeholders delay changes for Solid Waste Advisory Council after SWAC members object (12711)

2 Gathered July-December 2011 3 With assistance from Melissa Basile MCRP 2012 4 Gathered from the Star Ledger New Jersey County and Town Pages httpwwwnjcomlocal under ldquoMorerdquo ndashFor example httptopicsnjcomtagbernardspostshtml substituting the name of the town of interest

Appendix A News Snapshots

15

East Brunswick bull Young black bear is captured in East Brunswick after three-day trek VIDEO

(7611) bull East Brunswick landfill garbage to be used as power source (111311)

East Windsor Candidates unveil jobs plan (101311) Edison Edison teen wins silver medal in chemistry competition in Turkey Englishtown Boro Englishtown Council Censures and Reprimands Jayne Carr

Fanwood bull NJ organization wants a Scotch Plains-Fanwood merger (121111) bull Scotch-Plains Fanwood Girl Scouts boost recycling at their school

(121111)

Far Hills bull Somerset County Election Results (11811) bull Annual Far Hills Race Meeting brings families together for tailgating

tradition (102311)

Flemington Editorial Raritan Townships construction fee waiver for businesses can hurt neighbors including Flemington (12811)

Franklin Solar panels replace greenhouses in plan for Franklin Township land (112111)

Franklin Little snow in NJ but plenty of rain falls (12611)

Freehold Borough The 11th Annual Ride for Autism takes place this coming Saturday June 11 (6811)

Freehold Township bull Freehold entertainment center to offer arcade and rides in space that

used to host defunct shopping center (11911) bull Freehold opens first multiple sclerosis housing complex in NJ (103011)

Glen Gardner bull Pay to play ordinance on Glen Gardners agenda (9911) bull Plans to petition Route 31 intersection at Glen Gardner Borough Council

delayed tonights meeting rescheduled for Monday Nov 7 (11111)

Green Brook Be Red Cross Ready use this Hurricane Safety Checklist to prepare for Hurricane Irene (82511)

Hampton bull Hampton eyes lower speed limits for three streets (111511) bull Speed limit on three Hampton streets drops to 15 mph (112911)

Helmetta Santa to come in by train (12811) High Bridge High Bridge Boy Scouts walk the Gettysburg battlefield (12411)

Highland Park bull Highland Park student proposes municipal consolidation bull Highland Park along with other NJ communities does not want charter

schools

Hightstown

bull Hightstown officials in transition months after Hurricane Irene swept through the region (12111)

bull Projectrsquos foes forgo protest at Hightstown celebration (112611) bull Councilman considers civil rights suit against Hightstown (111611)

Hillsborough Enjoy a sunflower safari at Duke Farms in Hillsborough

Hopewell Boro

Hopewell Activists Want Citizens to have a Say in costly utility deal (121711)

Hopewell Twp

bull State historic panel criticizes county replacement plan for Jacobs Creek bridge (121611)

bull Mercer County saves with in-house bridge repair (11511) bull Historic designation raises hopes for Jacobs Creek Bridge activists in

16

Hopewell Township (121511) bull Mercer County saves with in-house bridge repair (11511)

Jamesburg bull Flooding in Jamesburg covers cars climbs up houses VIDEO (82811) bull Hurricane Irenes rising floodwaters bring new dangers to NJ residents (82911)

Lawrence bull Contracts awarded to repair firehouse mold in Lawrence (12811) bull Lawrence township takes action to improve water volume pressure

(112911)

Lebanon Borough

bull Clinton Township school board hires interim business administrator to serve Lebanon too (127111)

bull Tell us What are your favorite things to see at the Hunterdon County 4-H Fair (81511)

Lebanon Township Lebanon Twp agrees to put E-Ticket system in police cars Manalapan NJ to be spared from storm that was expected to cause flooding (92411)

Manville Political controversy over ldquodouble-dippingrdquo of pension funds Council members vote to increase property tax

Marlboro

bull JCPampL faces more heat from homeowners who lost power during Hurricane Irene (92711)

bull Marlboro natives passion for fossils leads to honor at site of numerous finds (71911)

Mendham Borough

bull 113000 NJ homes and businesses remain without power 4 days after October snowstorm (11311)

bull Mendham principals hair color on the line in upcoming 5K charity race (92611)

Mendham Township Metuchen

Middlesex Borough

bull Treating New Jerseys wastewater after it leaves your house Video (12111)

bull NJ residents wait for word on divvying of FEMA-issued $28M to buy out flood-ravaged towns (121011)

Millstone

bull 16th District races affect parts of Hunterdon Somerset Mercer and Middlesex counties (102411)

bull Somerset County and eight towns form flood control commission (101911)

bull Di Ionno A reflection on New Jerseys importance in the Revolutionary War (7411)

Millstone Community poised to help Franklin Food Bank promote canal and Millstone Valley (10411)

Milltown

bull Glimpse of History Raritan River Railroad station in Milltown had many lives (92511)

bull Braun Hurricane Irene makes Milltown residents rethink public power station (83111)

bull Aftermath of Hurricane Irene (11911)

Mine Hill bull Morris County recycling program reduces garbage collection (10911) bull Morris Habitat for Humanity 25th Anniversary Year A Great Success

(92011)

17

Monroe

bull Make a Wish opens wishing place castle in Middlesex County (12211) bull Water-logged NJ residents angered by post-hurricane scavengers

(91211) bull In Hurricane Irenes aftermath much praise but a few heart-felt

complaints as well in NJ (9411) bull Volunteers in Monroe Township in Middlesex County organize to bring

relief after Irene (9311)

Montgomery Update Somerset County announces road closings due to flooding (12811)

Mount Arlington Controversial plan to combine Mount Arlington Roxbury Township police force appears dead (91111)

Mount Olive NJ Land Conservancy purchases 208-acre farm in Mount Olive to protect drinking water mitigate flooding (112011)

Mountainside

bull National Flood Insurance officials extend deadline for flood insurance claims (12111)

bull In reversal Mountainside to now clear away fallen tree limbs downed by devastating October (11811)

New Brunswick

bull Inland NJ stocks up for Hurricane Irene Anti-flooding strategies weighed (82711)

bull New Brunswick tries to ban bulk garbage pickups (8311) bull Cityrsquos new transit village (EEK Architects 2010)

New Providence

bull New Providence to Give Old Generator to Senior Housing Complex (120511)

bull Graffiti is Damaging Property in New Providence (11811) bull New Providence Amateur Radio Club to talk about the next power failure

(112811)

North Brunswick

bull Half of Middlesex Countys 25 towns to have mayoral elections next month (102011)

bull Milltown Road on-ramp to Route 1 north to be closed in Middlesex County (9911)

bull Bicycling in Traffic Skills 101 course to be offered in North Brunswick on May 7 (5211)

North Plainfield Somerset County hosts Journey Through the Past at 21 sites (101011)

Old Bridge Quality care Magazine ranks four area hospitals among the best in the region (72911)

Peapack-Gladstone

bull September fun with the Upper Raritan Watershed Association (91311) bull Bernards Township in talks with Bernardsville for dispatch partnership

(62611) bull Gill St Bernards students take a turkey to school for food bank

(112110)

Pennington bull Hundreds in Mercer County still without power (11111) bull Lingering outages plague west Mercer (103111)

Perth Amboy

bull Glimpse of History Waterfront way station in Perth Amboy connects New York and Philadelphia (91111)

bull Perth Amboy declares state of emergency orders 8 pm curfew (82711) bull Hurricane Irene has Perth Amboy boat owners heading to the marina to

18

secure vessels (82711) bull Residents in low-lying areas of Middlesex County asked to voluntarily

evacuate ahead of Hurricane Irene (82611) Piscataway Water main break shuts down road (72811)

Plainfield bull Former Trenton official gets same role in Plainfield (112411) bull Union County historic buildings to get $500K for repairs (11611)

Plainsboro Opinion NJ towns should play strong role in determining where solar fields are built (112811) State Sens Greenstein and Turner easily win re-election (11811)

Princeton Borough Princetons working together to provide new voting districts

Princeton Township Historians speak out against proposed Institute housing plan on Princeton Battlefield (12911)

Randolph More than 17000 NJ residents still without power 6 days after October snowstorm (11411)

Raritan Boro Reenactment of the signing of the treaty that ended WWI

Raritan Twp bull South Branch and Upper Raritan watershed associations merging bull Raritan Township requests study to consider countywide school

consolidation (103111)

Readington Hunterdon Freeholder Will Mennen wants to fill Biondi seat in the state Assembly (12611)

Robbinsville Hopewell considers joining Robbinsville in complaint against JCPL (12211) Rocky Hill Bill of Fare Farm to fork menu is key to One 52s success (111111)

Roosevelt Boro Not available

Roxbury bull Post-snow notebook NJ residents recover after freak storm (103111) bull North Jersey plagued by post-Irene mosquitoes drawn to stagnant water

(91811)

Sayreville bull Residents concerned about dredged soil brought to National Lead site bull Sayreville proposing to construct a truck highway bypass

Scotch Plains Scotch Plains Brunner school now official Monarch butterfly waystation (112011)

Somerville Around The Towns Holidays celebrated in 18th century-style (121111)

South Amboy NJ residents flock to Raritan Bay for last look before Hurricane Irene hits (82711)

South Bound Brook Libertarian party organization chairrsquos lawsuit seeks file possibly involving South Bound Brook NJ mayorrsquos relative one year ago

South Brunswick South Brunswick Has Fifth Lowest Crime Rate in Middlesex County (112211)

South Plainfield Irene brings long hours and late summer bump for tree trimming companies (83011)

South River AIG sells NJ apartments some in South River (7711)

Spotswood Nearly 80 percent of NJ school budgets pass representing highest rate in a decade (42911)

Springfield Twp Red Cross shelter took in many area residents (91111)

Summit bull Summit council approves renegotiated garbage contract (121011) bull Summit parents can take free English as Second Language classes

(121211)

19

bull Scout Post New Venturing Crew seeking members (121112) bull Capitola Dickerson given Key to the City of Summit (12711)

Tewksbury

bull Elections 2011 Tewksbury Committee Seats Sought by Polito Van Doren (102511)

bull Tewksbury to Address False Alarms Tree Cutting (101711) bull NJ Highlands Act has saved thousands of acres from development but

continues to cause friction (102611)

Union (Hunterdon)

bull Main Street near Jutland in Union Twp open again after culvert repairs (121011)

bull Hunterdon Freeholders might amend SWAC bylaws to increase attendance give board more power (102011)

Warren

bull Somerset County under a Flood Watch rain might change to snow in Watchung Warren area (12611)

bull Somerset County offers workshops on how to access 2010 Census data (12611)

Washington (Morris)

bull October snowstorm gives NJ a white Halloween (103111) bull Morris County acquires 20-acre horse farm with preservation funds

(61611) bull Morris County preserves another Washington Township farm (121910)

Watchung bull In Somerset County RideWise promotes use of carpools (112911) bull Watchung Warren Rotary celebrates 50th anniversary (112011)

West Amwell

bull Editorial Property tax re-assessment in Hunterdon also makes sense during a recession (12911)

bull Hunterdon County updates list for shelter shower water cell-phone charging after snow storm cuts power (11211)

bull FFA chapter at South Hunterdon High chooses new officers (102911)

West Windsor County officials encounter opposition to bike route plan in West Windsor (12511)

Woodbridge

bull Di Ionno On NJ road the worlds dreams unfold (12411) bull Achievements Woodbridge honored by Sustainable Jersey (112011) bull In red-blue NJ a few towns have the power to swing elections (11611) bull Poverty rate growing in NJs working-class towns census data shows

(11311) bull Woodbridge votes to save Colonia Country Club from development

(102511)

20

Appendix B-1 Safety UCR Safety - Top of Form

TOP 100 Towns for crime5 Click town name for complete crime detail

RANK MUNICIPALITY COUNTY TOTAL 1 Newark ESSEX 15097 2 Jersey City HUDSON 10589 3 Camden CAMDEN 6515 4 Paterson PASSAIC 6037 5 Elizabeth UNION 5772 6 Atlantic City ATLANTIC 5357 7 Irvington ESSEX 4488 8 Trenton MERCER 3895 9 Vineland CUMBERLAND 3278

10 East Orange ESSEX 3181 11 Woodbridge MIDDLESEX 3117 12 Edison MIDDLESEX 2618 13 Cherry Hill CAMDEN 2376 14 Passaic PASSAIC 2325 15 New Brunswick MIDDLESEX 2276 16 Plainfield UNION 2189 17 Clifton City PASSAIC 2131 18 Hamilton Twp MERCER 2117 19 Dover OCEAN 2075 20 Union City HUDSON 1983 21 Lakewood OCEAN 1887 22 Orange ESSEX 1846 23 Gloucester Twp CAMDEN 1831 24 Millville CUMBERLAND 1775 25 Paramus BERGEN 1757 26 Union UNION 1744 27 Deptford GLOUCESTER 1549 28 Brick OCEAN 1545 29 Neptune Twp MONMOUTH 1540 30 Pennsauken CAMDEN 1507

5 Star Ledger Crime in New Jersey Report httpwwwstarledgercomstrindexpagecrimernkcrimeasp Crime in New Jersey from the NJ State Police Uniform Crime Report Rankings

31 Bloomfield ESSEX 1478 32 Linden UNION 1450 33 Bridgeton CUMBERLAND 1400 34 Perth Amboy MIDDLESEX 1373 35 Asbury Park MONMOUTH 1305 36 Wayne PASSAIC 1296 37 Egg Harbor Twp ATLANTIC 1290 38 Hackensack BERGEN 1237 39 Hoboken HUDSON 1194 40 Hamilton Twp ATLANTIC 1171 41 West Orange ESSEX 1159 42 Washington Twp GLOUCESTER 1104 43 Kearny HUDSON 1091 44 Montclair ESSEX 1090 45 Lawrence MERCER 1074 46 Ocean City CAPE MAY 1066 47 Bayonne HUDSON 1057 48 North Bergen HUDSON 1035 49 West New York HUDSON 1021 50 Winslow CAMDEN 1017 51 Belleville ESSEX 1014 52 Old Bridge MIDDLESEX 1000 53 Long Branch MONMOUTH 978 54 Freehold Twp MONMOUTH 957 55 Franklin Twp SOMERSET 950 56 North Brunswick MIDDLESEX 940 57 Parsippany-Troy Hills MORRIS 933 58 East Brunswick MIDDLESEX 932 59 Ewing MERCER 894 60 Piscataway MIDDLESEX 894 61 Sayreville MIDDLESEX 893 62 Voorhees CAMDEN 886 63 Middletown MONMOUTH 862 64 Galloway ATLANTIC 861 65 Middle Twp CAPE MAY 859 66 Lindenwold CAMDEN 839 67 Morristown MORRIS 835 68 Monroe GLOUCESTER 824 69 Glassboro GLOUCESTER 815

70 Mount Laurel BURLINGTON 814 71 Ocean MONMOUTH 800 72 Pleasantville ATLANTIC 795 73 Millburn ESSEX 781 74 Pemberton Twp BURLINGTON 774 75 Willingboro BURLINGTON 768 76 Evesham BURLINGTON 765 77 Hillside UNION 756 78 Berkeley OCEAN 735 79 Secaucus HUDSON 732 80 Woodbury GLOUCESTER 712 81 Rahway UNION 711 82 Teaneck BERGEN 688 83 Bridgewater SOMERSET 676 84 Englewood BERGEN 651 85 Lacey OCEAN 641 86 Eatontown MONMOUTH 639 87 South Brunswick MIDDLESEX 630 88 North Plainfield SOMERSET 615 89 Jackson OCEAN 612 90 Maplewood ESSEX 598 91 West Deptford GLOUCESTER 582 92 West Windsor MERCER 563 93 Livingston ESSEX 544 94 Howell MONMOUTH 535 95 Wildwood CAPE MAY 531 96 Garfield BERGEN 527 97 Lower Twp CAPE MAY 526 98 Wall MONMOUTH 522 99 South Plainfield MIDDLESEX 518

100 Stafford OCEAN 508

21

Safety ndash Lowest 100 Towns for crime6

RANK MUNICIPALITY COUNTY TOTAL 1 Pine Valley CAMDEN 0 2 Tavistock CAMDEN 0 3 Millstone SOMERSET 0 4 Rocky Hill SOMERSET 3 5 Walpack SUSSEX 3 6 Interlaken MONMOUTH 3 7 Roosevelt MONMOUTH 4 8 South Bound Brook SOMERSET 4 9 Stockton HUNTERDON 4 10 Califon HUNTERDON 5 11 Shrewsbury MONMOUTH 6 12 Ogdensburg SUSSEX 7 13 Fieldsboro BURLINGTON 7 14 Haworth BERGEN 8 15 Shiloh CUMBERLAN

D 8

16 Winfield UNION 8 17 Hardwick WARREN 8 18 Helmetta MIDDLESEX 8 19 Alpine BERGEN 9 20 Branchville SUSSEX 10 21 Harrington Park BERGEN 11 22 Rockleigh BERGEN 11 23 Greenwich CUMBERLAN

D 12

24 Andover Borough SUSSEX 12 25 Hopewell Borough MERCER 12 26 Far Hills SOMERSET 13 27 Milford HUNTERDON 13 28 Glen Gardner HUNTERDON 14 29 Bethlehem HUNTERDON 14 30 Ho-Ho-Kus BERGEN 14 31 Peapack-Gladstone SOMERSET 14 32 Mantoloking OCEAN 15 33 Oxford WARREN 15 34 Corbin City ATLANTIC 15

6 Star Ledger Crime in New Jersey Report httpwwwstarledgercomstrindexpagecrimernkcrimeasp Crime in New Jersey from the NJ State Police Uniform Crime Report Rankings

35 Essex Fells ESSEX 16 36 Lower Alloways

Creek SALEM 16

37 Englishtown MONMOUTH 16 38 Island Heights OCEAN 16 39 Harding MORRIS 17 40 Sea Bright MONMOUTH 17 41 Sandyston SUSSEX 17 42 Washington BURLINGTON 17 43 Loch Arbour

Village MONMOUTH 18

44 Allenhurst MONMOUTH 18 45 Frelinghuysen WARREN 19 46 Harvey Cedars OCEAN 19 47 Wenonah GLOUCESTER 19 48 Bloomsbury HUNTERDON 19 49 Kingwood HUNTERDON 19 50 Stow Creek CUMBERLAN

D 19

51 West Wildwood CAPE MAY 20 52 Holland Twp HUNTERDON 20 53 East Amwell HUNTERDON 20 54 Clinton HUNTERDON 20 55 Port Republic ATLANTIC 20 56 Belvidere WARREN 20 57 Harmony WARREN 20 58 Hope WARREN 21 59 New Hanover BURLINGTON 21 60 Lebanon Borough HUNTERDON 21 61 Alexandria HUNTERDON 22 62 Fredon SUSSEX 22 63 Green SUSSEX 22 64 Victory Gardens MORRIS 22 65 Pennington MERCER 23 66 Hi-nella CAMDEN 23 67 Estell Manor ATLANTIC 24 68 Allentown MONMOUTH 24 69 Chester Borough MORRIS 24 70 Liberty WARREN 24 71 Elsinboro SALEM 25 72 Saddle River BERGEN 25 73 Frenchtown HUNTERDON 26 74 Hampton HUNTERDON 26 75 Franklin Twp HUNTERDON 27 76 Pine Beach OCEAN 27 77 Lafayette SUSSEX 27

78 Farmingdale MONMOUTH 27 79 Boonton Twp MORRIS 28 80 Cresskill BERGEN 28 81 Weymouth ATLANTIC 29 82 Northvale BERGEN 29 83 Cape May Point CAPE MAY 29 84 Eagleswood OCEAN 29 85 West Amwell HUNTERDON 29 86 Stillwater SUSSEX 30 87 East Newark HUDSON 30 88 Wrightstown BURLINGTON 30 89 Teterboro BERGEN 31 90 Lavallette OCEAN 32 91 Alloway SALEM 32 92 Woodcliff Lake BERGEN 33 93 Audubon Park CAMDEN 33 94 Longport ATLANTIC 33 95 South Harrison GLOUCESTER 33 96 Newfield GLOUCESTER 34 97 Old Tappan BERGEN 34 98 Mannington SALEM 34 99 Spring Lake Heights MONMOUTH 34

100 Monmouth Beach MONMOUTH 35

22

Safety - Top 100 Towns for crime7

Click town name for complete crime detail

RR Rank Rank Municipality County Total

1 11 Woodbridge Middlesex 3117

2 12 Edison Middlesex 2618

3 15 New Brunswick Middlesex 2276

4 16 Plainfield Union 2189

5 26 Union Union 1744

6 34 Perth Amboy Middlesex 1373

7 45 Lawrence Mercer 1074

8 52 Old Bridge Middlesex 1000

9 54 Freehold Twp Monmouth 957

10 55 Franklin Twp Somerset 950

11 56 North Brunswick Middlesex 940

12 57 Parsippany-Troy Hills Morris 933

13 58 East Brunswick Middlesex 932

14 60 Piscataway Middlesex 894

15 61 Sayreville Middlesex 893

16 83 Bridgewater Somerset 676

17 87 South Brunswick Middlesex 630

18 88 North Plainfield Somerset 615

19 92 West Windsor Mercer 563

20 99 South Plainfield Middlesex 518

7 Star Ledger Crime in New Jersey Report httpwwwstarledgercomstrindexpagecrimernkcrimeasp Crime in New Jersey State Police Uniform Crime Report Rankings

Safety ndash LOWEST 100 Towns for Crime RR

Rank Rank Municipality County Total

1 3 Millstone Somerset 0 2 4 Rocky Hill Somerset 3 3 7 Roosevelt Monmouth 4 4 8 South Bound Brook Somerset 4 5 10 Califon Hunterdon 5 6 18 Helmetta Middlesex 8 7 25 Hopewell Borough Mercer 12 8 26 Far Hills Somerset 13 9 28 Glen Gardner Hunterdon 14

10 29 Bethlehem Hunterdon 14 11 31 Peapack-Gladstone Somerset 14 12 37 Englishtown Monmouth 16 13 53 East Amwell Hunterdon 20 14 60 Lebanon Borough Hunterdon 21 15 61 Alexandria Hunterdon 22 16 69 Chester Borough Morris 24 17 74 Hampton Hunterdon 26 18 75 Franklin Twp Hunterdon 27 19 85 West Amwell Hunterdon 29

23

Appendix B-2 Safety UCR Detail Safety Information

Safety information is listed for each community through the New Jersey State Police Uniform Crime Reports Star Ledger ldquoNew Jersey by the Numbersrdquo at httpwwwnjcomnewsbythenumbers

The following data for Peapack-Gladstone is an example of the information available

Peapack-Gladstone Overall Crime 1997

930 1998

1150 1999

1370 2000

860 2001

1030 2002

1230 2003

1050 2004

860 2005

360 2006

568 Crimes per 1000 population

Year-by-Year Detail Reported Crimes Click on Violent or Non-Violent for a crime breakdown

Index Offenses Crime Rates Personnel

Year Crimes Violent

Non-Violent

Rate Per 1000 pop

Violent Rate

Non Violent Rate Male Female Civilian Total

1997 21 0 21 930 000 930 8 0 1 9

1998 26 1 25 1150 040 1110 8 0 1 9

1999 32 1 31 1370 040 1330 9 0 1 10

2000 21 0 21 860 000 860 8 0 1 9

2001 25 0 25 1030 000 1030 8 0 1 9

2002 30 0 30 1230 000 1230 9 0 1 10

2003 26 2 24 1050 080 970 7 0 1 8

2004 21 0 21 860 000 860 7 0 1 8

2005 9 0 9 360 000 360 8 0 1 9

2006 14 1 13 568 041 527 9 0 1 10

Includes only offenses that make up official crime index Click on Violent or Non-Violent to see which offenses are included and to get numbers

24

Appendix C Maps

Map 1 Watershed Management Area 8 by FEMA Payout

25

Map 2 Watershed Management Area 9 by FEMA Payout

26

Map 3 Watershed Management Area 10 by FEMA Payout

27

Map 4 WMA 8 Planning Area

28

Map 5 WMA 9 Planning Area

29

Map 6 WMA 10 Planning Area

30

Appendix D Literature Economic Assessment Literature Review Chao P Floyd J and W Holliday (1998) ldquoEmpirical Studies of Effect of Flood Risk on

Housing pricesrdquo US Army Corps of Engineers Chatterton J Viavattene C Morris J Penning-Rowsell E Tapsel S (2010) ldquoThe Costs of

the Summer 2007 Floods in England Flood and Coastal Erosionrdquo Risk Management Research and Development Programme

Daniel Florax and Rietveld (2005) ldquoRiver Flooding and Housing Values An Economic

Assessment of Environmental Riskrdquo European Regional Science Association Dei-Tutut A (2002) ldquoFlood Hazards Insurance and House Prices-A Hedonic Property Price

Analysisrdquo East Carolina University Department of Economics Fridgen P and S Shultz (1999)The Influence of the Threat of Flooding on Housing Values

in Fargo North Dakota and Moorhead MinnesotaNorth Dakota State University Agricultural Economics Report No 417

Harrison D G Smersh and A Schwartz (2001)ldquoEnvironmental Determinants of Housing

Prices The Impact of Flood Zone Statusrdquo Journal of Real Estate Research 21 3-20 MacDonald D H White P Taube and W Huth (1990) ldquoFlood Hazard Pricing and Insurance

Premium Differentials Evidence from the Housing Marketrdquo Journal of Risk and Insurance 57 654-63

New Jersey Mitigation Task Force (2006) Report of Delaware River Flood Mitigation Pielke R Downton M and J Miller (2002) ldquoFlood Damage in the United States 1926-2000 A

reanalysis of National Weather Service Estimatesrdquo National Center for Atmospheric Research Pfurtscheller and Schwarze ldquoEstimating the Costs of Emergency Services During Flood

Events (PPT Only) Risk Management of Extreme Flood Event Shabman L and D Damianos (1976) ldquoFlood hazard effects on residential property valuesrdquo

Journal of the Water Resources Planning and Management Division 151-62 Simonovic and Carson (2003) ldquoFlooding in the Red River Basin-Lessons from Post Flood

Activitiesrdquo Natural Hazards 28 345-365 Speyer J and W R Ragas 1991 Housing Prices and Flood Risk An Examination using Spline

Regression Journal of Real Estate and Finance Economics 4 395-407 Turnbull G Zahirovic-Herbert V and Mothorpe C (2007) ldquoFlooding and Liquidity on the

Bayou The Capitalization of Flood Risk into House Value and Ease-of-Salerdquo University of Central Florida Dr P Phillips School of Real EstateWorking Paper 1107

31

Appendix E Guo CV Qizhong (George) Guo

Associate Professor Civil and Environmental Engineering qguorcirutgersedu

732-445-4444 CEB 109

Research and Specialty Areas

bull Hydraulics Hydrology and Water Quality bull Urban Stormwater Management bull Watershed and Water Environment Restoration bull Water Resources and Environmental Engineering

Current and Recent Research Projects Stormwater System Monitoring and Evaluation funded by NJDOT

Drainage System Identification and Analysis funded by NJDOT via NJIT

Automatic Vacuum Flushing Technology for Combined Sewer Solids funded by USEPA via WERF

High Volume Utilization of Fly Ash Containing Mercury-Impregnated Carbon funded by USDOE via Ceramatec Inc

Stormwater Management Rule Implementation Process funded by NJDOT

Correlation of Total Suspended Solids and Suspended Sediment Concentration Test Methods funded by NJDEP

Watershed Modeling to Support Post Brook Regional Stormwater Management Plan Development funded by USEPA-NJDEP via GSE LLC

Section 319H NPS Pollution Control and Management Implementation Grant Program Implementation of Stormwater Detention Basin Retrofitting Techniques funded by USEPA via NJDEP

Development of a Numerical Model to Assess the Impacts of Raw Water Quality on Conventional Drinking Water Treatment funded by NJDEP

Water Quality Modeling Study of Lower Maurice River Estuary funded by Commercial Township NJ

Publications Selected Refereed Journal Articles

Li Y and Guo Q (2012) ldquoAngular Velocity Formula for Turbulent Vortex Chamber Flowsrdquo Journal of Hydraulic Engineering in press

Roseen R M Fowler G D Ballestero T P and Guo Q (2011) Sediment Monitoring Bias by Autosampler in Comparison with Whole Volume Sampling for Parking Lot Runoff Journal of Irrigation and Drainage Engineering Vol 137 No 4 pp 251-257

Huang S Ng C and Guo Q (2007) ldquoExperimental investigation of the effect of flow turbulence and sediment transport on the adsorption of cadmium ions onto sediment particlesrdquo Journal of Environmental Sciences Vol 19 pp 696-703

25

Map 2 Watershed Management Area 9 by FEMA Payout

26

Map 3 Watershed Management Area 10 by FEMA Payout

27

Map 4 WMA 8 Planning Area

28

Map 5 WMA 9 Planning Area

29

Map 6 WMA 10 Planning Area

30

Appendix D Literature Economic Assessment Literature Review Chao P Floyd J and W Holliday (1998) ldquoEmpirical Studies of Effect of Flood Risk on

Housing pricesrdquo US Army Corps of Engineers Chatterton J Viavattene C Morris J Penning-Rowsell E Tapsel S (2010) ldquoThe Costs of

the Summer 2007 Floods in England Flood and Coastal Erosionrdquo Risk Management Research and Development Programme

Daniel Florax and Rietveld (2005) ldquoRiver Flooding and Housing Values An Economic

Assessment of Environmental Riskrdquo European Regional Science Association Dei-Tutut A (2002) ldquoFlood Hazards Insurance and House Prices-A Hedonic Property Price

Analysisrdquo East Carolina University Department of Economics Fridgen P and S Shultz (1999)The Influence of the Threat of Flooding on Housing Values

in Fargo North Dakota and Moorhead MinnesotaNorth Dakota State University Agricultural Economics Report No 417

Harrison D G Smersh and A Schwartz (2001)ldquoEnvironmental Determinants of Housing

Prices The Impact of Flood Zone Statusrdquo Journal of Real Estate Research 21 3-20 MacDonald D H White P Taube and W Huth (1990) ldquoFlood Hazard Pricing and Insurance

Premium Differentials Evidence from the Housing Marketrdquo Journal of Risk and Insurance 57 654-63

New Jersey Mitigation Task Force (2006) Report of Delaware River Flood Mitigation Pielke R Downton M and J Miller (2002) ldquoFlood Damage in the United States 1926-2000 A

reanalysis of National Weather Service Estimatesrdquo National Center for Atmospheric Research Pfurtscheller and Schwarze ldquoEstimating the Costs of Emergency Services During Flood

Events (PPT Only) Risk Management of Extreme Flood Event Shabman L and D Damianos (1976) ldquoFlood hazard effects on residential property valuesrdquo

Journal of the Water Resources Planning and Management Division 151-62 Simonovic and Carson (2003) ldquoFlooding in the Red River Basin-Lessons from Post Flood

Activitiesrdquo Natural Hazards 28 345-365 Speyer J and W R Ragas 1991 Housing Prices and Flood Risk An Examination using Spline

Regression Journal of Real Estate and Finance Economics 4 395-407 Turnbull G Zahirovic-Herbert V and Mothorpe C (2007) ldquoFlooding and Liquidity on the

Bayou The Capitalization of Flood Risk into House Value and Ease-of-Salerdquo University of Central Florida Dr P Phillips School of Real EstateWorking Paper 1107

31

Appendix E Guo CV Qizhong (George) Guo

Associate Professor Civil and Environmental Engineering qguorcirutgersedu

732-445-4444 CEB 109

Research and Specialty Areas

bull Hydraulics Hydrology and Water Quality bull Urban Stormwater Management bull Watershed and Water Environment Restoration bull Water Resources and Environmental Engineering

Current and Recent Research Projects Stormwater System Monitoring and Evaluation funded by NJDOT

Drainage System Identification and Analysis funded by NJDOT via NJIT

Automatic Vacuum Flushing Technology for Combined Sewer Solids funded by USEPA via WERF

High Volume Utilization of Fly Ash Containing Mercury-Impregnated Carbon funded by USDOE via Ceramatec Inc

Stormwater Management Rule Implementation Process funded by NJDOT

Correlation of Total Suspended Solids and Suspended Sediment Concentration Test Methods funded by NJDEP

Watershed Modeling to Support Post Brook Regional Stormwater Management Plan Development funded by USEPA-NJDEP via GSE LLC

Section 319H NPS Pollution Control and Management Implementation Grant Program Implementation of Stormwater Detention Basin Retrofitting Techniques funded by USEPA via NJDEP

Development of a Numerical Model to Assess the Impacts of Raw Water Quality on Conventional Drinking Water Treatment funded by NJDEP

Water Quality Modeling Study of Lower Maurice River Estuary funded by Commercial Township NJ

Publications Selected Refereed Journal Articles

Li Y and Guo Q (2012) ldquoAngular Velocity Formula for Turbulent Vortex Chamber Flowsrdquo Journal of Hydraulic Engineering in press

Roseen R M Fowler G D Ballestero T P and Guo Q (2011) Sediment Monitoring Bias by Autosampler in Comparison with Whole Volume Sampling for Parking Lot Runoff Journal of Irrigation and Drainage Engineering Vol 137 No 4 pp 251-257

Huang S Ng C and Guo Q (2007) ldquoExperimental investigation of the effect of flow turbulence and sediment transport on the adsorption of cadmium ions onto sediment particlesrdquo Journal of Environmental Sciences Vol 19 pp 696-703

26

Map 3 Watershed Management Area 10 by FEMA Payout

27

Map 4 WMA 8 Planning Area

28

Map 5 WMA 9 Planning Area

29

Map 6 WMA 10 Planning Area

30

Appendix D Literature Economic Assessment Literature Review Chao P Floyd J and W Holliday (1998) ldquoEmpirical Studies of Effect of Flood Risk on

Housing pricesrdquo US Army Corps of Engineers Chatterton J Viavattene C Morris J Penning-Rowsell E Tapsel S (2010) ldquoThe Costs of

the Summer 2007 Floods in England Flood and Coastal Erosionrdquo Risk Management Research and Development Programme

Daniel Florax and Rietveld (2005) ldquoRiver Flooding and Housing Values An Economic

Assessment of Environmental Riskrdquo European Regional Science Association Dei-Tutut A (2002) ldquoFlood Hazards Insurance and House Prices-A Hedonic Property Price

Analysisrdquo East Carolina University Department of Economics Fridgen P and S Shultz (1999)The Influence of the Threat of Flooding on Housing Values

in Fargo North Dakota and Moorhead MinnesotaNorth Dakota State University Agricultural Economics Report No 417

Harrison D G Smersh and A Schwartz (2001)ldquoEnvironmental Determinants of Housing

Prices The Impact of Flood Zone Statusrdquo Journal of Real Estate Research 21 3-20 MacDonald D H White P Taube and W Huth (1990) ldquoFlood Hazard Pricing and Insurance

Premium Differentials Evidence from the Housing Marketrdquo Journal of Risk and Insurance 57 654-63

New Jersey Mitigation Task Force (2006) Report of Delaware River Flood Mitigation Pielke R Downton M and J Miller (2002) ldquoFlood Damage in the United States 1926-2000 A

reanalysis of National Weather Service Estimatesrdquo National Center for Atmospheric Research Pfurtscheller and Schwarze ldquoEstimating the Costs of Emergency Services During Flood

Events (PPT Only) Risk Management of Extreme Flood Event Shabman L and D Damianos (1976) ldquoFlood hazard effects on residential property valuesrdquo

Journal of the Water Resources Planning and Management Division 151-62 Simonovic and Carson (2003) ldquoFlooding in the Red River Basin-Lessons from Post Flood

Activitiesrdquo Natural Hazards 28 345-365 Speyer J and W R Ragas 1991 Housing Prices and Flood Risk An Examination using Spline

Regression Journal of Real Estate and Finance Economics 4 395-407 Turnbull G Zahirovic-Herbert V and Mothorpe C (2007) ldquoFlooding and Liquidity on the

Bayou The Capitalization of Flood Risk into House Value and Ease-of-Salerdquo University of Central Florida Dr P Phillips School of Real EstateWorking Paper 1107

31

Appendix E Guo CV Qizhong (George) Guo

Associate Professor Civil and Environmental Engineering qguorcirutgersedu

732-445-4444 CEB 109

Research and Specialty Areas

bull Hydraulics Hydrology and Water Quality bull Urban Stormwater Management bull Watershed and Water Environment Restoration bull Water Resources and Environmental Engineering

Current and Recent Research Projects Stormwater System Monitoring and Evaluation funded by NJDOT

Drainage System Identification and Analysis funded by NJDOT via NJIT

Automatic Vacuum Flushing Technology for Combined Sewer Solids funded by USEPA via WERF

High Volume Utilization of Fly Ash Containing Mercury-Impregnated Carbon funded by USDOE via Ceramatec Inc

Stormwater Management Rule Implementation Process funded by NJDOT

Correlation of Total Suspended Solids and Suspended Sediment Concentration Test Methods funded by NJDEP

Watershed Modeling to Support Post Brook Regional Stormwater Management Plan Development funded by USEPA-NJDEP via GSE LLC

Section 319H NPS Pollution Control and Management Implementation Grant Program Implementation of Stormwater Detention Basin Retrofitting Techniques funded by USEPA via NJDEP

Development of a Numerical Model to Assess the Impacts of Raw Water Quality on Conventional Drinking Water Treatment funded by NJDEP

Water Quality Modeling Study of Lower Maurice River Estuary funded by Commercial Township NJ

Publications Selected Refereed Journal Articles

Li Y and Guo Q (2012) ldquoAngular Velocity Formula for Turbulent Vortex Chamber Flowsrdquo Journal of Hydraulic Engineering in press

Roseen R M Fowler G D Ballestero T P and Guo Q (2011) Sediment Monitoring Bias by Autosampler in Comparison with Whole Volume Sampling for Parking Lot Runoff Journal of Irrigation and Drainage Engineering Vol 137 No 4 pp 251-257

Huang S Ng C and Guo Q (2007) ldquoExperimental investigation of the effect of flow turbulence and sediment transport on the adsorption of cadmium ions onto sediment particlesrdquo Journal of Environmental Sciences Vol 19 pp 696-703

27

Map 4 WMA 8 Planning Area

28

Map 5 WMA 9 Planning Area

29

Map 6 WMA 10 Planning Area

30

Appendix D Literature Economic Assessment Literature Review Chao P Floyd J and W Holliday (1998) ldquoEmpirical Studies of Effect of Flood Risk on

Housing pricesrdquo US Army Corps of Engineers Chatterton J Viavattene C Morris J Penning-Rowsell E Tapsel S (2010) ldquoThe Costs of

the Summer 2007 Floods in England Flood and Coastal Erosionrdquo Risk Management Research and Development Programme

Daniel Florax and Rietveld (2005) ldquoRiver Flooding and Housing Values An Economic

Assessment of Environmental Riskrdquo European Regional Science Association Dei-Tutut A (2002) ldquoFlood Hazards Insurance and House Prices-A Hedonic Property Price

Analysisrdquo East Carolina University Department of Economics Fridgen P and S Shultz (1999)The Influence of the Threat of Flooding on Housing Values

in Fargo North Dakota and Moorhead MinnesotaNorth Dakota State University Agricultural Economics Report No 417

Harrison D G Smersh and A Schwartz (2001)ldquoEnvironmental Determinants of Housing

Prices The Impact of Flood Zone Statusrdquo Journal of Real Estate Research 21 3-20 MacDonald D H White P Taube and W Huth (1990) ldquoFlood Hazard Pricing and Insurance

Premium Differentials Evidence from the Housing Marketrdquo Journal of Risk and Insurance 57 654-63

New Jersey Mitigation Task Force (2006) Report of Delaware River Flood Mitigation Pielke R Downton M and J Miller (2002) ldquoFlood Damage in the United States 1926-2000 A

reanalysis of National Weather Service Estimatesrdquo National Center for Atmospheric Research Pfurtscheller and Schwarze ldquoEstimating the Costs of Emergency Services During Flood

Events (PPT Only) Risk Management of Extreme Flood Event Shabman L and D Damianos (1976) ldquoFlood hazard effects on residential property valuesrdquo

Journal of the Water Resources Planning and Management Division 151-62 Simonovic and Carson (2003) ldquoFlooding in the Red River Basin-Lessons from Post Flood

Activitiesrdquo Natural Hazards 28 345-365 Speyer J and W R Ragas 1991 Housing Prices and Flood Risk An Examination using Spline

Regression Journal of Real Estate and Finance Economics 4 395-407 Turnbull G Zahirovic-Herbert V and Mothorpe C (2007) ldquoFlooding and Liquidity on the

Bayou The Capitalization of Flood Risk into House Value and Ease-of-Salerdquo University of Central Florida Dr P Phillips School of Real EstateWorking Paper 1107

31

Appendix E Guo CV Qizhong (George) Guo

Associate Professor Civil and Environmental Engineering qguorcirutgersedu

732-445-4444 CEB 109

Research and Specialty Areas

bull Hydraulics Hydrology and Water Quality bull Urban Stormwater Management bull Watershed and Water Environment Restoration bull Water Resources and Environmental Engineering

Current and Recent Research Projects Stormwater System Monitoring and Evaluation funded by NJDOT

Drainage System Identification and Analysis funded by NJDOT via NJIT

Automatic Vacuum Flushing Technology for Combined Sewer Solids funded by USEPA via WERF

High Volume Utilization of Fly Ash Containing Mercury-Impregnated Carbon funded by USDOE via Ceramatec Inc

Stormwater Management Rule Implementation Process funded by NJDOT

Correlation of Total Suspended Solids and Suspended Sediment Concentration Test Methods funded by NJDEP

Watershed Modeling to Support Post Brook Regional Stormwater Management Plan Development funded by USEPA-NJDEP via GSE LLC

Section 319H NPS Pollution Control and Management Implementation Grant Program Implementation of Stormwater Detention Basin Retrofitting Techniques funded by USEPA via NJDEP

Development of a Numerical Model to Assess the Impacts of Raw Water Quality on Conventional Drinking Water Treatment funded by NJDEP

Water Quality Modeling Study of Lower Maurice River Estuary funded by Commercial Township NJ

Publications Selected Refereed Journal Articles

Li Y and Guo Q (2012) ldquoAngular Velocity Formula for Turbulent Vortex Chamber Flowsrdquo Journal of Hydraulic Engineering in press

Roseen R M Fowler G D Ballestero T P and Guo Q (2011) Sediment Monitoring Bias by Autosampler in Comparison with Whole Volume Sampling for Parking Lot Runoff Journal of Irrigation and Drainage Engineering Vol 137 No 4 pp 251-257

Huang S Ng C and Guo Q (2007) ldquoExperimental investigation of the effect of flow turbulence and sediment transport on the adsorption of cadmium ions onto sediment particlesrdquo Journal of Environmental Sciences Vol 19 pp 696-703

28

Map 5 WMA 9 Planning Area

29

Map 6 WMA 10 Planning Area

30

Appendix D Literature Economic Assessment Literature Review Chao P Floyd J and W Holliday (1998) ldquoEmpirical Studies of Effect of Flood Risk on

Housing pricesrdquo US Army Corps of Engineers Chatterton J Viavattene C Morris J Penning-Rowsell E Tapsel S (2010) ldquoThe Costs of

the Summer 2007 Floods in England Flood and Coastal Erosionrdquo Risk Management Research and Development Programme

Daniel Florax and Rietveld (2005) ldquoRiver Flooding and Housing Values An Economic

Assessment of Environmental Riskrdquo European Regional Science Association Dei-Tutut A (2002) ldquoFlood Hazards Insurance and House Prices-A Hedonic Property Price

Analysisrdquo East Carolina University Department of Economics Fridgen P and S Shultz (1999)The Influence of the Threat of Flooding on Housing Values

in Fargo North Dakota and Moorhead MinnesotaNorth Dakota State University Agricultural Economics Report No 417

Harrison D G Smersh and A Schwartz (2001)ldquoEnvironmental Determinants of Housing

Prices The Impact of Flood Zone Statusrdquo Journal of Real Estate Research 21 3-20 MacDonald D H White P Taube and W Huth (1990) ldquoFlood Hazard Pricing and Insurance

Premium Differentials Evidence from the Housing Marketrdquo Journal of Risk and Insurance 57 654-63

New Jersey Mitigation Task Force (2006) Report of Delaware River Flood Mitigation Pielke R Downton M and J Miller (2002) ldquoFlood Damage in the United States 1926-2000 A

reanalysis of National Weather Service Estimatesrdquo National Center for Atmospheric Research Pfurtscheller and Schwarze ldquoEstimating the Costs of Emergency Services During Flood

Events (PPT Only) Risk Management of Extreme Flood Event Shabman L and D Damianos (1976) ldquoFlood hazard effects on residential property valuesrdquo

Journal of the Water Resources Planning and Management Division 151-62 Simonovic and Carson (2003) ldquoFlooding in the Red River Basin-Lessons from Post Flood

Activitiesrdquo Natural Hazards 28 345-365 Speyer J and W R Ragas 1991 Housing Prices and Flood Risk An Examination using Spline

Regression Journal of Real Estate and Finance Economics 4 395-407 Turnbull G Zahirovic-Herbert V and Mothorpe C (2007) ldquoFlooding and Liquidity on the

Bayou The Capitalization of Flood Risk into House Value and Ease-of-Salerdquo University of Central Florida Dr P Phillips School of Real EstateWorking Paper 1107

31

Appendix E Guo CV Qizhong (George) Guo

Associate Professor Civil and Environmental Engineering qguorcirutgersedu

732-445-4444 CEB 109

Research and Specialty Areas

bull Hydraulics Hydrology and Water Quality bull Urban Stormwater Management bull Watershed and Water Environment Restoration bull Water Resources and Environmental Engineering

Current and Recent Research Projects Stormwater System Monitoring and Evaluation funded by NJDOT

Drainage System Identification and Analysis funded by NJDOT via NJIT

Automatic Vacuum Flushing Technology for Combined Sewer Solids funded by USEPA via WERF

High Volume Utilization of Fly Ash Containing Mercury-Impregnated Carbon funded by USDOE via Ceramatec Inc

Stormwater Management Rule Implementation Process funded by NJDOT

Correlation of Total Suspended Solids and Suspended Sediment Concentration Test Methods funded by NJDEP

Watershed Modeling to Support Post Brook Regional Stormwater Management Plan Development funded by USEPA-NJDEP via GSE LLC

Section 319H NPS Pollution Control and Management Implementation Grant Program Implementation of Stormwater Detention Basin Retrofitting Techniques funded by USEPA via NJDEP

Development of a Numerical Model to Assess the Impacts of Raw Water Quality on Conventional Drinking Water Treatment funded by NJDEP

Water Quality Modeling Study of Lower Maurice River Estuary funded by Commercial Township NJ

Publications Selected Refereed Journal Articles

Li Y and Guo Q (2012) ldquoAngular Velocity Formula for Turbulent Vortex Chamber Flowsrdquo Journal of Hydraulic Engineering in press

Roseen R M Fowler G D Ballestero T P and Guo Q (2011) Sediment Monitoring Bias by Autosampler in Comparison with Whole Volume Sampling for Parking Lot Runoff Journal of Irrigation and Drainage Engineering Vol 137 No 4 pp 251-257

Huang S Ng C and Guo Q (2007) ldquoExperimental investigation of the effect of flow turbulence and sediment transport on the adsorption of cadmium ions onto sediment particlesrdquo Journal of Environmental Sciences Vol 19 pp 696-703

29

Map 6 WMA 10 Planning Area

30

Appendix D Literature Economic Assessment Literature Review Chao P Floyd J and W Holliday (1998) ldquoEmpirical Studies of Effect of Flood Risk on

Housing pricesrdquo US Army Corps of Engineers Chatterton J Viavattene C Morris J Penning-Rowsell E Tapsel S (2010) ldquoThe Costs of

the Summer 2007 Floods in England Flood and Coastal Erosionrdquo Risk Management Research and Development Programme

Daniel Florax and Rietveld (2005) ldquoRiver Flooding and Housing Values An Economic

Assessment of Environmental Riskrdquo European Regional Science Association Dei-Tutut A (2002) ldquoFlood Hazards Insurance and House Prices-A Hedonic Property Price

Analysisrdquo East Carolina University Department of Economics Fridgen P and S Shultz (1999)The Influence of the Threat of Flooding on Housing Values

in Fargo North Dakota and Moorhead MinnesotaNorth Dakota State University Agricultural Economics Report No 417

Harrison D G Smersh and A Schwartz (2001)ldquoEnvironmental Determinants of Housing

Prices The Impact of Flood Zone Statusrdquo Journal of Real Estate Research 21 3-20 MacDonald D H White P Taube and W Huth (1990) ldquoFlood Hazard Pricing and Insurance

Premium Differentials Evidence from the Housing Marketrdquo Journal of Risk and Insurance 57 654-63

New Jersey Mitigation Task Force (2006) Report of Delaware River Flood Mitigation Pielke R Downton M and J Miller (2002) ldquoFlood Damage in the United States 1926-2000 A

reanalysis of National Weather Service Estimatesrdquo National Center for Atmospheric Research Pfurtscheller and Schwarze ldquoEstimating the Costs of Emergency Services During Flood

Events (PPT Only) Risk Management of Extreme Flood Event Shabman L and D Damianos (1976) ldquoFlood hazard effects on residential property valuesrdquo

Journal of the Water Resources Planning and Management Division 151-62 Simonovic and Carson (2003) ldquoFlooding in the Red River Basin-Lessons from Post Flood

Activitiesrdquo Natural Hazards 28 345-365 Speyer J and W R Ragas 1991 Housing Prices and Flood Risk An Examination using Spline

Regression Journal of Real Estate and Finance Economics 4 395-407 Turnbull G Zahirovic-Herbert V and Mothorpe C (2007) ldquoFlooding and Liquidity on the

Bayou The Capitalization of Flood Risk into House Value and Ease-of-Salerdquo University of Central Florida Dr P Phillips School of Real EstateWorking Paper 1107

31

Appendix E Guo CV Qizhong (George) Guo

Associate Professor Civil and Environmental Engineering qguorcirutgersedu

732-445-4444 CEB 109

Research and Specialty Areas

bull Hydraulics Hydrology and Water Quality bull Urban Stormwater Management bull Watershed and Water Environment Restoration bull Water Resources and Environmental Engineering

Current and Recent Research Projects Stormwater System Monitoring and Evaluation funded by NJDOT

Drainage System Identification and Analysis funded by NJDOT via NJIT

Automatic Vacuum Flushing Technology for Combined Sewer Solids funded by USEPA via WERF

High Volume Utilization of Fly Ash Containing Mercury-Impregnated Carbon funded by USDOE via Ceramatec Inc

Stormwater Management Rule Implementation Process funded by NJDOT

Correlation of Total Suspended Solids and Suspended Sediment Concentration Test Methods funded by NJDEP

Watershed Modeling to Support Post Brook Regional Stormwater Management Plan Development funded by USEPA-NJDEP via GSE LLC

Section 319H NPS Pollution Control and Management Implementation Grant Program Implementation of Stormwater Detention Basin Retrofitting Techniques funded by USEPA via NJDEP

Development of a Numerical Model to Assess the Impacts of Raw Water Quality on Conventional Drinking Water Treatment funded by NJDEP

Water Quality Modeling Study of Lower Maurice River Estuary funded by Commercial Township NJ

Publications Selected Refereed Journal Articles

Li Y and Guo Q (2012) ldquoAngular Velocity Formula for Turbulent Vortex Chamber Flowsrdquo Journal of Hydraulic Engineering in press

Roseen R M Fowler G D Ballestero T P and Guo Q (2011) Sediment Monitoring Bias by Autosampler in Comparison with Whole Volume Sampling for Parking Lot Runoff Journal of Irrigation and Drainage Engineering Vol 137 No 4 pp 251-257

Huang S Ng C and Guo Q (2007) ldquoExperimental investigation of the effect of flow turbulence and sediment transport on the adsorption of cadmium ions onto sediment particlesrdquo Journal of Environmental Sciences Vol 19 pp 696-703

30

Appendix D Literature Economic Assessment Literature Review Chao P Floyd J and W Holliday (1998) ldquoEmpirical Studies of Effect of Flood Risk on

Housing pricesrdquo US Army Corps of Engineers Chatterton J Viavattene C Morris J Penning-Rowsell E Tapsel S (2010) ldquoThe Costs of

the Summer 2007 Floods in England Flood and Coastal Erosionrdquo Risk Management Research and Development Programme

Daniel Florax and Rietveld (2005) ldquoRiver Flooding and Housing Values An Economic

Assessment of Environmental Riskrdquo European Regional Science Association Dei-Tutut A (2002) ldquoFlood Hazards Insurance and House Prices-A Hedonic Property Price

Analysisrdquo East Carolina University Department of Economics Fridgen P and S Shultz (1999)The Influence of the Threat of Flooding on Housing Values

in Fargo North Dakota and Moorhead MinnesotaNorth Dakota State University Agricultural Economics Report No 417

Harrison D G Smersh and A Schwartz (2001)ldquoEnvironmental Determinants of Housing

Prices The Impact of Flood Zone Statusrdquo Journal of Real Estate Research 21 3-20 MacDonald D H White P Taube and W Huth (1990) ldquoFlood Hazard Pricing and Insurance

Premium Differentials Evidence from the Housing Marketrdquo Journal of Risk and Insurance 57 654-63

New Jersey Mitigation Task Force (2006) Report of Delaware River Flood Mitigation Pielke R Downton M and J Miller (2002) ldquoFlood Damage in the United States 1926-2000 A

reanalysis of National Weather Service Estimatesrdquo National Center for Atmospheric Research Pfurtscheller and Schwarze ldquoEstimating the Costs of Emergency Services During Flood

Events (PPT Only) Risk Management of Extreme Flood Event Shabman L and D Damianos (1976) ldquoFlood hazard effects on residential property valuesrdquo

Journal of the Water Resources Planning and Management Division 151-62 Simonovic and Carson (2003) ldquoFlooding in the Red River Basin-Lessons from Post Flood

Activitiesrdquo Natural Hazards 28 345-365 Speyer J and W R Ragas 1991 Housing Prices and Flood Risk An Examination using Spline

Regression Journal of Real Estate and Finance Economics 4 395-407 Turnbull G Zahirovic-Herbert V and Mothorpe C (2007) ldquoFlooding and Liquidity on the

Bayou The Capitalization of Flood Risk into House Value and Ease-of-Salerdquo University of Central Florida Dr P Phillips School of Real EstateWorking Paper 1107

31

Appendix E Guo CV Qizhong (George) Guo

Associate Professor Civil and Environmental Engineering qguorcirutgersedu

732-445-4444 CEB 109

Research and Specialty Areas

bull Hydraulics Hydrology and Water Quality bull Urban Stormwater Management bull Watershed and Water Environment Restoration bull Water Resources and Environmental Engineering

Current and Recent Research Projects Stormwater System Monitoring and Evaluation funded by NJDOT

Drainage System Identification and Analysis funded by NJDOT via NJIT

Automatic Vacuum Flushing Technology for Combined Sewer Solids funded by USEPA via WERF

High Volume Utilization of Fly Ash Containing Mercury-Impregnated Carbon funded by USDOE via Ceramatec Inc

Stormwater Management Rule Implementation Process funded by NJDOT

Correlation of Total Suspended Solids and Suspended Sediment Concentration Test Methods funded by NJDEP

Watershed Modeling to Support Post Brook Regional Stormwater Management Plan Development funded by USEPA-NJDEP via GSE LLC

Section 319H NPS Pollution Control and Management Implementation Grant Program Implementation of Stormwater Detention Basin Retrofitting Techniques funded by USEPA via NJDEP

Development of a Numerical Model to Assess the Impacts of Raw Water Quality on Conventional Drinking Water Treatment funded by NJDEP

Water Quality Modeling Study of Lower Maurice River Estuary funded by Commercial Township NJ

Publications Selected Refereed Journal Articles

Li Y and Guo Q (2012) ldquoAngular Velocity Formula for Turbulent Vortex Chamber Flowsrdquo Journal of Hydraulic Engineering in press

Roseen R M Fowler G D Ballestero T P and Guo Q (2011) Sediment Monitoring Bias by Autosampler in Comparison with Whole Volume Sampling for Parking Lot Runoff Journal of Irrigation and Drainage Engineering Vol 137 No 4 pp 251-257

Huang S Ng C and Guo Q (2007) ldquoExperimental investigation of the effect of flow turbulence and sediment transport on the adsorption of cadmium ions onto sediment particlesrdquo Journal of Environmental Sciences Vol 19 pp 696-703

31

Appendix E Guo CV Qizhong (George) Guo

Associate Professor Civil and Environmental Engineering qguorcirutgersedu

732-445-4444 CEB 109

Research and Specialty Areas

bull Hydraulics Hydrology and Water Quality bull Urban Stormwater Management bull Watershed and Water Environment Restoration bull Water Resources and Environmental Engineering

Current and Recent Research Projects Stormwater System Monitoring and Evaluation funded by NJDOT

Drainage System Identification and Analysis funded by NJDOT via NJIT

Automatic Vacuum Flushing Technology for Combined Sewer Solids funded by USEPA via WERF

High Volume Utilization of Fly Ash Containing Mercury-Impregnated Carbon funded by USDOE via Ceramatec Inc

Stormwater Management Rule Implementation Process funded by NJDOT

Correlation of Total Suspended Solids and Suspended Sediment Concentration Test Methods funded by NJDEP

Watershed Modeling to Support Post Brook Regional Stormwater Management Plan Development funded by USEPA-NJDEP via GSE LLC

Section 319H NPS Pollution Control and Management Implementation Grant Program Implementation of Stormwater Detention Basin Retrofitting Techniques funded by USEPA via NJDEP

Development of a Numerical Model to Assess the Impacts of Raw Water Quality on Conventional Drinking Water Treatment funded by NJDEP

Water Quality Modeling Study of Lower Maurice River Estuary funded by Commercial Township NJ

Publications Selected Refereed Journal Articles

Li Y and Guo Q (2012) ldquoAngular Velocity Formula for Turbulent Vortex Chamber Flowsrdquo Journal of Hydraulic Engineering in press

Roseen R M Fowler G D Ballestero T P and Guo Q (2011) Sediment Monitoring Bias by Autosampler in Comparison with Whole Volume Sampling for Parking Lot Runoff Journal of Irrigation and Drainage Engineering Vol 137 No 4 pp 251-257

Huang S Ng C and Guo Q (2007) ldquoExperimental investigation of the effect of flow turbulence and sediment transport on the adsorption of cadmium ions onto sediment particlesrdquo Journal of Environmental Sciences Vol 19 pp 696-703

32

Sankararamakrishnan N and Guo Q (2005) ldquoChemical Tracers as Indicators of Human Fecal Coliforms at Storm Water Outfallsrdquo Environment International Vol 31 No 8 pp 1133-1140

Guo Q Fan C-Y Raghavan R and Field R (2004) ldquoGate and Vacuum Flushing of Sewer Sediment Laboratory Testingrdquo Journal of Hydraulic Engineering ASCE Vol130 No 5 pp 463-466

Decker T and Guo Q (2003) ldquoSite Assessment for a Stormwater Detention Basin Water Quality Retrofitrdquo Stormwater The Journal for Surface Water Quality Professionals Vol 4 No 3 pp 56-66

Guo Q (2002) ldquoApplicability of a Criterion for Onset of River Ice Breakuprdquo Journal of Hydraulic Engineering ASCE Vol 128 No 11 pp 1023-1026

Guo Q and Lordi P (2000) ldquoMethod for Quantifying Freshwater Input and Flushing Time in Estuariesrdquo Journal of Environmental Engineering ASCE Vol 126 No 7 pp 675-683

Suk N S Guo Q and Psuty N P (1999) ldquoSuspended Solids Flux at Estuary-Marsh Boundary A Long-Term Continuous Measurementrdquo Estuarine Coastal and Shelf Science Vol 49 pp 61-81

Guo Q Toomuluri P and Eckert J O (1998) ldquoLeachability of Regulated Metals from Cement-Mortar Liningsrdquo Journal American Water Works Association Vol 90 No 3 pp 62-73

Suk N Guo Q and Psuty N P (1998) ldquoFeasibility of Using Turbidimeter to Quantify Suspended Solids Concentration in a Tidal Saltmarsh Creekrdquo Estuarine Coastal and Shelf Science Vol 46 pp 383-391

Guo Q (1997) ldquoIncreases of Lead and Chromium in Drinking Water from Using Cement-Mortar Lined Pipes Initial Modeling and Assessmentrdquo Journal of Hazardous Materials Vol 56 pp 181-213

Guo Q (1997) ldquoSediment and Heavy Metal Accumulation in Dry Stormwater Detention Basinrdquo Journal of Water Resources Planning and Management ASCE Vol 123 No 5 pp 295-301

Guo Q and Psuty N P (1997) ldquoFlood-Tide Deltaic Wetlands Detection of their Sequential Spatial Evolutionrdquo Photogrammetric Engineering and Remote Sensing Vol 63 No 3 273-280

Guo Q and Song C (1991) ldquoDropshaft hydrodynamics under transient conditionsrdquo Journal of Hydraulic Engineering ASCE Vol 117 No 8 pp 1042-1055

Guo Q and Song C (1990) ldquoSurging in urban storm drainage systemsrdquo Journal of Hydraulic Engineering ASCE Vol 116 No 12 pp 1523-1537

Selected Conference Proceedings

Guo Q and Kim J (2011) Information for Inspection and Maintenance of Stormwater Manufactured Treatment Devices Proceedings of the ASCEEWRI World Environmental amp Water Resources Congress Palm Springs California May 22-26

Kim J and Guo Q (2011) Maintenance Interval for Stormwater Hydrodynamic Separators Proceedings of the ASCEEWRI World Environmental amp Water Resources Congress Palm Springs California May 22-26

Guo Q and Kim J (2010) Quantity and Quality of Stormwater Solids Trapped by Hydrodynamic Separators at Highway Sites Proceedings of the ASCEEWRI World Environmental amp Water Resources Congress Providence Rhode Island May 16-20

Fernnandez V and Guo Q (2009) ldquoWater Quality Design Storm for Hydrodynamic Separatorsrdquo Proceedings of the ASCEEWRI World Environmental amp Water Resources Congress May 17-21 Kansas City MO

33

Gulliver J S Guo Q Sansalone J J Williams G and Wu J S (2009) ldquoProposed scaling relations for manufactured stormwater BMPrdquo Proceedings of the ASCEEWRI World Environmental amp Water Resources Congress May 17-21 Kansas City MO

Guo Q (2008) ldquoUrban Flood Management in USArdquo Proceedings of the Advanced Symposium on New Technologies for Urban Flood Forecast and Warning Dongguan China November 19-21

Guo Q England G Johnston C E (2008) ldquoDevelopment of Certification Guidelines for Manufactured Stormwater BMPsrdquo Proceedings of the ASCEEWRI World Environmental amp Water Resources Congress May 12-16 Honolulu Hawaii

Hunt W F England G DeBruijn H Gee R Guo Q Lord W Miller M Mosheni O and Perry S (2008) ldquoInspection and Maintenance Guidance for Manufactured BMPsrdquo Proceedings of the ASCEEWRI World Environmental amp Water Resources Congress May 15-19 2007 Tampa Florida

Guo Q (2007) ldquoEffect of Particle Size on Difference between TSS and SSC Measurementsrdquo Proceedings of the ASCEEWRI World Environmental amp Water Resources Congress May 15-19 2007 Tampa Florida

Guo Q (2007) ldquoPerformance of Retrofitted Stormwater Extended Detention Wetlandsrdquo Proceedings of the ASCEEWRI World Environmental amp Water Resources Congress May 15-19 2007 Tampa Florida

Guo Q (2004) ldquoDevelopment of Flow-Based TMDLs for Fecal Coliformrdquo Proceedings of the World Water amp Environmental Resources Congress June 27-July 1 2004 Salt Lake City Utah ASCE-EWRI

Marcoon K B and Guo Q (2004) ldquoDetention Basin Retrofit Optimization between Water Quality Improvement and Flood Controlrdquo Proceedings of the World Water amp Environmental Resources Congress June 27-July 1 2004 Salt Lake City Utah ASCE-EWRI

Guo Q (2004) ldquoReengineering of Urban Storm Water Management Facilitiesrdquo IDS-Water Americas 2004 May 10 ndash 28 Internet Conference httpwwwidswatercom 4p

Li C and Guo Q (2003) ldquoIntegrated Point and Nonpoint Source Modeling of Fecal Coliform in Whippany Watershed New Jerseyrdquo Proceedings of the World Water amp Environmental Resources Congress June 22-26 Philadelphia PA ASCE-EWRI

Guo Q (2002) ldquoVariability of Net Mass Transport through Estuary-Ocean and Estuary-March Boundariesrdquo Proceeding of the 2nd International Workshop on Coastal Eutrophication Tianjin China November 21-24

Guo Q Psuty N P Pinheiro C G Sankararamkrishnan N and Pace J P (2001) ldquoThe Coastal Ocean as Source as Well as Sink of Nitrogen to an Estuaryrdquo Proceedings of the 3rd International Symposium on Environmental Hydraulics Tempe AZ December 5-8 CD-ROM 6p

Guo Q Agnoli N W Zhang N P and Hayes B D (2000) ldquoHydraulic and Water Quality Performance of Urban Storm Water Detention Basin before and after Outlet Modificationrdquo Proceedings of the 2000 Joint Conference on Water Resources Engineering and Water Resources Planning and Management Minneapolis MN July 30-August 2 ASCE CD-ROM 10p

34

Selected Technical Reports and Other Publications

Guo Q and Kim (2010) Stormwater System Monitoring and Evaluation Final Report Submitted to New Jersey Department of Transportation Bureau of Research and Federal Highway Administration December

Guo Q Nowicki P Stencel J R Rajagopal R Ulrich R Skupien J J and Maguire D (2007) Stormwater Management Rule Implementation Process Final Report FHWA-NJ-2007-023 Submitted to New Jersey Department of Transportation Bureau of Research and Federal Highway Administration December

Guo Q (2007) ldquoHow We Can Combat Floodingrdquo Op-Ed Article The Record April 30 Edition

Guo Q (2006) Correlation of Total Suspended Solids (TSS) and Suspended Sediment Concentration (SSC) Test Methods Final Report Submitted to New Jersey Department of Environmental Protection November

Guo Q (2006) Watershed Modeling to Support Post Brook Stormwater Management Plan Development Submitted to New Jersey Department of Environmental Protection September

Guo Q (2006) Implementation of Stormwater Detention Basin Retrofitting Techniques Final Report Submitted to New Jersey Department of Environmental Protection August

Guo Q Psuty N P Lordi G P Glenn S Mund M and Gastrich M D (2004) Hydrographic Study of Barnegat Bay Research Project Summary New Jersey Department of Environmental Protection Division of Science Research and Technology Trenton NJ July 4p Available from web site httpwwwstatenjusdepdsrresearchhydrographicpdf

Guo Q Wu H and Sankararamakrishnan N (2003) Development of a Numerical Model for Assessing the Impact of Raw Water Quality on Conventional Drinking Water Treatment Final Report Submitted to New Jersey Department of Environmental Protection Trenton NJ December 141p

Guo Q Kumaraswamy A and Saha S K (2002) Water Quality Modeling Study of Lower Maurice River Estuary Final Report Submitted to Township of Commercial Port Norris NJ January 121p

Guo Q Saha S K and Pace J (2001) Sewer and Tank Sediment Flushing Hydraulic Modeling Studies of Standpipe Vacuum Flushing Device Final Report Submitted to US Environmental Protection Agency Edison New Jersey November 46p

Guo Q and Psuty N P (2000) ldquoNitrogen Flux through Barnegat Inlet The Ocean as Source as well as Sinkrdquo The Jersey ShoreLine New Jersey Sea Grant College Program August Volume 19 Issue 4 pp 22-26

Guo Q Psuty N P Bongiovanni J (2000) Hydrographic Study of Barnegat Bay Year 3 Numerical Modeling Final Report Dept of Civil and Environmental Engineering Rutgers University Piscataway NJ May 252p

Guo Q Suk NS Ahn C H and Ho S H (1999) Water Quality Modeling Study of Maurice River Estuary Final Report Department of Civil and Environmental Engineering Rutgers University Piscataway NJ September Vol I 241p Vol II 138p

Guo Q Agnoli N Zhang N P and Hayes (1999) Retrofitting Stormwater Detention Basins Water Quality Performance Before and After Final Report Dept of Civil and Environmental Engineering Rutgers University Piscataway NJ September 131p

35

Guo Q Psuty N P Lordi G P and Tsai C S (1998) Hydrographic Study of Barnegat Bay Year 2 Data Analysis and Additional Data Collection Final Report Dept of Civil and Environmental Engineering Rutgers University Piscataway NJ December Vol I 81p Vol II 588p

Guo Q Psuty N P Lordi G Glenn S and Mund M (1995) Hydrographic Study of Barnegat Bay Year 1 Final Report Rutgers University New Brunswick NJ September Vol I138p Vol II 157p

Selected Service

Service to University College and Department

Member University Committee on Environmental Affairs Member New Brunswick Campus Faculty Advisory Council Rutgers University Advisor The Hubert H Humphrey Fellowship Program Edward J Bloustein School of Planning and Public Policy Courses of Study Committee School of Engineering Chair and Member Library Committee School of Engineering Chair and Member Director Rutgers Hydroinformatics Laboratory Department of Civil and Environmental Engineering Director Fluid MechanicsHydraulics Laboratory Department of Civil and Environmental Engineering

Service to Profession

Chair National Task Committee on Guidelines for Certification of Stormwater Manufactured Devices ASCEEWRI President Chinese American Water Resources Association Project Steering Committee Member Stormwater-Borne Solids Water Environment Research Foundation Peer Review Panelist Water Infrastructure Research Proposals US Environmental Protection Agency Office of Research and Development Honorary Theme Editor Encyclopedia of Life Support Systems (Sustainable Development) UNESCO Service to Society Member Advisory Council NJ Water Resources Research Institute Member Stream Corridors Subcommittee Delaware River Basin Commission Member Scientific and Technical Advisory Committee National Estuary Program for Barnegat Bay US Environmental Protection Agency Member Technical Advisory Committee Whippany Watershed Pilot Project New Jersey Department of Environmental Protection Interviewed by local and national media on water resources technical issues (MSNBC The New York Times Sand Francisco Chronicle The Star Ledger NJN News The Record Atlantic City Press Wired Magazine Omaha Herald etc)

Courses Taught at Rutgers bull 180331 Elements of Environmental Engineering bull 180387 Fluid Mechanics bull 180431 Design of Environmental Engineering Facilities bull 180563 Advanced Hydrology bull 180566 Sediment Transport bull 180567 Analysis of Receiving Water Quality bull 180568 Thermal Effects on Receiving Waters bull 180590 Coastal Engineering bull 180611 Advanced Topics in Environmental Engineering (Watershed Modeling)

36

Appendix F Robinson CV David A Robinson

Professor amp New Jersey State Climatologist

Department of Geography amp Office of the State Climatologist Rutgers University

54 Joyce Kilmer Ave Piscataway NJ 08854 voice 848-445-4741 fax 732-445-0006

email davidrobinsonrutgersedu web site httpclimaterutgersedu

EDUCATION PhD Geology 1984 Columbia University MS Geology 1981 Columbia University BS Geology 1977 Dickinson College Carlisle PA ACADEMIC POSITIONS 1995- Professor Department of Geography Rutgers University 1991- New Jersey State Climatologist 1996-2003 2005-2011 Chairman Dept Geography Rutgers 1994-98 01 Director Graduate Program Department of Geography Rutgers 1988-95 Assistant (lsquo88-lsquo91) amp Associate (lsquo91-lsquo95) Professor Dept Geography Rutgers 1988 Visiting Scientist United States National Climatic Data Center 1984-1988 Assoc Research Scientist Lamont-Doherty Earth Observatory Columbia U RESEARCH INTERESTS My primary research interests are in climate and climate change in particular state and regional climate and climate change issues hemispheric and regional snow cover dynamics interactions of snow cover with other climate elements the dynamics of solar and terrestrial radiative fluxes at and close to the surface of the earth and the collection and archiving of accurate climatic data RESEARCH GRANTS To date served as the principal investigator co-principal investigator or sponsor on 70 research grants totaling approximately $103 million Some active grants include

2011-2013 The NJ Mesonet Ongoing Operations And Maintenance NJ Department Of Environmental Protection 2011-2012 NOAA National Weather Service National Mesonet Expansion Global Science and Technology 2011-2012 Updates to Snow Cover Extent and Snow Depth Products NOAA 2011-2012 NJFireNet continuation USDA Forest Service 2010-2014 US Freshwater Resources in the Coming Decades an Integrated Climate-Hydrologic Modeling Study Environmental Protection Agency (Y Fan PI) 2008-2013 Development of Northern Hemisphere Snow Climate Data Records National Aeronautics and Space Administration (NASA) 2008-2012 Development of an Integrated Northern Hemisphere Snow and Ice Operational Climate Data Record National Oceanic and Atmospheric Administration (NOAA) PUBLICATIONS To date author or co-author of 90 refereed articles 16 book chapters 93 non-refereed articles and technical reports and 6 panel reports Some recent publications include Frei A M Tedesco S Lee J Foster DK Hall R Kelly amp DA Robinson (in press) A review of

current-generation satellite-based snow products Advances in Space Research

33

Gulliver J S Guo Q Sansalone J J Williams G and Wu J S (2009) ldquoProposed scaling relations for manufactured stormwater BMPrdquo Proceedings of the ASCEEWRI World Environmental amp Water Resources Congress May 17-21 Kansas City MO

Guo Q (2008) ldquoUrban Flood Management in USArdquo Proceedings of the Advanced Symposium on New Technologies for Urban Flood Forecast and Warning Dongguan China November 19-21

Guo Q England G Johnston C E (2008) ldquoDevelopment of Certification Guidelines for Manufactured Stormwater BMPsrdquo Proceedings of the ASCEEWRI World Environmental amp Water Resources Congress May 12-16 Honolulu Hawaii

Hunt W F England G DeBruijn H Gee R Guo Q Lord W Miller M Mosheni O and Perry S (2008) ldquoInspection and Maintenance Guidance for Manufactured BMPsrdquo Proceedings of the ASCEEWRI World Environmental amp Water Resources Congress May 15-19 2007 Tampa Florida

Guo Q (2007) ldquoEffect of Particle Size on Difference between TSS and SSC Measurementsrdquo Proceedings of the ASCEEWRI World Environmental amp Water Resources Congress May 15-19 2007 Tampa Florida

Guo Q (2007) ldquoPerformance of Retrofitted Stormwater Extended Detention Wetlandsrdquo Proceedings of the ASCEEWRI World Environmental amp Water Resources Congress May 15-19 2007 Tampa Florida

Guo Q (2004) ldquoDevelopment of Flow-Based TMDLs for Fecal Coliformrdquo Proceedings of the World Water amp Environmental Resources Congress June 27-July 1 2004 Salt Lake City Utah ASCE-EWRI

Marcoon K B and Guo Q (2004) ldquoDetention Basin Retrofit Optimization between Water Quality Improvement and Flood Controlrdquo Proceedings of the World Water amp Environmental Resources Congress June 27-July 1 2004 Salt Lake City Utah ASCE-EWRI

Guo Q (2004) ldquoReengineering of Urban Storm Water Management Facilitiesrdquo IDS-Water Americas 2004 May 10 ndash 28 Internet Conference httpwwwidswatercom 4p

Li C and Guo Q (2003) ldquoIntegrated Point and Nonpoint Source Modeling of Fecal Coliform in Whippany Watershed New Jerseyrdquo Proceedings of the World Water amp Environmental Resources Congress June 22-26 Philadelphia PA ASCE-EWRI

Guo Q (2002) ldquoVariability of Net Mass Transport through Estuary-Ocean and Estuary-March Boundariesrdquo Proceeding of the 2nd International Workshop on Coastal Eutrophication Tianjin China November 21-24

Guo Q Psuty N P Pinheiro C G Sankararamkrishnan N and Pace J P (2001) ldquoThe Coastal Ocean as Source as Well as Sink of Nitrogen to an Estuaryrdquo Proceedings of the 3rd International Symposium on Environmental Hydraulics Tempe AZ December 5-8 CD-ROM 6p

Guo Q Agnoli N W Zhang N P and Hayes B D (2000) ldquoHydraulic and Water Quality Performance of Urban Storm Water Detention Basin before and after Outlet Modificationrdquo Proceedings of the 2000 Joint Conference on Water Resources Engineering and Water Resources Planning and Management Minneapolis MN July 30-August 2 ASCE CD-ROM 10p

34

Selected Technical Reports and Other Publications

Guo Q and Kim (2010) Stormwater System Monitoring and Evaluation Final Report Submitted to New Jersey Department of Transportation Bureau of Research and Federal Highway Administration December

Guo Q Nowicki P Stencel J R Rajagopal R Ulrich R Skupien J J and Maguire D (2007) Stormwater Management Rule Implementation Process Final Report FHWA-NJ-2007-023 Submitted to New Jersey Department of Transportation Bureau of Research and Federal Highway Administration December

Guo Q (2007) ldquoHow We Can Combat Floodingrdquo Op-Ed Article The Record April 30 Edition

Guo Q (2006) Correlation of Total Suspended Solids (TSS) and Suspended Sediment Concentration (SSC) Test Methods Final Report Submitted to New Jersey Department of Environmental Protection November

Guo Q (2006) Watershed Modeling to Support Post Brook Stormwater Management Plan Development Submitted to New Jersey Department of Environmental Protection September

Guo Q (2006) Implementation of Stormwater Detention Basin Retrofitting Techniques Final Report Submitted to New Jersey Department of Environmental Protection August

Guo Q Psuty N P Lordi G P Glenn S Mund M and Gastrich M D (2004) Hydrographic Study of Barnegat Bay Research Project Summary New Jersey Department of Environmental Protection Division of Science Research and Technology Trenton NJ July 4p Available from web site httpwwwstatenjusdepdsrresearchhydrographicpdf

Guo Q Wu H and Sankararamakrishnan N (2003) Development of a Numerical Model for Assessing the Impact of Raw Water Quality on Conventional Drinking Water Treatment Final Report Submitted to New Jersey Department of Environmental Protection Trenton NJ December 141p

Guo Q Kumaraswamy A and Saha S K (2002) Water Quality Modeling Study of Lower Maurice River Estuary Final Report Submitted to Township of Commercial Port Norris NJ January 121p

Guo Q Saha S K and Pace J (2001) Sewer and Tank Sediment Flushing Hydraulic Modeling Studies of Standpipe Vacuum Flushing Device Final Report Submitted to US Environmental Protection Agency Edison New Jersey November 46p

Guo Q and Psuty N P (2000) ldquoNitrogen Flux through Barnegat Inlet The Ocean as Source as well as Sinkrdquo The Jersey ShoreLine New Jersey Sea Grant College Program August Volume 19 Issue 4 pp 22-26

Guo Q Psuty N P Bongiovanni J (2000) Hydrographic Study of Barnegat Bay Year 3 Numerical Modeling Final Report Dept of Civil and Environmental Engineering Rutgers University Piscataway NJ May 252p

Guo Q Suk NS Ahn C H and Ho S H (1999) Water Quality Modeling Study of Maurice River Estuary Final Report Department of Civil and Environmental Engineering Rutgers University Piscataway NJ September Vol I 241p Vol II 138p

Guo Q Agnoli N Zhang N P and Hayes (1999) Retrofitting Stormwater Detention Basins Water Quality Performance Before and After Final Report Dept of Civil and Environmental Engineering Rutgers University Piscataway NJ September 131p

35

Guo Q Psuty N P Lordi G P and Tsai C S (1998) Hydrographic Study of Barnegat Bay Year 2 Data Analysis and Additional Data Collection Final Report Dept of Civil and Environmental Engineering Rutgers University Piscataway NJ December Vol I 81p Vol II 588p

Guo Q Psuty N P Lordi G Glenn S and Mund M (1995) Hydrographic Study of Barnegat Bay Year 1 Final Report Rutgers University New Brunswick NJ September Vol I138p Vol II 157p

Selected Service

Service to University College and Department

Member University Committee on Environmental Affairs Member New Brunswick Campus Faculty Advisory Council Rutgers University Advisor The Hubert H Humphrey Fellowship Program Edward J Bloustein School of Planning and Public Policy Courses of Study Committee School of Engineering Chair and Member Library Committee School of Engineering Chair and Member Director Rutgers Hydroinformatics Laboratory Department of Civil and Environmental Engineering Director Fluid MechanicsHydraulics Laboratory Department of Civil and Environmental Engineering

Service to Profession

Chair National Task Committee on Guidelines for Certification of Stormwater Manufactured Devices ASCEEWRI President Chinese American Water Resources Association Project Steering Committee Member Stormwater-Borne Solids Water Environment Research Foundation Peer Review Panelist Water Infrastructure Research Proposals US Environmental Protection Agency Office of Research and Development Honorary Theme Editor Encyclopedia of Life Support Systems (Sustainable Development) UNESCO Service to Society Member Advisory Council NJ Water Resources Research Institute Member Stream Corridors Subcommittee Delaware River Basin Commission Member Scientific and Technical Advisory Committee National Estuary Program for Barnegat Bay US Environmental Protection Agency Member Technical Advisory Committee Whippany Watershed Pilot Project New Jersey Department of Environmental Protection Interviewed by local and national media on water resources technical issues (MSNBC The New York Times Sand Francisco Chronicle The Star Ledger NJN News The Record Atlantic City Press Wired Magazine Omaha Herald etc)

Courses Taught at Rutgers bull 180331 Elements of Environmental Engineering bull 180387 Fluid Mechanics bull 180431 Design of Environmental Engineering Facilities bull 180563 Advanced Hydrology bull 180566 Sediment Transport bull 180567 Analysis of Receiving Water Quality bull 180568 Thermal Effects on Receiving Waters bull 180590 Coastal Engineering bull 180611 Advanced Topics in Environmental Engineering (Watershed Modeling)

36

Appendix F Robinson CV David A Robinson

Professor amp New Jersey State Climatologist

Department of Geography amp Office of the State Climatologist Rutgers University

54 Joyce Kilmer Ave Piscataway NJ 08854 voice 848-445-4741 fax 732-445-0006

email davidrobinsonrutgersedu web site httpclimaterutgersedu

EDUCATION PhD Geology 1984 Columbia University MS Geology 1981 Columbia University BS Geology 1977 Dickinson College Carlisle PA ACADEMIC POSITIONS 1995- Professor Department of Geography Rutgers University 1991- New Jersey State Climatologist 1996-2003 2005-2011 Chairman Dept Geography Rutgers 1994-98 01 Director Graduate Program Department of Geography Rutgers 1988-95 Assistant (lsquo88-lsquo91) amp Associate (lsquo91-lsquo95) Professor Dept Geography Rutgers 1988 Visiting Scientist United States National Climatic Data Center 1984-1988 Assoc Research Scientist Lamont-Doherty Earth Observatory Columbia U RESEARCH INTERESTS My primary research interests are in climate and climate change in particular state and regional climate and climate change issues hemispheric and regional snow cover dynamics interactions of snow cover with other climate elements the dynamics of solar and terrestrial radiative fluxes at and close to the surface of the earth and the collection and archiving of accurate climatic data RESEARCH GRANTS To date served as the principal investigator co-principal investigator or sponsor on 70 research grants totaling approximately $103 million Some active grants include

2011-2013 The NJ Mesonet Ongoing Operations And Maintenance NJ Department Of Environmental Protection 2011-2012 NOAA National Weather Service National Mesonet Expansion Global Science and Technology 2011-2012 Updates to Snow Cover Extent and Snow Depth Products NOAA 2011-2012 NJFireNet continuation USDA Forest Service 2010-2014 US Freshwater Resources in the Coming Decades an Integrated Climate-Hydrologic Modeling Study Environmental Protection Agency (Y Fan PI) 2008-2013 Development of Northern Hemisphere Snow Climate Data Records National Aeronautics and Space Administration (NASA) 2008-2012 Development of an Integrated Northern Hemisphere Snow and Ice Operational Climate Data Record National Oceanic and Atmospheric Administration (NOAA) PUBLICATIONS To date author or co-author of 90 refereed articles 16 book chapters 93 non-refereed articles and technical reports and 6 panel reports Some recent publications include Frei A M Tedesco S Lee J Foster DK Hall R Kelly amp DA Robinson (in press) A review of

current-generation satellite-based snow products Advances in Space Research

34

Selected Technical Reports and Other Publications

Guo Q and Kim (2010) Stormwater System Monitoring and Evaluation Final Report Submitted to New Jersey Department of Transportation Bureau of Research and Federal Highway Administration December

Guo Q Nowicki P Stencel J R Rajagopal R Ulrich R Skupien J J and Maguire D (2007) Stormwater Management Rule Implementation Process Final Report FHWA-NJ-2007-023 Submitted to New Jersey Department of Transportation Bureau of Research and Federal Highway Administration December

Guo Q (2007) ldquoHow We Can Combat Floodingrdquo Op-Ed Article The Record April 30 Edition

Guo Q (2006) Correlation of Total Suspended Solids (TSS) and Suspended Sediment Concentration (SSC) Test Methods Final Report Submitted to New Jersey Department of Environmental Protection November

Guo Q (2006) Watershed Modeling to Support Post Brook Stormwater Management Plan Development Submitted to New Jersey Department of Environmental Protection September

Guo Q (2006) Implementation of Stormwater Detention Basin Retrofitting Techniques Final Report Submitted to New Jersey Department of Environmental Protection August

Guo Q Psuty N P Lordi G P Glenn S Mund M and Gastrich M D (2004) Hydrographic Study of Barnegat Bay Research Project Summary New Jersey Department of Environmental Protection Division of Science Research and Technology Trenton NJ July 4p Available from web site httpwwwstatenjusdepdsrresearchhydrographicpdf

Guo Q Wu H and Sankararamakrishnan N (2003) Development of a Numerical Model for Assessing the Impact of Raw Water Quality on Conventional Drinking Water Treatment Final Report Submitted to New Jersey Department of Environmental Protection Trenton NJ December 141p

Guo Q Kumaraswamy A and Saha S K (2002) Water Quality Modeling Study of Lower Maurice River Estuary Final Report Submitted to Township of Commercial Port Norris NJ January 121p

Guo Q Saha S K and Pace J (2001) Sewer and Tank Sediment Flushing Hydraulic Modeling Studies of Standpipe Vacuum Flushing Device Final Report Submitted to US Environmental Protection Agency Edison New Jersey November 46p

Guo Q and Psuty N P (2000) ldquoNitrogen Flux through Barnegat Inlet The Ocean as Source as well as Sinkrdquo The Jersey ShoreLine New Jersey Sea Grant College Program August Volume 19 Issue 4 pp 22-26

Guo Q Psuty N P Bongiovanni J (2000) Hydrographic Study of Barnegat Bay Year 3 Numerical Modeling Final Report Dept of Civil and Environmental Engineering Rutgers University Piscataway NJ May 252p

Guo Q Suk NS Ahn C H and Ho S H (1999) Water Quality Modeling Study of Maurice River Estuary Final Report Department of Civil and Environmental Engineering Rutgers University Piscataway NJ September Vol I 241p Vol II 138p

Guo Q Agnoli N Zhang N P and Hayes (1999) Retrofitting Stormwater Detention Basins Water Quality Performance Before and After Final Report Dept of Civil and Environmental Engineering Rutgers University Piscataway NJ September 131p

35

Guo Q Psuty N P Lordi G P and Tsai C S (1998) Hydrographic Study of Barnegat Bay Year 2 Data Analysis and Additional Data Collection Final Report Dept of Civil and Environmental Engineering Rutgers University Piscataway NJ December Vol I 81p Vol II 588p

Guo Q Psuty N P Lordi G Glenn S and Mund M (1995) Hydrographic Study of Barnegat Bay Year 1 Final Report Rutgers University New Brunswick NJ September Vol I138p Vol II 157p

Selected Service

Service to University College and Department

Member University Committee on Environmental Affairs Member New Brunswick Campus Faculty Advisory Council Rutgers University Advisor The Hubert H Humphrey Fellowship Program Edward J Bloustein School of Planning and Public Policy Courses of Study Committee School of Engineering Chair and Member Library Committee School of Engineering Chair and Member Director Rutgers Hydroinformatics Laboratory Department of Civil and Environmental Engineering Director Fluid MechanicsHydraulics Laboratory Department of Civil and Environmental Engineering

Service to Profession

Chair National Task Committee on Guidelines for Certification of Stormwater Manufactured Devices ASCEEWRI President Chinese American Water Resources Association Project Steering Committee Member Stormwater-Borne Solids Water Environment Research Foundation Peer Review Panelist Water Infrastructure Research Proposals US Environmental Protection Agency Office of Research and Development Honorary Theme Editor Encyclopedia of Life Support Systems (Sustainable Development) UNESCO Service to Society Member Advisory Council NJ Water Resources Research Institute Member Stream Corridors Subcommittee Delaware River Basin Commission Member Scientific and Technical Advisory Committee National Estuary Program for Barnegat Bay US Environmental Protection Agency Member Technical Advisory Committee Whippany Watershed Pilot Project New Jersey Department of Environmental Protection Interviewed by local and national media on water resources technical issues (MSNBC The New York Times Sand Francisco Chronicle The Star Ledger NJN News The Record Atlantic City Press Wired Magazine Omaha Herald etc)

Courses Taught at Rutgers bull 180331 Elements of Environmental Engineering bull 180387 Fluid Mechanics bull 180431 Design of Environmental Engineering Facilities bull 180563 Advanced Hydrology bull 180566 Sediment Transport bull 180567 Analysis of Receiving Water Quality bull 180568 Thermal Effects on Receiving Waters bull 180590 Coastal Engineering bull 180611 Advanced Topics in Environmental Engineering (Watershed Modeling)

36

Appendix F Robinson CV David A Robinson

Professor amp New Jersey State Climatologist

Department of Geography amp Office of the State Climatologist Rutgers University

54 Joyce Kilmer Ave Piscataway NJ 08854 voice 848-445-4741 fax 732-445-0006

email davidrobinsonrutgersedu web site httpclimaterutgersedu

EDUCATION PhD Geology 1984 Columbia University MS Geology 1981 Columbia University BS Geology 1977 Dickinson College Carlisle PA ACADEMIC POSITIONS 1995- Professor Department of Geography Rutgers University 1991- New Jersey State Climatologist 1996-2003 2005-2011 Chairman Dept Geography Rutgers 1994-98 01 Director Graduate Program Department of Geography Rutgers 1988-95 Assistant (lsquo88-lsquo91) amp Associate (lsquo91-lsquo95) Professor Dept Geography Rutgers 1988 Visiting Scientist United States National Climatic Data Center 1984-1988 Assoc Research Scientist Lamont-Doherty Earth Observatory Columbia U RESEARCH INTERESTS My primary research interests are in climate and climate change in particular state and regional climate and climate change issues hemispheric and regional snow cover dynamics interactions of snow cover with other climate elements the dynamics of solar and terrestrial radiative fluxes at and close to the surface of the earth and the collection and archiving of accurate climatic data RESEARCH GRANTS To date served as the principal investigator co-principal investigator or sponsor on 70 research grants totaling approximately $103 million Some active grants include

2011-2013 The NJ Mesonet Ongoing Operations And Maintenance NJ Department Of Environmental Protection 2011-2012 NOAA National Weather Service National Mesonet Expansion Global Science and Technology 2011-2012 Updates to Snow Cover Extent and Snow Depth Products NOAA 2011-2012 NJFireNet continuation USDA Forest Service 2010-2014 US Freshwater Resources in the Coming Decades an Integrated Climate-Hydrologic Modeling Study Environmental Protection Agency (Y Fan PI) 2008-2013 Development of Northern Hemisphere Snow Climate Data Records National Aeronautics and Space Administration (NASA) 2008-2012 Development of an Integrated Northern Hemisphere Snow and Ice Operational Climate Data Record National Oceanic and Atmospheric Administration (NOAA) PUBLICATIONS To date author or co-author of 90 refereed articles 16 book chapters 93 non-refereed articles and technical reports and 6 panel reports Some recent publications include Frei A M Tedesco S Lee J Foster DK Hall R Kelly amp DA Robinson (in press) A review of

current-generation satellite-based snow products Advances in Space Research

35

Guo Q Psuty N P Lordi G P and Tsai C S (1998) Hydrographic Study of Barnegat Bay Year 2 Data Analysis and Additional Data Collection Final Report Dept of Civil and Environmental Engineering Rutgers University Piscataway NJ December Vol I 81p Vol II 588p

Guo Q Psuty N P Lordi G Glenn S and Mund M (1995) Hydrographic Study of Barnegat Bay Year 1 Final Report Rutgers University New Brunswick NJ September Vol I138p Vol II 157p

Selected Service

Service to University College and Department

Member University Committee on Environmental Affairs Member New Brunswick Campus Faculty Advisory Council Rutgers University Advisor The Hubert H Humphrey Fellowship Program Edward J Bloustein School of Planning and Public Policy Courses of Study Committee School of Engineering Chair and Member Library Committee School of Engineering Chair and Member Director Rutgers Hydroinformatics Laboratory Department of Civil and Environmental Engineering Director Fluid MechanicsHydraulics Laboratory Department of Civil and Environmental Engineering

Service to Profession

Chair National Task Committee on Guidelines for Certification of Stormwater Manufactured Devices ASCEEWRI President Chinese American Water Resources Association Project Steering Committee Member Stormwater-Borne Solids Water Environment Research Foundation Peer Review Panelist Water Infrastructure Research Proposals US Environmental Protection Agency Office of Research and Development Honorary Theme Editor Encyclopedia of Life Support Systems (Sustainable Development) UNESCO Service to Society Member Advisory Council NJ Water Resources Research Institute Member Stream Corridors Subcommittee Delaware River Basin Commission Member Scientific and Technical Advisory Committee National Estuary Program for Barnegat Bay US Environmental Protection Agency Member Technical Advisory Committee Whippany Watershed Pilot Project New Jersey Department of Environmental Protection Interviewed by local and national media on water resources technical issues (MSNBC The New York Times Sand Francisco Chronicle The Star Ledger NJN News The Record Atlantic City Press Wired Magazine Omaha Herald etc)

Courses Taught at Rutgers bull 180331 Elements of Environmental Engineering bull 180387 Fluid Mechanics bull 180431 Design of Environmental Engineering Facilities bull 180563 Advanced Hydrology bull 180566 Sediment Transport bull 180567 Analysis of Receiving Water Quality bull 180568 Thermal Effects on Receiving Waters bull 180590 Coastal Engineering bull 180611 Advanced Topics in Environmental Engineering (Watershed Modeling)

36

Appendix F Robinson CV David A Robinson

Professor amp New Jersey State Climatologist

Department of Geography amp Office of the State Climatologist Rutgers University

54 Joyce Kilmer Ave Piscataway NJ 08854 voice 848-445-4741 fax 732-445-0006

email davidrobinsonrutgersedu web site httpclimaterutgersedu

EDUCATION PhD Geology 1984 Columbia University MS Geology 1981 Columbia University BS Geology 1977 Dickinson College Carlisle PA ACADEMIC POSITIONS 1995- Professor Department of Geography Rutgers University 1991- New Jersey State Climatologist 1996-2003 2005-2011 Chairman Dept Geography Rutgers 1994-98 01 Director Graduate Program Department of Geography Rutgers 1988-95 Assistant (lsquo88-lsquo91) amp Associate (lsquo91-lsquo95) Professor Dept Geography Rutgers 1988 Visiting Scientist United States National Climatic Data Center 1984-1988 Assoc Research Scientist Lamont-Doherty Earth Observatory Columbia U RESEARCH INTERESTS My primary research interests are in climate and climate change in particular state and regional climate and climate change issues hemispheric and regional snow cover dynamics interactions of snow cover with other climate elements the dynamics of solar and terrestrial radiative fluxes at and close to the surface of the earth and the collection and archiving of accurate climatic data RESEARCH GRANTS To date served as the principal investigator co-principal investigator or sponsor on 70 research grants totaling approximately $103 million Some active grants include

2011-2013 The NJ Mesonet Ongoing Operations And Maintenance NJ Department Of Environmental Protection 2011-2012 NOAA National Weather Service National Mesonet Expansion Global Science and Technology 2011-2012 Updates to Snow Cover Extent and Snow Depth Products NOAA 2011-2012 NJFireNet continuation USDA Forest Service 2010-2014 US Freshwater Resources in the Coming Decades an Integrated Climate-Hydrologic Modeling Study Environmental Protection Agency (Y Fan PI) 2008-2013 Development of Northern Hemisphere Snow Climate Data Records National Aeronautics and Space Administration (NASA) 2008-2012 Development of an Integrated Northern Hemisphere Snow and Ice Operational Climate Data Record National Oceanic and Atmospheric Administration (NOAA) PUBLICATIONS To date author or co-author of 90 refereed articles 16 book chapters 93 non-refereed articles and technical reports and 6 panel reports Some recent publications include Frei A M Tedesco S Lee J Foster DK Hall R Kelly amp DA Robinson (in press) A review of

current-generation satellite-based snow products Advances in Space Research

36

Appendix F Robinson CV David A Robinson

Professor amp New Jersey State Climatologist

Department of Geography amp Office of the State Climatologist Rutgers University

54 Joyce Kilmer Ave Piscataway NJ 08854 voice 848-445-4741 fax 732-445-0006

email davidrobinsonrutgersedu web site httpclimaterutgersedu

EDUCATION PhD Geology 1984 Columbia University MS Geology 1981 Columbia University BS Geology 1977 Dickinson College Carlisle PA ACADEMIC POSITIONS 1995- Professor Department of Geography Rutgers University 1991- New Jersey State Climatologist 1996-2003 2005-2011 Chairman Dept Geography Rutgers 1994-98 01 Director Graduate Program Department of Geography Rutgers 1988-95 Assistant (lsquo88-lsquo91) amp Associate (lsquo91-lsquo95) Professor Dept Geography Rutgers 1988 Visiting Scientist United States National Climatic Data Center 1984-1988 Assoc Research Scientist Lamont-Doherty Earth Observatory Columbia U RESEARCH INTERESTS My primary research interests are in climate and climate change in particular state and regional climate and climate change issues hemispheric and regional snow cover dynamics interactions of snow cover with other climate elements the dynamics of solar and terrestrial radiative fluxes at and close to the surface of the earth and the collection and archiving of accurate climatic data RESEARCH GRANTS To date served as the principal investigator co-principal investigator or sponsor on 70 research grants totaling approximately $103 million Some active grants include

2011-2013 The NJ Mesonet Ongoing Operations And Maintenance NJ Department Of Environmental Protection 2011-2012 NOAA National Weather Service National Mesonet Expansion Global Science and Technology 2011-2012 Updates to Snow Cover Extent and Snow Depth Products NOAA 2011-2012 NJFireNet continuation USDA Forest Service 2010-2014 US Freshwater Resources in the Coming Decades an Integrated Climate-Hydrologic Modeling Study Environmental Protection Agency (Y Fan PI) 2008-2013 Development of Northern Hemisphere Snow Climate Data Records National Aeronautics and Space Administration (NASA) 2008-2012 Development of an Integrated Northern Hemisphere Snow and Ice Operational Climate Data Record National Oceanic and Atmospheric Administration (NOAA) PUBLICATIONS To date author or co-author of 90 refereed articles 16 book chapters 93 non-refereed articles and technical reports and 6 panel reports Some recent publications include Frei A M Tedesco S Lee J Foster DK Hall R Kelly amp DA Robinson (in press) A review of

current-generation satellite-based snow products Advances in Space Research

37

Decker SG amp DA Robinson (in press) Unexpected high winds in northern New Jersey A downslope windstorm in the presence of modest topography Weather and Forecasting

Brown RD amp DA Robinson 2011 Northern Hemisphere spring snow cover variability and change over 1922-2010 including an assessment of uncertainty The Cryosphere 5 219ndash229 wwwthe-cryospherenet52192011doi105194tc-5-219-2011

Ghatak D A Frei G Gong J Stroeve amp D Robinson (2010) On the emergence of an Arctic amplification signal in terrestrial Arctic snow extent Journal of Geophysical Research ndash Atmospheres 115 D24105 doi1010292010JD014007

Davi N G Jacoby K Fang J LI R DArrigo N Baatarbileg amp D Robinson (2010) Reconstructing drought variability for Mongolia based on a large-scale tree-ring network 1520-1993 Journal of Geophysical Research - Atmospheres 115 D22103 doi1010292010JD013907

DeAngelis A F Dominguez Y Fan A Robock MD Kustu amp D Robinson (2010) Evidence of enhanced precipitation due to irrigation over the Great Plains of the United States Journal of Geophysical Research 115 D15115 doi1010292010JD013892

Choi G DA Robinson amp S Kang (2010) Changing Northern Hemisphere snow seasons Journal of Climate 23 5305-5310

Kunkel KE M Palecki L Ensor D Easterling KG Hubbard D Robinson amp K Redmond (2009) Trends in 20th century US extreme snowfall seasons using a quality-controlled data set Journal of Climate 22 6204-6216

Yang D Y Zhao R Armstrong amp DA Robinson (2009) Yukon river streamflow response to seasonal snowcover changes Hydrological Processes 23 109-121

Kunkel KE L Ensor M Palecki D Easterling D Robinson KG Hubbard amp K Redmond (2009) A new look at lake-effect snowfall trends in the Laurentian Great Lakes using a temporally homogenous data set Journal of Great Lakes Research 35 23-29

Kunkel KE M Palecki L Ensor KG Hubbard DA Robinson K Redmond amp D Easterling (2009) Trends in twentieth-century US snowfall using a quality-controlled data set Journal of Atmospheric and Oceanic Technology 26 33-44

Foster J D Hall DA Robinson amp T Estilow (2008) Spring snow melt timing and changes over Arctic lands Polar Geography 31 145-157

Ye H D Yang amp DA Robinson (2008) Winter rain on snow and its association with air temperature in northern Eurasia Hydrological Processes 22 2728-2736

Kunkel KE MA Palecki KG Hubbard DA Robinson KT Redmond amp DR Easterling (2007) Trend identification in twentieth-century US snowfall the challenges Journal of Atmospheric and Oceanic Technology 24 64-73

PROFESSIONAL ACTIVITY Appointedelected positions (currently active and selected positions since 2000)

2011- Member National Academy of Sciences Board on Atmospheric Sciences and Climate 2010- Member Climate Adaptation Working Group Sustainable Jersey and the NJDEP 2010- Member Climate and Atmospheric Sciences Standing Committee NJDEP 2007- Member Learning and Teaching Advisory Committee Liberty Science Center 2006- Member NOAA Science Advisory Board Climate Working Group 2006- Member National Integrated Drought Information System Implementation Team

2002- Executive committee American Association of State Climatologists (past president) 2001- Advisory Board National Weather Service StormReady Community Program 1999- New Jersey Department of Environmental Protection Internal Drought Task Force 1999- Contributor Intergovernmental Panel on Climate Change

2009-2011 Member National Academy of Sciences Climate Research Committee 2006-2008 Chair National Research Council (NRC) Committee Archiving and Accessing

Environmental and Geospatial Data at NOAA 2004-2008 Member AMS Applied Climatology Committee 2003-2005 Chair NRC Committee Climate Data Records from Operational Satellite 1997-2000 Chair AMS Committee on Polar Meteorology and Oceanography RECENT AWARDS 2008 ldquoEnvironmental Herordquo National Oceanic and Atmospheric Administration 2008 Lifetime Achievement Award Climate Specialty Group Association of American Geographers 2012 Fellow American Meteorological Society (forthcoming in January)

• Introduction
• Task One Community Assessment
• Task Two Economic Modeling
• Task Three Ideas on Risk Assessment for Flooding on the Raritan
• Appendix A News Snapshots
• Appendix B-1 Safety UCR
• Appendix B-2 Safety UCR Detail
• Appendix C Maps
• Appendix D Literature
• Appendix E Guo CV
• Appendix F Robinson CV