Capital Region Particulate Matter Air Modelling Assessment … · 2016-03-16 · Capital Region...

Capital Region Particulate Matter Air Modelling Assessment

Final Report

Prepared for:

David Lyder Environment and Sustainable

Resource Development Policy Division

9th Floor, Oxbridge Place 9820 – 106 Street

Edmonton, Alberta Canada T5K 2J6

Prepared by:

Uarporn Nopmongcol, Jaegun Jung, Justin Zagunis,

Tejas Shah, and Ralph Morris ENVIRON International Corporation

773 San Marin Drive, Suite 2115 Novato, California, 94945

and Ted Pollock and

William Allan ENVIRON (EC) Canada Inc.

7070 Mississauga Road, Suite 140 Mississauga, ON, LFN 7G2

www.environcorp.com F-415-899-0707

and Xin Qiu, Nick Walters, and Fuquan Yang

Novus Environmental 150 Research Lane, Suite 105

Guelph, ON N1G 4T2

March 2014

i

CONTENTS

1.0 INTRODUCTION ............................................................................................................. 1

1.1 Background ...................................................................................................................... 1

1.2 Overview of Approach ..................................................................................................... 2

1.2.1 Task 1: Base Case Emission Inputs ........................................................................ 4

1.2.2 Task 2: Meteorological Inputs............................................................................... 5

1.2.3 Task 3: Photochemical Grid Modelling and Model Performance Evaluation ............................................................................................................. 5

1.2.4 Task 4: Conduct Scenarios .................................................................................... 6

1.3 Report Organization ........................................................................................................ 6

2.0 UNDERSTANDING PM ISSUES IN THE CAPITAL REGION ................................................... 7

2.1 Measurement Data.......................................................................................................... 7

2.2 Particulate Matter Measurement Technologies ............................................................. 8

2.2.1 Correlations of PM2.5 Measurement Technologies ............................................. 11

2.3 PM2.5 Problem in the Capital Region ............................................................................. 14

2.4 Modelling Winter Secondary PM2.5 ............................................................................... 16

2.4.1 Conversion of NOx to HNO3 ................................................................................ 16

2.4.2 Availability of NH3 Concentrations to form Particulate NO3 .............................. 18

2.4.3 Equilibrium between Particulate NO3 and Gaseous HNO3 ................................. 18

3.0 BASE CASE EMISSIONS INPUT ....................................................................................... 19

3.1 Industrial Point Sources ................................................................................................. 19

3.1.1 Data Consolidation .............................................................................................. 20

3.1.2 Updates to ESRD 2008 Industrial Survey ............................................................ 21

3.1.3 Inventory Harmonization .................................................................................... 23

3.1.4 Province-Wide Inventory Finalization ................................................................. 26

3.2 Non-Point Sources ......................................................................................................... 28

3.2.1 Comparison of Available Road Networks and Emissions ................................... 28

3.2.2 Combining CALMOB6 with Province Wide Transportation Emissions ............... 31

3.3 Natural Sources ............................................................................................................. 38

3.3.1 Biogenic Emissions .............................................................................................. 38

3.3.2 Fire Emissions ...................................................................................................... 39

3.4 SMOKE Emissions Modeling .......................................................................................... 39

ii

3.4.1 Spatial Surrogates Development ........................................................................ 40

3.4.2 SMOKE Processing and Merging ......................................................................... 45

3.4.3 Emissions Summaries .......................................................................................... 46

4.0 METEOROLOGICAL INPUTS........................................................................................... 48

4.1 36/12/4/1.33 km WRF Nesting Modelling .................................................................... 52

4.2 4 km WRF Model Evaluation ......................................................................................... 55

4.3 4 km WRF Physics Options Tests ................................................................................... 56

4.4 WRF Sensitivity Tests for 1.33 km Domain .................................................................... 57

4.5 Vertical Temperature and Wind Profiles Comparison .................................................. 61

4.6 Summary ........................................................................................................................ 70

5.0 PHOTOCHEMICAL GRID MODELLING AND MODEL PERFORMANCE EVALUATION................................................................................................................ 71

5.1 CMAQ model configuration ........................................................................................... 71

5.2 Model inputs.................................................................................................................. 72

5.2.1 Initial and Boundary Conditions (ICs/BCs) .......................................................... 72

5.2.2 Meteorology (MCIP) ........................................................................................... 72

5.2.3 Photolysis tables ................................................................................................. 72

5.3 CMAQ MODEL EVALUATION ......................................................................................... 73

5.3.1 CMAQ Model Evaluation Methodology .............................................................. 73

5.4 Diagnostic Tests and Sensitivity Analyses ..................................................................... 76

5.4.1 Modelling Episodes ............................................................................................. 76

5.4.2 CMAQ Diagnostic and Sensitivity Simulations .................................................... 77

5.4.3 Conclusion of Diagnostic Analyses ...................................................................... 99

5.5 Final Base Case .............................................................................................................. 99

6.0 SOURCE APPORTIONMENT MODELLING ..................................................................... 109

6.1 PM Source Apportionment .......................................................................................... 109

6.2 Zero-Out Source Apportionment Simulations............................................................. 109

6.3 CMAQ Zero-Out Results and Analyses ........................................................................ 111

6.4 Source Contributions at Monitoring Stations ............................................................. 120

7.0 SUMMARY AND RECOMMENDATIONS ....................................................................... 125

7.1 Development of Modelling Inputs .............................................................................. 125

7.1.1 Base Case Emissions Inputs .............................................................................. 125

iii

7.1.2 Meteorological Inputs ....................................................................................... 125

7.2 Diagnostic Evaluation .................................................................................................. 126

7.2.1 Sensitivity Test Results ...................................................................................... 126

7.2.2 Final Base Case .................................................................................................. 127

7.3 Source Apportionment Modelling Results .................................................................. 127

7.4 Uncertainties and Limitations ..................................................................................... 128

7.5 Conclusions and Recommendations ........................................................................... 129

8.0 REFERENECES ............................................................................................................. 132

APPENDICES Appendix A: Selection of PM2.5 Episodes for 2010 Sensitivity Test Modelling and

2008-2009 Modelling using the 1.33 km Capital Region Domain

Appendix B. Province-Wide Point Source Anthropogenic Emissions Summary Table

TABLES

Table 1-1 Definition of the Lambert Conformal Projection (LCP) 36/12/4/1.33 km domains used in the CMAQ photochemical and SMOKE emissions modelling of the Capital Region. ............................................................................. 2

Table 2-1. Capital Regional continuous PM2.5 monitoring sites. ............................................ 10

Table 3-1. Province-wide point-source emissions summary by inventory. ........................... 20

Table 3-2. Summary of stakeholder updates from 2008 Industrial Survey to 2010 for point sources. .................................................................................................. 21

Table 3-3. Capital Region point source anthropogenic emissions summary from updated province wide total. ................................................................................ 23

Table 3-4. NSRP2006 inventory noSurvey source divisions. .................................................. 25

Table 3-5. Typical January weekday total emissions comparison for Edmonton. ................. 30

Table 3-6. CALMOB6 vehicle types used in each temporal/spatial profile. ........................... 34

Table 3-7. Weekly traffic profiles summary. .......................................................................... 38

Table 3-8. Recommended Correction Factor (CF) (%) for Five Land Cover Types (source: Pace, 2005). ............................................................................................. 45

Table 3-9. Land use categories and transport factor. ............................................................ 45

Table 3-10. Summary of geographic regions and source sectors for SMOKE modelling. .............................................................................................................. 46

iv

Table 3-11. Emissions summary by pollutant and by source sector (tonne per month). .................................................................................................................. 47

Table 4-1. WRF model domain configurations. ...................................................................... 48

Table 4-2. Initial WRF physics options used in the Capital Region PM Modelling Study. ..................................................................................................................... 52

Table 4-3. Definition of WRF 40 vertical levels (39 vertical layers) and mapping to the 22 vertical layers used in the CMAQ Chemical Transport Model. Heights (m) are geopotential heights above ground level, actual layer thicknesses will be shallower in areas above. ....................................................... 54

Table 4-4. The 4 km WRF modeling evaluation Summary for temperature, wind speed and moisture content across 41 weather stations in the 4 km domain. .................................................................................................................. 55

Table 4-5. Differences in Novus recommended physics options and the 4 km WRF test run. ................................................................................................................. 56

Table 4-6. The 4 km WRF modeling performance evaluation comparison between the recommended WRF physics and ASHRAE-WRF physics in temperature and wind speed. ............................................................................... 57

Table 4-7. 1.33 km WRF modeling evaluation comparison among sensitivity tests a) to d) with 7 weather stations for temperature. ................................................ 59

Table 4-8. 1.33 km WRF modeling evaluation comparison among sensitivity tests a) to d) with 7 weather stations for wind speed. .................................................. 60

Table 5-1. CMAQ CTM model configuration. ......................................................................... 71

Table 5-2. Statistical model performance evaluation measure definitions. .......................... 75

Table 5-3. Model performance goals and criteria for PM. ..................................................... 75

Table 5-4. Maximum at any monitoring site and McIntyre daily PM2.5 concentrations in the 1st highest ranked 2010 episode for the Capital Region (Episode#2 – January 26 – February 4, 2010). .......................................... 77

Table 5-5. Maximum at any monitoring site and McIntyre daily PM2.5 concentrations in the 2nd highest ranked 2010 episode for the Capital Region (Episode#1 – January 17 – 21, 2010). ....................................................... 77

Table 5-6a. Total 24-hour PM2.5 Mass model performance metrics for the CMAQ WRF-CFSR sensitivity test and Episode#1. ............................................................ 79

Table 5-6b. Total 24-hour PM2.5 Mass model performance metrics for the CMAQ WRF-NARR sensitivity test and Episode#1. ........................................................... 79

Table 5-7a. Total 24-hour PM2.5 Mass model performance metrics for the CMAQ WRF-CFSR sensitivity test and Episode#2. ............................................................ 80

v

Table 5-7b. Total 24-hour PM2.5 Mass model performance metrics for the CMAQ WRF-NARR sensitivity test and Episode#2. ........................................................... 80

Table 5-8. CMAQ WRF-CFSR and WRF-NARR Episode#2 Fractional Bias and Error model performance for 24-hour speciated PM2.5 at the Edmonton McIntyre monitoring site. ..................................................................................... 81

Table 5-9. Hourly total PM2.5 mass model performance for CMAQ WRF-CFSR corrected emissions sensitivity test (Test#2), the 1.33 km Capital Region domain and Episode#2. ............................................................................. 82

Table 5-10. 24-hour speciated PM2.5 (µg/m3) model performance for CMAQ WRF-CFSR corrected emissions sensitivity test (Test#2), the 1.33 km Capital Region domain and Episode#2. ............................................................................. 87

Table 5-11. 24-hour speciated PM2.5 (µg/m3) model performance for CMAQ no primary sulphate emissions sensitivity test (Test#3), the 1.33 km Capital Region domain and Episode#2. ................................................................. 90

Table 5-12. Hourly total PM2.5 mass model performance for CMAQ limit vertical diffusion sensitivity test (Test#4), the 1.33 km Capital Region domain and Episode#2. ...................................................................................................... 91

Table 5-13. 24-Hour speciated PM2.5 model performance at Edmonton McIntyre for CMAQ limit vertical diffusion sensitivity test (Test#4), the 1.33 km Capital Region domain and Episode#2. ................................................................. 92

Table 5-14. 24-Hour speciated PM2.5 model performance at Edmonton McIntyre for CMAQ N2O5 heterogeneous sensitivity test (Test#8), the 1.33 km Capital Region domain and Episode#2. ................................................................. 97

Table 5-15. 24-Hour speciated PM2.5 model performance at Edmonton McIntyre for CMAQ CALMOB6 sensitivity test (Test#9), the 1.33 km Capital Region domain and Episode#2. ............................................................................. 98

Table 5-16. Hourly total PM2.5 mass model performance for CMAQ base case, the 1.33 km Capital Region domain and Jan-Feb period........................................... 104

Table 5-17. 24-hour speciated PM2.5 (µg/m3) model performance for base case, the 1.33 km Capital Region domain and Jan-Feb period. ................................... 107

Table 5-18. Hourly ozone model performance for CMAQ base case, the 1.33 km Capital Region domain and Jan-Feb period. ....................................................... 108

vi

FIGURES

Figure 1-1. 36/12/4 km CMAQ modelling domains used for the North Saskatchewan Region CMAQ modelling study that will also be used for the Capital Region PM modelling study (Nopmoingcol et al., 2011). ..................... 3

Figure 1-2. 4 km NSR and 1.33 km Capital Region domains used in the Capital Region PM Modelling Study. ................................................................................... 4

Figure 2-1. Locations of CASA PM2.5 monitoring sites within the Capital Region. .................... 7

Figure 2-2. Locations of NAPS PM2.5 monitoring sites within the 4 km Capital Region domain. ....................................................................................................... 8

Figure 2-3. Comparison of different PM2.5 measurement technologies for 24-hour PM2.5 concentrations at the Edmonton McIntyre monitoring site. ...................... 11

Figure 2-4a. Comparison of TEOM FDMS versus Met One E-BAM hour PM2.5

concentration measurements at McIntyre during 2008-2010. ............................ 12

Figure 2-4b. Comparison of TEOM FDMS versus TEOM @ 30C hourly PM2.5


Figure 2-4c. Comparison of TEOM FDMS versus BAM @ 35RH hourly PM2.5


Figure 2-4d. Comparison of TEOM FDMS versus Dichot 24-hourl PM2.5


Figure 2-5. Composition of 24-hour PM2.5 concentrations at Edmonton McIntyre site (Station 090132) for six winter days that exceeded the PM2.5 CWS during 2008-2010. ................................................................................................. 15

Figure 3-1. NOx emissions >250 tonne/yr around Capital Region. Yellow stars indicate air quality monitoring stations in Edmonton. ......................................... 22

Figure 3-2. SO2 emissions >250 tonne/yr around Capital Region. Yellow stars indicate air quality monitoring stations in Edmonton. ......................................... 23

Figure 3-3. Imperial Oil Strathcona facility plotted spatially with GIS software. NSRP2006 inventory points shown in yellow, AAEI2008+ shown in blue. Red circle indicates the actual location of the facility's main stack. (Imagery © 2014 Google) ........................................................................... 27

Figure 3-4. Edmonton road networks from CanVec 2010 (left) and EC 2006 (right) .............. 28

Figure 3-5. Red Deer road networks from CanVec 2010 (left) and EC 2006 (right). ............... 29

Figure 3-6. CanVec 2010 road network (left) and CALMOB6 Links (right). ............................. 29

Figure 3-7. Direct comparison of CALMOB6 (blue) and CanVec (gray) road networks in residential areas. ............................................................................... 30

vii

Figure 3-8. NOx emissions distribution around City of Edmonton from SAOS/CanVec (left) and CALMOB6 emissions model (right). ............................... 31

Figure 3-9. Discarded CALMOB6 data along the Yellowhead Highway. ................................. 32

Figure 3-10. Residential areas used for scaling CanVec + SMOKE emissions based on CALMOB6 emissions. ............................................................................................. 33

Figure 3-11. Traffic count locations provided by the Transportation Operations Division. (Map data © 2014 Google) .................................................................... 34

Figure 3-12. Weekday diurnal traffic profile from City of Edmonton traffic counts. ............... 35

Figure 3-13. Weekend diurnal traffic profile from City of Edmonton traffic counts. .............. 36

Figure 3-14. Edmonton Regional External Truck/Commodity Survey diurnal truck distribution. ........................................................................................................... 37

Figure 3-15. HDV emissions distribution for NOx (green), PM2.5 (blue), and SO2 (purple). ................................................................................................................. 41

Figure 3-16. Light duty spatial surrogate (left) and heavy duty spatial surrogate (right). .................................................................................................................... 41

Figure 3-17. Province-wide Canvec paved road network......................................................... 42

Figure 3-18. Heavy construction dust re-allocation. Green circles represent mining and extraction areas, dark blue squares are grid cells which received the re-allocated emissions. ................................................................................... 44

Figure 4-1. WRF domain extents for 36/12/4/1.33 km domains as D1, D2, D3, and D4 respectively. ..................................................................................................... 49

Figure 4-2. WRF domain extents for 12/4/1.33 km domains as D2, D3, and D4 respectively. .......................................................................................................... 50

Figure 4-3. WRF domain extents for 4/1.33 km domains as D3 and D4 respectively. .......................................................................................................... 51

Figure 4-4. Locations of surface meteorological monitoring sites around the Capital used in the WRF surface model performance evaluation. ....................... 58

Figure 4-5. 1.33 km WRF Temperature time-series comparison between Test a) NARR-WRF (blue line) and Test d) CFSR-WRF (red line) with observation data (green dots) in January 2010. ................................................... 61

Figure 4-6. Temperature comparison for 2010 Episode 1: Jan 17-21, 2010: Jan 18, 19 and 20. Red line is modeled 4 km temperature, blue line is modeled 1.33km temperature and green dot is upper air sounding data. ......................... 64

Figure 4-7. Wind speed comparison for 2010 Episode 1: Jan 17-21, 2010: Jan 18, 19 and 20. Red line is modeled 4 km wind speed, blue line is modeled 1.33km wind speed and green dot is upper air sounding data. ........................... 65

viii

Figure 4-8. Temperature comparison for 2010 Episode 2: Jan 26 – Feb 4, 2010: Jan 28 and 29, Feb 02. Red line is modeled 4 km temperature, blue line is modeled 1.33km temperature and green dot is upper air sounding data. ....................................................................................................................... 66

Figure 4-9. Wind speed comparison for 2010 Episode 2: Jan 26 – Feb 4, 2010: Jan 28 and 29, Feb 02. Red line is modeled 4 km wind speed, blue line is modeled 1.33km wind speed and green dot is upper air sounding data. ............ 67

Figure 4-10. Temperature comparison for 2010 Episode 5: Feb 20 – Mar 8, 2010: Feb 24, Mar 01 and 03. Red line is modeled 4 km temperature, blue line is modeled 1.33km temperature and green dot is upper air sounding data. ....................................................................................................... 68

Figure 4-11. Wind speed comparison for 2010 Episode 5: Feb 20 – Mar 8, 2010: Feb 24, Mar 01 and 03. Red line is modeled 4 km wind speed, blue line is modeled 1.33km wind speed and green dot is upper air sounding data. ............ 69

Figure 5-1a. Time series of predicted (blue) and observed (red) hourly PM2.5

concentrations (µg/m3) for sensitivity Test#2 and monitoring sites in Edmonton. ............................................................................................................. 83

Figure 5-1b. Time series of predicted (blue) and observed (red) hourly PM2.5

concentrations (µg/m3) for sensitivity Test#2 and Edmonton South (top left) and five monitoring sites northeast of Edmonton. ................................ 84

Figure 5-1c. Time series of predicted (blue) and observed (red) hourly PM2.5

concentrations (µg/m3) for sensitivity Test#2 and four monitoring sites southwest of Edmonton. ....................................................................................... 85

Figure 5-2a. Time series of predicted (blue) and observed (red) 24-hour speciated PM2.5 concentrations (µg/m3) for sensitivity Test#2 and monitoring sites in Edmonton. ................................................................................................. 88

Figure 5-2b. Time series of predicted (blue) and observed (red) 24-hour speciated PM2.5 concentrations (µg/m3) for sensitivity Test#2 and monitoring sites in Edmonton. ................................................................................................. 89

Figure 5-3. Time series and performance statistics for 24-hour PM2.5 concentrations for 4 km Test#5 (NODUST) and Test#6 (NODUST_BC) at several monitoring sites during 2010 Quarter 1. .................................................. 94

Figure 5-4a. Time series and performance statistics for 24-hour speciated PM2.5 concentrations at Edmonton McIntyre monitoring site during 2010 Quarter 1 for 4 km Test#5 (NODUST) and Test#6 (NODUST_BC). ........................ 95

Figure 5-4b. Time series and performance statistics for hourly ozone concentrations at Edmonton East (left) and Elk Island (right)

ix

monitoring sites during Episodes#2 (top), #3 (middle) and #3 (bottom) for 4 km Test#5 (NODUST) and Test#6 (NODUST_BC). ......................................... 96

Figure 5-5a. Time series of predicted (blue) and observed (red) 24-hour PM2.5

concentrations (µg/m3) for base case and monitoring sites in Edmonton. ........................................................................................................... 101

Figure 5-5b. Time series of predicted (blue) and observed (red) 24-hour PM2.5

concentrations (µg/m3) for sensitivity Test#2 and Edmonton South (top left) and five monitoring sites northeast of Edmonton. .............................. 102

Figure 5-5c. Time series of predicted (blue) and observed (red) 24-hour PM2.5

concentrations (µg/m3) for base case and four monitoring sites southwest of Edmonton. ..................................................................................... 103

Figure 5-6a. Time series of predicted (blue) and observed (red) 24-hour speciated PM2.5 concentrations (µg/m3) for base case at Edmonton McIntyre site. .......... 105

Figure 5-6b. Time series of predicted (blue) and observed (red) 24-hour speciated PM2.5 concentrations (µg/m3) for base case at Edmonton McIntyre site. .......... 106

Figure 5-7. Ozone (left) and NOx (right) scattered plots for base case, the 1.33 km Capital Region domain and Jan-Feb period. ....................................................... 107

Figure 6-1. Monthly anthropogenic emissions for the Capital Region by source sector (averaged over January-March, 2010). .................................................... 110

Figure 6-2. CMAQ-estimated average PM2.5 concentrations (µg/m3) for base case ............ 111

Figure 6-3. CMAQ-estimated average speciated PM2.5 concentrations (µg/m3) for base case (Note different maximum scales). ...................................................... 112

Figure 6-4. Difference in CMAQ-estimated average PM2.5 concentrations during January-February for (a) On-road mobile source, (b) EGU, (c) other point source, and (d) all anthropogenic source zero-out simulations (Base – Scenario). ............................................................................................... 114

Figure 6-5. Difference in CMAQ-estimated average sulphate concentrations during January-February for (a) On-road mobile source, (b) EGU, (c) other point source, and (d) all anthropogenic source zero-out simulations (Base - Scenario). ............................................................................. 115

Figure 6-6. Difference in CMAQ-estimated average nitrate concentrations during January-February for (a) On-road mobile source, (b) EGU, (c) other point source, and (d) all anthropogenic source zero-out simulations (Base - Scenario). ................................................................................................. 116

Figure 6-7. Difference in CMAQ-estimated average ammonium concentrations during January-February for (a) On-road mobile source, (b) EGU, (c)

x

other point source, and (d) all anthropogenic source zero-out simulations (Base - Scenario). ............................................................................. 117

Figure 6-8. Difference in CMAQ-estimated average elemental carbon concentrations during January-February for (a) On-road mobile source, (b) EGU, (c) other point source, and (d) all anthropogenic source zero-out simulations (Base - Scenario). ....................................................................... 118

Figure 6-9. Difference in CMAQ-estimated average organic aerosols concentrations during January-February for (a) On-road mobile source, (b) EGU, (c) other point source, and (d) all anthropogenic source zero-out simulations (Base - Scenario). ....................................................................... 119

Figure 6-10. Difference in CMAQ-estimated average soil concentrations during January-February for (a) On-road mobile source, (b) EGU, (c) other point source, and (d) all anthropogenic source zero-out simulations (Base - Scenario). ................................................................................................. 120

Figure 6-11. Source Contributions to average PM2.5 at monitoring stations within the Capital Region. .............................................................................................. 121

Figure 6-12. Source Contributions to average sulphate at monitoring stations within the Capital Region. ................................................................................... 122

Figure 6-13. Source Contributions to average nitrate at monitoring stations within the Capital Region (negative contributions are indicative of nitrate increase due to elimination of source emissions). .............................................. 122

Figure 6-14. Source Contributions to average ammonium at monitoring stations within the Capital Region. ................................................................................... 123

Figure 6-15. Source Contributions to average elemental carbon at monitoring stations within the Capital Region. ..................................................................... 123

Figure 6-16. Source Contributions to average organic aerosols at monitoring stations within the Capital Region. ..................................................................... 124

Figure 6-17. Source Contributions to average soil at monitoring stations within the Capital Region. ..................................................................................................... 124

1

1.0 INTRODUCTION

1.1 Background

Alberta’s Capital Region consists of many municipalities that surround and include the City of Edmonton. Air quality issues within the Capital Region are summarized in the Capital Region Air Quality Management Framework for Nitrogen Dioxide (NO2), Sulphur Dioxide (SO2), Fine Particulate Matter (PM2.5) and Ozone (O3)1. The Clean Air Strategic Alliance Particulate Matter and Ozone Management Framework defines a series of action trigger levels for fine particulate matter and ozone to help assure that the Canada Wide Standards (CWS) are not exceeded. For PM2.5, the CWS has a threshold of 30 µg/m3 to be achieved by 2010 based on the 98th percentile of the 24-hour PM2.5 concentrations averaged over three consecutive years. Based on recent PM2.5 measurements, the Edmonton Central and Edmonton East monitoring sites exceed the Mandatory Plan trigger level (30 µg/m3) and the Province is required to develop a plan for reducing the PM2.5 concentrations to below the CWS threshold.

ENVIRON International Corporation and Novus Environmental performed the Capital Region Particulate Matter Air Modelling Assessment Study for the Alberta Environmental and Sustainable Resources Development (ESRD). The objective of the study is to develop a Photochemical Grid Model (PGM) modelling database for the Capital Region, which includes Edmonton and surrounding communities, that reproduces the observed winter elevated fine particulate matter (PM2.5) concentrations sufficiently well that it can be a reliable tool for analyzing source contributions to elevated PM2.5 concentrations and evaluating the effects of alternative emission control strategies on elevated PM2.5 concentrations. To achieve these objectives, the study was conducted in four tasks:

Task 1: Base Case Emission Inputs

Task 2: Meteorological Inputs

Task 3: Photochemical Grid Modelling and Model Performance Evaluation

Task 4: Conduct Scenarios

The scope of work in Task 1 (Base Case Emission Inputs) was extended to include an integration of on-road mobile emissions developed by City of Edmonton using CALMOB6 (Calibrated MOBILE6) emissions factor model. As part of the Task 1 work effort, we prepared a draft Modelling Plan dated November 25, 2013 that was designed to be a summary of the specific procedures of the modelling approach. The modelling plan presents the selection of winter elevated PM2.5 episodes in the Capital Region (Appendix A) with the highest ranked ones in 2010 used in the sensitivity modelling to determine an optimal model configuration.

In Task 3, we performed several sensitivity tests to determine the optimal modelling configuration by evaluating the results against measurement data. In earlier sensitivity

1 http://environment.gov.ab.ca/info/library/8593.pdf

2

simulations, we found that CMAQ over-predicted PM at all monitors within the Capital Region. A comparison against the speciated data at McIntyre site suggested that both primary and secondary PMs (especially sulphate) were over-estimated. After discussing with ESRD, we revised the scope of work to focus primarily on the model performance based on the January-March, 2010, period that included the most days with exceedances of the PM2.5 CWS.

1.2 Overview of Approach

The Capital Region PM Modelling Study used three main models:

The Sparse Matrix Operator Kernel Emissions (SMOKE) model to generate the hourly gridded speciated emission inputs needed by the PGM.

The Weather Research Forecast (WRF) meteorological model used to generate the PGM meteorological inputs.

The Community Multiscale Air Quality (CMAQ) PGM modelling system.

The modelling domains used are the same 36 km southwestern Canada (SWCAN) and northwest U.S., 12 km Alberta Province and 4 km North Saskatchewan Region (NSR) modelling domains (see Figure 1-1) as used in the North Saskatchewan Region (NSR) CMAQ Modelling Study (Nopmongcol et al., 2011). In addition, a 1.33 km CMAQ modelling domain focused on the Capital Region was used as shown in Figure 1-2. These domains use a Lambert Conformal Projection using the projection parameters given in Table 1-1.

Table 1-1 Definition of the Lambert Conformal Projection (LCP) 36/12/4/1.33 km domains used in the CMAQ photochemical and SMOKE emissions modelling of the Capital Region.

LCP Projection Parameters Central Longitude Meridian -121.0 degrees Latitude Origin 49.0 degrees 1

st Standard Parallel 30.0 degrees

2nd

Standard Parallel 60.0 degrees

36 km SWCAN Domain SW Corner (-828 km, -936 km) 59 x 74

12 km Alberta Domain SW Corner (12 km, -12 km) 70 x 110

4 km NSR Domain SW Corner (84 km, 168 km) 162 x 123

1.33 km Capitol Region Domain SW Corner (372 km, 448 km) 156 x 99

3

SW Corner: (-828, -936) 59 x 74 cells 36 km SW Corner: ( 12, -12) 70 x 110 cells 12 km

SW Corner: ( 84, 168) 162 x 123 cells 4 km

Figure 1-1. 36/12/4 km CMAQ modelling domains used for the North Saskatchewan Region CMAQ modelling study that will also be used for the Capital Region PM modelling study (Nopmoingcol et al., 2011).

4

Figure 1-2. 4 km NSR and 1.33 km Capital Region domains used in the Capital Region PM Modelling Study.

The overall approach for achieving the project objectives are summarized by task below.

1.2.1 Task 1: Base Case Emission Inputs

2010 emissions database were developed for Alberta and the Capital Region by combining the emissions from the ESRD North Saskatchewan Regional Plan (NSRP), South Saskatchewan Regional Plan (SSRP) and South Athabasca Oil Sands Area (SAOS) studies with additional data (e.g., survey data, new Environmental Assessments and City of Edmonton mobile source emissions).

In Alberta, construction operations are a significant source of dust emissions, and can have a substantial impact on regional air quality. In the previous NSRP and SSRP CMAQ modelling studies, we experienced high concentration of PM in large populated areas and concluded that the dust emissions were not well-represented. This issue was addressed by two major improvements introduced in this study:

5

Applied transport fractions2 using revised land use/land cover

Used CanVec3 data to update industrial construction locations to provide better spatial distribution of construction and off-road dust emissions

The SMOKE emissions modelling system was used to generate the hourly, gridded, speciated CMAQ model-ready emissions inputs for January-March, 2010 period.

1.2.2 Task 2: Meteorological Inputs

This study applied the latest version of the Weather Research Forecast (WRF) meteorological model (Version 3.5.1 released September 23, 2013) for meteorological modelling. The input data used in the WRF simulation consists of gridded NARR (North American Regional Reanalysis at 32 km grid resolution) data produced and distributed by National Centers for Environmental Prediction (NCEP) as well as upper air and surface observational data which are used to “nudge” the WRF fields to obtain a better representation of the meteorology. WRF sensitivity tests were conducted using the NCEP Climate Forecast System Reanalysis (CFSR) and the WRF fine grid performance was compared between NARR and CFSR. The WRF 36/12/4/1.33 km domains have been defined slightly bigger than the CMAQ modelling domains in order to eliminate any boundary artifacts.

Several sensitivity tests using the 1.33 km domain were conducted to identify an optimal performing WRF model configuration. The selected WRF simulation outputs were converted to CMAQ input format using the MCIP program.

1.2.3 Task 3: Photochemical Grid Modelling and Model Performance Evaluation

An initial CMAQ V5.0.1 base case simulation was performed for high PM2.5 episodic periods selected for analysis under Task 1. Based on these results, we performed a series of sensitivity tests designed to better simulate the winter elevated PM2.5 concentrations in the Capital Region with particular emphasis on secondary PM model performance. The model performance evaluation procedures follow the recommendations in USEPA’s (USEPA, 2007) and Alberta’s (Idriss and Spurrell, 2009) air quality modelling guidance.

Our sensitivity tests were limited by the project schedule. The best performing WRF/CMAQ configuration was selected as the base case and the CMAQ simulations were performed for all modelling domains covering 2010 January-March winter period.

2 Fugitive dust transport fractions represent the fraction of dust emissions that are not locally deposited so are

transported downwind. They are land use dependent with more local deposition in forested areas than barren areas (http://www.epa.gov/ttnchie1/emch/dustfractions/). 3 CanVec is a digital cartographic reference product produced by Natural Resources Canada (NRCan). CanVec

originates from the best available data sources covering Canadian territory and offers quality topographic information in vector format that complies with international geomatics standards. (http://ftp2.cits.rncan.gc.ca/pub/canvec/)

http://www.epa.gov/ttnchie1/emch/dustfractions/

6

1.2.4 Task 4: Conduct Scenarios

Four (4) emission zero-out simulations were conducted using the 2010 CMAQ modelling setup for 1.33 km domain. The purpose of these source apportionment model simulations is to investigate and assess the relative importance of specific emission from various sectors on PM in the Capital region. The specific simulations were selected in order to address ESRD’s desire to ascertain the air quality impacts in the Capital region due to local production. The four zero-out source apportionment simulations are defined as follows:

1. On-road mobile sources;

2. Stationary point sources excluding electrical generating units (EGU) and upstream oil & gas (UOG) sources;

3. Coal-fired Power Plants (EGU); and

4. Total Anthropogenic sources

In scenario 1 through 3, entire source emissions from specified sector within the Capital Region are eliminated (i.e., zeroed-out) within the CMAQ-ready emissions inputs. In Scenario 4, all anthropogenic sources including stationary point and area, and mobile sources within the Capital Region are eliminated. This simulation facilitates assessment of impact from outside of the Capital Region. Natural emissions throughout the modelling domain, including biogenic and fire emissions in all regions, were kept at 2010 base case levels for all four scenarios.

The results of each zero-out simulation were compared with the base case simulation results to evaluate the impacts of PM concentrations due to local source emissions within the Capital Region.

1.3 Report Organization

This report provides detailed air quality modelling for the Capital Region. Chapter 2 describes winter PM issues in the Capital Region. Chapter 3 describes the base case emissions development. Chapter 4 presents WRF modelling setup and model performance evaluation at the monitoring stations within the Capital Region. Chapter 5 presents CMAQ modelling setup, sensitivity analyses, and model performance evaluation at the monitoring stations in the 1.33 km modelling domain. The assessment of air quality impacts from four source sectors is presented in Chapter 6. Chapter 7 provides the project recommendations with respect to improving model performance evaluation and summarizes the remaining project tasks to be considered in future work.

7

2.0 UNDERSTANDING PM ISSUES IN THE CAPITAL REGION

In this chapter, we describe PM2.5 issues based on available observational data in the Capital Region.

2.1 Measurement Data

Continuous hourly PM2.5 measurements have been collected in the Capital Region using the Tapered Element Oscillating Microbalance (TEOM) measurement technology as a well as several other technologies for several years. The hourly PM2.5 measurement data for the year 2008-2010 is available from the Alberta Clean Air Strategic Alliance (CASA) data warehouse4 and Environment Canada National Air Pollution Surveillance (NAPS) program website5. Figures 2-1 and 2-2 displays the locations of the monitoring sites in the Capital Region with more details provided in Table 2-1. There was numerous overlap in the data from CASA and NAPS, so NAPS monitoring sites with data duplicated at the CASA sites were eliminated; this resulted in three NAPS site remaining, 90601 in the Capital Region just to the northeast of Edmonton and 92801 and 93901 that reside just outside of the western boundary of the Capital Region (Figure 2-2).

Figure 2-1. Locations of CASA PM2.5 monitoring sites within the Capital Region.

4 http://www.casadata.org/

5 http://www.ec.gc.ca/rnspa-naps/

8

Figure 2-2. Locations of NAPS PM2.5 monitoring sites within the 4 km Capital Region domain.

2.2 Particulate Matter Measurement Technologies

One issue associated with measuring PM2.5 is that it is defined by the measurement technology used. Each measurement technology has its own artifacts, including evaporation of some volatile compounds, inclusion or exclusion of water mass, the use of blank corrections and other issues. Using different measurement technologies can produce very different PM2.5

observations. These measurement artifacts need to be accounted for when evaluating atmospheric chemistry models.

Table 2-1 lists the monitoring sites in the Capital Region that were operating during the 2008-2010 period. There are several different types of continuous (i.e., reporting hourly concentrations) and 24-hour average PM2.5 monitoring technologies used in the Capital Region during the 2008-2010 time period, each with its own measurement artifacts. The TEOM

9

technique is the most widely used PM2.5 technology in the Capital Region with several different types of TEOM technologies used. The TEOM particulate sampler operates by drawing air through a filter attached at the tip of a glass tube. An electrical circuit places the tube into oscillation, and the resonant frequency of the tube is proportional to the square root of the PM2.5 mass on the filter. The original version of the TEOM operated at a filter temperature of 50 °C (TEOM @ 50C) to avoid condensation of humidity and satisfy the need to keep the filter at a constant temperature. However, heating the filter results in loss of some of the semi-volatile particulate matter, such as ammonium nitrate that is the largest component of PM2.5 during PM2.5 episodes in the Capital Region (discussed below). Field studies in Utah and Pennsylvania found the TEOM @ 50C only measured from 50 to 85 percent of the total PM2.5 mass. As seen in Table 2-1, many of the Capital Region continuous PM2.5 monitoring sites employ TEOM @ 40C and TEOM @ 30C monitoring devices that will have a similar semi-volatile PM2.5 loss as the TEOM @ 50C. The TEOM with a filter dynamics measurement system (TEOM FDMS) overcomes some of the semi-volatile PM2.5 loss through a self-referencing system taking samples every 6 minutes using two different methodologies and correcting for the particulate matter loss through volatilization. Over the years, monitoring sites in the Capital Region have been migrating to the TEOM FDMS technology, which has resulted in higher observed PM2.5 concentrations in the Capital Region for the more recent years. Other PM2.5 measurement technologies used in the Capital Region include the Beta Attenuation Monitor (BAM) continuous and a Dichot and PM2.5 speciation 24-hour average filter based monitoring sites that are co-located at McIntyre monitoring site. The speciated PM2.5 measurements at McIntyre are summed to obtain Reconstructed Fine Mass (RCFM) total PM2.5 concentrations that are compared against the Dichot total PM2.5 mass concentrations as part of the quality assurance (QA) process. When the speciated PM2.5 RCFM deviates more than 5 percent from the Dichot total PM2.5 mass concentration, the speciated PM2.5 data are deemed invalid and not used in our modeling analysis.

10

Table 2-1. Capital Regional continuous PM2.5 monitoring sites. Stn ID Name Method

EDMC TEOM4 Edmonton Central TEOM @ 40C


EDMC FDMS Edmonton Central TEOM @ 30C with FDMS (self-referencing)

EDME TEOM4 Edmonton East TEOM @ 40C


EDME FDMS Edmonton East TEOM @ 30C with FDMS (self-referencing)

MCIN BAM1 Edmonton McIntyre E-BAM (Met1)

MCIN TEOM3 Edmonton McIntyre TEOM @ 30C

MCIN FDMS Edmonton McIntyre TEOM @ 30C with FDMS (self-referencing)

MCIN BAM3 Edmonton McIntyre BAM @ 35RH

EDMS TEOM4 Edmonton South TEOM @ 40C


EDMS FDMS Edmonton South TEOM @ 30C with FDMS (self-referencing)

ELKI TEOM4 Elk Island TEOM @ 40C

FTSA TEOM3 Fort Saskatchewan TEOM @ 40 C

FTSA SHARP Fort Saskatchewan Sharp (hybrid nephelometer/BAM sys) with data reported at actual ambient conditions

GENE TEOM4 Genesee TEOM @ 40C

LAMA BAM Lamont Met One BAM 1020

REDW TEOM4 Redwater Industrial TEOM @ 40C

TOMA TEOM4 Tomahawk TEOM @ 40C

90601 SES NAPS 09601 SES

90601 TEOM NAPS 09601 TEOM

Figure 2-3 displays a time series of 24-hour average PM2.5 concentrations measured using six PM2.5 monitoring methods: (a) TEOM @ 30C; (b) TEOM @ FDMS (self-referencing); (c) BAM Met 1; (d) BAM @ 35RH; (e) 24-hour DICHOT mass; and (f) 24-hour speciated PM sampler (RCFM). Among the six methods, it is quite clear that PM2.5 levels measured by TEOM FDMS PM2.5 is usually higher than the other methods followed by E-BAM, BAM @ 35H and TEOM @ 30C. Figure 2-3 also illustrates there is sometimes a disagreement between DICHOT and RCFM. For example, on January 17 and 20, 2010 there is a large disagreement between the two methods resulting in the speciated PM2.5 concentrations on these two days being declared invalid.

11

Figure 2-3. Comparison of different PM2.5 measurement technologies for 24-hour PM2.5

concentrations at the Edmonton McIntyre monitoring site.

2.2.1 Correlations of PM2.5 Measurement Technologies

As discussed in the previous section, it appears that the most accurate continuous PM2.5

measurement technology deployed in the Capital Region is the TEOM FDMS. When evaluating air quality modeling using different PM2.5 measurement technologies it is important to understand their artifacts. The correlations and bias between the TEOM FDMS and other measurement technologies that were co-located at the Edmonton McIntyre monitoring site are shown in Figure 2-4. TEOM FDMS tends to measure PM2.5 concentrations that are approximately 30 percent higher than the TEOM @ 30C (Figure 2-4b). The agreement between the TEOM FDMS and MetOne E-BAM is not as good with the regression equation suggested E-BAM is higher than TEOM FDMS, which is the opposite impression from the time series in Figure 2-3. There is also a lot of scatter between the TEOM FDMS and BAM @ 35RH with the BAM being approximately 10 percent higher than TEOM FDMS. Good agreement is seen between the TEOM FDMS and Dichot 24-hour PM2.5 concentration measurements (Figure 2-4d). This is due in part to the longer averaging time that produces less variation. Of the total PM2.5 mass measurement technologies analyzed, the Dichot is the only one that directly measures the PM2.5 mass on a filter rather than inferring it.

These results indicate that, as expected, the TEOM @30C and TEOM @ 40C underrepresent actual ambient PM2.5 atmospheric concentration levels. The fact that the TEOM FDMS is lower than the BAM may indicate that it still suffers some loss of volatile PM2.5 in the sampling process, or possibly the BAM is overstating PM2.5.

12

Figure 2-4a. Comparison of TEOM FDMS versus Met One E-BAM hour PM2.5 concentration measurements at McIntyre during 2008-2010.

Figure 2-4b. Comparison of TEOM FDMS versus TEOM @ 30C hourly PM2.5 concentration measurements at McIntyre during 2008-2010.

y = 0.76062x + 5.11727R² = 0.61554

0

20

40

60

80

100

120

140

0 20 40 60 80 100 120 140 160

Teo

m @

30

C w

ith

FD

MS

E-BAM (Met1)

TEOM FDMS v.s. E-BAM (Met1)

y = 1.30845x + 4.00417R² = 0.74335

0

20

40

60

80

100

120

140

0 20 40 60 80 100 120

Teo

m @

30

C w

ith

FD

MS

TEOM @ 30C

TEOM FDMS v.s. TEOM @ 30C

13

Figure 2-4c. Comparison of TEOM FDMS versus BAM @ 35RH hourly PM2.5 concentration measurements at McIntyre during 2008-2010.

Figure 2-4d. Comparison of TEOM FDMS versus Dichot 24-hourl PM2.5 concentration measurements at McIntyre during 2008-2010.

y = 0.90534x + 3.17681R² = 0.71961

0

20

40

60

80

100

120

140

0 20 40 60 80 100 120 140

Teo

m @

30

C w

ith

FD

MS

BAM @ 35RH

TEOM FDMS v.s. BAM @ 35RH

y = 1.07555x + 0.75948R² = 0.91861

0

10

20

30

40

50

60

70

80

0 10 20 30 40 50 60 70

Teo

m @

30

C w

ith

FD

MS

DICHOT

TEOM FDMS v.s. DICHOT

14

2.3 PM2.5 Problem in the Capital Region

Appendix A presents a discussion of the observed total PM2.5 mass concentrations in the Capital Region during 2008-2010 and identifies episodes when the PM2.5 CWS is exceeded. Table A-12 lists the 24-hour average observed PM2.5 concentration at each monitoring site and each day from 2008-2010 color coding the observed values yellow if they exceed the CWS 30 µg/m3 threshold and blue if they exceed the 20 µg/m3 trigger level, with potential modeling episodes identified by the red box. There are many more exceedances of the PM2.5 CWS in 2010 than in previous years, which is likely due in part to the higher deployment of the TEOM FDMS measurement technology in 2010 than previous years.

Since PM consists of numerous primary and secondary organic and inorganic species, it is important for the model to accurately reproduce the major components of PM2.5 in addition to total PM2.5 mass, in order to have an adequate modelling system for the evaluation of alternative emission control strategies for reducing PM2.5 concentrations. We analyzed speciated 24-hour PM2.5 measurements from the Edmonton McIntyre monitoring site during 2008-2010 and found that, with the exception of a few days when smoke from wildfires impacted the Capital Region during August 2010, PM2.5 concentration that exceeded the CWS always occurred during the winter. These winter PM2.5 CWS exceedance days were characterized by higher than typical secondary PM2.5 concentrations. Figure 2-5 displays the components of the PM2.5 concentrations for six of the highest PM2.5 days during 2008-2010 at the Edmonton McIntyre monitoring site, all of which exceeded the CWS standard and occurred during the winter months. The observed PM2.5 concentrations during these six days ranged from 30.9 to 61.4 µg/m3. Note that a higher 24-hour PM2.5 concentration of 64.6 µg/m3 occurred at the Edmonton McIntyre monitoring site on August 21, 2010, however this was during a period when Edmonton was inundated with smoke from wildfires so is excluded from consideration when comparing to the CWS or trigger level thresholds. As shown in Figure 2-5, a majority (58% to 72%) of the PM2.5 mass on these CWS exceedance days come from nitrate (NO3), ammonium (NH4) and sulphate (SO4) that is primarily secondary in nature (i.e., it is formed in the atmosphere from emissions of gaseous NOx, NH3 and SO2). And of these species, ammonium nitrate [NH3NO3] makes up the largest component of the secondary PM2.5 on these high PM2.5 days contributing 36 to 62 percent of the total PM2.5 mass. Note that some Organic Carbon (OC) may also be secondary in nature (i.e., Secondary Organic Aerosol or SOA). However, there is also a lot of primary organic carbon (POC) emitted by many sources (e.g., mobile sources, biomass burning, etc.) and we would expect a vast majority of the OC in Edmonton during the winter to be POC.

The correct simulation of wintertime secondary ammonium nitrate formation is critically important in this study. The identification of the source sectors that contribute most to the CWS PM2.5 exceedances requires the proper simulation of secondary PM.

The six elevated PM2.5 days in Edmonton shown in Figure 2-5 are all associated with very slow to stagnant wind speeds (< 2 m/s). Thus, to simulate these winter PM events in Edmonton the meteorological modelling will be critically important to simulate the buildup of ammonium

15

nitrate precursors as well as the thermodynamic conditions that favor ammonium nitrate formation (i.e., cooler and wetter).

Figure 2-5. Composition of 24-hour PM2.5 concentrations at Edmonton McIntyre site (Station 090132) for six winter days that exceeded the PM2.5 CWS during 2008-2010.

SO4 24%

NO3 29%

NH4 17%

EC5%

OC11%

Pb: MDL Method

0%

Cd: MDL Method

0%

Salt2%

Soil: MDL Method

2% TEO: MDL Method

0% Δ10%

EDM 1/29/10 61.4 ug/m3

SO4 13%

NO3 28%

NH4 12%

EC4%

OC11%

Pb: MDL Method

0%

Cd: MDL Method

0%

Salt1%

Soil: MDL Method

1%

TEO: MDL Method

0% Δ30%

EDM 12/27/09 47.0 ug/m3

SO4 5%

NO3 41%

NH4 12%EC

6%

OC20%Pb: MDL

Method0%

Cd: MDL Method

0%

Salt1%

Soil: MDL Method

2%

TEO: MDL Method

0% Δ13%

EDM 12/7/10 42.0 ug/m3SO4 5%

NO3 48%

NH4 14%

EC2%

OC6%

Pb: MDL Method

0%

Cd: MDL Method

0%

Salt4%

Soil: MDL Method

2%TEO: MDL Method

0%

Δ19%

EDM 2/21/08 35.9 ug/m3

SO4 16%

NO3 40%

NH4 16%

EC4%

OC7%

Pb: MDL Method

0%

Cd: MDL Method

0%

Salt1%

Soil: MDL Method

1%TEO: MDL Method

0% Δ15%

EDM 2/9/09 31.5 ug/m3

SO4 13%

NO3 35%

NH4 14%

EC6%

OC13%

Pb: MDL Method

0%

Cd: MDL Method

0%

Salt1%

Soil: MDL Method

2%

TEO: MDL Method

0% Δ16%

EDM 1/10/08 30.9 ug/m3

16

2.4 Modelling Winter Secondary PM2.5

Elevated secondary PM2.5 concentrations are typically associated with either summer sulphate episodes, when there is lots of photochemical activity to convert the SO2 to sulphate as occurs in the eastern U.S. and southeastern Canada. However, elevated secondary PM2.5 concentrations can also occur in the winter or cooler seasons when the colder temperatures result in aerosol thermodynamics that favors particulate nitrate formation over gaseous nitric acid and there is less competition by sulphate for the ammonia so that particulate ammonium nitrate can form. Because sulphate is a stronger acid than nitrate, ammonia will preferentially bond with sulphate over nitric acid. The formation of elevated particulate ammonium nitrate concentrations in the Capital Regions requires the following chemical processes to occur:

Conversion of the primary emitted NOX (NO and NO2) to gaseous nitric acid (HNO3);

Availability of ammonia (NH3 or other basic compound) to bind with gaseous HNO3 to form particulate nitrate (NO3); and

Meteorological conditions (i.e., cooler and moister) so that the thermodynamic equilibrium favors particulate ammonium nitrate (NO3NH4) over gaseous HNO3 and NH3.

Thus, keys to simulating the Capital Region high winter PM2.5 episodes will be simulating the conversion of NOX to nitric acid, the availability of ammonia making the ammonia inventory critically important and the meteorological model simulation of winds, temperature and humidity.

2.4.1 Conversion of NOx to HNO3

The availability of NOx emission to form HNO3 is the first step for simulating elevated winter particulate NO3 in the Capital Region. As discussed in Chapter 3, there are numerous sources of NOx emissions within the Capital Region, including mobile sources, industrial sources and even power generating units to the west. Thus, there is ample availability of NOx to be converted to HNO3.

There are several pathways to convert NOx to HNO3. In this study we used the CB05 chemical mechanism (Yarwood et al., 20056) within the CMAQ PGM model. Most NOx is emitted as NO (Nitric Oxide) that gets converted to NO2 (Nitrogen Dioxide), which in turn can form HNO3 (Nitric Acid) or gets converted to NO3

- (Nitrate Radical) or NTR (Organic Nitrate, RNO3) that in turn can form HNO3. Several of these reactions are as follows:

NO2 + OH --> HNO3

NO3 + HO2 --> HNO3

NTR + OH --> HNOS + other products

6 http://www.camx.com/publ/pdfs/cb05_final_report_120805.pdf

http://www.camx.com/publ/pdfs/cb05_final_report_120805.pdf

17

The three HNO3 formation reactions above involve the hydroxyl (OH) or hydroperoxyl (HO2) radicals that are produced by photochemistry. Photochemistry is the main source of ozone formation that involves VOC and NOx precursors in the presence of sunlight. During the Capital Region winter PM2.5 episodes, sunlight will be scarce and radical concentrations would be expected to be low and depend on the availability of incoming ozone concentrations.

Another gas-phase HNO3 formation reaction is the reaction with aldehydes [formaldehyde (FORM) and acetaldehyde (ALD2)]:

FORM + NO3- --> HNO3 + HO2 + CO

ALD2 + NO3- --> C2O3 + HNO3

One of the primary sources of aldehydes in the Capital Region will be from mobile sources, especially with the cold temperature conditions resulting in the catalysts taking longer to warm up and be more effective at reducing the mobile source VOC emissions. Another source of aldehydes is refineries and petrochemical plants, such as occur to the northeast of Edmonton.

At night, HNO3 formation occurs through N2O5 (Dinitrogen Pentoxide), which will disassociate in the presence of sunlight. N2O5 is first formed through a reaction between NO2 and NO3

-, and then in the gas-phase forms HNO3 through reaction with water vapor (H2O):

NO2 + NO3- --> N2O5

N2O5 + H2O --> HNO3

N2O5 + H2O + H2O --> HNO3

Thus, the correct simulation of water vapor concentrations by the meteorological model becomes important. If the WRF meteorological fields are too dry, that may inhibit HNO3 formation.

The availability of NO3- radical is a key component in the nighttime HNO3 formation pathway, as

well as the pathway through aldehydes. NO3- radical is formed through reactions involving NO2

and ozone. As we would not expect there to be very much ozone formation in the Capital Region during the winter PM2.5 episodes, the simulation of ozone transport into the region and the ability to simulate the observed ozone concentrations is very important in simulating the elevated particulate NO3 levels in the region.

The final HNO3 formation pathway in the CMAQ chemistry modules is the heterogeneous reaction probability (Ƴ) of N2O5 that is a function of temperature, relative humidity (RH), particle composition and phase state. As described by Davis, Bhave and Foley (2008), the reaction probabilities are used for defining aqueous ammonium sulphate and ammonium nitrate formation. We believe that the heterogeneous particulate NO3 formation through the Ƴ reaction probabilities will be a key particulate NO3 formation pathway for the Capital Region winter NO3 episodes.

18

2.4.2 Availability of NH3 Concentrations to form Particulate NO3

In 2011, 91% of the ammonia emissions in Canada came from the agricultural sector that is due mainly to livestock and fertilizer that are large source categories in Alberta resulting in Alberta accounting for 26% of the total ammonia emissions in Canada7. According to the NPRI, there are also some industrial facilities in Edmonton that emit ammonia as well as ammonia emitted from the oil sands region to the northeast of the Capital Region. Ammonia emissions are also emitted from mobile sources with 3-way catalyst control technology. Ammonia emissions are more uncertain than other criteria air contaminants (e.g., NOx and SO2). There are very few if any NH3 measurements in the Capital Region so verification of the NH3 emissions inventory will be difficult and this may be a subject of sensitivity tests.

2.4.3 Equilibrium between Particulate NO3 and Gaseous HNO3

As noted above, the WRF simulation of the temperature and absolute humidity conditions during the Capital Region PM2.5 episodes will be important for both the gas-phase and heterogeneous-phase particulate nitrate formation pathways, as well as the equilibrium between particulate NO3 and gaseous HNO3.

7 http://www.ec.gc.ca/indicateurs-indicators/default.asp?lang=en&n=FE578F55-1

http://www.ec.gc.ca/indicateurs-indicators/default.asp?lang=en&n=FE578F55-1

19

3.0 BASE CASE EMISSIONS INPUT

The development of the harmonized emission inventory for air quality simulation of the Capital Region is documented in this Chapter. The inventory data sources and inventory development methodology are described, including the approach for reconciliation of emissions among the various databases used, and the various updates to the emissions modelling system and databases. Summary tables of the final emission inventories are also provided.

3.1 Industrial Point Sources

A number of emissions inventories in Alberta have been updated and refined since 2011, providing a solid base for developing a complete, province-wide emissions inventory. Harmonization of these databases to eliminate double counting, remove historical/decommissioned sources and updated stack parameters and/or emissions were performed as part of this study. The following point source emissions inventories were consolidated to develop the final inventory.

a. Stakeholder updates from 2008 ESRD Industrial Survey to 2010 – These updates were provided specifically for the project to provide the most up-to-date emissions in the Capital Region and Red Deer areas. Updates were received from individual stakeholders and used to replace 2008 emissions from the Industrial Survey.

b. CEMA / LAR area 2010 emissions inventory from the SAOS modelling – Emissions were updated from ESRD Southern Alberta Oil Sands (SAOS) project performed in 2013. The majority of the inventory is within the boundary of the Cumulative Effects Management Association (CEMA) and Lower Athabasca Region (LAR) with limited sources extending just outside of the CEMA boundary.

c. 2008 ESRD Industrial Survey – The province wide industrial survey (AAEI) was updated with emissions for 2010 when provided in the Capital Region and Red Deer areas.

d. 2006 NSRP Inventory – Emissions were updated using data from the ESRD North Saskatchewan Region Plan (NSRP) modelling project. This inventory covers Alberta north of SSRP region and outside of CEMA/SAOS region. Emissions from this inventory are based on the 2006 ESRD Industrial Survey and the Environment Canada 2006 Emissions Inventory.

e. 2008 SSRP/CRAZ Inventories – Emissions data from the ESRD South Saskatchewan Regional Plan (SSRP) modelling project and Calgary Regional Airshed Zone (CRAZ) were used to update the inventory for the SSRP area in southern Alberta. Emissions from this inventory are based on the 2008 ESRD Industrial Survey and the Environment Canada 2006 Emissions Inventory.

In order to validate the point source emissions inventories used, the supplementary data sources listed below were included in the harmonization process. While no emissions from these databases were used they did provide valuable input in terms of both source locations and current operations.

20

a. 2006 – 2010 National Pollutant Release Inventories (NPRI) – Full facilities emissions totals and locations are provided. Multiple years of NPRI data were used based on the emission original emission inventory dates and the model year desired.

b. Alberta Energy Regulator (AER) ST50 Gas Plant List – Another publicly available document, the AER gas plant list details all gas plants in the province and provides locations, operators, licensees, and production figures. This tool is very valuable for instances where facilities may have changed ownership/operators.

Table 3-1 provides a brief summary of the point source emissions inventories used and the final province wide emissions inventory totals. A more detailed emissions summary is available in Appendix B and details the breakdown of each inventory by file.

Table 3-1. Province-wide point-source emissions summary by inventory.

Emission Inventory NOx

[tonne/yr]

SO2

[tonne/yr] NH3

[tonne/yr] PM2.5

[tonne/yr]

PM10

[tonne/yr]

CO [tonne/yr]

VOC [tonne/yr]

Alberta Environment Industrial Survey 2008 + 2010 updates (province-wide)

151,444 240,867 5,712 5,372 9,517 51,381 8,524

CEMA / LAR area 2010 66,365 115,790 0 3,797 0 46,110 2,777

SSRP 2008 Inventory 91,629 16,825 411 1,533 1,844 79,844 47,570

NSRP 2006 Inventory 169,711 20,817 528 3,747 4,253 176,332 129,657 Province Wide Total 477,743 394,299 6,651 14,449 15,614 353,667 188,527

3.1.1 Data Consolidation

Emission updates for 2010 were received from Individual stakeholders as updates to the 2008 ESRD Industrial Survey spreadsheet. Each submission was reviewed for changes to facility stack parameters and stack emissions from the stakeholder reporting. Any changes in emissions or parameters were replaced in the 2008 Industrial Survey with the changes to each stack noted separately as well. This process created the updated Industrial Survey database, referred to as AAEI2008+ in this report.

All emissions inventories were first converted to excel spreadsheet/comma separated value files so that all sources could be evaluated using the same methods. Once converted, all files/point-sources were plotted spatially using GIS software with stack parameters, emissions, and source properties retained for comparison. This produced a full, un-harmonized inventory which facilitated directly comparison of any sources with a spatial reference. Geographic boundaries, road networks, and satellite images were also incorporated into the GIS database to provide additional context if required during evaluation.

It should be noted that earlier SMOKE file formats such as IDA and ORL imposed character limits on most fields. This resulted in truncation of source latitude and longitudes, as well as a lack of source information with respect to operator, source type, and source name.

21

All emissions inventories were plotted as multiple groups by maintaining any file splits or divisions previously used, such as sector sub-groups. This allowed for all components of each inventory to be easily compared and harmonized as individual elements or as an entire inventory.

3.1.2 Updates to ESRD 2008 Industrial Survey

Anthropogenic emissions in the Capital Region are very important for PM modelling. The stakeholder updates to the 2008 ESRD Industrial Survey were focused in this area and provided the most up to date emissions possible for the model year 2010.Table 3-2 provides a summary of these updates for important CAC point source emissions.

Table 3-2. Summary of stakeholder updates from 2008 Industrial Survey to 2010 for point sources.

NOx SO2 NH3 PM2.5 PM10 VOC

# of sources updated 144 84 32 151 157 153

Province-wide 2008 Survey emissions prior to stakeholder updates (t/y)

193,549.2 358,696.7 7,021.9 6,946.2 128,11.4 103,70.0

Total emissions changes by stakeholder updates to 2010 (t/y)

-909.7 -4,943.2 -261.9 -11.9 -17.8 -15.1

% Difference in province-wide survey emissions

-0.47% -1.39% -3.80% -0.17% -0.14% -0.17%

The spatial distribution and relative contributions from these industrial point sources is also important for understanding emissions in the area. Figure 3-1 and 3-2 illustrate the location of NOx and SO2 emission sources from AAEI2008+ around Capital Region and their relative emission levels. Sources with annual emissions >250 tonne/yr are labeled with company and facility name. It should be noted that the Transalta Wabamun Generating Plant was decommissioned during 2010. Modelling carried out during the beginning of 2010 includes this source which was shut down mid-year.

For both NOx and SO2 the major contributors in Capital Region are the thermal electric power generating plants to the west of the city, followed by the industrial heartland and the Strathcona area to the northeast of Edmonton.

22

Figure 3-1. NOx emissions >250 tonne/yr around Capital Region. Yellow stars indicate air quality monitoring stations in Edmonton.

23

Figure 3-2. SO2 emissions >250 tonne/yr around Capital Region. Yellow stars indicate air quality monitoring stations in Edmonton.

Table 3-3 provides a summary of emissions within the Capital Region alone and compares these emissions to the province-wide anthropogenic point-source total. It should be noted that point source emissions in the Capital Region may account for an inflated percentage of the actual province wide total for some pollutants due to increased data quality of these sources. For example, the Industrial Survey focuses largely on the Capital Region and includes both NH3 and PM10 in the inventory. Emissions from the CEMA/LAR database do not provide these emissions however, increasing the impact from the Capital Region on the province wide total.

Table 3-3. Capital Region point source anthropogenic emissions summary from updated province wide total. NOx [t/y] SO2 [t/y] NH3 [t/y] PM2.5 [t/y] PM10 [t/y] CO [t/y] VOC [t/y]

Capital Region 71,058 81,831 2,683 2,841 5,360 17,721 1,647 % of Province-Wide Total 14.87% 18.91% 40.04% 19.32% 33.40% 4.93% 0.87%

3.1.3 Inventory Harmonization

Once all inventories were consolidated and plotted spatially the harmonization followed a careful procedure to eliminate any double counting and valid emissions sources. This section outlines the process that was used and provides specific examples of some situations that required consideration.

24

3.1.3.1 Establish Emission Inventory Hierarchy

In order to develop the province-wide inventory, a data hierarchy was developed to allow overlapping or conflicting sources to be resolved using the data for what is believed to be the most accurate emissions database for the final harmonized database. The hierarchy used is as follows:

a. Stakeholder updates from 2008 ESRD Industrial Inventory to 2010 b. CEMA/SAOS 2010 emissions c. 2008 ESRD Industrial Survey Inventory d. 2006 NSRP Inventory e. 2008 SSRP/CRAZ Inventories

In general, the ESRD Industrial Survey with 2010 updates (AAEI2008+) was selected as the top tier data source. The inventory contains full stack parameters for all sources listed and was updated specifically for this project for the most impactful sources. The only exception to this rule in the data hierarchy was inside of CEMA/LAR region where industrial emissions were updated through CEMA studies and ESRD’s SAOS project. The main focus of the Industrial Survey is Capital Region and Red Deer. While a number of sources from AAEI2008+ are included inside of the CEMA boundary, the ESRD SAOS project in 2013 provides a more detailed emissions inventory of existing (2010) emissions for the entire CEMA area. This inventory was selected as the only source of emissions within the CEMA/LAR region.

For any sources that are outside of the CEMA boundary, the Industrial Survey was preferred and the top tiered database. The NSRP2006 and SSRP2008 inventories were considered to be secondary data sources for the province-wide inventory.

3.1.3.2 Survey Source Harmonization

The NSRP Inventory and SRRP/CRAZ region inventory both included previous versions of the ESRD Industrial Survey as well as the EC2006 inventory as their major data sources. As such, all emission sources in these inventories are divided into “Survey” sources, sources adapted from the ESRD Industrial Survey, and “noSurvey,” with sources from the EC2006 Inventory which could not be updated from that time.

Given that the “Survey” sources were originally from the ESRD Industrial Survey, these sources could be updated with the most recent Industrial Survey. While most of these sources are identical, survey sources were updated on a facility wide basis. This mean any sources included in the 2006 survey but no longer existing in the 2008+ survey were not kept for the province wide inventory. Any emissions from removed sources are assumed to be included in the most current survey from other stacks or are decommissioned.

3.1.3.3 noSurvey Source Harmonization

Emissions sources in the NSRP and SSRP/CRAZ Inventories that were developed from the EC2006 Inventory were handled separately from the Survey emissions. These sources typically

25

did not contain detailed records, with many missing stack parameters, source descriptions, or usable source names/ids. However, these emissions are a major component of the province wide total and are derived from credible sources that need to be included.

Since these sources are fairly outdated, considerable effort was made to validate sources for their inclusion in the final inventory. Given the large number of sources in each of these inventories the sources needed to be sub divided to allow for a more detailed approach. Four major groups were selected;

noSurvey small - sources with all emissions <1 ton/year

noSurvey medium - sources with emissions 1-100 ton/year,

noSurvey VOC noLarge - VOC only sources <100 ton/yr, and

noSurvey Large - sources with any emissions >100 ton/yr.

Table 3-4 provides a summary of these divisions for the NSRP inventory only. The same divisions were applied to the SSRP inventory but are not detailed here. NH3 was not covered in the noSurvey sources and as such is left out of this summary table.

Table 3-4. NSRP2006 inventory noSurvey source divisions.

noSurvey Inventory # of

Sources NOx

[tonne/yr] SO2

[tonne/yr]

PM2.5

[tonne/yr]

PM10

[tonne/yr]

CO [tonne/yr]

VOC [tonne/yr]

noSurvey small 3,473 176 46 91 92 298 18

noSurvey medium 1,844 19,227 5,937 1,196 1,200 25,672 445

noSurvey VOC noLarge 41,856 40,117

noSurvey Large 510 91,870 6,012 511 513 93,173 25,141

As shown, the majority of emissions are from a limited number of sources outlined in the “noSurvey Large” division. It was decided that these sources warranted further investigation in order to validate emissions and prevent double counting of sources which could have a large impact on the province wide total. This was done by comparing the noSurvey Large sources to the NPRI 2006-2010 inventories, the AER ST50 Gas Plant List, and the ESRD Industrial Survey using GIS software.

AAEI2008+ - The first step was to compare noSurvey Large sources to the 2008+ ESRD Industrial Survey. This was first done on a spatial basis in GIS, followed by comparing stack parameters or emissions if applicable, and then on a google-link basis. If it could be reasonably assumed that the noSurvey source was representative of a facility included in the Industrial Survey then it was recommended these sources be eliminated from the province wide total. Using the Industrial Survey as the top tier data sources indicates that the emissions reported represent all relevant emissions from the entire facility. Including noSurvey emissions which can be reasonably assumed to be from an Industrial Survey facility would contradict this data hierarchy.

26

NPRI 2006-2010 - Once it was confirmed that these EC sources were not accounted for in the Industrial Survey they were next compared to the publicly available NPRI datasets. The NPRI facilities were all plotted spatially as well to compare to the noSurvey Large sources. If a noSurvey source could be matched with confidence to both the 2006 and 2010 NPRI datasets it could be confirmed that this source was present in both 2006 when the EC Inventory was developed, and remained active in 2010, the modelling year for this study. Conversely, if a noSurvey source could be matched with confidence to the 2006 NPRI but was not present again in 2010, it could be reasonably assumed that this specific source location was no longer active. These sources were removed from the noSurvey inventory.

AER ST50 - Remaining noSurvey sources were finally compared to the AER ST50 Gas Plant List. Many of these facilities do not directly report to the NPRI, with their emissions likely grouped into another reported facility from the same operator. The AER list allowed includes additional information regarding not only the facility name, but also the licensee and operator. In cases where plants have changed ownership, this additional information allowed a number of sources to be correctly confirmed.

Any noSurvey sources which could not be matched with the NPRI datasets, AER ST50, or AAEI were grouped as independent sources. With no method to validate or eliminate these sources they were kept in the final province wide inventory.

3.1.3.4 Non-Point Source Harmonization

Included in the ESRD Industrial Surveys are non-point source emissions, often reported as lumped fugitive emissions for an entire facility. These are included in the NSRP and SSRP/CRAZ inventories from the 2006 Industrial Survey. Using the same logic as for the point-source survey sources, non-point source emissions from the updated Industrial Survey were used to replace the NSRP and SSRP/CRAZ inventory non-point source emissions.

In the Industrial Survey, non-point source emissions are often reported using either 4 co-ordinates, 2 co-ordinates and an area, or 1 co-ordinate and an area. To facilitate the inclusion of these sources in SMOKE however it was decided that they would be treated as virtual point sources. This simplifies the modelling of these sources as well as the conversion into SMOKE ready format.

3.1.4 Province-Wide Inventory Finalization

Spatially plotting all of the emission sources for the province wide inventory is in itself a preliminary QA of sources. It allows for a quick comparison of sources that can help identify potential overlapping or double counting sources. With the limited data available due to truncation from .ida files this can be essential in identifying issues with the data. Sources could be evaluated by first comparing spatial locations, then using source names, SCC codes, and finally stack parameters if necessary. Figure 3-3 below provides one example of spatial differences between the existing NSRP inventory and the updated inventory. Plotting both sets of data together allows for easy identification of repeated facilities even with different numbers

27

of sources and source names. The red circle also indicates the accuracy of the AAEI2008+ data without truncation, with the main stack emissions plotted directly on the main stack of the facility.

Figure 3-3. Imperial Oil Strathcona facility plotted spatially with GIS software. NSRP2006 inventory points shown in yellow, AAEI2008+ shown in blue. Red circle indicates the actual location of the facility's main stack. (Imagery © 2014 Google)

In both the NSRP and SSRP/CRAZ inventories there is a separate grouping of sources listed as “matchSurvey_noSCCmatched” which is derived from the EC2006 inventory. When these inventories were originally developed these sources matched the Industrial Survey by location and name but did not have matching SCC codes. The emissions from these sources are therefore irrefutably from the facilities covered by the Industrial Survey. It was concluded that the updated Industrial Survey accounted for these sources, as the stakeholders are required to report all emissions from each facility. Including these files in the final province-wide inventory would contradict the use of the Industrial Survey as the top tier data source. With this consideration, these sources were removed from the province wide inventory for both point and non-point sources.

NSRP inventory includes small UOG (SUOG) sources that are not included in the Survey data. ESRD advised that their emissions be included in the Capital Region modelling. The SUOG dataset has high uncertainties (locations and emission rates) and many known issues (e.g. based on out-dated equipment list, potential duplication of NPRI sources, inclusion of facilities no longer operating, exclusion of newer sources, etc.). However, their emissions are significant

28

(14% of NOx and 58% of VOC at provincial level) and this warrants the inclusion of these sources.

3.2 Non-Point Sources

The non-point source emissions inventory is based on the SAOS 2010 inventory with some updates to on-road mobile emissions. The updates include improvements of spatial allocation and refinement of Edmonton on-road mobile emissions based on information provided by the City of Edmonton.

3.2.1 Comparison of Available Road Networks and Emissions

The SAOS 2010 on-road mobile emissions were based on the EC 2006 on-road inventory projected to the year 2010 for the entire province. Specifically, the growth in activity as vehicle kilometers travelled (VKT) was based using an extrapolation of the Statistics Canada CANSIM 2002-2009 data out to 2010. The effect of changes in fleet mix and other parameters for was accounted for using the MOBILE6.2C model. In the SAOS study, the EC 2006 road network was used as spatial surrogates. In order to improve the spatial distribution of these emissions, the road network was updated using the most recent CanVec road network available from Natural Resources Canada. CanVec is a digital cartographic reference product produced by Natural Resources Canada (NRCan). CanVec originates from the best available data sources covering Canadian territory and offers quality topographic information in vector format that complies with international geomatics standards.

The CanVec updated shapefiles provide detailed information for every road in Alberta including all major highways and interior roads. A comparison of the 2006 EC road network and the updated Canvec road network for Edmonton and Red Deer are provided in Figure 3-4 and Figure 3-5.

Figure 3-4. Edmonton road networks from CanVec 2010 (left) and EC 2006 (right)

29

Figure 3-5. Red Deer road networks from CanVec 2010 (left) and EC 2006 (right).

While the updated CanVec road network is a significant improvement to the spatial allocation of transportation emissions, a secondary emissions inventory was available for use inside of the City of Edmonton. This additional data was provided by the City of Edmonton’s Transportation Operations division who used the CALMOB6 model (Calibrated MOBILE6), which their fuel consumption and emissions program used for the transportation master plan. CALMOB6 is a custom tool for the City of Edmonton that is based on the US EPA’s MOBILE6 model. It uses real traffic counts from inside the city along with urban travel forecasting models, and local parameters such as road grade and ambient weather conditions. A comparison of the road networks used for each model is provided in Figure 3-6.

Figure 3-6. CanVec 2010 road network (left) and CALMOB6 Links (right).

Upon initial inspection it appears that the CALMOB6 network lacks the spatial distribution and road coverage of the CanVec road network. As shown in Table 3-5, the CALMOB6 emissions for the City of Edmonton were typically much lower than the emissions extracted from the SAOS inventory that were spatially allocated using the CanVec road network. A direct comparison of road network densities is also provided in Figure 3-7 and shows the apparently limited coverage in the CALMOB6 data.

30

Table 3-5. Typical January weekday total emissions comparison for Edmonton.

NOx [tonne/d]

PM 2.5 [tonne/d]

SO2 [tonne/d]

CO [tonne/d]

Canvec + SMOKE 11.34 0.35 0.25 140.16

CALMOB6 7.63 0.16 0.06 148.69

Figure 3-7. Direct comparison of CALMOB6 (blue) and CanVec (gray) road networks in residential areas.

However, consultation with the Transportation Operations division indicated that the CALMOB6 link based network does include the traffic and emissions from smaller residential streets by grouping them in with larger links. This results in a road network that appears to have lower resolution than required but actually accounts for all traffic within the city. Further comparison of Edmonton emissions was done by plotting daily total emissions spatially using the 1.33 km modelling grid cells. Figure 3-8 illustrates the distribution of emissions from both the CALMOB6 data and the SAOS/CanVec data.

31

Figure 3-8. NOx emissions distribution around City of Edmonton from SAOS/CanVec (left) and CALMOB6 emissions model (right).

Since the CALMOB6 emissions data is specifically tailored to the City of Edmonton it is expected to provide a better spatial representation of emissions than the CanVec province-wide road network. CALMOB6 show a large fraction of emissions in the downtown core and along major highway and truck routes. Conversely, the CanVec road network results in a higher distribution in residential areas and parts of the city with a higher road density. While the CanVec road network works very well for distributing emissions on a large scale, the CALMOB6 distribution is more spatially accurate for the City of Edmonton. This spatial distribution helps explain the large difference in total emissions between the two methods. The high density of roads within the City of Edmonton likely draws a larger portion of the province wide total than it should. The surrogate tool only takes into account length of these residential roads and no input of actual traffic levels. This is why the SAOS/CanVec method shows higher emission rates in residential areas (Figure 3-8) which are likely over estimated.

3.2.2 Combining CALMOB6 with Province Wide Transportation Emissions

Once CALMOB6 was selected as the source of transportation emissions within the City of Edmonton, a method had to be developed to merge this data with the SAOS/CanVec emissions for the remainder of the province. This presented a number of technical challenges in terms of both converting the emissions into a SMOKE usable format, as well as harmonizing the data smoothly without introducing possible spatial or temporal dissonance.

32

CALMOB6 emissions data was generated using the City of Edmonton boundary. However, the city boundary often cuts directly through the 1.33 km modelling grid imposed on the Capital Region. To simplify the merging of the data sources, grid cells needed to be assigned entirely to either the CALMOB or the CanVec method. In many cases this meant discarding CALMOB6 emissions when they could not account for all of the roads in a given 1.33 km grid cell. Figure 3-9 illustrates a series of grid cells where this was required.

Figure 3-9. Discarded CALMOB6 data along the Yellowhead Highway.

Once grid cells were properly distributed between the two emission sources, the roads used for the SMOKE spatial surrogate outside of Edmonton were re-processed and re-run. This included the removal of all roads from inside the CALMOB6 grid cells, as well as the removal of the North-West portion of Anthony Henday Dr., the ring road around the city. This is due to the fact that this section of the road was not completed and opened to the public until 2012, while the model year being evaluated is 2010. Further discussion of the spatial surrogates used is provided in Section 3.4.1.

The final step in re-processing the SMOKE emissions outside the city was to scale down emissions from residential areas on the city outskirts. Both St. Albert and Sherwood Park have residential road densities similar to residential areas inside the city. When comparing CALMOB6 to CanVec + SMOKE data it was evident that these areas were being allocated a higher share of the province wide totals than expected. Since CALMOB6 data is not available for St. Albert or Sherwood Park a scaling factor was developed based on the ratio of Canvec + SMOKE based emissions for residential areas to CALMOB6 emissions for the same area. Figure 3-10 shows the representative residential area from inside city boundary (green) that was used to scale the residential areas outside (blue).

33

Figure 3-10. Residential areas used for scaling CanVec + SMOKE emissions based on CALMOB6 emissions.

3.2.2.1 Development of CALMOB6 Temporal Profiles

While it was shown that the CALMOB6 data provides a better spatial distribution than the CanVec + SMOKE data, the temporal distribution provided is very coarse, dividing daily emissions into 4 separate 1 hour periods, and one 20 hour emission rate which is evenly divided over the remainder of the day (off-peak time). This results in too high overnight emissions and too low daytime emissions. It is necessary to re-evaluate temporal allocation of the CALMOB6 data not only to provide more accurate emissions in the city core, but also to prevent any edge effects from impacting the results. If two completely contrasting temporal profiles were to be used on the edge of the city it could produce discontinuities at the boundaries of the two methods.

Temporal profiles for emissions within the City of Edmonton were developed as two separate classes; heavy and light vehicles. Emissions from the CALMOB6 data were provided by vehicle class that can be easily grouped into these two categories. Table 3-6 lists the vehicles classes that were included in each. All buses were grouped into the light duty vehicle class based solely on their spatial distribution. Specifically, bus traffic followed the light duty spatial profile more closely than heavy duty vehicles, with no emissions coming from highways.

34

Table 3-6. CALMOB6 vehicle types used in each temporal/spatial profile. Light Profile Heavy Profile

LDV Mini Bus-SS HDV2b

LDV Economy BUS-SL HDV3

LDV Large BUS-TN HDV4

LDT1 BUS-TO HDV5 LDT2 BUS-TL HDV6

LDT3 BUS-TS HDV7

LDT4 HDV8a

HDV8b

To develop the light duty vehicle temporal profile, traffic counts from the City of Edmonton’s Transportation Operations division were used. Traffic counts from 5 different major streets were provided which were selected to give a representative profile for the city. The streets selected were;

75th St. North of 86 Avenue

118 Avenue West of 78th St.

178th St. North of 89 Avenue

87 Avenue West of 178th St.

Capilano Bridge

The locations of these traffic counts are provided in Figure 3-11. It should be noted that two of these counts are for a similar geographic location, both bordering the West Edmonton Mall.

Figure 3-11. Traffic count locations provided by the Transportation Operations Division. (Map data © 2014 Google)

35

While these traffic counts did not provide vehicle distributions, the data from CALMOB6 indicated that 94% of total vehicles accounted for in the model are light duty vehicles. Therefore it was assumed that the profile developed from these traffic counts could be applied to the light duty vehicle class as a whole.

To develop the temporal profiles, the hourly traffic across the winter months (Jan-Mar, Oct-Dec) was averaged across all five roads. This was done for both weekday and weekends to develop two distinct diurnal profiles that are based on real data from the City of Edmonton. The overall weekday diurnal light duty traffic profile is shown in Figure 3-12 while the weekend diurnal light duty traffic profile is shown in Figure 3-13. Both of these profiles agree very well with the US EPA standard profiles that were used for areas outside of Edmonton. This eliminates any possible boundary or edge effects that may have occurred if conflicting profiles were implemented.

Figure 3-12. Weekday diurnal traffic profile from City of Edmonton traffic counts.

36

Figure 3-13. Weekend diurnal traffic profile from City of Edmonton traffic counts.

A separate profile for heavy duty emissions was developed using the Edmonton Region External Truck/Commodity Survey8 results. While this survey was limited to major highways entering and exiting the city, the spatial distribution of heavy duty emissions from CALMOB6 shows the majority of emissions are from these roads. Figure 3-14 provides the diurnal distribution reported in this survey for weekday periods only. Without a weekend profile available, it was determined that the heavy duty traffic weekday diurnal profile could also be applied to the weekend periods. The final heavy duty profile applied for modelling in SMOKE slightly modified this profile to smooth out hourly transitions.

8 Ishani, M., 2004. Edmonton Region External Truck/Commodity Survey: Alberta Transportation & The City of

Edmonton.

37

Figure 3-14. Edmonton Regional External Truck/Commodity Survey diurnal truck distribution.

With new diurnal profiles established, it was also necessary to consider weekly variations in traffic volume. The City of Edmonton traffic counts were once again used to develop a light duty traffic weekly profile by comparing Saturday and Sunday emissions to weekday emissions. This comparison could not be done on a full seven day basis due to the fact that CALMOB6 emissions were for a “typical” weekday, not a specific day of the week. While all 5 traffic counts were used to develop diurnal profiles, this could not be done for weekly profiles due to the influence of West Edmonton Mall on two of the counts. These counts provided inflated weekend traffic and were excluded from developing the weekday to weekend comparisons.

Due to the limited information on vehicle distributions, a heavy duty weekly traffic profile could not be developed from the data available. Instead, the US EPA standard heavy duty weekly traffic distribution was applied. This profile is the standard for traffic modelling and is based on extensive traffic counts in the US.

The weekly profiles used for both light duty and heavy duty are summarized in Table 3-7.

38

Table 3-7. Weekly traffic profiles summary.

Day of the Week

% of Average Weekday Traffic

Light Duty Profile Heavy Duty Profile

Weekday 100% 100%

Saturday 89.4% 93.8%

Sunday 72.8% 31.3%

The CALMOB6 data used for modelling was provided by the City of Edmonton for a typical January weekday. Additional data was provided to produce a full month to month emissions profile, however due to time limitations of the modelling process this profile was not implemented.

3.2.2.2 Preparation of CALMOB6 Emissions for SMOKE Processing

Since CALMOB6 emissions are only provided as daily totals for a “typical” day, the weekly and monthly profiles need to be included in the calculation when determining an annual total to input to SMOKE. Given the method which SMOKE uses to process emissions from an annual total, the typical weekday emissions could not simply be multiplied by 365 to obtain emissions for the year. The weekly profiles selected needed to be included by accounting for the number of Saturdays and Sundays in the given time period. For example, 2010 light duty annual emissions were calculated as;

) ))

) ))

This was done for both light duty and heavy duty vehicles, with the resulting emissions converted to .FF10 format. Once the files were processed by SMOKE the weekday and weekend emissions for heavy and light vehicles were extracted and validated.

3.3 Natural Sources

3.3.1 Biogenic Emissions

The inventory of biogenic emissions was generated using the Model of Emissions of Gases and Aerosols from Nature (MEGAN) version 2.1. The MEGAN model estimates net emission of gases and aerosols from terrestrial ecosystems into the atmosphere. MEGAN driving variables include weather data, Leaf Area Index (LAI), plant functional type (PFT) cover and compound-specific emission factors that are based on plant species composition. Global distributions of land cover variables such as emission factors, leaf area index and plant functional types are available for spatial resolutions ranging from ~1 to 100 km (downloadable files from http://cdp.ucar.edu).

The MEGAN model was applied using the daily meteorology (temperature and solar radiation) from the WRF model outputs to generate day-specific biogenic emissions for the 3 month

39

modelling period in 2010 on the 36/12/4/1.33 km domains. To prepare for the MEGAN modelling, ArcGIS software will be applied to generate MEGAN input files.

3.3.2 Fire Emissions

Fire emissions were based on the Fire Inventory from NCAR (FINN) version 1 dataset, which can be downloaded from http://bai.acd.ucar.edu/Data/fire/. The global dataset contained daily emissions for each satellite pixel, which represented an area of approximately 1 km2. Emission species included NO, NO2, PM2.5, CO, and NMOC’s speciated into MOZART-4 species for six fire types – tropical, temperate, and boreal forests, cropland, shrublands, and grasslands. Fire points within 5 km of one another are assumed to be part of the same fire and assigned properties of a larger fire.

The daily fire emissions were then processed for the winter modelling period using an updated version of EPS3 version 3.20 (developed and maintain by ENVIRON). EPS3 incorporated the Western Regional Air Partnership (WRAP) methodology to temporally and vertically allocate the fire emissions. Temporally, the same diurnal profile can be applied to all fires such that emissions are highest in the early afternoon and lowest at night. Vertically, a fraction of each hour’s emissions are assigned to the lowest layer; the rest is distributed into multiple point sources directly above with one point assigned to each CMAQ layer between the plume bottom and plume top, weighted by the thickness of each layer. The fraction in layer 1 and the plume bottom and top are all dependent on the hour of the day and size of the fire. Emission outputs were formatted to CMAQ in-line emissions format. Note that fire emissions were not important contributors to the winter elevated PM2.5 episodes in the Capital Region.

3.4 SMOKE Emissions Modeling

Emissions modelling for the project utilized the Sparse Matrix Operator Kernel Emissions (SMOKE) modelling system version 3.1 (released September 2012). The overall approach for the SMOKE modelling conducted for the project is described below, including the spatial allocation of inventory data, chemical speciation of VOC emissions sources and merging of the inventory database described previously. The general emissions processing steps for the preparation of emission inventories for air quality modelling include:

Input and QA of EI data.

Chemical speciation: Emission estimates of criteria pollutants must be speciated for the particular chemical mechanism employed in the air quality model. The Capital Region CMAQ modeling used the CB05 chemical mechanism.

Temporal allocation: Annual or seasonal, emission estimates are resolved hourly for air quality modelling. These allocations are generally determined from the particular source category, specified by SCC codes. Monthly, weekly and diurnal profiles are cross-referenced to SCC codes to provide the appropriate temporal resolution.

Spatial allocation: Regional or province level emission estimates must be spatially resolved to the modelling grid cells (e.g., 1.33 km for the Capital Region) for air quality

40

modelling. The spatial allocation is generally accomplished using surrogates cross-referenced to source categories.

Output of air-quality model-ready files.

QA/QC of emissions modelling.

The Capital Region inventories were converted to the SMOKE-ready FF10 and IDA formats. SMOKE uses ancillary data to perform temporal, spatial and chemical allocation of emissions according to source category. This study used the SAOS SMOKE ancillary data and setup, and generated emissions inputs for CMAQ version 5.0.1 with Carbon Bond 05 (CB05) chemical mechanism species. The point sources emissions were generated for in-line plume rise calculation in CMAQ.

3.4.1 Spatial Surrogates Development

As defaults in the SMOKE model, non-point source emissions are spatially allocated to modelling grid cells using gridded spatial surrogates based on provincial level population, socioeconomic and other appropriate data. These surrogates are usually developed using U.S. EPA’s Spatial Allocation Tool which combines GIS-based data (shapefiles) and modelling domain definitions to generate the appropriate gridded surrogate data sets for use with SMOKE. The provincial-level emission estimates are assigned specific surrogates for gridding by cross-referencing Source Classification Codes (SCCs).

Environment Canada’s 2006 socioeconomic shapefiles were used to generate the basic spatial surrogates for this study, with an update to transportation surrogates (paved and unpaved roads) using the new CanVec data and an update to construction dust surrogates. Spatial surrogate tool was used to generate gridded surrogates for 36/12/4/1.33 km domains. For 1.33km domain, only areas outside Edmonton were applied.

3.4.1.1 Edmonton On-Road Spatial Surrogates

Inside of the City of Edmonton, new spatial surrogates needed to be developed from the CALMOB6 data to integrate the emissions with SMOKE. This needed to be done for each of the vehicle classes to properly distribute emissions from each SCC.

Spatial surrogates were handled first by assigning to each grid cell the percentage of the total emissions for the area from each vehicle class. While there are slight variations from pollutant to pollutant in terms of emission distribution, it was found that the overall profile for heavy duty and light duty remained fairly consistent. As shown in Figure 3-15, the distribution of heavy duty emissions is uniform across pollutants.

41

Figure 3-15. HDV emissions distribution for NOx (green), PM2.5 (blue), and SO2 (purple).

To generate the final surrogates, the emission distribution for NOx, PM2.5, and SO2 were averaged together. Figure 3-16 shows the final spatial surrogates for light duty and heavy duty emissions. These surrogates effectively communicate the patterns produced by the CALMOB6 emissions, with heavy duty contributions coming largely from the highways and the downtown core, while light duty emissions are well distributed, especially in residential areas.

Figure 3-16. Light duty spatial surrogate (left) and heavy duty spatial surrogate (right).

42

3.4.1.2 Outside of Edmonton On-Road Spatial Surrogates

Figure 3-17Figure 3-17 provides the province-wide CanVec paved road network, which provides the most updated on-road network information up to 2010. CanVec road network contains different road types, including highway/freeway, arterial, collector, ramps, local, alleyway, and service roads. Based on Edmonton’s real traffic emissions studies, we applied ratios of 85% highway and 15% other roads as the best representative emissions surrogates of all traffic.

Figure 3-17. Province-wide Canvec paved road network.

3.4.1.3 Dust Transport Factor Corrections and Surrogate Updates

In Canada, construction operations are a significant source of dust emissions, and can have a substantial impact on regional air quality. Previous CMAQ modeling studies (NSRP and SSRP) experienced high concentration of PM in large populated areas and concluded that the dust emissions were not well-represented. These studies used the dust emissions and spatial surrogates from the 2006 EC inventory. EC has acknowledged the issue in their modelling database and attributed the inaccuracies to the following causes (Sassi et al., 2012).

43

The Canadian fugitive dust emissions from road dust and construction sectors are reported as annual area sources on a provincial level.

The transport fraction needs to be recalculated using revised land use/land cover to account for the local deposition of fugitive dust emissions that does not get transported downwind.

Spatial surrogates, which are based on socio-economic data, need to be adapted to each emission source.

Temporal allocation based mostly on default profiles need to be revised to account for the reduction in construction activities in the winter.

To address the over-estimated dust issue, transportation and construction surrogates were updated using CanVec data. Re-allocation of dust emissions is important for ensuring that dust emissions in large populated areas are well represented. One area where this was improved is for heavy industrial construction, which allocates some portion of these construction emissions to city centres. This was corrected by modifying the construction dust surrogate using CanVec data.

For all city boundaries in the province, any heavy industrial construction emissions which would have been allocated to those grid cells modelling were removed and set to zero. Following that, the CanVec data set was used to identify any major mining or extraction sites. These sites were considered good candidates to attribute the heavy industrial construction emissions to. The total dust emissions removed from the cities were equally divided among these sites and added back into the surrogate based on the grid cell where each mining site occurs. An example of this redistribution in and around the capital region is provided in Figure 3-18 below. This re-allocation was done for the heavy industrial construction surrogate in all 4 modelling domains.

44

Figure 3-18. Heavy construction dust re-allocation. Green circles represent mining and extraction areas, dark blue squares are grid cells which received the re-allocated emissions.

In addition to the surrogate updates, the project team corrected dust emissions by applying dust transport factors based on dominant land-use category from the MCIP meteorological data (GRIDCRO2D file). The transport factor is the fraction of fugitive dust emissions not captured and deposited by the surrounding land cover. We adjusted the fugitive and road dust emissions as a post-processing step after the emissions data were output from SMOKE. The values of the transport factor associated with each land cover category are available from US EPA (Pace, 2005). The dust transport correction factor (CF) refers to the correction applied to account for the fraction of dust captured. Table 3-8 shows the dust transport correction factor for five land cover types from EPA. Table 3-9 provides land use categories found in the MCIP files and the associated transport factor.

45

Table 3-8. Recommended Correction Factor (CF) (%) for Five Land Cover Types (source: Pace, 2005).

Table 3-9. Land use categories and transport factor. UNQ LY Description Transport Factor (TF)

1 Urban and Built-Up Land 0.5

2 Dryland Cropland and Pasture 0.75

3 Irrigated Cropland and Pasture 0.75

5 Cropland/Grassland Mosaic 0.75

6 Cropland/Woodland Mosaic 0.75

7 Grassland 0.75

8 Shrubland 0.75

11 Deciduous Broadleaf Forest 0

14 Evergreen Needleleaf Forest 0

15 Mixed Forest 0

18 Wooded Wetland 1

19 Barren or Sparsely Vegetated 1

21 Wooded Tundra 1

22 Mixed Tundra 1

24 Snow or Ice 1

3.4.2 SMOKE Processing and Merging

The SMOKE emissions processing was conducted separately for different source sectors and geographic regions across the modelling domains. The separate SMOKE emission processing streams were defined for ease of performing source apportionment (zero-out) CMAQ simulations and facilitated quality assurance of the resulting modelling emission files. The geographic regions and source sectors modelled separately with SMOKE are presented in Table 3-10.

46

Table 3-10. Summary of geographic regions and source sectors for SMOKE modelling. Geographic Region Source Sectors

LARP Oil Sands Fugitives

Stationary Area

Capital Region/Alberta Power Plants

Upstream Oil & Gas

Industrial - Others

Commercial and residential heating

Agriculture

On-road Mobile

Off-Road Mobile

Canada (except Alberta) Power Plants

Upstream Oil & Gas

Industrial - Others

Commercial and residential heating

Agriculture

Transportation

United States Stationary Point

Non-Point

Domain-Wide Biogenics (MEGAN)

Fires (NCAR)

The separation of industrial sources was done first using SCC codes and then based on facility types and names. Classification based on SCC code alone could be problematic as individual emission sources at a given facility may not be grouped into the same grouping as the facility itself. It was decided that any sources at a UOG facility should be grouped into Industrial – UOG, including any electric power generation sources at UOG facilities. Similarly, any sources at a primarily electric power generation facility were appointed as EGU. Any sources which could not be grouped into either of these categories was left as Industrial – Other.

SMOKE was set up to separately process the emissions for different regions and source sectors to generate “pre-merged emissions.” The pre-merged 2-D emissions were then merged with each other and with the biogenic emission to create CMAQ-ready emissions inputs of all source categories. The point source in-line plume rise emissions files can be input to the CMAQ model separately and do not require merging with other emissions data.

3.4.3 Emissions Summaries

Table 3-11 summarizes the Capital Region 2010 emissions by pollutant in tons per month (averaged for January-March period).

47

Table 3-11. Emissions summary by pollutant and by source sector (tonne per month).

NOx VOC TOG CO NH3 SO2 PM25 PM10

Industrial – upstream oil and gas 428 354 815 550 0 140 7 7

Industrial – electric power generation 4,810 42 82 675 8 5,010 144 290

Industrial – others 1,325 2,387 3,394 2,193 229 1,973 146 282

Transportation: Onroad 1,193 953 1,100 13,252 39 16 25 55

Transportation: Offroad 2,797 625 703 8,137 6 14 182 191

Commercial and residential heating 437 264 291 1,392 5 74 246 249

Agriculture 0 396 611 7 503 0 23 65

48

4.0 METEOROLOGICAL INPUTS

The latest version of the Weather Research Forecast (WRF) meteorological model (Version 3.5.1 released September 23, 2013) was used for meteorological modelling of the Capital Region to develop the CMAQ meteorological inputs. The input data used in the WRF simulation consists of gridded NARR (North American Regional Reanalysis at 32 km grid resolution) data produced and distributed by National Centers for Environmental Prediction (NCEP) as well as upper air and surface observational data which are used to “nudge” the WRF fields to a better representation of the meteorology. The NCEP Climate Forecast System Reanalysis (CFSR) was used to compare the WRF fine grid performance between NARR and CFSR. The WRF 36/12/4/1.33 km domains have been defined slightly bigger than the CMAQ modelling domains in order to eliminate any boundary artifacts resulting from the “relaxation” techniques employed by WRF. Figure 4-1 to Figure 4-3 show the four WRF domains respectively with domain configuration parameters as described in Table 4-1.

Table 4-1. WRF model domain configurations. Science Options Configurations

Model Codes WRF v3.5.1, ARW core

Horizontal Grid Mesh 36/12/4/1.33 km

36 km grid 85 x 109 (SWCAN)

12 km grid 85 x 133 (Alberta)

4 km grid 175 x 193 (NSR)

1.33 km grid 187 x 133 (Capital Region)

Vertical Grid Mesh 40 vertical layers

Initial Conditions NARR/CFSR

Boundary Conditions NARR/CFSR

49

Figure 4-1. WRF domain extents for 36/12/4/1.33 km domains as D1, D2, D3, and D4 respectively.

50

Figure 4-2. WRF domain extents for 12/4/1.33 km domains as D2, D3, and D4 respectively.

51

Figure 4-3. WRF domain extents for 4/1.33 km domains as D3 and D4 respectively.

Note that one change from our proposal is the use of a 1.33 km grid spacing for the Capital Region modelling domain instead of a 1 km resolution. Originally our plan was to do a WRF nest-down (ndown) simulation of the 1 km Capital Region domain whereby the WRF hourly 4 km output is post-processed to provide initial conditions (IC) and hourly boundary conditions (BCs) for the WRF 1 km ndown simulation. Although the WRF user’s guide indicates that this ndown approach for the 1 km domain could be performed, both ENVIRON and Novus have had poor experience with WRF using the ndown feature. For modelling winter time elevated ozone concentrations in Wyoming, ENVIRON used the WRF ndown approach to perform a 1.33 km WRF run with 4 km WRF output and obtained poor WRF performance in the 1.33 km domain. However, when a 36/12/4/1.33 WRF run was performed using one-way nesting (no feedback), much better WRF model performance was obtained within the 1.33 km domain. We believe this is because when doing a 1.33 km ndown WRF run the BCs from the 4 km WRF run are

52

provided on only an hourly basis and don’t include the microphysics variables, whereas using one-way nesting the BCs are provided to the 1.33 km domain at each integration time step and include the microphysics variables. Given that the meteorological conditions that lead to winter elevated ozone events in Wyoming are similar to those for the winter high PM2.5 days in the Capital Region, and the need to obtain the best WRF meteorological solution possible for this study, we don’t recommend using the WRF ndown modelling approach and instead use the WRF one-way nested approach. As WRF nesting requires an odd nesting ratio then we propose a 3:1 ratio so that the fine Capital Region domain will have a 1.33 km grid resolution (1.33 = 4 / 3).

4.1 36/12/4/1.33 km WRF Nesting Modelling

Table 4-2 summarizes the physics options used in the initial WRF modelling. These options were partially modified in the sensitivity test modelling after evaluation of the preliminary WRF performance for the 4 km and 1.33 km domains.

Table 4-2. Initial WRF physics options used in the Capital Region PM Modelling Study. WRF Treatment Option Selected Comments Microphysics Thompson scheme New with WRF 3.1.

Longwave Radiation RRTMG Rapid Radiative Transfer Model for GCMs includes random cloud overlap and improved efficiency over RRTM

Shortwave Radiation RRTMG Same as above, but for shortwave radiation.

Land Surface Model (LSM) NOAH Four-layer scheme with vegetation and sub-grid tiling

Planetary Boundary Layer (PBL) scheme YSU Yonsie University (Korea) Asymmetric Convective Model with non-local upward mixing and local downward mixing.

Explicit Moisture Scheme WSM6 WRF single-moment WSM6 package

Cumulus parameterization Kain-Fritsch in the 36 and 12 km domains, with KF trigger option 2 or 3. None in the 4 and 1.33 km domains.

4 km and 1.33 km domains can explicitly simulate cumulus convection so parameterization not needed

Grid Nesting Two-way grid nesting for 36 km domain

No feedback between domains

Analysis nudging Nudging applied to winds, temperature and moisture in the 36 km

Temperature and moisture nudged above PBL only

Observation Nudging Nudging applied to both surface wind and temperature for all four domains.

If surface temperature and moisture observation nudging introduces instabilities, we will only nudge wind fields

Initialization Dataset NARR/CFSR North American Regional Reanalysis or Climate Forecast System Reanalysis

53

The WRF vertical layer structure in terms of sigma levels9 was defined to focus on the lowest part of the atmosphere within the planetary boundary layer (PBL) in wintertime. Thus, 23 vertical levels are allocated under 2,500 m height above ground level (AGL) in order to adequately simulate the winter stable atmosphere and a total of 40 vertical levels have been applied to WRF modelling (see Table 4-3). Note that most of the meteorological variables, such as temperature, wind and moisture content are output at the half sigma level (i.e., in between two sigma levels). For example, when the first sigma level is 1.0 at the ground, and the second sigma level is 0.9975 with equivalent height at 19 m above ground, meteorological parameters such as temperature and wind are generated at the height of 9.5 m above ground.

Running CMAQ with 39 vertical layers would be computationally demanding, so the WRF vertical layers were collapsed for CMAQ modelling concentrating on air quality issues within the PBL. Table 4-3 shows a WRF layer collapsing scheme for CMAQ modelling from 39 to 22 vertical layers. No layers are collapsed within the first 500 m AGL, and each CMAQ layer below ~2,000 m AGL comes from collapsing two WRF layers.

Terrain and land cover were based on the U.S. Geological Survey (USGS) geophysical data commonly used in WRF modelling. For 1.33 km WRF modelling, terrain and land cover data with a horizontal grid spacing of 30 arc seconds (approximately 900 meters) were used.

To improve the quality of WRF modelling results, upper air and surface observational data were used to “nudge” the WRF fields to the observations to obtain a better representation of the meteorology. We utilized observed temperature and wind data from Environment Canada weather stations and Alberta’s Clean Air Strategic Alliance (CASA) air quality monitoring stations in the nudging process to improve the representation of meteorological parameters from the 4 km and 1.33 km WRF simulations in the Capital Region.

The 36/12/4 km WRF modelling was completed for 2010 winter (6 months + 2 x half month spin-up period).

9 http://en.wikipedia.org/wiki/Sigma_coordinate_system

54

Table 4-3. Definition of WRF 40 vertical levels (39 vertical layers) and mapping to the 22 vertical layers used in the CMAQ Chemical Transport Model. Heights (m) are geopotential heights above ground level, actual layer thicknesses will be shallower in areas above.

WRF CMAQ

Layer Sigma Pressure

(mb) Height

(m) Depth

(m) Layer Sigma Pressure

(mb) Height

(m) Depth

(m)

40 0.0000 100 15685 1190 22 0 100 15685 5206

39 0.0280 125.2 14495 1229

38 0.0620 155.8 13266 1514

37 0.1120 200.8 11752 1273

36 0.1620 245.8 10479 1104 21 0.162 245.8 10479 2961

35 0.2120 290.8 9375 977

34 0.2620 335.8 8397 879

33 0.3120 380.8 7518 800 20 0.312 380.8 7518 2216

32 0.3620 425.8 6718 735

31 0.4120 470.8 5983 681

30 0.4620 515.8 5302 635 18 0.462 515.8 5302 1569

29 0.5120 560.8 4667 479

28 0.5520 596.8 4189 456

27 0.5920 632.8 3733 350 17 0.592 632.8 3733 933

26 0.6240 661.6 3383 317

25 0.6540 688.6 3067 266

24 0.6800 712.0 2800 259 16 0.68 712 2800 739

23 0.7060 735.4 2541 233

22 0.7300 757.0 2308 247

21 0.7560 780.4 2061 223 15 0.756 780.4 2061 440

20 0.7800 802.0 1838 218

19 0.8040 823.6 1621 178 14 0.804 823.6 1621 353

18 0.8240 841.6 1443 175

17 0.8440 859.6 1268 155 13 0.844 859.6 1268 307

16 0.8620 875.8 1113 152

15 0.8800 892.0 961 125 12 0.88 892 961 241

14 0.8950 905.5 836 115

13 0.9090 918.1 720 106 11 0.909 918.1 720 203

12 0.9220 929.8 614 97

11 0.9340 940.6 517 88 10 0.934 940.6 517 88

10 0.9450 950.5 429 79 9 0.945 950.5 429 79

9 0.9550 959.5 350 71 8 0.955 959.5 350 71

8 0.9640 967.6 279 63 7 0.964 967.6 279 63

7 0.9720 974.8 216 62 6 0.972 974.8 216 62

6 0.9800 982.0 154 46 5 0.98 982 154 46

5 0.9860 987.4 108 46 4 0.986 987.4 108 46

4 0.9920 992.8 61 23 3 0.992 992.8 61 23

3 0.9950 995.5 38 19 2 0.995 995.5 38 19

2 0.9975 997.8 19 19 1 0.9975 997.8 19 19

1 1.0000 1000.0 0 0

55

4.2 4 km WRF Model Evaluation

The 4 km WRF modeling performance for the 6-month winter period in 2010 was evaluated by using WRF-MET program (http://www.dtcenter.org/met/users/index.php). Table 4-4 summarizes the model evaluation results across all weather stations within the 4 km domain (41 stations total). The model evaluations applied the commonly used performance parameters: correlation coefficient (R), Mean Error Bias (MEB) and Root Mean Square Error (RMSE). In addition, the mean and the 10th, 50th and 90th percentiles of the values are also calculated to indicate the model performance density distributions across all 41 weather stations.

Table 4-4. The 4 km WRF modeling evaluation Summary for temperature, wind speed and moisture content across 41 weather stations in the 4 km domain.

Temperature (K) Correlation R Mean Error Bias RMSE Mean 0.98 -0.11 2.18

10th percentile 0.96 -0.78 1.76

50th percentile 0.98 0.04 2.04


Wind Speed (m/s) Correlation R Mean Error Bias RMSE Mean 0.57 -0.66 2.12

10th percentile 0.46 -1.70 1.47

50th percentile 0.58 -0.62 1.84


Wind Direction (deg) -- Mean Error Bias --

Mean -- -7.5 --

10th percentile -- -26.3 --

50th percentile -- -5.5 --

90th percentile -- 9.4 --

Moisture cont. (kg/kg) Correlation R Mean Error Bias RMSE Mean 0.91 0.00027 0.00065

10th percentile 0.86 0.00010 0.00048

50th percentile 0.92 0.00025 0.00063

90th percentile 0.96 0.00046 0.00083

The temperature performance is very good across the 4 km domain. Temperature mean and 50th percentile MEBs are low, only -0.1 deg and 0.04 deg, respectively. Temperature correlation is very good, mostly above 0.96. Moisture performance is generally good with slight over-predictions. Wind speed is relatively imperfect largely due to low wind speed modelling accuracies, but the mean and 50th percentile MEBs are only around -0.6 m/s, and the 50th percentile RMSE is less than 1.8 m/s. Wind direction is predicted well with slight counter-clockwise rotations (-7.5 deg and -5.5 deg for mean and 50th percentile MEBs , respectively). For

56

moisture content, the model performed well with R values mostly above 0.9. In summary, WRF 4 km performance is sufficient to support CMAQ modelling.

4.3 4 km WRF Physics Options Tests

In addition to the recommended WRF physics options presented in Table 4-2, the project team also tested another set of physics options, which were applied in the ASHRAE (American Society of Heating, Refrigeration, and Air-Conditioning Engineers) Climatic Research Project in WRF 4 km North America domains (mostly in the U.S.). The difference is that the ASRAE setup used WSM6 in microphysics and Duhdia in radiation scheme. 41 weather stations were used to evaluate the 2010 WRF model 6-month performance based on two sets of different microphysics and radiation schemes. Table 4-5 summarizes the differences in physics options for the 4 km WRF tests.

Table 4-5. Differences in Novus recommended physics options and the 4 km WRF test run. WRF Tests Microphysics Radiation

4 km WRF recommended Thompson RRTM

4 km WRF ASHRAE WSM6 Dudhia

Table 4-6summarizes the model evaluation comparison between the WRF recommended physics option and the tested ASHRAE WRF run. The recommended WRF run outperformed the ASHRAE WRF run for temperature for the R, Mean Error Bias (MEB) and RMSE statistical metrics. While wind speed performance from the two runs are relatively close to each other, the WRF recommended physics run is slightly better. The recommended WRF physics may perform better than the ASHRAE physics because the ASHRAE physics were mainly designed and tested in the U.S., under presumably different meteorological regimes than what is typically observed in Alberta and the Capital Region during the winter. This test suggests that our recommended WRF physics are appropriate for winter meteorological modelling in the study area.

57

Table 4-6. The 4 km WRF modeling performance evaluation comparison between the recommended WRF physics and ASHRAE-WRF physics in temperature and wind speed.

Recommended WRF 4 km


10th percentile 0.96 -0.78 1.76



Tested ASHRAE WRF 4 km


10th percentile 0.94 -1.05 1.81

50th percentile 0.97 -0.19 2.06


Recommended WRF 4 km


10th percentile 0.46 -1.70 1.47

50th percentile 0.58 -0.62 1.84


Tested ASHRAE WRF 4 km


10th percentile 0.49 -1.73 1.40

50th percentile 0.57 -0.59 1.87


4.4 WRF Sensitivity Tests for 1.33 km Domain

The 36/12/4/1.33 km WRF simulations were initially applied to the highest ranked elevated PM2.5 episodes during 2010 as part of the WRF/CMAQ sensitivity test modelling. The definition of high PM2.5 episode is a consecutive period of time/dates when 24-hr PM2.5 level exceeded 30 µg/m3 (see Appendix A).

The 1.33 km WRF modelling results were evaluated by using all available weather station data, including upper air observation data from Stony Plain from west of Edmonton and surface wind, temperature and mixing ratio (humidity) observations from Environment Canada and CASA air quality monitoring stations. WRF-MET model evaluation program was used to evaluate WRF modelling results at the surface as well as aloft. Figure 4-4 displays the locations of the surface meteorological stations within the Capital Region that were used for the WRF model performance evaluation.

58

Figure 4-4. Locations of surface meteorological monitoring sites around the Capital used in the WRF surface model performance evaluation.

In order to test and investigate the 1.33km domain WRF model performance, we have performed several tests as described below:

Test a) Original Design Initial boundary conditions:

NARR Physics options:

Boundary layer: YSU

Microphysics: Thompson

Longwave and shortwave schemes: RRTM

Cumulus: 36/12 km KF, 4/1.33 km None

Land Surface: Noah Nudging:

Analysis nudging for 36 km only, both upper-air and surface

Observation nudging for 36/12/4/1.33 km, surface only

59

Test b) Change in Geophysics Input The difference from test a) is that the static geophysics data used in test a) in 1.33 km domain is in 30s resolution, while that in b) is 2m ( the same as 4 km WRF geophysics input). This test checks the correctness of 30s geographical data in 1.33 km. Test c) Experimental Upper Air Obs Nudging The difference from test a) is that the observation nudging in this run is for 36/12/4 km and for both surface and upper-air, but no observation nudging is applied for 1.33 km domain. Test d) Change in Reanalysis Input to CFSR The difference from test a) is that CFSR data is used as initial and boundary input instead of NARR used in Test a).

Sensitivity tests were conducted for the first 3-months of 2010, that included several winter PM2.5 episodes. Tables 4-7 and 4-8 summarize 1.33 km WRF modeling evaluation comparison among Test a) to d) with 7 weather stations in the 1.33 km domain for temperature and wind speed. For comparison purpose with previous 4 km WRF modeling evaluation, 4 km WRF performance in the same period was also presented in these two Tables.

Table 4-7. 1.33 km WRF modeling evaluation comparison among sensitivity tests a) to d) with 7 weather stations for temperature.

Test a) Original NARR WRF 1.33 km

Temperature (K) Correlation R Mean Error Bias RMSE Mean 0.92 0.18 3.07

10th percentile 0.92 -0.59 2.85



Test b) Change in Geophysics Input WRF 1.33 km


10th percentile 0.90 -0.87 2.59



Test c) Experimental Upper Air Obs Nudging WRF 1.33 km


10th percentile 0.67 -0.04 3.12



Test d) Change in Reanalysis Input to CFSR WRF 1.33 km





60

Original Recommended WRF 4 km


10th percentile 0.96 -0.67 1.89



Table 4-8. 1.33 km WRF modeling evaluation comparison among sensitivity tests a) to d) with 7 weather stations for wind speed.

Test a) Original NARR WRF 1.33 km


10th percentile 0.57 -0.49 1.41



Test b) Change in Geophysics Input WRF 1.33 km

Wind Speed (m/s) Correlation R Mean Error Bias RMSE Mean 0.61 0.02 1.28

10th percentile 0.55 -0.34 1.16



Test c) Experimental Upper Air Obs Nudging WRF 1.33 km


10th percentile 0.53 -0.48 1.14



Test d) Change in Reanalysis Input to CFSR WRF 1.33 km

Wind Speed (m/s) Correlation R Mean Error Bias RMSE Mean 0.62 0.02 1.55

10th percentile 0.56 -0.35 1.42



Original Recommended WRF 4 km


10th percentile 0.50 -0.89 1.35

50th percentile 0.63 -0.05 1.49


The comparison suggests that for 1.33 km domain, Test a) and d) outperformed Test b) and c) in correlation R (the worst is Test c) and temperature (Table 4-7). Test d) CFSR-WRF performed slightly better than Test a) NARR-WRF in correlation coefficient R and RMSE, bur slightly worse (warmer) in mean error bias. Although wind speed performance is similar for all simulations, Test c) performed marginally better than the rest of the tests. For temperature, 4 km WRF

61

preformed slightly better than any one of the 1.33 km tests in temperature, and performed similar to 1.33 km WRF in wind speed prediction.

Between Test a) and d), taking a close look at the time-series of temperature from Edmonton downtown station CYXD (Figure 4-5), we found that Test d) CFSR-WRF performed better than Test a) NARR-WRF. For example, from January 1 to 8, CFSR–WRF could replicate temperature very well while NARR-WRF predicted strong temperature drops during nighttime resulting in large under-estimation bias. We tested CMAQ with both Test a) NARR-WRF and test d) CFSR-WRF output for the selected 2010 episodes. The test results are described in Chapter 5.

The CFSR WRF output of 36/12/4/1.33km for year 2010 was the configuration used for the final CMAQ modelling.

Figure 4-5. 1.33 km WRF Temperature time-series comparison between Test a) NARR-WRF (blue line) and Test d) CFSR-WRF (red line) with observation data (green dots) in January 2010.

4.5 Vertical Temperature and Wind Profiles Comparison

Upper-air meteorological variables are very important to air dispersion and transport. There is one upper air sounding station located in Stony Plain, about 35 km west of the City of Edmonton. Two standard rawinsonde observations are launched every day at 0000 UTC and 1200 UTC, which correspond to 1700 MST the previous day and 0500 MST respectively. 1.33 km and 4 km WRF model output of Test d), the CFSR-WRF, were extracted at Stony Plain station up to 3,000 m above ground level.

62

Temperature vertical profiles are commonly used to assess planetary boundary layer (PBL) heights and stability classes. PBL inversion and strength of inversion are key elements impacting air dispersion and transport. In this study, both vertical wind and temperature profiles from WRF 1.33 km and 4 km modelling were compared with the rawinsonde upper air observation data.

Three elevated PM2.5 episodes were selected to assess whether the WRF modelling accurately simulated the vertical structure of temperature. Temperature profiles from the worst three days (i.e., the highest 24-hour PM2.5 concentrations were observed in the 3 days) per episode were plotted in Figure 4-6 to Figure 4-11. The comparison days in the three episodes are derived from Capital Region PM2.5 episode analysis provided in Appendix A:

2010 Episode 1: Jan 17-21, 2010: Jan 18, 19 and 20

2010 Episodes 2: Jan 26 – Feb 4, 2010: Jan 28 and 29, Feb 02

2010 Episode 5: Feb 20 – Mar 8, 2010: Feb 24, Mar 01 and 03

For 2010 Episodes 1 (Figure 4-6), both 1.33 km and 4 km WRF predicted very strong inversions (~ 4 degC /100 m up to 300 m) which matched sounding data very well for Jan 18 and 19. On Jan 20, WRF model produced lower surface temperature than the observations, which over-estimated inversion strength. However, the model predicted inversion top heights accurately.

For wind speed of 2010 Episodes 1 (Figure 4-7), both 1.33 km and 4 km WRF predicted correctly the overall wind structure inside inversion layer and also matched sounding data very well for Jan 18 and 19. On Jan 20 00Z, WRF model produced weaker wind speed than the observations, which under-estimated inversion strength.

For 2010 Episode 2 (Figure 4-8), similar to Episode 1, both 1.33 km and 4 km WRF also predicted very strong inversions (~ 5 degC /100 m up to 300 m) which matched with sounding data very well for Jan 28 and 29. On Feb 2, the WRF model produced a weaker inversion than the observed sounding at 00Z, but performed better at 12Z.

For wind speed during 2010 Episode 2 (Figure 4-9), similar with Episode 1, both 1.33 km and 4 km WRF predicted correctly the overall wind structure inside inversion layer and also matched sounding data very well for Jan 28 and 29. On Jan 29, WRF model produced faster wind speed, but captured the inversion height correctly.

For 2010 Episode 5 (Figure 4-10), WRF simulated inversions accurately at Feb 24 12Z and Mar 03 00Z, but under-estimated inversion strengths at Feb 24 00Z, Mar 01 00Z and 12Z and Mar 03 12Z. However, both 1.33 km and 4 km WRF simulations predicted inversion heights accurately.

For wind speed of 2010 Episode 5 (Figure 4-11), as in Episodes 1 and 2, both 1.33 km and 4 km WRF predicted correctly the overall wind structure inside inversion layer. However, WRF model produced weaker wind speed than the observations at Feb 24 12Z, Mar 01 12Z and Mar 03 00Z and 12Z.

63

Overall, both 1.33 km and 4 km WRF predicted inversion heights well throughout most of the worst episodes and produced reasonably good temperature and wind speed profiles. We observed that WRF may have slower response to those fast warm up phenomenon as we can see WRF generally produced slightly cooler temperature within the PBL.

64

Figure 4-6. Temperature comparison for 2010 Episode 1: Jan 17-21, 2010: Jan 18, 19 and 20. Red line is modeled 4 km temperature, blue line is modeled 1.33km temperature and green dot is upper air sounding data.

65

Figure 4-7. Wind speed comparison for 2010 Episode 1: Jan 17-21, 2010: Jan 18, 19 and 20. Red line is modeled 4 km wind speed, blue line is modeled 1.33km wind speed and green dot is upper air sounding data.

66

Figure 4-8. Temperature comparison for 2010 Episode 2: Jan 26 – Feb 4, 2010: Jan 28 and 29, Feb 02. Red line is modeled 4 km temperature, blue line is modeled 1.33km temperature and green dot is upper air sounding data.

67

Figure 4-9. Wind speed comparison for 2010 Episode 2: Jan 26 – Feb 4, 2010: Jan 28 and 29, Feb 02. Red line is modeled 4 km wind speed, blue line is modeled 1.33km wind speed and green dot is upper air sounding data.

68

Figure 4-10. Temperature comparison for 2010 Episode 5: Feb 20 – Mar 8, 2010: Feb 24, Mar 01 and 03. Red line is modeled 4 km temperature, blue line is modeled 1.33km temperature and green dot is upper air sounding data.

69

Figure 4-11. Wind speed comparison for 2010 Episode 5: Feb 20 – Mar 8, 2010: Feb 24, Mar 01 and 03. Red line is modeled 4 km wind speed, blue line is modeled 1.33km wind speed and green dot is upper air sounding data.

70

4.6 Summary

Based on 4 km and 1.33 km WRF model performance evaluation and sensitivity tests, we recommended a set of physics options for the Capital Region in order to provide an accurate representation of the meteorology during periods of elevated PM2.5 concentrations. Sensitivity tests indicated Test d) CFSR-WRF performed slightly better than Test a) NARR-WRF. The vertical temperature profile comparison using Stony Plain station data indicated that the CFSR-WRF model reproduced PBL height and inversion strengths reasonably well. Both CFSR-WRF and NARR-WRF outputs were used with CMAQ as part of the sensitivity analyses (see Chapter 5).

Perhaps an important finding in this task was that higher resolution does not always guarantee better representation of meteorological phenomena. It may be true that running WRF at 1.33 km cannot be fully appreciated without a dense observation network for model evaluation. While this study had a reliable and robust source of surface data, the study was still unable to show the value added in improved meteorological fields from increasing the resolution from 4 km to 1.33 km. The high-resolution advantages may be hidden or still not fully understood in very small-scale wind flows, or a sharp horizontal temperature gradient along a sloping terrain of the river valley. The higher resolution may also be needed to reproduce more refined chemical regimes in the CMAQ modeling that are not evident in the meteorological model performance evaluation.

71

5.0 PHOTOCHEMICAL GRID MODELLING AND MODEL PERFORMANCE EVALUATION

5.1 CMAQ model configuration

CMAQ Version 5.0.1 (released July 2012) was exercised for the January-March, 2010, using the 2010 base case inputs developed under Tasks 1 and 2 that were discussed in Chapters 1 through 4. The CMAQ modelling domains are shown in Figure 1-1. Table 5-1 summarizes the CMAQ Chemical Transport Model (CTM) model configuration for the CMAQ simulation.

Table 5-1. CMAQ CTM model configuration. Science Options Configurations

Model Code CMAQ Version 5.0.1 (July 2012)

Horizontal Grid Mesh 36/12/4/1.33 km

Vertical Grid Mesh 22 Layers up to 100 mb

Initial Conditions 15 days full spin-up

Boundary Conditions MOZART(2010)

Emissions Processing SMOKE Version 3.1

Gas-Phase Chemistry CB05

Aerosol Chemistry AE5 (With Sea Salt Emissions)

Secondary Organic Aerosols SORGAM

Meteorological Processor MCIP Version 4.1

Horizontal Transport PPM

Horizontal Diffusion K-theory spatially varying

Vertical Advection Scheme Yamartino

Vertical Eddy Diffusivity Scheme ACM2

Diffusivity Lower Limit Kzmin = 0.5 to 2.0 m2/s (PURB option) (KZMIN set to true)

Deposition Scheme M3dry

72

5.2 Model inputs

5.2.1 Initial and Boundary Conditions (ICs/BCs)

The CMAQ CTM (CCTM) requires Boundary Conditions (BC) inputs to specify the assumed concentrations along the outer lateral edges of the 36 km modelling domain (see Figure 1-1) that are in the CCTM BCON input file. Initial Conditions (ICs) are also needed to be specified for the first day of the model simulation. The 12 km, 4 km, and 1.33 km domains are nested within the 36 km grid using one-way grid nesting, which means that the nested domain are run after the coarse domain and there is no feedback from the fine nest to the coarse domain. The BCs for the 12 km Alberta CMAQ modelling domain were obtained by processing the CMAQ CTM 36 km domain output using the CMAQ BCON processor to generate an hourly 12 km BC input file. The ICs for the 12 km domain were obtained from the 36 km CCTM modelling results. Similarly, the BCs/ICs for the 4 km and 1.33 km modelling domains were obtained from the 12 km and 4 km CCTM modelling results, respectively.

For the 36 km domain, we extracted boundary conditions from 6-hour, 2.8 degree resolution model concentrations from the MOZART-4/GEOS-5 global 3-D chemical transport model (Emmons et al., 2010) using ENVIRON software that interpolates three-dimensional concentration fields horizontally and vertically to the CMAQ boundary grid definition. It then maps MOZART gas species to the compounds required by CMAQ. The initial concentrations for the 36 km and 12 km CMAQ CTM simulations were based on clean concentrations. The 36/12 km modelling used a 15 day spin up period and the 4/1.33 km modelling used initial concentrations from the 12 km and 4 km CMAQ output, respectively, and a 3 day spin up period prior to January 1, 2010 to initialize the model.

5.2.2 Meteorology (MCIP)

The CMAQ Chemical Transport Model (CTM) meteorological inputs were generated by processing the WRF meteorological model output (from Task 2 and discussed in Chapter 4) using the CMAQ Meteorological-Chemistry Interface Program (MCIP). The latest MCIP Version 4.1 (released July 2012) was used to extract 36/12/4/1.33 km fields from WRF simulation outputs. The 36/12/4/1.33 km WRF domains are shown in Figure 4-1. Table 4-3 shows the vertical layer mapping of the 39 WRF layers to 22 CMAQ layers.

5.2.3 Photolysis tables

The CMAQ JPROC processor was used to calculate clear-sky photolysis rates (or J-values) for each date. JPROC uses default values for total aerosol loading and date-specific data for total ozone column from Total Ozone Mapping Spectrometer (TOMS) satellites. TOMS data for the year 2010 are available daily from http://toms.gsfc.nasa.gov/eptoms/ep.html. The photolysis input table for the 12/4/1.33 km modeling was the same as the 36-km modeling.

The current publicly available versions of CMAQ does not treat the effects snow cover albedo has on photolysis rates. Winter ozone studies in Wyoming and Utah have found that the presence of snow cover can almost double photolysis rates, however this effect was not

http://toms.gsfc.nasa.gov/eptoms/ep.html

73

implemented in CMAQ correctly. We obtained a correction from EPA and implemented in the version of CMAQ used in the Capital Region CMAQ modelling.

5.3 CMAQ MODEL EVALUATION

5.3.1 CMAQ Model Evaluation Methodology

The CMAQ evaluation conducted for the project focuses primarily on the operational and diagnostic model evaluation of the air quality model’s performance with respect to fine particulate matter (PM2.5).

5.3.1.1 Evaluation Approach

The U.S. EPA’s integrated ozone, PM2.5 and regional haze modelling guidance calls for a comprehensive, multi-layered approach to model performance testing, consisting of the four major components: operational, diagnostic, mechanistic (or scientific) and probabilistic (EPA, 2007). The Alberta Environment Air Quality Modelling Guideline references the EPA SCRAM website where EPA’s modelling guidance resides (Idriss and Spurrell, 2009). The CMAQ model performance evaluation effort for PM2.5 discussed in this task focused on the first two components of the EPA’s recommended evaluation approach, namely:

Operational Evaluation: Tests the ability of the model to estimate PM2.5 mass concentrations and the components of PM2.5, sulphate, nitrate, ammonium, organic aerosol, elemental carbon, and other inorganic PM2.5. This evaluation examines whether the measurements are properly represented by the model predictions but does not necessarily ensure that the model is getting “the right answer for the right reason”; and

Diagnostic Evaluation: Tests the ability of the model to predict visibility and extinction, PM chemical composition including ozone and PM precursors (e.g., SOx, NOx, VOC and NH3) and associated oxidants (e.g., nitric acid); PM size distribution; temporal variation; spatial variation; mass fluxes; and components of light extinction (i.e., scattering and absorption).

The diagnostic evaluation also may include the performance of diagnostic sensitivity tests to better understand model performance and identify potential flaws in the modelling system that can be corrected. Such diagnostic sensitivity tests were conducted as part of this study.

As in any model performance evaluation, the evaluation is limited by the availability of observed concentration data that can be compared with the model estimates. For the model evaluation presented in this report, observed data for PM precursors and total PM2.5 mass were available. Observed data for PM component species (e.g., sulphate, nitrate, organic aerosol, elemental carbon, etc.) were also available but limited to one site in the Capital Region.

Ideally, the model should be separately evaluated for each component of PM2.5 as the evaluation for just total PM2.5 mass can be misleading due to the introduction of compensating errors. For example, a model could predict the same PM2.5 mass as observed but over-predict

74

the SO4 component that is compensating an under-prediction of the OA component of the PM2.5.

The mapping of the CMAQ modeled species versus those measured in the monitoring networks is fairly straight forward for the inorganic gaseous species. However, for the PM species the measured total PM2.5 mass is composed of numerous CMAQ species. The following CMAQ species mapping for the PM2.5 concentrations were used:

PM2.5 = ASO4J + ASO4I + ANO3J + ANO3I + ANH4J + ANH4I + AORGAT + AORGPAJ + AORGPAI + AORGBT + AORGCT + AECJ + AECI + A25 (Eq.2-1)10

5.3.1.2 Performance Statistics and Goals

To quantify model performance, several statistical measures were calculated and evaluated for all monitors and at individual monitors within the CMAQ 1.33 km domain. Table 5-2 lists the definitions of several statistical performance measures that were used in model performance evaluation discussed below. The statistical measures selected were based on the recommendations outlined in section 18.4 of the USEPA’s Guidance on the Use of Models and Other Analyses for Demonstrating Attainment of Air Quality Goals for Ozone, PM2.5, and Regional Haze11 (EPA, 2007).

10 ASO4=sulphate; ANO3=nitrate; ANH4=ammonium; AORGAT=anthropogenic organic aerosols; AORGPA=primary organic aerosols; AORGBT=biogenic organic aerosols; AEC=elemental carbon; A25=particulate others;I represents Aitken mode and J represents Accumulation mode. More specific listing can be found in spec_def.conc file that is part of the CMAQ evaluation utility package. 11

http://www.epa.gov/scram001/guidance/guide/final-03-pm-rh-guidance.pdf

75

Table 5-2. Statistical model performance evaluation measure definitions.

Statistical Measure Short hand Notation Mathematical Expression Units

Normalized Mean Bias NMB

N

i

i

N

i

ii

O

OP

1

1

Percent

Normalized Mean Error NME

N

i

i

N

i

ii

O

OP

1

1

Percent

Mean Normalized Bias MNB

N

i i

ii

O

OP

N 1

1

Reported as %

Mean Normalized Gross Error MNE

N

i i

ii

O

OP

N 1

1

Reported as %

Mean Fractionalized Bias (Fractional Bias)

MFB or FB

N

i ii

ii

OP

OP

N 1

2

Reported as %

Mean Fractional Error MFE or FE

N

i ii

ii

OP

OP

N 1

2

Reported as %

The U.S. Regional Planning Organizations (RPOs) have established model performance goals and criteria for PM2.5, PM10 and components of fine particle mass based on previous model performance for ozone and fine particles (e.g., Boylan and Russell, 2006; Morris et al., 2004a,b; 2009a,b). Table 5-3 summarizes EPA’s ozone performance goals (EPA, 1991) and lists the model performance goals and criteria developed by the RPOs for PM to assist in interpreting the evaluating regional model performance for PM species.

Table 5-3. Model performance goals and criteria for PM. Fractional Bias Fractional Error Comment

≤±15% ≤35% Goal for PM model performance based on ozone model performance, considered excellent performance

1

≤±30% ≤50% Goal for PM model performance, considered good performance.2

≤±60% ≤75% Criteria for PM model performance, considered average performance. Exceeding this level of performance indicates fundamental concerns with the modelling system and triggers diagnostic evaluation.

2

1The ozone performance goals were originally developed for hourly ozone (EPA, 1991) but have also been shown to be useful

for 8-hour ozone and 24-hour PM. Although we would not expect a model’s PM performance to achieve this goal very often as measurement artifacts can be greater than this goal. 2The PM performance goals and criteria were developed for 24-hour PM concentrations.

76

5.4 Diagnostic Tests and Sensitivity Analyses

To understand the issues of winter PM in the Capital Region, it is important to have a good performing modelling setup. Our initial CMAQ simulations based on the 2010 meteorology conditions showed some promise with elevated secondary PM2.5 concentrations, but there were also several issues and concerns. We discussed the initial CMAQ results with the ESRD and agreed to focus on diagnostic evaluation to improve CMAQ model performance rather than performing extended base case simulations with questionable model performance results. This section documents diagnostic and sensitivity tests conducted in this study.

5.4.1 Modelling Episodes

Two elevated PM2.5 episodes from the 2010 winter months were selected for focused WRF/CMAQ 4 km and 1.33 km domain sensitivity test modelling. As discussed in Chapter 2, one of the main differences between elevated and non-elevated PM2.5 days in the Capital Region is due to much higher secondary PM2.5 concentrations on the elevated PM2.5 days. Thus, it is important to have speciated PM2.5 measurements during our sensitivity test periods to be sure that we are simulating the high elevated secondary PM2.5 concentrations. However, 24-hour average speciated PM2.5 concentrations are only collected at the McIntyre monitoring site on a 1:3 day sampling frequency so are not as available on a spatial or temporal basis as the hourly total PM2.5 mass measurements. Appendix A discusses the PM2.5 episodes in the Capital Region during 2010 and ranks the episodes for use in the sensitivity test modelling. We have identified 9 multiday episodes in 2010 that, with one exception, includes all of the days in 2010 that the observed 24-hour PM2.5 concentrations exceeded the 30 µg/m3 Canada Wide Standard (CWS) as well as many days that exceeded the 20 µg/m3 Management Plan Trigger Level. The exception is for the August 19-22, 2010 PM2.5 episode that had the highest PM2.5 concentrations in 2010 when the Capital Region was inundated by smoke from wildfires. Of the nine 2010 multiday PM2.5 episodes identified in Appendix A, four have been chosen for use in the WRF/CMAQ sensitivity modelling that have high 24-hour PM2.5 concentrations in the Capital Region and 24-hour speciated PM2.5 observations available at the McIntyre 1:3 day speciated PM2.5 monitoring site. Due to time constraints, only the top two ranked episodes were used in the sensitivity modelling. The highest ranked episode was Episode#2 that spanned the January 26 through February 4, 2010 10-day period and included five CWS exceedance days including one day at the McIntyre speciated PM monitoring site (Table 5-4). The second highest ranked episode was the January 17-21, 2010 Episode#1 whose 3 of 5 days were CWS exceedance days including one day at McIntyre (Table 5-5). More details on the 2010 PM2.5 episodes, as well as episodes in 2008 and 2009, are provided in Appendix A. Note that after episode selection analysis, the quality assurance of the McIntyre speciated PM2.5 observations identified two of the samples during Episode#1 as being invalid due to discrepancies with the Dichot PM2.5 measurements that limited its usefulness for evaluating the CMAQ diagnostic sensitivity tests.

77

Table 5-4. Maximum at any monitoring site and McIntyre daily PM2.5 concentrations in the 1st highest ranked 2010 episode for the Capital Region (Episode#2 – January 26 – February 4, 2010).

JDay Year Month Day Max McIntyr

26 2010 Jan 26 9.7 11.5

27 2010 Jan 27 24.6

28 2010 Jan 28 58.0

29 2010 Jan 29 74.4 61.4

30 2010 Jan 30 25.3

31 2010 Jan 31 8.7

32 2010 Feb 1 30.7 25.4

33 2010 Feb 2 40.5

34 2010 Feb 3 37.9

35 2010 Feb 4 24.8 18.9

Table 5-5. Maximum at any monitoring site and McIntyre daily PM2.5 concentrations in the 2nd highest ranked 2010 episode for the Capital Region (Episode#1 – January 17 – 21, 2010).


17 2010 Jan 17 7.5 3.2

18 2010 Jan 18 37.5

19 2010 Jan 19 57.0

20 2010 Jan 20 51.0 40.6

21 2010 Jan 21 17.0

5.4.2 CMAQ Diagnostic and Sensitivity Simulations

CMAQ was first run for the 36/12/4 km domains using meteorological output from WRF-NARR sensitivity test. The CMAQ 4 km outputs were processed using the BCON processor to generate BCs inputs for the 1.33 km domain that covers the Capital Region and vicinity. Most of the sensitivity analyses focused on the 1.33 km domain, although some test used the 4 km domain. Model performance statistics for 24-hour PM2.5 were computed for Episode#1 and #2. PM speciated data at McIntyre sites were not valid during Episode#1, thus model performance statistics for PM species were only calculated for Episode#2.

Although model performance statistics were calculated for three types of bias and error (fractional, normalized mean and mean normalized), we focus the discussion on the fractional bias and error (FB and FE) since that was the form that the PM Performance Goals and Criteria were developed for (Table 5-3). Furthermore, in a recent paper by U.S. EPA (Simon, Philips and Baker, 2012) they preferred the FB/FE metrics because they were balanced treating high and low observations equally and bounded; FB by -200 to +200 percent and FE by 0 to 200 percent.

5.4.2.1 Test# 1: CFSR versus NARR

Given the importance of meteorology for simulating PM concentrations, the first sensitivity test compared PM model performance of the CMAQ model using WRF-CFSR and WRF-NARR

78

meteorology. These WRF simulations only differ in the meteorological analysis fields used as initial and boundary conditions and in the analysis nudging (data assimilation). The WRF-CFSR and WRF-NARR total 24-hour PM2.5 mass model performance statistics of bias and error for Episodes#1 and Episode#2 are shown in, respectively, Tables 5-3 and 5-4. For Episode#1 using all PM2.5 monitoring sites in the Capital region the CMAQ 1.33 km Fractional Bias (FB) for WRF-CSFR and WRF-NARR are 58 and 80 percent, respectively, indicating an overestimation bias. Both FBs fail to achieve the PM bias Performance Goal (≤±30%), but the CMAQ WRF-CFSR FB (58%) achieves the PM Performance Criteria. As discussed in Section 2.2.1, the different PM measurements technologies have different artifacts with many of them failing to fully capture semi-volatile species such as ammonium nitrate that are the biggest PM2.5 component during the winter episodes. The TEOM FDMS technology is a self-referencing system that does not experience as much volatilization of the PM as some of the other measurement technologies. Thus, in the Episode#1 statistics in Table 5-6 we also present model performance averaged across the four TEOM FDMS monitoring as well as at each individual monitoring site. The Episoe#1 CMAQ WRF-CFSM sensitivity test FB (20%) and FE (46%) achieve the PM Performance Goal but the CMAQ WRF-NARR FB (45%) does not using the TEOM FDMS observations.

For Episode#2, the two CMAQ WRF sensitivity tests exhibit an over-prediction bias that exceeds the PM Performance Criteria even when restricted to just the TEOM FDMS monitoring sites. The CMAQ WRF-CFSR FB (68%) is slightly worse than WRF NARR (61%). To understand the reasons for this over-prediction bias we examined the speciated PM modelling performance at the McIntyre site where the CMAQ WRF-CFSR and WRF-NARR Episode#2 FB using TEOM FDMS was 55 and 47 percent, respectively.

79

Table 5-6a. Total 24-hour PM2.5 Mass model performance metrics for the CMAQ WRF-CFSR sensitivity test and Episode#1.

CFSR

Episode Pollutant Site N

Bias Metrics Error Metrics

FB NMB MNB FE NMGE MNGE

Performance Goal

≤±30% ≤±30% ≤±30% ≤50% ≤50% ≤50%

Performance Criteria

≤±60% ≤±60% ≤±60% ≤75% ≤75% ≤75%

1 PM2.5 092801-SES 5 136 545 590 136 545 590

1 PM2.5 093901-SES 5 111 199 321 111 199 321

1 PM2.5 EDCEN-FDMS 5 72 92 166 72 92 166

1 PM2.5 EDEAST-FDMS 5 -24 -33 -14 43 38 39

1 PM2.5 EDMCIN-BAM35 5 38 19 103 50 37 113

1 PM2.5 EDMCIN-EBAM 5 7 -4 18 32 26 39

1 PM2.5 EDMCIN-FDMS 5 -3 -16 8 33 28 38

1 PM2.5 EDMCIN-SPEC 2 48 1 170 81 53 198

1 PM2.5 EDSTH-FDMS 5 37 28 63 38 30 65

1 PM2.5 ELKIS-TEOM40 5 71 62 1009 103 122 1032

1 PM2.5 GENE-TEOM40 5 79 58 298 89 89 307

1 PM2.5 All Sites 77 58 62 304 77 90 318

1 PM2.5 All TEOM FDMS Sites 20 20 18 56 46 47 77

Table 5-6b. Total 24-hour PM2.5 Mass model performance metrics for the CMAQ WRF-NARR sensitivity test and Episode#1.

NARR




Performance Goal

≤±30% ≤±30% ≤±30% ≤50% ≤50% ≤50%


≤±60% ≤±60% ≤±60% ≤75% ≤75% ≤75%

1 PM2.5 092801-SES 5 132 453 467 132 453 467

1 PM2.5 093901-SES 5 123 265 358 123 265 358

1 PM2.5 EDCEN-FDMS 5 90 137 230 90 137 230

1 PM2.5 EDEAST-FDMS 5 6 -6 17 38 31 42

1 PM2.5 EDMCIN-BAM35 5 60 51 153 67 62 160

1 PM2.5 EDMCIN-EBAM 5 31 22 50 43 40 61

1 PM2.5 EDMCIN-FDMS 5 21 6 38 43 36 56

1 PM2.5 EDMCIN-SPEC 2 58 15 207 79 49 225

1 PM2.5 EDSTH-FDMS 5 61 64 106 61 64 106

1 PM2.5 ELKIS-TEOM40 5 99 126 1153 99 126 1153

1 PM2.5 GENE-TEOM40 5 98 101 293 98 101 293

1 PM2.5 All Sites 77 80 99 367 87 112 373

1 PM2.5 All TEOM FDMS Sites 20 45 50 97 58 67 108

80

Table 5-7a. Total 24-hour PM2.5 Mass model performance metrics for the CMAQ WRF-CFSR sensitivity test and Episode#2.

CFSR




Performance Goal

≤±30% ≤±30% ≤±30% ≤50% ≤50% ≤50%


≤±60% ≤±60% ≤±60% ≤75% ≤75% ≤75%

2 PM2.5 092801-SES 10 129 413 468 129 413 468

2 PM2.5 093901-SES 10 97 202 234 97 202 234

2 PM2.5 EDCEN-FDMS 10 98 192 228 98 192 228

2 PM2.5 EDEAST-FDMS 10 46 83 97 55 88 104

2 PM2.5 EDMCIN-BAM35 10 77 138 154 77 138 154

2 PM2.5 EDMCIN-EBAM 10 77 135 176 77 135 176

2 PM2.5 EDMCIN-FDMS 10 55 86 98 58 87 101

2 PM2.5 EDMCIN-SPEC 4 75 138 196 75 138 196

2 PM2.5 EDSTH-FDMS 10 74 125 137 74 125 137

2 PM2.5 ELKIS-TEOM40 10 107 300 300 107 300 300

2 PM2.5 GENE-TEOM40 10 89 164 190 89 164 190

2 PM2.5 All Sites 150 91 197 323 93 199 325

PM2.5 All TEOM FDMS Sites 40 68 122 140 71 123 143

Table 5-7b. Total 24-hour PM2.5 Mass model performance metrics for the CMAQ WRF-NARR sensitivity test and Episode#2.

NARR




Performance Goal

≤±30% ≤±30% ≤±30% ≤50% ≤50% ≤50%


≤±60% ≤±60% ≤±60% ≤75% ≤75% ≤75%

2 PM2.5 092801-SES 10 126 422 494 126 422 494

2 PM2.5 093901-SES 10 78 143 171 79 144 172

2 PM2.5 EDCEN-FDMS 10 93 173 222 93 173 222

2 PM2.5 EDEAST-FDMS 10 42 70 90 47 73 94

2 PM2.5 EDMCIN-BAM35 10 70 111 142 70 111 142

2 PM2.5 EDMCIN-EBAM 10 68 108 169 68 108 169

2 PM2.5 EDMCIN-FDMS 10 47 64 90 49 67 92

2 PM2.5 EDMCIN-SPEC 4 61 91 176 61 91 176

2 PM2.5 EDSTH-FDMS 10 64 99 120 64 99 120

2 PM2.5 ELKIS-TEOM40 10 92 228 247 92 229 247

2 PM2.5 GENE-TEOM40 10 69 100 150 69 100 150

2 PM2.5 All Sites 150 81 158 283 84 160 286

2 All TEOM FDMS Sites 40 61 101 130 63 103 132

A comparison of the CMAQ WRF sensitivity test results against the speciated PM2.5 data at McIntyre site suggest that, with one exception, both primary and secondary PM (especially sulphate) are overestimated, with the exception being that nitrate is underestimated. We examined large SO2 sources in the emissions inventory and found two issues in the point source

81

inventory. First, there were some duplicates of large non-EGU point sources in the Industrial Heartland areas. Secondly, SMOKE inputs were incorrectly formatted resulting in erroneous stack parameters that could result in too low plume rise in some cases. Corrections to these issues in the emissions inventory were made for the next sensitivity test.

Table 5-8. CMAQ WRF-CFSR and WRF-NARR Episode#2 Fractional Bias and Error model performance for 24-hour speciated PM2.5 at the Edmonton McIntyre monitoring site.

Fractional Bias Fractional Error

Episode Species N CFSR NARR CFSR NARR

2 EC 2 127 114 127 114

2 NH4 2 76 60 80 65

2 NO3 2 -64 -54 87 94

2 OC 2 122 107 122 107

2 SO4 2 132 116 132 116

2 SOIL 2 188 184 188 184

2 TCM 2 123 109 123 109

5.4.2.2 Test# 2: CFSR-WRF with Corrected Emissions

The second CMAQ sensitivity test was run for just Episode#2 using the WRF-CFSR meteorological inputs, the 1.33 km Capital Region domain and corrected emissions to eliminate double counting and erroneous stack parameters that were used in Test#1. Episode#1 was not run due to the additional computing time and the fact that there are no valid speciated PM2.5 observations during Episode#1, which makes it difficult to interpret model performance.

Table 5-9 displays the total PM2.5 mass model performance statistics for the CMAQ WRF-CFSR sensitivity test with the corrected emissions (Test#2). Note that Table 5-9 differs from the performance statistics for Test#1 above in that the hourly continuous PM2.5 observations were used rather than turning them into 24-hour PM2.5 averages. Over all sites, CMAQ sensitivity Test#2 overestimates PM2.5 by almost a factor of 3 during Episode#2. Across the TEOM FDMS the overestimation bias is a factor of 2.2, with average predicted and observed values of 65 and 30 µg/m3, respectively, resulting in a FB of 66 percent that fails to achieve the PM Performance Criteria.

Figure 5-1 displays the hourly time series of predicted and observed PM2.5 concentrations at the 16 monitoring sites in and near the Capital Region and CMAQ Test#2. Figure 5-1a shows six time series for Edmonton. The top two panels are for Edmonton Central and Edmonton East using the TEOM FDMS technology where the predicted PM2.5 rises and falls with the observed values only the predictions are too high. The bottom four panels in Figure 5-1a are for the Edmonton McIntyre monitoring using four different measurement technologies. Again the model follows the rise and fall of the observed PM2.5 well, only it is too high. The McIntyre BAM and TEOM FDMS PM2.5 observations are similar, but the TEOM30 is much lower. These results suggest that we should place more weight on the TEOM FDMS and BAM observations and expect the model to over-predict the TEOM30 and TEOM40 measurements.

82

Figure 5-1b displays time series results at the Edmonton South monitor and five monitors to the northeast of Edmonton. The results for EDSTH-FDMS and LAMNT-BAM1020 are not bad with the modelling hourly PM2.5 peaks being too high. But the results for the other four sites in Figure 5-1b not as good due to a large overestimation. This is due in part because these sites used the TEOM40 measurement technique that doesn’t measure all of the PM2.5 mass. The time series plots for the four sites to the southwest of Edmonton are shown in Figure 5-1b where the model predicts some very large peaks on some days (e.g., January 30) that are not present in the observations. The modeled concentrations tend to always be higher than the observed values, especially at site 92801 that lies just southwest of the Capital Region (see Figure 2-2).

Table 5-9. Hourly total PM2.5 mass model performance for CMAQ WRF-CFSR corrected emissions sensitivity test (Test#2), the 1.33 km Capital Region domain and Episode#2.

Episode Species Station N Average

Obs Average Model

Ratio P/O FB FE

2 PM25 090601-SES 240 11.1 69.8 6.3 150.1 150.5

2 PM25 092801-SES 239 11.9 60.3 5.1 119.0 119.4

2 PM25 093901-SES 240 8.6 24.2 2.8 93.1 96.4

2 PM25 EDCEN-FDMS 240 29.6 85.4 2.9 95.3 95.4

2 PM25 EDEAST-FDMS 238 30.8 54.4 1.8 42.4 58.8

2 PM25 EDMCIN-BAM35 240 24.1 57.1 2.4 75.8 78.0

2 PM25 EDMCIN-EBAM 240 24.3 57.1 2.3 76.0 81.2

2 PM25 EDMCIN-FDMS 240 30.8 57.1 1.9 52.5 60.5

2 PM25 EDMCIN-TEOM30 239 14.7 57.0 3.9 128.7 128.9

2 PM25 EDSTH-FDMS 238 28.8 64.3 2.2 73.4 76.6

2 PM25 ELKIS-TEOM40 240 7.2 25.9 3.6 103.1 106.2

2 PM25 FTSASK-TEOM40 240 11.1 69.8 6.3 150.6 151.0

2 PM25 GENE-TEOM40 240 11.2 29.4 2.6 81.3 85.8

2 PM25 LAMNT-BAM1020 240 16.6 25.1 1.5 23.7 55.3

2 PM25 REDWIN-TEOM40 157 9.4 69.4 7.4 107.6 108.3

2 PM25 TOMAHK-TEOM40 240 8.5 21.1 2.5 75.9 83.2

2 PM25 All Sites 3751 17.6 51.3 2.9 90.2 95.7

2 PM25 All TEOM FDMS 956 30.0 65.3 2.2 65.9 72.8

83

Figure 5-1a. Time series of predicted (blue) and observed (red) hourly PM2.5 concentrations (µg/m3) for sensitivity Test#2 and monitoring sites in Edmonton.

84

Figure 5-1b. Time series of predicted (blue) and observed (red) hourly PM2.5 concentrations (µg/m3) for sensitivity Test#2 and Edmonton South (top left) and five monitoring sites northeast of Edmonton.

85

Figure 5-1c. Time series of predicted (blue) and observed (red) hourly PM2.5 concentrations (µg/m3) for sensitivity Test#2 and four monitoring sites southwest of Edmonton.

The CMAQ Test#2 model performance statistics the Edmonton McIntyre speciated monitoring

site is shown in Table 5-10 with time series shown in Figure 5-2. Note that the time series for

the speciated PM2.5 performance only consists of four points during Epsiode#2 due to the 1:3

day sampling frequency of 24-hour average speciated PM2.5 concentrations.

With the exception of NO3, all species are overestimated. There are four carbon species

presented:

Elemental Carbon (EC) that is overestimated by a factor of 3.8.

The model produces organic aerosol (OA) but the measurement technique only

measures the Organic Carbon (OC) portion of the OA (i.e., missing oxygen and other

species in the OA). Thus, the evaluation is conducted two ways, comparing the

measured OC to the modeled OA that is labeled OC and adjusting the measured OC to

OA using a 1.4 scaling factor and comparing that to the modeled OA (labeled OA).

86

Because some carbon measurement techniques have difficulty in distinguishing

between EC and OC, we also evaluate using total carbon mass (TCM) species that is

EC+OC for the observed and EC+OA for the predicted.

In any event, the four carbon species are overestimated by factors of 2.1 to 3.8 for

CMAQ Test#2 and Episode#2.

Sulphate (SO4) is overestimated by almost factor of 5 with average observed and predicted

values for the four speciated PM2.5 measurement days during Episode#2 of 5.3 and 25.8 µg/m3,

respectively. We believe this sulphate overestimation may be due in part to errors in wind

direction that are bringing emissions from the Industrial Heartland area northeast of Edmonton

into Edmonton. With SO4 overestimated it is not surprising that ammonium (NH4) is also

overestimated.

Nitrate (NO3) is underestimated by over a factor of 2 with average observed and predicted

values of 10.3 and 4.3 µg/m3, respectively. The species TNO3 is total nitrate that is the sum of

particulate nitrate (NO3) plus nitric acid (HNO3) and it is compared with just the observed NO3

since observed HNO3 is not available. Similarly, the species TNH4 is modeled total ammonia

(NH4+NH3) that is compared against just the observed NH4 because there are no NH3

measurements. So the model performance evaluation statistics for TNO3 and TNH4 in Table 5-

10 are not very meaningful, but the results do provide some valuable information:

The average modeled NO3 (4.3 µg/m3) is almost identical to TNO3 (4.4 µg/m3) indicating

that there is very little leftover gaseous HNO3 that has not been converted to particulate

NO3.

The average modeled TNH4 (15.2 µg/m3) is 4.5 µg/m3higher than the NH4 (10.7 µg/m3)

indicating that there is left over ammonia concentrations (~6 ppb).

The above results suggest that NO3 formation in the CMAQ model is nitric acid limited

and not ammonia limited.

Note that the conversion of the HNO3 and NH3 from ppm to µg/m3 was performed using

standard temperature and pressure conditions (1 atm and 25°C) and conditions in Edmonton

are approximately 30-40°C colder than that, which would increase the concentrations as µg/m3

by 10-15%, so is not enough to affect the above conclusions.

Soil is grossly overestimated by the model by factor of 26.2 (0.5 µg/m3 observed and 12.2

µg/m3 predicted). Part of this is measurement and modeled incommensurability where the

measured Soil is built up by elements but the modeled Soil is all the PM emissions that are not

explicitly speciated as SO4, NO3, NH4, EC and OA. However, the large Soil overestimation bias is

87

a cause for concern and is contributing to the general PM2.5 overestimation bias in CMAQ

Test#2.

Table 5-10. 24-hour speciated PM2.5 (µg/m3) model performance for CMAQ WRF-CFSR corrected emissions sensitivity test (Test#2), the 1.33 km Capital Region domain and Episode#2.


Obs Average Model

Ratio P/O

FB

FE

2 EC EDMCIN 4 1.5 5.6 3.8 127.6 127.6

2 NH4 EDMCIN 4 4.6 10.7 2.3 72.8 76.3

2 NO3 EDMCIN 4 10.3 4.3 0.4 -58.9 90.6

2 OA EDMCIN 4 4.4 9.0 2.1 101.2 101.2

2 OC EDMCIN 4 3.1 9.0 2.9 121.6 121.6

2 SO4 EDMCIN 4 5.3 25.8 4.9 129.8 129.8

2 SOIL EDMCIN 3 0.5 12.2 26.2 188.1 188.1

2 TCM EDMCIN 4 4.6 14.6 3.2 123.6 123.6

2 TNH4 EDMCIN 4 4.6 15.2 3.3 102.3 102.3

2 TNO3 EDMCIN 4 10.3 4.4 0.4 -55.8 88.1

88

Figure 5-2a. Time series of predicted (blue) and observed (red) 24-hour speciated PM2.5

concentrations (µg/m3) for sensitivity Test#2 and monitoring sites in Edmonton.

89

Figure 5-2b. Time series of predicted (blue) and observed (red) 24-hour speciated PM2.5

concentrations (µg/m3) for sensitivity Test#2 and monitoring sites in Edmonton.

90

5.4.2.3 Test# 3: Exclusion of Primary Sulphate Emissions

The third CMAQ sensitivity test was designed to investigate the reasons for the over-prediction of sulphate (SO4) during Episode#2 CMAQ Test#2. The SO4 overestimation could be part of the cause of the problem with NO3 underestimation with the SO4 binding NH3 that would no longer be available to bind with HNO3 to form NO3. Although indications are that NO3 formation is more NH3 than HNO3 limited.

Typically most of the particulate SO4 is formed in the atmosphere from gaseous SO2 emissions. High sulphate concentrations are more frequently associated with the warmer months when there is more photochemical activity to convert the SO2 to SO4. However, if in the emissions modelling a lot of the PM emissions were speciated to SO4 that could cause a winter SO4 overestimation. CMAQ sensitivity Test#3 was designed to determine whether overstated primary SO4 emissions could be contributing to the SO4 overestimation. In Test#3 all of the primary SO4 emissions were set to zero in the 1.33 km domain. Table 5-11 lists the model performance statistics at the Edmonton McIntyre speciated PM monitor for CMAQ Test#3 and Epsiode#2. The average modeled SO4 value for Test#3 (25.2 µg/m3) is approximately 2 percent (0.6 µg/m3) less that Test#2 (25.8 µg/m3) indicating that overstated primary SO4 emissions are not the cause of the SO4 overestimation bias.

Table 5-11. 24-hour speciated PM2.5 (µg/m3) model performance for CMAQ no primary sulphate emissions sensitivity test (Test#3), the 1.33 km Capital Region domain and Episode#2.


Obs Average Model

Ratio P/O

FB

FE

2 EC EDMCIN 4 1.5 5.6 127.5 127.5 2

2 NH4 EDMCIN 4 4.6 10.5 70.9 75.5 2

2 NO3 EDMCIN 4 10.3 4.3 -58.7 90.4 2

2 OA EDMCIN 4 4.4 9.0 101.1 101.1 2

2 OC EDMCIN 4 3.1 9.0 121.5 121.5 2

2 SO4 EDMCIN 4 5.3 25.2 128.1 128.1 2

2 SOIL EDMCIN 3 0.5 12.2 188.1 188.1 2

2 TCM EDMCIN 4 4.6 14.6 123.6 123.6 2

2 TNH4 EDMCIN 4 4.6 15.2 102.1 102.1 2

2 TNO3 EDMCIN 4 10.3 4.4 -55.9 88.0 2

91

5.4.2.4 Test# 4: Limit Vertical Diffusion

A major portion of the SO2 emissions in the greater Capital Region are from large point sources (e.g., EGUs to the west and industrial facilities to the northeast). The Capital Region winter PM events occur during periods of low wind speeds and limited vertical mixing allowing pollutants to build up. If there is too much vertical mixing that could be bringing too much SO4 from the elevated point sources to the ground contributing to the SO4 overestimation bias and mixing too much surface NOx in upper layers contributing to the NO3 underestimation bias. CMAQ vertical mixing is controlled by vertical diffusion coefficients (Kv) for these winter time periods (there is also a convective component that will not be an issue for these winter episodes). In the CMAQ sensitivity Test#4 the vertical diffusion coefficient were set to a low value (0.01 m2/s) to limit vertical mixing. Table 5-12 displays the total PM2.5 mass and Table 5-13 displays the speciated PM2.5 model performance statistics for CMAQ vertical mixing sensitivity Test#4 that can be compared with the CMAQ Test#2 results in Tables 5-9 and 5-10. Limiting the vertical diffusion increases the ground level PM2.5 concentrations by over a factor of two. For example, averaged across the four TEOM FDMS monitoring sites the average modeled PM2.5 concentration during Episodes#2 increases from 65.3 µg/m3 in Test#2 to 205.0 µg/m3 in the Test#4 limited vertical mixing sensitivity test. Examining the speciated PM2.5 concentrations, all components are higher in the limited vertical mixing Test#4 versus Test#2 except NO3 which is slightly lower. In particular, SO4 is higher in the limited vertical mixing Test#4 (53.8 µg/m3) than Test#2 (25.8 µg/m3).

Table 5-12. Hourly total PM2.5 mass model performance for CMAQ limit vertical diffusion sensitivity test (Test#4), the 1.33 km Capital Region domain and Episode#2.


Obs Average Model

Ratio P/O FB FE

2 PM25 090601-SES 240 11.1 104.8 9.4 166.8 166.8

2 PM25 EDCEN-FDMS 240 29.6 340.2 11.5 164.4 166.1

2 PM25 EDEAST-FDMS 238 30.8 103.6 3.4 96.2 97.7

2 PM25 EDMCIN-BAM35 240 24.1 150.5 6.2 140.0 140.0

2 PM25 EDMCIN-EBAM 240 24.3 150.5 6.2 139.1 139.1

2 PM25 EDMCIN-FDMS 240 30.8 150.5 4.9 126.2 126.2

2 PM25 EDMCIN-TEOM30 239 14.7 150.5 10.2 167.6 167.6

2 PM25 EDSTH-FDMS 238 28.8 225.9 7.9 152.1 152.1

2 PM25 ELKIS-TEOM40 240 7.2 34.4 4.8 121.4 121.7

2 PM25 FTSASK-TEOM40 240 11.1 104.8 9.5 167.1 167.1

2 PM25 GENE-TEOM40 240 11.2 45.9 4.1 113.0 113.0

2 PM25 LAMNT-BAM1020 240 16.6 32.9 2.0 50.5 61.6

2 PM25 REDWIN-TEOM40 157 9.4 44.7 4.7 112.8 112.8

2 PM25 TOMAHK-TEOM40 240 8.5 27.8 3.3 98.8 100.2

2 PM25 All Sites 3751 17.6 114.9 6.5 130.3 131.4

2 PM25 All TEOM FDMS 956 30.0 205.0 6.8 134.7 135.5

92

Table 5-13. 24-Hour speciated PM2.5 model performance at Edmonton McIntyre for CMAQ limit vertical diffusion sensitivity test (Test#4), the 1.33 km Capital Region domain and Episode#2.


Obs Average Model

Ratio P/O FB FE

2 EC EDMCIN 4 1.5 22.8 15.5 177.8 177.8

2 NH4 EDMCIN 4 4.6 21.3 4.6 118.5 118.5

2 NO3 EDMCIN 4 10.3 4.0 0.4 -66.2 98.6

2 OA EDMCIN 4 4.4 36.8 8.4 162.2 162.2

2 OC EDMCIN 4 3.1 36.8 11.7 171.7 171.7

2 SO4 EDMCIN 4 5.3 53.8 10.2 163.8 163.8

2 SOIL EDMCIN 3 0.5 34.8 74.8 195.2 195.2

2 TCM EDMCIN 4 4.6 59.5 12.9 174.1 174.1

2 TNH4 EDMCIN 4 4.6 43.7 9.4 163.4 163.4

2 TNO3 EDMCIN 4 10.3 4.0 0.4 -66.2 98.6

5.4.2.5 Test# 5 and Test#6: January through March 2010 with Increased Ozone BCs

The next two CMAQ sensitivity tests were designed to address several issues:

Some of the PM2.5 overestimation bias is believed to be due in part to overstated dust emissions. Construction is a major source of dust emissions in the Capital Region. However, we would not expect there to be any construction dust emissions in the winter. Thus dust emissions were set to zero in sensitivity Test#5 and Test#6.

The SO4 overestimation issue may be is associated with wind direction issues during Episode#2 bringing emissions from the Industrial Heartland into Edmonton. Thus, we need to look at other episodes to see whether they are also characterized by SO4 overestimation.

The NO3 underestimation is probably the biggest concern in the CMAQ simulations performed to date. As discussed in Chapter 2, NO3 formation depends on generation of radicals (e.g., NO3

-) that will be highly dependent on ozone concentrations. Transport is the main source of ozone in the Capital Region in the winter so Test#6 investigated the sensitivity of the CMAQ model predictions to increases in the amount of ozone transport.

Sensitivity Test#5 and Test#6 were run for the January through March 2010 period using the 4 km domain that includes Episode#1 through Episode#5. Both sensitivity tests were run with no dust emissions. The differences in the two runs were that Test#5 used ozone boundary conditions (BCs) from the CMAQ 12 km Alberta domain simulation and in Test#6 the ozone BCs were increased by 20 ppb.

Figure 5-3 displays the time series of predicted and observed 24-hour PM2.5 concentrations during 2010 Quarter 1 at six monitoring sites for Test#5 and #6. The model is still

93

overestimating PM2.5 concentrations even with no dust emissions. At the TEOM FDMS sites, the model is mostly achieving the PM Performance Criteria, but not the PM Performance Goals, with an overestimation bias. It is encouraging that the model has spikes on the days when the observed PM2.5 spikes (e.g., January 29, 2010). The PM2.5 is slightly higher for the increased PM2.5 sensitivity test (Test#6).

The 2010 Quarter 1 Test#5 and #6 modelling results for speciated PM2.5 at the Edmonton McIntyre site are shown in Figure 5-4. The top two panels show results for NO3 and TNO3

(NO3+HNO3 predicted and NO3 observed), that are almost identical indicating there is very little free HNO3 available in the model. The 20 ppb higher BCs in the CMAQ 4 km run results in increased NO3 concentrations that is especially apparent on the observed high NO3 days. This suggest that incoming ozone concentrations are very important and the too low NO3 is due to insufficient radical generation in the model.

The middle two panels in Figure 5-4 are for NH4 and TNH4 (NH4+NH3 predicted and NH4 observed). NH4 is overestimated due to SO4 being overestimated. TNH4 is noticeably higher than NH4 indicating that there is excess NH3 available to form ammonium nitrate if the model were able to oxidize more of the NOx to HNO3. However, there is not enough NH3 available to neutralize sufficient HNO3 to NO3 to match the very highest observed NO3 days. To obtain sufficient NH3, the SO4 overestimation also needs to be addressed so that would free up NH3 that could then bind with HNO3.

The bottom two panels in Figure 5-4 display SO4 and Soil model performance. As noted above, SO4 is overestimated by a large margin throughout 2010 Quarter 1 so this is not just an Episode#2 issue. SO2 also tends to be overestimated at most sites with higher values at monitors to the northeast of Edmonton and lower values at the monitors to the southeast of Edmonton except for occasional large spikes. Soil is still overestimated even with no dust emissions (Figure 5-4, lower right), although there are still discrepancies in what makes up this species in the observations and in the model.

Figure 5-5 displays example predicted and observed ozone hourly time series at Edmonton East and Elk Island monitoring sites and Episodes#2, #3 and #4 and Tests#5 and Test#6. Test#5 using the CMAQ 12 km BCs underestimates ozone in the Capital Region and better ozone performance is seen when the ozone BCs are increased by 20 ppb in Test#6. As noted above, increased ozone BCs also improve NO3 model performance.

94

Figure 5-3. Time series and performance statistics for 24-hour PM2.5 concentrations for 4 km Test#5 (NODUST) and Test#6 (NODUST_BC) at several monitoring sites during 2010 Quarter 1.

95

Figure 5-4a. Time series and performance statistics for 24-hour speciated PM2.5 concentrations at Edmonton McIntyre monitoring site during 2010 Quarter 1 for 4 km Test#5 (NODUST) and Test#6 (NODUST_BC).

96

Figure 5-4b. Time series and performance statistics for hourly ozone concentrations at Edmonton East (left) and Elk Island (right) monitoring sites during Episodes#2 (top), #3 (middle) and #3 (bottom) for 4 km Test#5 (NODUST) and Test#6 (NODUST_BC).

97

5.4.2.6 Test# 7: Default photolysis rates

Sensitivity Test#7 was similar to Test#2 but using CMAQ with default photolysis rates. EPA identified a problem in the treatment of the albedo effects due to snow cover in CMAQ version 5 and provided us with a fix that was implemented in CMAQ Version 5.0.1 for the Capital Region simulations. We noted that the CMAQ modelling of the Capital Region performed under the North Saskatchewan Regional Plan (NSRP) using the default snow cover albedo effects did not have the same ozone underestimation issue. Sensitivity Test#7 used the default photolysis rate approach without the fix provided by EPA with everything else exactly the same as Test#2. The resultant CMAQ PM2.5 model performance for Test#7 ended up being identical to Test#2, whose results are shown in Table 5-10.

5.4.2.7 Test# 8: Enhanced N2O5 Heterogeneous Reactions

The heterogeneous reaction of N2O5 is believed to be the most important HNO3 formation pathway for the Capital Region winter PM2.5 episodes. The rate of this reaction increases with lower temperature and moister conditions. However, there is a cap on this rate based on the laboratory studies whose temperatures only went down to 0°C and RH only went up to 70 percent. As colder and moisture conditions occur during the Capital region winter PM2.5 episodes, we increased that cap on this reaction for frozen particle by a factor of 3. This resulted in small increases (0.1 µg/m3) in NO3 concentrations.

Table 5-14. 24-Hour speciated PM2.5 model performance at Edmonton McIntyre for CMAQ N2O5 heterogeneous sensitivity test (Test#8), the 1.33 km Capital Region domain and Episode#2.


Obs Average Model

Ratio P/O FB FE

2 EC EDMCIN 4 1.5 5.6 3.8 127.6 127.6

2 NH4 EDMCIN 4 4.6 10.7 2.3 72.9 76.3

2 NO3 EDMCIN 4 10.3 4.4 0.4 -57.5 91.2

2 OA EDMCIN 4 4.4 9.0 2.1 101.1 101.1

2 OC EDMCIN 4 3.1 9.0 2.9 121.6 121.6

2 OPM EDMCIN 3 4.6 59.8 12.9 153.8 153.8

2 SO4 EDMCIN 4 5.3 25.7 4.9 129.8 129.8

2 SOIL EDMCIN 3 0.5 12.2 26.2 188.1 188.1

2 TCM EDMCIN 4 4.6 14.6 3.2 123.6 123.6

2 TNH4 EDMCIN 4 4.6 15.2 3.3 102.3 102.3

98

5.4.2.8 Test# 9 and #10: Integration of CALMOB6 emissions with and without dust emissions

The next two tests were designed to address the impacts of CALMOB6 emissions integration. The CALMOB6 emissions are lower than the projected 2010/CANVEC emissions for most species (see Table 3-5). In Edmonton, on-road NOx emissions from CALMOB6 is 7.6 tonne/day compared to 11.3 tonne/day in the 2010/CANVEC inventory. CALMOB6 does not include NH3 emissions.

Sensitivity Test#9 and Test#10 used the CALMOB6 emissions with everything else exactly the same as Test#2. The differences in the two runs were that Test#9 included dust emissions while Test#10 excluded all surface crustal dust emissions in Alberta. The resultant CMAQ PM2.5 model performance for Test#9 is similar to Test#2 whose results are shown in Table 5-10. Slightly better performance is seen for SO4, EC and NH4 compared to Test#2. Excluding dust in Test#10 helped lower OC and SOIL performance biases. Sensitivity Test#10 has the best overall PM performance compared to all of the sensitivity tests conducted under this study.

Table 5-15. 24-Hour speciated PM2.5 model performance at Edmonton McIntyre for CMAQ CALMOB6 sensitivity test (Test#9), the 1.33 km Capital Region domain and Episode#2.

Episode Species Station N

Average Obs

CALMOB6

CALMOB6 without dust

Average Model FB FE

Average Model FB FE

2 EC EDMCIN 4 1.5 5.6 126 126 5.4 126 126

2 NH4 EDMCIN 4 4.6 10.7 70 75 10.1 69 74

2 NO3 EDMCIN 4 10.3 4.4 -59 91 4.2 -60 91

2 OA EDMCIN 4 4.4 9.0 101 101 8.5 97 97

2 OC EDMCIN 4 3.1 9.0 121 121 8.5 118 118

2 OPM EDMCIN 3 4.6 59.8 153 153 56.4 152 152

2 SO4 EDMCIN 4 5.3 25.7 128 128 24.2 127 127

2 SOIL EDMCIN 3 0.5 12.2 188 188 6.9 179 179

2 TCM EDMCIN 4 4.6 14.6 123 123 13.9 121 121

2 TNH4 EDMCIN 4 4.6 15.2 98 98 14.1 97 97

5.4.2.9 Test#11: APT

Plume-in-Grid (PinG) modelling refines the near-source transport and chemistry of point source plumes. Lower bias in sulfate predictions is expected when using PinG. Nonetheless, monitoring stations may only sometimes be impacted by plumes from upwind point sources and the similarity in model performance with and without PinG treatment has been noted in previous modelling studies with CMAQ.

The beta versions of CMAQ V5.0.1 with the Advanced Plume Treatment (APT) is the only current version of CMAQ with a “plume-in-grid” treatment and is available to the ENVIRON Team for use. The CMAQ APT is only available for Aerosol Module version 6 (AE6). To date, EPA have not released the public version with APT.

99

We selected top 20 NOx emitters from EGU and non-EGU sectors for APT treatment. Our initial CMAQ setup led to failing CMAQ simulation. Further testing was successful only if using one computer processor, which would require too much computer time for the Capital Region CMAQ simulations. Due to the time constraints of this study, we did not conduct further APT testing and turned off APT in our base case simulation.

5.4.3 Conclusion of Diagnostic Analyses

The initial CMAQ setup showed some promise with elevated secondary PM2.5 concentrations, but there were also several issues and concerns. Specifically, all PM species except nitrate were overstated.

Sensitivity tests suggested that under-estimation of nitrate is related to HNO3-limited not NH3-limited condition. More radicals are needed to aid HNO3 formation leading to improved nitrate performance. Ozone is one of the important radical sources and higher ozone transported through boundary helped improve the nitrate performance. Another pathway of HNO3 formation is through heterogeneous N2O5 chemistry and the CMAQ parameters for this reaction may need to be revised for Capital Region. The meteorological conditions that occur during the Capital Region winter PM2.5 episodes occur outside the range of conditions that the CMAQ heterogeneous N2O5 chemistry module was developed for.

Over-estimation of sulfate is dominated by secondary sulfate (conversion of SO2). Limited vertical diffusion only worsened the sulfate model performance. Removal of fugitive dust emissions helps improve model performance for most species and total PM2.5.

5.5 Final Base Case

Final base case setup is the same as Test#10. Specifically, the final base case emissions incorporated CALMOB6 Edmonton on-road data and excluded all fugitive dust emissions in Alberta. The CALMOB6 data offered better resolved on-road emissions in Edmonton, thus were incorporated in the base case. The fugitive dust sources are construction, agriculture, and unpaved road dust, all of which are expected to be insignificant during wintertime. Other CMAQ modelling inputs are as described in section 5.2 with no further modifications. The CFSR-WRF was the selected meteorology for the final base case since it slightly outperformed the NARR-WRF. CMAQ modelling was conducted for 36, 12, 4 and 1.33 km domains for January and February, 2010, using a model configuration shown in Table 5-1.

Table 5-16 displays the hourly total PM2.5 mass model performance statistics for the CMAQ base case. Over all sites, CMAQ overestimates PM2.5 by a factor of 2 during January-February, 2010. Across the TEOM FDMS the overestimation bias is a factor of 1.5, with average predicted and observed values of 32 and 21 µg/m3, respectively, and resulting FB of 32 percent that achieves the Performance Criteria but fails to achieve the Performance Goals. Although the CMAQ setup was very similar to Test#10, the statistics appear more satisfactory. This is because the statistics in Table 5-16 includes non-episode days when CMAQ could better replicate observed PM concentrations compared to episode days.

100

Figure 5-5 displays the 24-hour time series of predicted and observed PM2.5 concentrations at the 16 monitoring sites in and near the Capital Region and CMAQ final base case. Generally, the predicted 24-hour PM2.5 rise and fall with the observed values only the predictions are too high. CMAQ replicates the PM observations well on some non-episode days, especially at Edmonton East and Edmonton McIntyre sites (Figure 5-5a; top right and bottom left, respectively). The worst performing sites with higher than 100% FB are 090601-SES, EDMCIN-TEOM30, and FTSASK-TEOM40. Similar magnitude bias and error statistics at these sites indicate that the underestimation trends are consistent temporally. This is due in part because these sites used the TEOM30 or TEOM40 measurement techniques that don’t measure all of the PM2.5.

The CMAQ base case model performance statistics the Edmonton McIntyre speciated monitoring site as shown in Table 5-17 with time series shown in Figure 5-6. With the exception of NO3, all species are overestimated with similar bias and error statistics. Sulphate and soil have the worst performances with FE more than 150%.

Sulphate is overestimated by a factor of 6 with average observed and predicted values of 2.3 and 13.6 µg/m3, respectively. Again, this sulphate overestimation may be due in part to errors in wind direction that are bringing emissions from the Industrial Heartland area northeast of Edmonton into Edmonton. However, given that the total PM statistics show consistent overestimations across all monitors, the SO2 inventory maybe overstated during these winter months. With sulphate overestimated it is not surprising that ammonium (NH4) is also overestimated. Note, however, that predicted total ammonia (TNH4) is higher than predicted NH4, indicating that there is left over ammonia concentrations.

Nitrate is underestimated by a factor of 2 with average observed and predicted values of 8 and 3.7 µg/m3, respectively. The predicted total nitrate (TNO3) is 3.9 µg/m3, suggesting that nitrate formation is nitric acid limited and not ammonia limited. A better nitrate performance will require more radicals in the atmosphere to convert NOx to nitric acid.

Elemental carbon (EC) and soil are emitted as primary specie and have no chemistry pathways in the CMAQ model. FB for EC and soil are 80.5% and 172%, respectively. Large overestimation bias for primary species is a cause for concern. This is measurement and modeled incommensurability of soil as noted previously. Nonetheless, given that the total PM statistics show consistent overestimations across all monitors, it is possible that primary PMs emissions may be overstated during the modelling period and/or vertical mixing is understated. Two major source sectors contributing to primary PMs are on-road and off-road mobile sources and commercial/residential heating.

101

Figure 5-5a. Time series of predicted (blue) and observed (red) 24-hour PM2.5

concentrations (µg/m3) for base case and monitoring sites in Edmonton.

102

Figure 5-5b. Time series of predicted (blue) and observed (red) 24-hour PM2.5

concentrations (µg/m3) for sensitivity Test#2 and Edmonton South (top left) and five monitoring sites northeast of Edmonton.

103

Figure 5-5c. Time series of predicted (blue) and observed (red) 24-hour PM2.5

concentrations (µg/m3) for base case and four monitoring sites southwest of Edmonton.

104

Table 5-16. Hourly total PM2.5 mass model performance for CMAQ base case, the 1.33 km Capital Region domain and Jan-Feb period.


Obs Average Model

Ratio P/O FB FE

Jan-Feb PM25 090601-SES 1401 7.9 30.9 3.9 113.7 119.7

Jan-Feb PM25 092801-SES 1408 9.1 29.4 3.2 94.0 95.4

Jan-Feb PM25 093901-SES 1402 5.5 9.2 1.7 56.5 83.6

Jan-Feb PM25 EDCEN-FDMS 1403 23.5 49.5 2.1 64.4 74.4

Jan-Feb PM25 EDEAST-FDMS 1388 21.7 23.5 1.1 -4.8 54.5

Jan-Feb PM25 EDMCIN-BAM35 1406 14.7 26.2 1.8 56.5 71.7

Jan-Feb PM25 EDMCIN-EBAM 1363 15.7 26.5 1.7 59.5 82.9

Jan-Feb PM25 EDMCIN-FDMS 1404 20.0 26.1 1.3 22.7 53.4

Jan-Feb PM25 EDMCIN-TEOM30 1401 8.7 26.2 3.0 110.3 115.3

Jan-Feb PM25 EDSTH-FDMS 1380 18.9 30.0 1.6 46.1 66.6

Jan-Feb PM25 ELKIS-TEOM40 1394 4.8 10.6 2.2 67.7 95.4

Jan-Feb PM25 FTSASK-TEOM40 1401 7.9 30.9 3.9 114.2 120.2

Jan-Feb PM25 GENE-TEOM40 1401 6.4 9.2 1.4 44.4 80.1

Jan-Feb PM25 LAMNT-BAM1020 1408 11.9 11.5 1.0 -14.8 75.5

Jan-Feb PM25 REDWIN-TEOM40 1163 8.1 26.3 3.2 67.4 90.8

Jan-Feb PM25 TOMAHK-TEOM40 1405 5.5 8.7 1.6 47.3 76.9

Jan-Feb PM25 All Sites 22128 11.9 23.4 2.0 59.0 84.7

Jan-Feb PM25 All TEOM FDMS 5575 21.0 32.3 1.5 32.1 62.2

105

Figure 5-6a. Time series of predicted (blue) and observed (red) 24-hour speciated PM2.5

concentrations (µg/m3) for base case at Edmonton McIntyre site.

106

Figure 5-6b. Time series of predicted (blue) and observed (red) 24-hour speciated PM2.5

concentrations (µg/m3) for base case at Edmonton McIntyre site.

107

Table 5-17. 24-hour speciated PM2.5 (µg/m3) model performance for base case, the 1.33 km Capital Region domain and Jan-Feb period.


Obs Average Model

Ratio P/O FB FE

Jan-Feb EC EDMCIN 12 1.5 3.6 2.4 80.5 80.5

Jan-Feb NH4 EDMCIN 12 2.9 6.0 2.1 72.2 75.8

Jan-Feb NO3 EDMCIN 12 8.0 3.7 0.5 -56.0 76.5

Jan-Feb OA EDMCIN 12 3.8 6.2 1.6 61.8 62.4

Jan-Feb OC EDMCIN 12 2.7 6.2 2.3 88.8 88.8

Jan-Feb SO4 EDMCIN 12 2.3 13.6 6.0 150.1 150.1

Jan-Feb SOIL EDMCIN 11 0.3 4.3 12.5 171.6 171.6

Jan-Feb TCM EDMCIN 12 4.3 9.9 2.3 85.2 85.2

Jan-Feb TNH4 EDMCIN 12 2.9 9.2 3.2 109.4 109.4

Jan-Feb TNO3 EDMCIN 12 8.0 3.9 0.5 -50.8 71.5

Figure 5-7 displays scatter plots of predicted and observed hourly ozone and NOx concentrations for the CMAQ 1.33 km base case simulation. The ozone scatter plot is slightly off centered on the 1:1 line of perfect agreement, with a lot of scatter. The CMAQ model reproduces NOx concentrations relatively well with NMB of 7.7%. While ozone performance is not a focus in this study, the ozone under-estimations, especially at Edmonton sties (Table 5-18), indicate inadequate radicals in the model which are needed for NOx-HNO3 conversion.

Figure 5-7. Ozone (left) and NOx (right) scattered plots for base case, the 1.33 km Capital Region domain and Jan-Feb period.

108

Table 5-18. Hourly ozone model performance for CMAQ base case, the 1.33 km Capital Region domain and Jan-Feb period.


Obs Average Model

Ratio P/O NMB NME

Jan-Feb O3 090601 1402 15.5 11.2 0.7 -28.0 56.3

Jan-Feb O3 EDCEN 1401 9.8 5.5 0.6 -43.8 62.4

Jan-Feb O3 EDEAST 1393 16.0 11.9 0.7 -25.7 51.6

Jan-Feb O3 EDSTH 1401 13.9 8.8 0.6 -37.0 60.9

Jan-Feb O3 ELKIS 1337 23.2 21.2 0.9 -8.5 37.0

Jan-Feb O3 FTSASK 1402 15.5 11.2 0.7 -27.7 56.3

Jan-Feb O3 GENE 1341 20.2 20.3 1.0 0.8 42.5

Jan-Feb O3 LAMNT 1320 26.8 21.1 0.8 -21.2 35.3

Jan-Feb O3 TOMAHK 1342 22.7 18.7 0.8 -17.7 42.7

Jan-Feb O3 All Sites 12339 18.1 14.3 0.8 -20.8 47.0

Jan-Feb NOx 090601 1393 41.9 48.2 1.2 15.2 82.4

Jan-Feb NOx 093901 1338 10.5 13.1 1.3 25.8 84.9

Jan-Feb NOx EDCEN 1391 65.5 75.8 1.2 15.7 54.2

Jan-Feb NOx EDEAST 1393 50.0 40.7 0.8 -18.6 61.0

Jan-Feb NOx EDSTH 1394 42.7 50.0 1.2 17.1 73.4

Jan-Feb NOx ELKIS 1394 6.2 12.0 1.9 93.0 151.8

Jan-Feb NOx GENE 1339 12.8 14.5 1.1 13.2 79.2

Jan-Feb NOx LAMNT 1332 10.9 12.4 1.1 14.3 96.7

Jan-Feb NOx REDWIN 1122 30.3 23.3 0.8 -23.1 65.9

Jan-Feb NOx TOMAHK 1342 13.6 14.4 1.1 5.9 62.7

Jan-Feb NOx All Sites 13438 28.7 30.9 1.1 7.7 70.1

109

6.0 SOURCE APPORTIONMENT MODELLING

This chapter describes four (4) emission zero-out simulations conducted using the January-February, 2010 CMAQ modeling setup for 1.33 km domain with the 4 km base case simulation results as initial and boundary conditions. The purpose of these source apportionment model simulations is to investigate and assess the relative importance of specific emissions from various sectors on particulate matter in the Capital Region.

6.1 PM Source Apportionment

We used the “brute force” approach to perform source apportionment analysis. In this approach, a “base case” model simulation is performed and then emissions from a particular source are eliminated (“zero-out”). The importance of that source is assessed by evaluating the change in ambient air quality as the difference in pollutant concentration between the base case (with the source) minus the sensitivity scenario (without the source). A separate model zero-out sensitivity run is required for each source for which contributions are desired.

Note that the Capital Region brute force zero-out run is a sensitivity analysis, not a true source apportionment. In a source apportionment method (e.g., The Ozone Source Apportionment Technology [OSAT] and PM Source Apportionment Technology [PSAT] mass tracking source apportionment techniques in the CAMx model: www.camx.com), the sum of the contributions from all sources adds up to the total concentration in the base case simulation. However, neither of these source apportionment approaches is available in publicly released versions of CMAQ. Since the brute force zero-out runs have traditionally been referred to as a form of source apportionment so we continue to do so in this report.

6.2 Zero-Out Source Apportionment Simulations

In order to evaluate the impacts and contributions to winter PM due to various emission sectors, four zero-out emission sensitivity simulations were performed. The specific simulations were selected in order to address AESRD’s desire to ascertain the PM air quality impacts in the Capital Region due to local productions. The four zero-out source apportionment simulations are defined as follows:

1. On-road mobile sources: For this simulation, all on-road mobile source emissions within the Capital Region are eliminated (i.e., zeroed-out) within the CMAQ-ready emissions inputs. All other sources throughout the modelling domain, including stationary sources, biogenic and fire emissions in all regions, were kept at 2010 base case levels. As shown in Figure 6-1, on-road sources contribute to 51% and 11% of CO and NOx emissions in the Capital Region, respectively.

2. Power Plants (EGU): For this simulation, all power source (coal-fired and gas) emissions within the Capital Region are eliminated within the CMAQ-ready emission inputs. This sector is the main contributor of SO2 (69%) and NOx (44%) emissions in the Capital Region.

http://www.camx.com/

110

3. Other stationary point sources: For this simulation, all stationary point sources except power plants and UOG within the Capital Region are eliminated within the CMAQ-ready emission inputs. Major sources in this sector are refineries. This sector is a major contributor of VOC emissions (48%) and, to a lesser extent, SO2 emissions (27%) in the Capital Region.

4. All anthropogenic Sources: For this simulation, all anthropogenic sources including stationary point and area, on-road and off-road mobile sources and agricultural emissions within the Capital Region are eliminated. This simulation facilitates assessment of impact from outside of the Capital Region.

The source apportionment simulations were conducted on the 1.33 km modelling domain using the 4 km base case simulation results as initial and boundary conditions. The results of each zero-out simulation were compared with the base case simulation results to evaluate the impacts of PM due to local source emissions within the Capital Region.

Figure 6-1. Monthly anthropogenic emissions for the Capital Region by source sector (averaged over January-March, 2010).

NOx VOC TOG CO NH3 SO2 PM25 PM10

Agriculture 0 437 673 8 555 0 26 72

Commercial and residential heating 482 291 321 1,534 6 82 271 274

Transportation: Offroad 3,083 689 775 8,970 6 16 201 210

Transportation: Onroad 1,315 1,051 1,212 14,608 43 18 28 61

Industrial – others 1,461 2,631 3,741 2,418 252 2,175 161 311

Industrial – electric power generation 5,303 46 91 744 8 5,522 158 319

Industrial – upstream oil and gas 472 390 899 607 0 154 8 8

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

% c

on

trib

uti

on

s

Capital Region Emissions (monthly averaged Jan-Mar, 2010)

tonne/month

111

6.3 CMAQ Zero-Out Results and Analyses

Figure 6-2 presents the CMAQ-estimated two-month average PM2.5 for the base case simulation from January through February period. Elevated average PM2.5 concentrations appear around Edmonton extending to Industrial Heartland to the northeast. The highest predicted average PM2.5 (52 µg/m3) occurs near Edmonton. Most of these elevated PM2.5 concentrations are sulphate which peaks at the same location (Figure 6-3a). As expected, sulphate is generally high (>12 µg/m3) near large SO2 sources including Imperial Oil Strathcona refinery east of Edmonton and Shell Scotford Upgrader in the Industrial Heartland. Nitrate appears low with the maximum concentration of 3 µg/m3 predicted in the city core (Figure 6-3b). Ammonium has similar spatial patterns as sulphate but smaller magnitude (Figure 6-3e). Elemental carbon and organic aerosols are highest in the city core with concentrations over 3 and 6 µg/m3, respectively.

Figure 6-2. CMAQ-estimated average PM2.5 concentrations (µg/m3) for base case

112

(a) Sulphate

(b) Nitrate

(c) Elemental Carbon (d) Organic Aerosols

(e) Ammonium

(f) Soil

Figure 6-3. CMAQ-estimated average speciated PM2.5 concentrations (µg/m3) for base case (Note different maximum scales).

Figures 6-4 through 6-10 (a-d) display the results of the four zero-out CMAQ source apportionment simulations for scenarios 1 through 4, respectively that are expressed as differences in concentrations from the base case simulation. High PM appears to originate from local sources within the Capital Region as shown in Figure 6-5(d). Much of these contributions are attributable to sulphate (Figure 6-4d). Other point sector appears to be the major sulphate contributor; whereas other anthropogenic sources (e.g., agriculture and off-road) dominate contributions to all other PM species. Agriculture (in other anthropogenic source group) is the

113

main contributor to ammonia emissions in the Capital Region. Eliminating emissions from this sector resulted in less available ammonia to form ammonium nitrate.

Contributions of on-road mobile to the average PM2.5 concentrations are small (less than 1 µg/m3). Spatial distributions of nitrate contributions from on-road mobile sources (Figure 6-6a) generally follow highway networks with an exception to the city core. The nitrate contributions in the city core are potentially understated because the CALMOB6 model does not include ammonia.

Contributions of EGU sources to PM2.5 are generally less than 1 µg/m3 (Figure 6-4b). The highest EGU impact of 2.5 µg/m3 is seen near Sundance power plant. Outside of the Capital Region, sulphate contributions from EGU of about 1 µg/m3 are predicted close to large non-EGU ammonia sources (Figure 6-5b). While sulphate is reduced with the elimination of EGU sources, nitrate is predicted to increase in most areas due to more ammonium become available. However, these increases are small, generally less than 0.1 µg/m3 (Figure 6-6b).

Other point source sector gives highest PM impacts (up to 29 µg/m3) with large PM contributions close to Edmonton and Industrial Heartlands. This sector dominates contributions to total PM2.5 and several PM species including sulphate, ammonium, and soil in the Capital Region.

114

(a) On-road mobile

(b) EGU

(c) Other point sources

(d) All Anthropogenic

Figure 6-4. Difference in CMAQ-estimated average PM2.5 concentrations during January-February for (a) On-road mobile source, (b) EGU, (c) other point source, and (d) all anthropogenic source zero-out simulations (Base – Scenario).

115

(a) On-road mobile

(b) EGU

(c ) Other point sources


Figure 6-5. Difference in CMAQ-estimated average sulphate concentrations during January-February for (a) On-road mobile source, (b) EGU, (c) other point source, and (d) all anthropogenic source zero-out simulations (Base - Scenario).

116

(a) On-road mobile

(b) EGU



Figure 6-6. Difference in CMAQ-estimated average nitrate concentrations during January-February for (a) On-road mobile source, (b) EGU, (c) other point source, and (d) all anthropogenic source zero-out simulations (Base - Scenario).

117

(a) On-road mobile (b) EGU



Figure 6-7. Difference in CMAQ-estimated average ammonium concentrations during January-February for (a) On-road mobile source, (b) EGU, (c) other point source, and (d) all anthropogenic source zero-out simulations (Base - Scenario).

118

(a) On-road mobile (b) EGU



Figure 6-8. Difference in CMAQ-estimated average elemental carbon concentrations during January-February for (a) On-road mobile source, (b) EGU, (c) other point source, and (d) all anthropogenic source zero-out simulations (Base - Scenario).

119

(b) On-road mobile

(b) EGU


(d)All Anthropogenic

Figure 6-9. Difference in CMAQ-estimated average organic aerosols concentrations during January-February for (a) On-road mobile source, (b) EGU, (c) other point source, and (d) all anthropogenic source zero-out simulations (Base - Scenario).

120

(a) On-road mobile

(b) EGU


(d)All Anthropogenic

Figure 6-10. Difference in CMAQ-estimated average soil concentrations during January-February for (a) On-road mobile source, (b) EGU, (c) other point source, and (d) all anthropogenic source zero-out simulations (Base - Scenario).

6.4 Source Contributions at Monitoring Stations

Evaluation of source contributions at monitoring sites provides a closer look at local contributions from each source sector. In this analysis, the monitoring sites are roughly grouped as western monitors (TOMAHK and GENE), urban monitors (four monitors within Edmonton), industrial monitors (FTSASK, 90601, REDWIN), and eastern monitors (LAMNT, ELKIS). Generally, similar trends are seen among monitors in the same group for all three pollutants examined as discussed below.

As shown in Figure 6-11, the results of the source apportionment simulations show that high PM concentrations in the Capital Region are mostly from local sources (up to 45 µg/m3 at EDCEN). Contributions from EGU are low (up to 1 µg/m3) even at the western monitors located closer to EGU facilities. PM concentrations at both TAMAHK and GENE monitors are influenced mainly by sources outside of the Capital Region, likely due to UOG sources in the west. PM contributions from on-road mobile are also low with highest impacts seen at the industrial and Edmonton monitors. However, the on-road contributions at Edmonton monitors are understated due to lack of ammonia emissions from the CALMOB6 data. Edmonton and

121

industrial monitors are mostly affected by other anthropogenic sources. Impacts from other point sources at monitor stations vary depending on their downwind distance from sources. It appears that highest contributions of other point sources to the average PM2.5 concentrations (15 µg/m3) are seen at REDWIN. Contributions from sources outside of the region are about 4-6 µg/m3 across all monitors.

Industrial sources are major SO2 emitters, thus the impacts of other point sources (mainly refineries) to average sulphate concentrations are seen at near-by monitors of these sources (Figure 6-12). However, sulphate concentrations at the western monitors are mainly influenced by sources outside of the Capital Region instead of near-by EGU sources.

In contrast to sulphate, industrial sources have small impacts to nitrate (Figure 6-13). In fact, nitrate disbenefits (increases) occur at most sites when eliminating local EGU sources. But these disbenefits are small, less than 0.1 µg/m3. Other anthropogenic sources yield highest impacts (up to 2.5 µg/m3) to nitrate at all monitors due mainly to agriculture ammonia sources. Eliminating emissions from the agriculture sector resulted in less available ammonia to form ammonium nitrate. Figure 6-14 shows source contributions to ammonium that are similar to contributions to sulphate shown in Figure 6-12. This is expected as most of the predicted ammonium binds with sulphate.

Contributions to elemental and organic aerosols at Edmonton and industrial sites are mainly from other anthropogenic sources. Almost all organic aerosols predicted are primary. Commercial and residential heating and off-road sectors dominate primary PM emissions in the Capital Region. Since these emissions are spatially allocated near city centre, highest contributions are seen at EDCEN site.

Figure 6-11. Source Contributions to average PM2.5 at monitoring stations within the Capital Region.

0

10

20

30

40

50

60

mg/

m3

Natural + non-CR

Other anthro

On-road mobile

EGU

Other Point

122

Figure 6-12. Source Contributions to average sulphate at monitoring stations within the Capital Region.

Figure 6-13. Source Contributions to average nitrate at monitoring stations within the Capital Region (negative contributions are indicative of nitrate increase due to elimination of source emissions).

0

2

4

6

8

10

12

14

16m

g/m

3

Natural + non-CR

Other anthro

On-road mobile

EGU

Other Point

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

mg/

m3

Natural + non-CR

Other anthro

On-road mobile

EGU

Other Point

123

Figure 6-14. Source Contributions to average ammonium at monitoring stations within the Capital Region.

Figure 6-15. Source Contributions to average elemental carbon at monitoring stations within the Capital Region.

0.0

1.0

2.0

3.0

4.0

5.0

6.0m

g/m

3

Natural + non-CR

Other anthro

On-road mobile

EGU

Other Point

0.0

1.0

2.0

3.0

4.0

5.0

6.0

mg/

m3

Natural + non-CR

Other anthro

On-road mobile

EGU

Other Point

124

Figure 6-16. Source Contributions to average organic aerosols at monitoring stations within the Capital Region.

Figure 6-17. Source Contributions to average soil at monitoring stations within the Capital Region.

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0m

g/m

3

Natural + non-CR

Other anthro

On-road mobile

EGU

Other Point

0.0

2.0

4.0

6.0

8.0

10.0

12.0

mg/

m3

Natural + non-CR

Other anthro

On-road mobile

EGU

Other Point

125

7.0 SUMMARY AND RECOMMENDATIONS

ENVIRON International Corporation and Novus Environmental performed the Capital Region Particulate Matter Air Modelling Assessment Study for the Alberta Environmental and Sustainable Resources Development (ESRD). The objective of the study is to develop a Photochemical Grid Model (PGM) modelling database for the Capital Region, which includes Edmonton and surrounding communities, that reproduces the observed winter elevated fine particulate matter (PM2.5) concentrations sufficiently well that it can be a reliable tool for analyzing source contributions to elevated PM2.5 concentrations and evaluating the effects of alternative emission control strategies on elevated PM2.5 concentrations. Modelling inputs are a key to good model performance. In this study, modelling inputs were developed for the Community Multiscale Air Quality (CMAQ) modelling system. We used this CMAQ input database to conduct air quality model simulations.

Several sensitivity tests were performed to determine the optimal modelling configuration by evaluating the results against measurement data. After finalizing the base case modelling, the CMAQ model was applied to estimate the contributions of local production sources to PM concentrations in the Capital region. The resultant modelling technology will be provided to Alberta Environment for use in assisting them in air quality management.

7.1 Development of Modelling Inputs

7.1.1 Base Case Emissions Inputs

2010 emissions database were developed for Alberta and the Capital Region by combining the best current available inventories and augmented with industrial survey data and new Environmental Assessments. The study also introduced improvements to non-stationary sources including: (1) refinement of Edmonton on-road mobile emissions based on information provided by the City of Edmonton (CALMOB6 data) and (2) use of CANVEC data to improve spatial distributions of on-road and fugitive dust emissions.

The SMOKE emissions modelling system was used to generate the hourly, gridded, speciated CMAQ model-ready emissions inputs for January-March, 2010 period

7.1.2 Meteorological Inputs

This study applied the WRF meteorological model for meteorological modelling. The input data used in the WRF simulation consists of gridded NARR (North American Regional Reanalysis at 32 km grid resolution) data as well as upper air and surface observational data which are used to nudge the WRF fields to a better representation of the meteorology. Three sensitivity tests on the 1.33 km domain were conducted to identify an optimal performing WRF model inputs and configuration. One of the sensitivity tests used the NCEP Climate Forecast System Reanalysis (CFSR). The WRF fine grid performance was comparable between NARR and CFSR during high PM episode days in 2010. Both NARR and CFSR were converted to CMAQ input format for further sensitivity test.

126

7.2 Diagnostic Evaluation

Our initial CMAQ simulations based on the 2010 meteorology conditions showed some promise with elevated secondary PM2.5 concentrations, but there were also several issues and concerns. We discussed the initial CMAQ results with the ESRD and agreed to focus on diagnostic evaluation to improve CMAQ model performance rather than performing extended base case simulations with questionable model performance results. We performed several sensitivity tests to determine the optimal modelling configuration by evaluating the results against measurement data. The sensitivity tests focus primarily on the model performance based on the January-March, 2010, period that included the most days with exceedances of the PM2.5

CWS.

7.2.1 Sensitivity Test Results

Approximately ten sensitivity simulations were conducted to find an optimal CMAQ setup for winter PM modelling. These sensitivity tests were designed to investigate specific issues associated with the CMAQ model performance from the initial simulations. The CMAQ model performance during the winter PM episodes in the Capital Region was characterized by an underestimation of nitrate (NO3) and an overestimate of other species, especially sulphate (SO4) and Soil. The sensitivity test modelling identified the following issues:

The WRF meteorological inputs were very important and sensitivity modeling using different analysis field for boundary conditions and data assimilation identified the WRF-CSFR as performing better than WRF-NARR.

Dust emissions were overstated in the initial CMAQ runs. We would expect very little dust emissions during these winter episodes so they were set to zero in the Capital Region.

The SO4 overestimation bias is not due to primary SO4 emissions but rather gaseous SO2 emissions that are converted to SO4 in the atmosphere. We suspect that the WRF too moist meteorological conditions are contributing to the SO4 overestimation bias by overstating the amount of SO4 formation through the aqueous-phase chemistry pathway.

The SO4 overestimation bias is contributing to the NO3 underestimating by binding ammonia (NH3) that is no longer available to convert gaseous nitric acid (HNO3) to particulate NO3.

Ozone is generally underestimated in the Capital Region for these winter periods that causes to few radicals to be available so understates the conversion of NOx to HNO3. An increased ozone Boundary Conditions (BC) sensitivity tests resulted in increased NO3 concentrations.

We believe the most important pathway for HNO3/NO3 formation from NOx is the heterogeneous reaction of N2O5 for these winter PM episodes (Davis, Bhave and Foley, 2008). This reaction increases with colder temperatures and moister conditions. However, the reaction rate is capped based on laboratory measurements at much

127

warmer temperature and drier conditions than occurs during the Capital Region winter PM episodes. A sensitivity test raising this cap resulted in more NO3 formation.

7.2.2 Final Base Case

The final base case incorporated CALMOB6 Edmonton on-road emissions and excluded all fugitive dust emissions in Alberta. These fugitive dust emissions are from construction, agriculture, and unpaved road dust, and are expected to be insignificant during wintertime. The meteorology inputs were based on CFSR-WRF simulation.

Over all sites, CMAQ overestimates PM2.5 by a factor of 2 during January-February, 2010. Across the TEOM FDMS the overestimation bias is a factor of 1.5, with average predicted and observed values of 32 and 21 µg/m3, respectively, and resulting FB of 32 percent that achieves the Performance Criterion (≤±60 percent) but fails the Performance Goal (≤±30 percent). With the exception of nitrate, all species are overestimated with similar bias and error statistics. Sulphate and soil have the worst performances with FE more than 150%. Nitrate is underestimated by a factor of 2 with average observed and predicted values of 8 and 3.7 µg/m3, respectively. Better nitrate performance will require more radicals in the atmosphere to convert NOx to nitric acid. It is likely that either primary PM emissions are overstated and/or vertical mixing is understated during the modelling periods. Two major source sectors contributing to primary PMs are mobile and commercial/residential heating.

7.3 Source Apportionment Modelling Results

In order to evaluate the impacts and contributions to air quality due to various emission sectors, four zero-out simulations were performed for January-February, 2010 period. The four source apportionment simulations were defined as follows:

1. On-road mobile sources; 2. Electric power Plants (EGUs); 3. Other stationary point sources; and 4. All anthropogenic Sources.

On-road mobile sources are major contributor of CO emissions (51%). VOC and NOx contributions from this sector are 19% and 11%, respectively. Electric power sector is the main contributor of both SO2 (69%) and NOx (44%) emissions in the Capital Region 1.33 km domain. Other point sources, mainly refineries, are the second largest SO2 and NH3 contributors in the region (27% and 29%, respectively). Agriculture dominates ammonia emissions (64%). UOG sources are minor in the Capital Region, contributing to less than 5% of NOx and SO2. Off-road sources are the second largest NOx contributors in the region (25%) and their emissions are located mainly in the Edmonton city.

PM predicted in the Capital Region appears to originate from local sources. Sulphate is a key component of high winter PM2.5 predicted in the Capital Region (more than 8 µg/m3 with a maximum of 25 µg/m3 in Edmonton) and much of this sulphate is attributable to the other

128

stationary point sources. The results of the source apportionment simulations showed that local on-road mobile and EGUs had a small effect on the CMAQ-estimated winter averaged PM2.5 concentrations, generally less than 1 µg/m3. Other anthropogenic sources (e.g., agriculture and off-road sources) dominate contributions to nitrate in the region, which is due in part to the agricultural source category that contains most of the ammonia emissions so when ammonia is eliminated nitrate cannot form. Contributions to elemental and organic carbons are mainly from major primary PM emitters including commercial/residential heating and off-road sources.

7.4 Uncertainties and Limitations

There are several limitations and sources of uncertainties associated with the modelling results of this study, including the meteorology, the emission inventory estimates and the air quality modelling.

Regarding the meteorology, while the overall 2010 WRF model performance (based on wind speed, temperature and PBL) appears reasonable and similar to what has been used for previous air quality modelling studies, the model predicted poor moisture content (i.e., relative humidity close to 100%) especially during PM episode days. The over-predicted wetness resulted from the underlying initial and boundary inputs from both NARR and CFSR analysis fields that are also used in the analysis nudging. The too wet conditions may be contributing to the sulphate overestimation bias through too much formation through aqueous-phase chemistry. The high PM biases occur across the region and through January-March modelling period implying that there may be some systematic issues with the simulation.

Emissions are a critical component of the air quality modelling. We highlight some of the key limitations and uncertainties with regards to the emission inputs including the following:

Emission inventory development – Emission inputs used in this study were based on the 2010 SAOS emissions inventory with updated industrial data. Although a quantitative analysis of the uncertainties of the inventory data was beyond the scope of the present study, it is not uncommon for emission inventory data to exhibit levels of uncertainty in the range of ±30% depending on the specific source sector. While much of the effort in this study focus on large NOx and SO2 sources, ammonia emissions in the underlying data were not updated. In addition, lacking of ammonia emissions in the CALMOB6 data for the on-road mobile sector resulted in understated nitrate contributions in the Edmonton city core.

Emission modelling – Uncertainties and limitations in the emissions modelling for the study include the speciation of VOC emissions as well as the temporal and spatial allocation of the emission inventory. In particular, the spatial allocation of regional emission estimates introduces additional limitations with respect to the modelling inventories used for the study. In the non-stationary sources, estimates at the Provincial level were allocated to grid cells using spatial surrogates developed for the entire province. For certain emission source sectors, this could result in allocation of

129

these emissions to erroneous areas of the modelling domain and/or artificially spreading these emissions across for broader regions than is realistic.

Some of the industrial point sources will have episodic emissions (e.g., fugitives) that are not captured by the average emissions used in the modelling

As is true for any grid-based chemical transport model, such as CMAQ used in the current study, a number of inherent limitations and uncertainties are present. These include uncertainties in the chemical speciation of input data and inherent assumptions with respect to the chemical mechanism implementation in the model, as well as uncertainties and biases associated with meteorology, transport and deposition of modelled pollutants. Moreover, there remains uncertainty in the treatment of aqueous chemistry, aerosol thermodynamics, and rates of removal of PM and gaseous species by wet and dry deposition. The winter PM episode temperature conditions fall outside of the range that the CMAQ parameterization of HNO3 formation through heterogeneous N2O5 reactions. As this reaction rate increases with reduced temperature and increased moisture, it is likely understated in the CMAQ simulations. There are also errors and biases in the monitoring data that were used in the MPE, which in some cases may be as large as, or larger than, the modelled biases for certain species, particularly PM species.

7.5 Conclusions and Recommendations

The PM model performance reveals systematic high PM biases in all species by NO3 that is underestimated. Sulphate is predicted to make up the largest component of secondary PM2.5, but this prediction is largely over-stated. On the contrary, nitrate is under-estimated by the CMAQ model.

The development of a photochemical modelling database usually involves the performance of diagnostic model simulations designed to refine the model inputs and configuration to improve model performance. In this study, a series of diagnostic tests were performed, but further examinations are required for better simulation results. The following recommendations are made with respect to improvements to the CMAQ 2010 base case modelling database.

Meteorological Inputs:

In this study, the model performance was affected by too high moisture content as input from NARR or CFSR. However, we believe that WRF modelling results could be improved from a number of aspects:

o The latest OBSGRID program (aligning with WRF v3.5.1) should be able to provide observed moisture field and other met fields for OBS nudging. In the current study, we used the default setting to process OBS data with limitation.

o It is possible that Four-Dimensional Data Assimilation (FDDA) moisture field be corrected based on the observations, particularly on the surface or in the PBL, to help improve moisture content predictions. However, it is important to be aware of the strength and depth of WRF interior grid nudging (i.e., analysis nudging, or

130

FDDA). Bowden et. al (2012) indicated that, the influence of FDDA nudging throughout the atmospheric column, particularly near the PBL, where nudging too strongly toward coarse input fields could dampen the RCM’s ability to generate important mesoscale features near the surface. It is recommended that FDDA turned off in PBL in the outer domain to allow the inter-domain PBL physics to develop undistracted.

o Use MESOnet as OBS nudging input for enhanced PBL met observations Emission Inventory:

Conversion of the primary emitted NOx to gaseous nitric acid (HNO3) relies on reactivity of the atmosphere. Specifically, the model will need more active radicals from VOC and/or background ozone. VOC speciation profiles from on-road mobile and refinery should be reviewed to ensure proper allocation of VOC to model species.

Forming nitrate requires more availability of ammonia emissions to bind with HNO3. The availability of ammonia makes the ammonia inventory critically important. The ammonia emissions in Alberta came from the agricultural sector that is due mainly to livestock and fertilize. There are also some industrial facilities in Edmonton that emit ammonia as well as ammonia emitted from the oil sands region to the northeast. The base case inventory should be reviewed to ensure that the all ammonia sources are accounted for. For example, use provincial level ammonia emissions for on-road sector since CALMOB6 data does not include these emissions. We note, however, that ammonia emissions are more uncertain than other criteria air contaminants.

Hot spots generally occur near emitting sources such as high SO2 concentrations occurring near refineries sources. Emissions of these facilities should be reviewed to ensure correctness of the emissions inputs.

Primary PM emissions and their spatial/temporal allocations should also be reviewed. Despite the removal of all fugitive dust emissions, the model still over-estimated primary PM species (e.g., elemental carbon).

SO2 emissions need to be reviewed to see whether they may be overstated.

Diagnostic Sensitivity Tests: The same winter period should be investigated further through PGM to identify model options that could improve PM model performance, such as:

Emission sensitivity tests, such as increasing anthropogenic VOC and NH3 emissions;

Analysis of the boundary conditions including comparison of the GEOS-CHEM BCs to other global model outputs (e.g., MOZART) , particularly on background ozone concentrations;

Reductions in the WRF moisture fields used as input to CMAQ to investigate their effect on the sulphate overestimation issue.

131

Additional vertical mixing tests are recommended as the overestimation of all species (except nitrate), even primary emitted species (e.g., EC), may indicate insufficient vertical mixing. The effects of the urban heat island on vertical mixing in the Capital Region is likely not fully account for in the WRF/CMAQ modelling.

Further analysis of the heterogeneous N2O5 reaction probability is needed. Removing the current cap on these reactions that occurs a 0°C and 70 percent RH will allow it to increase the HNO3 formation rate from N2O5 under below freezing conditions as occurs during winter PM episodes.

The CMAQ default chemistry coupling time step is likely limiting the throughput of HNO3 formation through N2O5 and possible other reactions. The gas-phase and aerosol-phase chemistry are coupled only every 15 minutes and lowering this coupling time step may increase NO3 formation. The gas-phase is converting NO2 to N2O5 using ~30 s time steps, with part of the N2O4 disassociating back to NO2. However, the gas-phase chemistry and aerosol/heterogeneous chemistry are only coupled every 15 minutes in CMAQ. If these processes were allowed to interact every ~30 s, that could form more HNO3 and NO3.

132

8.0 REFERENECES

Boylan, J.W. and A.G. Russell. 2006. PM and Light Extinction Model Performance Metrics, goals and Criteria for Three-Dimensional Air Quality Models. Atmos. Env. 40:4946-4959.

Davis, J.M., P.V., Bhave, and K.M., Foley. 2008. Parameterization of N2O5 Reaction Probabilities on the Surface of Particles Containing Ammonium, Sulphate and Nitrate. Atmos. Chem. Phys., 8, 5295-5311. http://www.atmos-chem-phys.net/8/5295/2008/acp-8-5295-2008.html

Idriss, A. and F. Spurrell. 2009. Air Quality Model Guideline. Alberta Environment, Climate Change, Air and Land Policy Branch, Edmonton, Alberta. (http://environment.gov.ab.ca/info/library/8151.pdf). May.

Morris, R.E., B. Koo, B. Wang, G. Stella, D. McNally and C. Loomis. 2009a. Technical Support Document for VISTAS Emissions and Air Quality Modelling to Support Regional Haze State Implementation Plans ENVIRON International Corporation, Novato, CA and Alpine Geophysics, LLC, Arvada, CO. March.

Morris, R.E., B. Koo, S. Lau, D. McNally, T.W. Tesche, C. Loomis, G. Stella, G. Tonnesen and C-J. Chien. 2004b. “VISTAS Phase II Emissions and Air Quality Modelling – Task 4a Report: Evaluation of the Initial CMAQ 2002 Annual Simulation” Prepared by ENVIRON International Corporation, Alpine Geophysics, LLC, and the University of California, Riverside (CE-CERT). Prepared for the VISTAS Technical Analysis Committee. September.

Morris, R.E., B. Koo, S. Lau, T.W. Tesche, D. McNally, C. Loomis, G. Stella, G. Tonnesen and Z. Wang. 2004a. “VISTAS Emissions and Air Quality Modelling – Phase I Task 4cd Report: Model Performance Evaluation and Model Sensitivity Tests for Three Phase I Episodes”. Prepared by ENVIRON International Corporation, Alpine Geophysics, LLC, and the University of California, Riverside (CE-CERT). Prepared for the VISTAS Technical Analysis Committee.

Morris, R.E., B. Koo, T. Sakulyanontvittaya, G. Stella, D. McNally, C. Loomis and T.W. Tesche. 2009b. Technical Support Document for the Association for Southeastern Integrated Planning (ASIP) Emissions and Air Quality Modelling to Support PM2.5 and 8-Hour Ozone State Implementation Plans. ENVIRON International Corporation, Novato, CA and Alpine Geophysics, LLC, Arvada, CO. March 24.

Nopmongcol, U., T. Shah, J. Johnson, T. Sakulyanontvittaya, P. Piyachaturawat, R. Morris and T. Pollock. 2011. Air Quality Modelling Exercise using Community Multiscale Air Quality (CMAQ) Model for North Saskatchewan Regional Planning (NSRP). ENVIRON EC (Canada), Inc., Mississauga, ON. Prepared for Alberta Environment, Northern Region, Edmonton, AB. November 31.

Pace, 2005: “Methodology to Estimate the Transportable Fraction (TF) of Fugitive Dust Emissions for Regional and Urban Scale Air Quality Analyses”. US Environmental Protection Agency.

http://www.atmos-chem-phys.net/8/5295/2008/acp-8-5295-2008.html

http://www.atmos-chem-phys.net/8/5295/2008/acp-8-5295-2008.html

http://environment.gov.ab.ca/info/library/8151.pdf

133

http://www.epa.gov/ttn/chief/emch/dustfractions/transportable_fraction_080305_rev.pdf.

Sassi, M., Leroux, A. and Boucher, L. 2012. Improving spatial allocation of construction emissions in Canada. Presented at 20th International Emission Inventory Conference: Emission Inventories - Meeting the Challenges Posed by Emerging Global, National, Regional and Local Air Quality Issues. Tampa, Florida, August 13-16, 2012.

USEPA. 1991. "Guidance for Regulatory Application of the Urban Airshed Model (UAM)." Office of Air Quality Planning and Standards, U.S. Environmental Protection Agency, Research Triangle Park, NC.

USEPA. 2007. “Guidance on the Use of Models and Other Analyses for Demonstrating Attainment of Air Quality Goals for Ozone, PM2.5 and Regional Haze.” U.S. Environmental Protection Agency, Research Triangle Park, NC. EPA-454/B-07-002. April.

APPENDIX A

Selection of PM2.5 Episodes for 2010 Sensitivity Test Modelling and 2008-2009 Modelling using the 1.33 km Capital Region Domain

A-1

Appendix A: Selection of PM2.5 Episodes for 2010 Sensitivity Test Modelling and 2008-2009 Modelling using the 1.33 km Capital Region Domain

INTRODUCTION In the modelling plan, the CMAQ photochemical grid model would be exercised on the 36 km SWCA, 12 km Alberta and 4 km NSR modelling domains for the six winter months (Jan-Mar and Oct-Dec) in 2008, 2009 and 2010. For the 2010 year, CMAQ would be applied using the 1.33 km Capital Region domain for the entire six month period. However, for the 2008 and 2009 years the 1.33 km domain would only be used for elevated PM2.5 days (e.g., above the CWS standard) with at least one day of spin-up. We also planned on using elevated PM2.5 episodes from 2010 for the WRF and CMAQ sensitivity test modelling that is being used to determine an optimal performing model configuration. This Appendix discusses the selection of the elevated PM2.5 episodes in the Capital Region for sensitivity test modelling (2010) and 1.33 km Capital Region domain modelling (2008 and 2009). Note, however, that due to changes in the scope of work regarding concerning CMAQ performances we only conducted CMAQ modelling for three winter months (Jan-Feb) in 2010.

Approach

Episodes were defined to include all winter days during 2008-2010 that had observed 24-hour PM2.5 concentrations that exceeded the 30 µg/m3 Canada Wide Standard (CWS). Episodes were extended forward or backward in time to include days that also exceeded the 20 µg/m3 Management Plan Trigger Level, but episodes were not developed for isolated days with just exceedances of the 20 µg/m3 threshold.

The continuous (hourly) PM2.5 measurement data were downloaded from the CASA and NAPS websites. Figures A-1 and A-2 displays the locations of the monitoring sites with more details on the monitoring sites provided in Table A-1. There was numerous overlap in the data from CASA and NAPS so NAPS monitoring sites with data duplicated at the CASA sites were eliminated; this resulted in only one NAPS site remaining (90601, see Figure 2). Table A-1 lists the monitoring sites in the Capital Region that were analyzed. There are several different types of continuous PM2.5 monitoring technologies used in the Capital Region during this time period each with its own measurement artifacts. No attempt was made to make any corrections to the PM2.5

measurements at this time. In 2008, the Edmonton sites used the TEOM @ 30C and TEOM @ 40C continuous PM2.5 monitors. In 2009 the TEOM @ 40C was dropped and during the year the TEOM @ 30C FDMS (self-referencing) device was added. Some sites also included a BAM continuous analyzed. Speciated 24-hour PM2.5 measurements were just collected at the McIntyre site on a 1:3 day sampling frequency. Given that a big focus of this study is on the ability of the CMAQ model to simulate the elevated winter PM2.5 events that have high secondary PM2.5 concentrations, the availability of the speciated PM2.5 measurements are critically important for the model performance evaluation and sensitivity modelling.

The hourly PM2.5 measurements were averaged to obtain 24-hour PM2.5 values for comparisons with the 30 µg/m3 threshold for the episode selection. A day had to have at least 75% of its

A-2

hourly data (18 hours) to make a valid 24-hour PM2.5 measurement. If there were less than 18 hours of valid hourly data the 24-hour PM2.5 measurement was set equal to the -777 missing data flag. This is in contrast to the -999 missing data flag when there was no data available.

There were 9 CWS exceedance days in the Capital Region during 2008 all of which occurred during the six-month winter period. During 2009 there were only 7 CWS exceedance days with 2 of the days falling outside of the winter period. However, in 2010 there were 41 CWS exceedance days in the Capital Region. The very highest 24-hour PM2.5 days in the Capital Region during 2010 were August 19 (349.8 µg/m3), 20 (109.3 µg/m3), 21 (76.0 µg/m3) and 22 (48.9 µg/m3) that was a period with large amounts of wildfires whose smoke plumes impacted the Capital Region so were not considered for modelling leaving 37 winter CWS exceedance days in 2010.

Figure A-1. Locations of CASA PM2.5 monitoring sites within the Capital Region.

A-3

Figure A-2. Locations of NAPS PM2.5 monitoring sites within the 4 km Capital Region domain.

A-4

Table A-1. Capital Regional continuous PM2.5 monitoring sites. Stn ID Name Method



EDMC FDMS Edmonton Central TEOM @ 30C with FDMS (self-referencing)



EDME FDMS Edmonton East TEOM @ 30C with FDMS (self-referencing)

MCIN BAM1 Edmonton McIntyre E-BAM (Met1)

MCIN TEOM3 Edmonton McIntyre TEOM @ 30C

MCIN FDMS Edmonton McIntyre TEOM @ 30C with FDMS (self-referencing)

MCIN BAM3 Edmonton McIntyre BAM @ 35RH



EDMS FDMS Edmonton South TEOM @ 30C with FDMS (self-referencing)

ELKI TEOM4 Elk Island TEOM @ 40C

FTSA TEOM3 Fort Saskatchewan TEOM @ 40 C

FTSA SHARP Fort Saskatchewan Sharp (hybrid nephelometer/BAM sys) with data reported at actual ambient conditions

GENE TEOM4 Genesee TEOM @ 40C

LAMA BAM Lamont Met One BAM 1020

REDW TEOM4 Redwater Industrial TEOM @ 40C

TOMA TEOM4 Tomahawk TEOM @ 40C

90601 SES NAPS 09601 SES

90601 TEOM NAPS 09601 TEOM

2010 EPISODES 2010 episodes were defined so that we could use the highest ranked ones for the WRF and CMAQ sensitivity modelling. Nine episodes were identified in 2010 that included all 37 of the winter CWS exceedance days, plus spin-up days and adjacent days that exceed the 20 µg/m3 trigger threshold. Table A-14 at the end of this Appendix lists the observed daily PM2.5 values at all monitoring sites for the entire 2010 year with CWS exceedance days highlighted in yellow, values above the 20 µg/m3 Management Trigger Threshold (but below the CWS) highlighted in blue and the nine 2010 episodes are shown by red boxes. Table A-14 shows that all winter CWS exceedance events were captured by the nine 2010 episodes, as well as most of the days that exceeded the 20 µg/m3 threshold.

A-5

2010 Episodes#1: Jan 17-21, 2010

2010 Episode#1 spans the January 17-21 5-day period and includes three CWS exceedance days including a 40.6 µg/m3 exceedance at the McIntyre PM2.5 speciation monitoring site (Table A-2 and Figure A-3). There is a clean spin-up day on January 17 (maximum PM2.5 of 7.5 µg/m3) and a spin-down day on January 21. Because of the three high CWS exceedance days and high value at McIntyre speciation monitor, Episode#1 was the second highest ranked 2010 episode for sensitivity modelling.

Table A-2. Observed Capital Region maximum 24-hour PM2.5 concentration and value at McIntyre PM2.5 speciation site for 2010 Episode#1.


17 2010 Jan 17 7.5 3.2

18 2010 Jan 18 37.5

19 2010 Jan 19 57.0

20 2010 Jan 20 51.0 40.6

21 2010 Jan 21 17.0

Figure A-3. 24-hour PM2.5 observations during 2010 Episode#1 (January 17-21, 2010) at monitoring sites in the Capital Region with the 20 and 30 µg/m3 thresholds indicated by the dotted and sold lines, respectively.

0

10

20

30

40

50

60

17 18 19 20 21

Jan

2010

Episode 1 090601: Method: SES

090601: Method: TEOM

Edmonton Central: Method: Teom @ 30C

Edmonton Central: Method: Teom @ 30C with FDMS (self referencing)


Edmonton East: Method: Teom @ 30C

Edmonton East: Method: Teom @ 30C with FDMS (self referencing)


Edmonton McIntyre (speciated): Total Adjusted PM2.5

Edmonton McIntyre: Method: BAM @ 35RH

Edmonton McIntyre: Method: E-BAM (Met1)

Edmonton McIntyre: Method: Teom @ 30C

Edmonton McIntyre: Method: Teom @ 30C with FDMS (self referencing)

Edmonton South: Method: Teom @ 30C

Edmonton South: Method: Teom @ 30C with FDMS (self referencing)


Elk Island: Method: Teom @ 40C

Fort Saskatchewan: Method: Sharp (hybrid nephelometer/BAM sys) with data reported at actualambient conditionsFort Saskatchewan: Method: Teom @ 40C

Genesee: Method: Teom @ 40C

Lamont: Method: Met One Bam 1020

Redwater Industrial: Method: Teom @ 40C

Tomahawk: Method: Teom @ 40C

Linear (Canada Wide Standard (CWS))

Linear (Alberta Management Plan Trigger Level)

Station and Method

Year Month Day

Average of Value

A-6

2010 Episodes#2: Jan 26 – Feb 4, 2010

Episode#2 spans the 10-day period of January 26 through February 4, 2010 and includes 5 CWS exceedance days. This includes January 29th that was the highest PM2.5 day in the Capital Region (outside of the August wildfire days) with a maximum value of 74.3 µg/m3 and exceedances of the CWS at all four Edmonton monitoring sites. January 29th also has PM2.5 speciation data at the McIntyre site that was above the CWS. For these reasons it was the highest ranked 2010 episode for sensitivity modelling. Table A-3 and Figure A-4 summarize the observed PM2.5 data during Episode#2.



26 2010 Jan 26 9.7 11.5

27 2010 Jan 27 24.6

28 2010 Jan 28 58.0

29 2010 Jan 29 74.4 61.4

30 2010 Jan 30 25.3

31 2010 Jan 31 8.7

32 2010 Feb 1 30.7 25.4

33 2010 Feb 2 40.5

34 2010 Feb 3 37.9

35 2010 Feb 4 24.8 18.9

Figure A-4. 24-hour PM2.5 observations during 2010 Episode#2 (January 26 – February 4, 2010) at monitoring sites in the Capital Region with the 20 and 30 µg/m3 thresholds indicated by the dotted and sold lines, respectively.

0

10

20

30

40

50

60

70

80

26 27 28 29 30 31 1 2 3 4

Jan Feb

2010

























Station and Method

Year Month Day

Average of Value

Combined Date

A-7

2010 Episode#3: February 9-13, 2010

The five day Episode#3 includes two CWS exceedance days on February 11 and 12, with no exceedances at the McIntyre PM2.5 speciation monitoring site. Details on this episode are shown in Table A-4 and Figure A-5.



40 2010 Feb 9 25.7

41 2010 Feb 10 21.8 13.6

42 2010 Feb 11 39.5

43 2010 Feb 12 39.3

44 2010 Feb 13 15.8 8.0

Figure A-5. 24-hour PM2.5 observations during 2010 Episode#3 (February 9-13, 2010) at monitoring sites in the Capital Region with the 20 and 30 µg/m3 thresholds indicated by the dotted and sold lines, respectively.

0

5

10

15

20

25

30

35

40

45

9 10 11 12 13

Feb

2010

























Station and Method

Year Month Day

Average of Value

A-8

2010 Episode#4: Feb 15-17, 2010

Episode#4 is a short 3-day episode with one CWS exceedance day on February 16, 2010 (Table A-5 and Figure A-6).



46 2010 Feb 15 21.7

47 2010 Feb 16 33.5 16.4

48 2010 Feb 17 24.6

Figure A-6. 24-hour PM2.5 observations during 2010 Episode#4 (February 15-17, 2010) at monitoring sites in the Capital Region with the 20 and 30 µg/m3 thresholds indicated by the dotted and sold lines, respectively.

2010 Episode#5: Feb 20 – Mar 8, 2010

Episode#5 is the longest PM2.5 episode in 2010 that is 17 days long with 11 CWS exceedance days. It could have been split up into two episodes, but the days between the CWS exceedance days were still fairly high and above the 20 µg/m3 threshold. However, there were no CWS exceedances at the McIntyre speciated PM2.5 monitoring site, although the PM2.5 was elevated there and slightly above the 20 µg/m3 threshold. Given the many high PM2.5 days during Episode#5, it was the fourth highest ranked episode in 2010.

0

5

10

15

20

25

30

35

40

15 16 17

Feb

2010

























Station and Method

Year Month Day

Average of Value

A-9



51 2010 Feb 20 26.8

52 2010 Feb 21 33.7

53 2010 Feb 22 31.7 18.7

54 2010 Feb 23 36.8

55 2010 Feb 24 59.1

56 2010 Feb 25 40.8 21.7

57 2010 Feb 26 26.5

58 2010 Feb 27 26.8

59 2010 Feb 28 44.4 23.3

60 2010 Mar 1 55.9

61 2010 Mar 2 29.2

62 2010 Mar 3 40.8 21.3

63 2010 Mar 4 47.5

64 2010 Mar 5 39.7

65 2010 Mar 6 28.9 22.1

66 2010 Mar 7 37.6

67 2010 Mar 8 24.5

Figure A-7. 24-hour PM2.5 observations during 2010 Episode#5 (February 20 - March 8, 2010) at monitoring sites in the Capital Region with the 20 and 30 µg/m3 threshold indicated by the dotted and sold lines, respectively.

0

10

20

30

40

50

60

70

20 21 22 23 24 25 26 27 28 1 2 3 4 5 6 7 8

Feb Mar

2010


















Fort Saskatchewan: Method: Sharp (hybrid nephelometer/BAM sys) with data reported at actualambient conditions

Fort Saskatchewan: Method: Teom @ 40C







Station and Method

Year Month Day

Average of Value

Combined Date

A-10

2010 Episode#6: Mar 25-27, 2010

Episode#6 is a short 3-day period covering March 26-27, 2010 (Table A-7) with one CWS exceedance day where the exceedance occurred at only one monitoring site (Figure A-8).

Table A-7. Observed Capital Region maximum 24-hour PM2.5 concentration and value at McIntyre PM2.5 speciation site for 2010 Episode#6. Year Month Day Max McIntyr

84 2010 Mar 25 26.2

85 2010 Mar 26 33.3

86 2010 Mar 27 24.4 9.8

Figure A-8. 24-hour PM2.5 observations during 2010 Episode#6 (March 25-27, 2010) at monitoring sites in the Capital Region with the 20 and 30 µg/m3 threshold indicated by the dotted and sold lines, respectively.

0

5

10

15

20

25

30

35

25 26 27

Mar

2010


















Fort Saskatchewan: Method: Sharp (hybrid nephelometer/BAM sys) with data reported at actualambient conditions

Fort Saskatchewan: Method: Teom @ 40C







Station and Method

Year Month Day

Average of Value

A-11

2010 Epsiode#7: Nov 30 – Dec 2, 2010

Episode#7 was a short 4-day episode with two CWS exceedance days (Table A-8 and Figure A-9).



334 2010 Nov 30 13.4

335 2010 Dec 1 32.1 10.6

336 2010 Dec 2 34.3

337 2010 Dec 3 23.7

Figure A-9. 24-hour PM2.5 observations during 2010 Episode#7 (November 30 – December 3,

2010) at monitoring sites in the Capital Region with the 20 and 30 µg/m3 threshold indicated by the dotted and sold lines, respectively.

0

5

10

15

20

25

30

35

40

30 1 2 3

Nov Dec

2010

























Station and Method

Year Month Day

Average of Value

Combined Date

A-12

2010 Episode#8: Dec 5-9, 2010

Episode#8 spans 5 days in early December (Table A-9) with three CWS exceedance days with the CWS exceedances occurring at multiple sites in the Capital Region (Figure A-10). This includes exceedances at the McIntyre PM2.5 speciation monitoring site on December 7, 2010. For this reason, Episoe#8 was the third highest ranked episode during 2010.



339 2010 Dec 5 20.5

340 2010 Dec 6 57.1

341 2010 Dec 7 58.1 42.0

342 2010 Dec 8 38.8

343 2010 Dec 9 12.8

Figure A-10. 24-hour PM2.5 observations during 2010 Episode#8 (December 5-9, 2010) at monitoring sites in the Capital Region with the 20 and 30 µg/m3 threshold indicated by the dotted and sold lines, respectively.

0

10

20

30

40

50

60

70

5 6 7 8 9

Dec

2010

























Station and Method

Year Month Day

Average of Value

A-13

2010 Epsiode#9: Dec 21-27, 2010

The last 2010 episode (Episode#9) is seven days long (Table A-10) with four CWS exceedance days of which three of the days has exceedances of the CWS at multiple monitoring sites (Figure A-11). Although there are no CWS exceedances at the McIntyre PM2.5 speciation site, the PM2.5 concentrations are high and are just below the CWS. Episode#9 was considered as one of the top four episodes for use in the sensitivity modelling, but we were concerned about the representativeness of the emissions during this holiday period.



355 2010 Dec 21 13.5

356 2010 Dec 22 33.9 24.5

357 2010 Dec 23 37.4

358 2010 Dec 24 16.3

359 2010 Dec 25 38.5 29.8

360 2010 Dec 26 37.5

361 2010 Dec 27 20.1

Figure A-11. 24-hour PM2.5 observations during 2010 Episode#9 (December 21-27, 2010) at monitoring sites in the Capital Region with the 20 and 30 µg/m3 threshold indicated by the dotted and sold lines, respectively.

0

5

10

15

20

25

30

35

40

45

21 22 23 24 25 26 27

Dec

2010

Episode 9090601: Method: SES
























Station and Method

Year Month Day

Average of Value

A-14

2009 EPISODES Four winter elevated PM2.5 episodes were identified in 2009. These four episodes included all winter days that measured exceedances of the 24-hour PM2.5 CWS. Table A-13 at the end of this Appendix lists the observed 24-hour PM2.5 concentrations in the Capital Region during 2009 with CWS exceedances colored in yellow, exceedances of the 20 µg/m3 threshold colored in blue and the four episodes indicated by the boxed red areas. Because CMAQ modelling is run using the GMT time zone and the 24-hour PM2.5 observations are collected using MST, two full spin-up days are used on the 1.33 km domain before the first day of interest. This gives us over 40 hours of spin-up on the Capital Region 1.33 km domain. One day is added to the end of the episode after the last day of interest to see whether the model can accurately reproduce the clean out period. Although there were no CWS exceedances at the McIntyre speciated PM2.5 monitor during the first three episodes, on the fourth episode there was a CWS exceedance on December 27, 2009 that was the worst PM2.5 day during 2009 in the Capital Region.


Episode#1 had one CWS exceedance day where the 30 µg/m3 threshold was exceeded at three monitoring sites in the Capital Region. On February 13th there was also an exceedance of the 20 µg/m3 threshold.

2009 Episode#2: February 28 – March 4, 2009

The CWS was exceeded on March 3 at one site during Episode#2, with another site exceeding the 20 µg/m3 threshold on that day. The 20 µg/m3 threshold was also exceeded at one site on March 1.

2009 Episode#3: December 14-18, 2009

Episode#3 had one site that exceeded the CWS on December 16, 2009 and 5 sites that exceeded the 20 µg/m3 threshold. December 17 also had five sites that exceeded the 20 µg/m3 threshold but no CWS exceedances.

2009 Episode#4: December 23-29, 2009

Episode#4 was the most severe PM2.5 episode in the Capital Region during 2009 with two days exceeding the CWS. December 27, 2009 was the worst day in 2009 with 8 exceedances of the CWS at four different monitoring sites. December 27 was also one of the 1:3 day McIntyre speciated PM2.5 measurement days where a 47.0 µg/m3 value was observed. During December 24-28, 2009 there was at least one site exceeding the 20 µg/m3 threshold. Note that when modelling 2009 Episode#4, the 1.33 km domain should also be used for December 30-31, 2009 so that it can be used to spin-up the model for the 2010 1.33 km modelling of the 6-month winter period.

A-15

2008 EPISODES There were four episodes identified for 1.33 km modelling during 2008 that included all of the CWS exceedance days. These episodes are highlighted in red in Table A-12.

2008 Episode#1: January 6-14, 2008

There were two days (January 9 and 11) during 2008 Episode#1 that measured exceedances of the CWS with five days having exceedances of the 20 µg/m3 threshold at multiple sites. In fact, January 8-11 had measured values of 29 µg/m3 or higher with January 10 also having high speciated PM2.5 observations at McIntyre.

2008 Episode#2: January 31 – February 5, 2008

February 3 and 4 were CWS exceedance days at the McIntyre monitoring site using the BAM method, but not the other PM2.5 measurement methods, although the PM2.5 concentrations were elevated using the other methods. There were three days with PM2.5 observations above 20 µg/m3 at multiple monitoring sites. The speciated PM2.5 at McIntyre on February 3 is also above the CWS.


Episode#3 had three days above the CWS with February 22, 2008 having CWS exceedances at 5 monitoring sites. Four of the days during Episode#3 had measured PM2.5 concentrations above 20 µg/m3 at multiple sites. The PM2.5 at the McIntyre speciated PM sites was high on February 21st.

2008 Episode#4: October 26-30, 2008

The fourth 2008 episode had two days above the CWS at the McIntyre BAM monitoring with elevated PM2.5 also measured at the other monitoring methods at McIntyre as well as at other monitoring sites. There were no high measured PM2.5 concentrations at the 1:3 day speciated PM2.5 McIntyre monitor.

A-16

Table A-12. Measured 24-Hour PM2.5 at monitoring sites in the Capital Region during 2008 with values above the CWS (> 30 µg/m3) highlighted in yellow, values above 20 µg/m3 highlighted in blue and PM2.5 modelling episodes for the 1.33 km fine-scale Capital Region modelling domain shown by the red boxes.

JD Y M D EDMC TEOM4

EDMC TEOM3

EDME TEOM4

EDME TEOM3

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMC TEOM3

EDMS TEOM4

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

1 2008 Jan 1 1.8 -999.0 2.1 -999.0 1.8 1.8 7.3 4.8 -999.0 2.4 1.7 2.0 1.5 6.1 1.3 1.2 1.9

2 2008 Jan 2 2.4 -999.0 8.2 -999.0 5.1 5.0 12.0 7.5 -999.0 2.8 2.5 4.0 4.0 6.0 5.7 1.8 5.1

3 2008 Jan 3 1.2 -999.0 0.4 -999.0 1.8 0.2 4.9 1.8 -999.0 0.1 0.3 2.1 3.0 1.2 6.2 0.5 0.2

4 2008 Jan 4 3.3 -999.0 3.7 -999.0 3.6 3.0 9.9 6.2 -999.0 2.7 0.5 3.6 4.3 2.0 6.7 0.9 3.1

5 2008 Jan 5 5.5 -999.0 8.3 -999.0 10.4 7.8 18.3 14.7 -999.0 7.8 6.9 7.4 9.6 -777.0 5.1 4.2 7.8

6 2008 Jan 6 5.9 -999.0 7.9 -999.0 5.6 5.9 13.4 9.5 -999.0 5.9 7.4 8.1 4.0 -777.0 6.6 3.7 5.9

7 2008 Jan 7 5.6 -999.0 6.2 -999.0 5.1 5.7 10.5 9.4 -999.0 5.9 3.5 4.5 6.4 5.4 2.8 6.5 5.8

8 2008 Jan 8 15.2 -999.0 20.1 -999.0 18.3 18.9 29.2 28.9 -999.0 15.4 3.5 10.6 8.1 7.9 9.2 4.6 19.0

9 2008 Jan 9 17.8 -999.0 23.7 -999.0 15.3 19.7 33.0 31.8 -999.0 18.6 7.5 16.4 15.3 12.0 7.6 8.5 19.8

10 2008 Jan 10 16.2 -999.0 18.7 -999.0 11.3 16.4 29.7 27.7 -999.0 18.5 4.6 13.4 27.5 11.0 4.0 12.6 16.4

11 2008 Jan 11 15.9 -999.0 21.2 -999.0 13.0 16.3 30.2 27.1 -999.0 18.6 12.0 22.1 14.6 22.6 -777.0 10.6 16.3

12 2008 Jan 12 4.6 -999.0 6.2 -999.0 6.1 6.4 16.2 11.7 -999.0 5.6 8.0 13.2 9.3 9.5 -777.0 7.7 6.4

13 2008 Jan 13 3.5 -999.0 3.8 -999.0 4.9 4.0 11.8 8.7 -999.0 4.0 4.0 11.9 6.8 7.5 10.0 2.3 4.0

14 2008 Jan 14 9.4 -999.0 19.9 -999.0 11.5 12.5 23.6 20.6 -999.0 7.6 1.9 8.7 9.1 4.4 8.0 5.8 12.5

15 2008 Jan 15 1.9 -999.0 1.8 -999.0 1.2 1.9 3.4 2.8 -999.0 0.7 1.8 1.1 10.0 2.1 -777.0 1.1 2.0

16 2008 Jan 16 0.8 -999.0 2.6 -999.0 2.5 1.7 4.2 3.0 -999.0 0.5 1.9 2.2 1.7 2.1 1.8 0.8 1.8

17 2008 Jan 17 3.5 -999.0 6.1 -999.0 4.7 6.8 8.7 9.8 -999.0 4.5 2.5 3.4 2.5 1.8 1.6 2.2 6.8

18 2008 Jan 18 6.3 -999.0 7.9 -999.0 5.6 6.9 12.7 13.3 -999.0 7.1 3.1 7.6 3.9 4.0 9.5 1.7 6.9

19 2008 Jan 19 7.8 -999.0 8.4 -999.0 10.4 9.7 17.9 17.1 -999.0 8.7 4.2 6.4 6.8 7.2 4.7 5.8 9.7

20 2008 Jan 20 3.6 -999.0 5.9 -999.0 1.5 5.2 9.6 8.4 -999.0 5.7 7.2 5.9 3.3 5.2 4.8 3.3 5.3

21 2008 Jan 21 2.2 -999.0 4.5 -999.0 -777.0 3.5 8.0 5.7 -999.0 2.0 3.1 3.0 1.3 3.4 3.8 1.2 3.6

22 2008 Jan 22 1.3 -999.0 1.6 -999.0 1.7 1.7 5.4 2.1 -999.0 1.3 1.8 1.4 4.2 1.6 1.8 1.6 1.7

23 2008 Jan 23 3.9 -999.0 8.5 -999.0 7.7 7.3 15.9 14.5 -999.0 5.8 3.8 7.7 2.0 6.6 5.6 1.4 7.3

24 2008 Jan 24 4.7 -999.0 4.6 -999.0 4.8 3.9 9.9 7.7 -999.0 4.5 2.1 3.6 4.5 2.8 4.9 4.6 3.8

A-17

JD Y M D EDMC TEOM4

EDMC TEOM3

EDME TEOM4

EDME TEOM3

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMC TEOM3

EDMS TEOM4

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

25 2008 Jan 25 11.2 -999.0 12.2 -999.0 10.7 11.5 20.8 19.6 -999.0 12.1 9.7 10.5 11.6 10.8 13.5 11.8 11.5

26 2008 Jan 26 11.9 -999.0 11.6 -999.0 12.6 10.3 23.0 20.2 -999.0 11.4 7.8 11.8 8.0 15.2 19.9 7.1 10.4

27 2008 Jan 27 4.6 -999.0 5.6 -999.0 -777.0 5.5 11.7 11.1 -999.0 5.9 9.3 5.0 7.5 9.0 9.0 3.9 5.4

28 2008 Jan 28 2.8 -999.0 2.8 -999.0 -999.0 4.1 -777.0 4.0 -999.0 4.0 7.0 2.5 2.1 7.0 5.6 2.5 4.2

29 2008 Jan 29 5.5 -999.0 4.9 -999.0 -999.0 4.9 -999.0 7.8 -999.0 5.5 6.5 6.6 3.6 2.4 4.1 3.7 5.0

30 2008 Jan 30 4.1 -999.0 3.3 -999.0 -999.0 3.2 -999.0 4.4 -999.0 3.3 3.8 2.2 3.2 1.7 2.4 4.0 3.3

31 2008 Jan 31 7.5 -999.0 7.5 -999.0 -999.0 8.3 -999.0 11.5 -999.0 7.3 3.3 4.8 5.5 1.8 3.7 5.8 8.3

32 2008 Feb 1 12.0 -999.0 10.7 -999.0 -999.0 10.2 -999.0 16.4 -999.0 11.5 4.3 11.3 7.6 3.6 7.4 6.5 10.1

33 2008 Feb 2 22.4 -999.0 22.8 -999.0 -777.0 17.6 -999.0 28.6 -999.0 20.9 8.6 18.2 11.9 15.2 18.8 9.6 17.7

34 2008 Feb 3 23.1 -999.0 19.8 -999.0 -777.0 28.3 -999.0 42.8 -999.0 26.2 8.9 14.0 22.3 13.7 6.8 17.8 28.4

35 2008 Feb 4 19.4 -999.0 20.9 -999.0 -777.0 20.6 -999.0 35.1 -999.0 18.6 8.1 24.2 17.6 20.5 14.8 17.0 20.6

36 2008 Feb 5 3.2 -999.0 3.0 -999.0 3.9 3.0 -999.0 11.3 -999.0 3.8 2.5 9.4 1.5 6.3 8.1 9.3 3.1

37 2008 Feb 6 1.9 -999.0 3.7 -999.0 2.4 1.6 -999.0 7.9 -999.0 0.6 4.0 4.0 2.6 5.1 3.5 3.0 1.6

38 2008 Feb 7 4.8 -999.0 5.5 -999.0 -777.0 6.5 -999.0 11.5 -999.0 5.2 1.8 3.8 2.7 1.5 3.6 3.7 6.5

39 2008 Feb 8 3.5 -999.0 2.5 -999.0 -999.0 3.0 -999.0 5.3 -999.0 2.5 3.3 2.6 3.6 1.5 1.9 3.3 3.0

40 2008 Feb 9 3.1 -999.0 2.6 -999.0 -999.0 3.0 -999.0 5.6 -999.0 2.7 5.3 2.5 2.9 1.7 1.8 2.9 3.0

41 2008 Feb 10 2.0 -999.0 3.0 -999.0 -999.0 2.1 -999.0 2.7 -999.0 2.0 2.1 2.3 2.1 1.0 1.0 4.0 2.2

42 2008 Feb 11 5.0 -999.0 9.7 -999.0 -777.0 7.8 -999.0 -777.0 -999.0 5.7 4.9 6.1 3.0 4.5 6.5 4.1 7.7

43 2008 Feb 12 4.6 -999.0 2.9 -999.0 4.3 4.9 -999.0 8.4 -999.0 3.3 3.2 2.3 3.6 3.5 4.9 4.4 5.0

44 2008 Feb 13 6.1 -999.0 4.6 -999.0 2.5 7.2 -999.0 9.8 -999.0 5.1 4.3 4.7 6.7 1.8 6.5 7.3 7.1

45 2008 Feb 14 10.8 -999.0 10.9 -999.0 14.9 14.1 -999.0 21.0 -999.0 11.4 3.8 11.9 1.3 4.1 5.8 3.1 14.1

46 2008 Feb 15 1.5 -999.0 4.0 -999.0 4.6 2.8 -999.0 7.9 -999.0 0.9 1.5 4.7 0.7 2.1 7.3 1.7 2.8

47 2008 Feb 16 0.9 -999.0 1.5 -999.0 1.0 1.3 -999.0 1.4 -999.0 0.4 1.5 0.3 1.7 1.9 7.7 1.8 1.5

48 2008 Feb 17 2.2 -999.0 2.7 -999.0 2.0 2.5 -999.0 2.2 -999.0 1.5 2.0 1.5 0.9 2.6 3.5 1.6 2.4

49 2008 Feb 18 6.0 -999.0 8.8 -999.0 6.2 7.0 -999.0 10.0 -999.0 5.1 3.2 7.1 1.3 6.3 11.7 2.5 7.0

50 2008 Feb 19 5.3 -999.0 6.2 -999.0 8.9 4.4 -999.0 12.0 -999.0 4.7 3.0 3.7 4.5 7.7 6.5 5.9 4.4

51 2008 Feb 20 14.1 -999.0 16.8 -999.0 21.4 14.3 -999.0 28.7 -999.0 9.9 6.4 15.6 7.1 15.2 23.5 9.0 14.2

A-18

JD Y M D EDMC TEOM4

EDMC TEOM3

EDME TEOM4

EDME TEOM3

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMC TEOM3

EDMS TEOM4

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

52 2008 Feb 21 14.1 -999.0 17.0 -999.0 19.4 15.3 -999.0 32.1 -999.0 13.6 10.0 22.1 7.2 19.1 29.6 7.7 15.3

53 2008 Feb 22 25.7 -999.0 37.7 -999.0 -777.0 27.4 -777.0 53.1 -999.0 22.5 20.1 30.5 10.5 47.1 49.3 10.4 27.3

54 2008 Feb 23 12.6 -999.0 18.7 -999.0 -999.0 17.2 22.7 36.8 -999.0 14.9 15.0 13.6 9.0 36.7 23.9 9.0 17.2

55 2008 Feb 24 2.9 -999.0 2.8 -999.0 -999.0 3.5 7.9 8.9 -999.0 4.7 0.2 0.7 5.8 2.0 -777.0 5.2 3.6

56 2008 Feb 25 8.1 -999.0 8.1 -999.0 -777.0 9.2 15.2 22.1 -999.0 8.4 2.3 10.1 6.1 12.9 -777.0 6.0 9.3

57 2008 Feb 26 4.7 -999.0 7.8 -999.0 -777.0 8.2 13.3 19.2 -999.0 5.1 3.9 7.3 3.1 16.5 10.0 3.9 8.3

58 2008 Feb 27 5.6 -999.0 7.7 -999.0 -999.0 6.4 11.6 15.5 -999.0 4.7 3.2 8.1 2.7 11.8 11.9 3.3 6.4

59 2008 Feb 28 2.2 -999.0 1.2 -999.0 -999.0 1.9 4.5 4.7 -999.0 1.6 1.1 1.5 1.2 2.0 1.9 1.4 2.0

60 2008 Feb 29 5.9 -999.0 3.4 -999.0 -999.0 2.8 8.8 8.7 -999.0 4.5 0.6 2.6 1.6 5.5 5.2 2.8 2.9

61 2008 Mar 1 2.5 -999.0 2.8 -999.0 -999.0 2.8 6.6 7.0 -999.0 2.3 3.5 4.1 2.9 8.3 5.1 3.5 3.0

62 2008 Mar 2 3.8 -999.0 3.3 -999.0 -999.0 3.9 3.8 6.4 -999.0 3.7 3.0 3.4 2.7 2.7 7.1 3.5 4.0

63 2008 Mar 3 8.2 -999.0 13.9 -999.0 -777.0 9.7 10.6 16.9 -999.0 7.5 4.3 10.3 3.0 10.6 13.4 4.5 9.7

64 2008 Mar 4 3.9 -999.0 4.8 -999.0 3.9 5.3 6.3 9.1 -999.0 4.0 4.6 5.1 2.8 5.8 7.1 3.0 5.4

65 2008 Mar 5 -777.0 -999.0 4.4 -999.0 4.9 3.3 6.1 6.5 -999.0 -777.0 4.0 3.0 1.5 7.0 9.0 2.3 3.4

66 2008 Mar 6 6.2 -999.0 5.4 -999.0 7.3 4.5 8.8 10.3 -999.0 4.2 2.5 3.5 2.3 5.0 5.7 3.7 4.5

67 2008 Mar 7 5.6 -999.0 4.0 -999.0 6.0 4.3 7.2 9.2 -999.0 -777.0 2.5 3.2 1.7 5.2 2.7 2.6 4.3

68 2008 Mar 8 7.2 -999.0 8.6 -999.0 9.0 7.5 10.5 14.6 -999.0 5.9 3.9 7.3 2.8 9.6 10.3 3.3 7.5

69 2008 Mar 9 4.8 -999.0 5.8 -999.0 5.6 4.7 7.1 7.9 -999.0 4.3 2.1 5.8 2.3 4.7 5.1 3.3 4.7

70 2008 Mar 10 7.3 -999.0 8.6 -999.0 7.7 5.9 9.5 11.1 -999.0 7.8 2.6 7.5 2.3 8.6 8.9 2.8 6.0

71 2008 Mar 11 5.2 -999.0 7.1 -999.0 7.0 -777.0 -777.0 9.3 -999.0 3.4 2.3 2.8 1.3 9.0 8.4 2.6 -777.0

72 2008 Mar 12 2.5 -999.0 1.5 -999.0 3.3 -999.0 4.4 3.4 -999.0 0.9 0.5 0.5 0.5 -777.0 0.4 1.7 -999.0

73 2008 Mar 13 4.5 -999.0 3.2 -999.0 5.0 -999.0 5.7 5.6 -999.0 3.2 1.7 1.5 1.8 -777.0 2.9 2.3 -999.0

74 2008 Mar 14 4.1 -999.0 3.5 -999.0 6.0 -999.0 6.5 6.2 -999.0 3.6 2.1 3.6 2.3 -777.0 17.7 5.3 -999.0

75 2008 Mar 15 4.9 -999.0 5.1 -999.0 7.3 -999.0 7.3 9.7 -999.0 4.7 2.0 2.2 5.2 6.8 6.0 4.3 -999.0

76 2008 Mar 16 3.3 -999.0 2.5 -999.0 1.7 -999.0 3.5 4.4 -999.0 2.3 2.5 2.1 2.1 6.3 3.0 2.7 -999.0

77 2008 Mar 17 4.7 -999.0 6.8 -999.0 6.5 -777.0 7.9 9.9 -999.0 4.0 2.2 3.4 2.5 6.8 3.3 3.3 -777.0

78 2008 Mar 18 4.8 -999.0 7.5 -999.0 4.1 -777.0 6.0 7.4 -999.0 3.8 1.7 4.0 1.9 5.2 4.8 2.3 -777.0

A-19

JD Y M D EDMC TEOM4

EDMC TEOM3

EDME TEOM4

EDME TEOM3

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMC TEOM3

EDMS TEOM4

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

79 2008 Mar 19 -777.0 -999.0 6.8 -999.0 5.9 -777.0 7.3 8.2 -999.0 4.7 2.4 4.4 2.9 7.3 5.4 3.8 -777.0

80 2008 Mar 20 -999.0 -999.0 6.9 -999.0 6.5 6.3 8.0 10.3 -999.0 7.9 3.3 5.0 4.2 5.8 15.0 5.0 6.4

81 2008 Mar 21 -999.0 -999.0 8.5 -999.0 9.9 8.2 9.4 12.4 -999.0 8.5 6.2 7.1 6.5 12.7 12.0 11.2 8.4

82 2008 Mar 22 -999.0 -999.0 23.3 -999.0 17.8 13.4 14.7 22.1 -999.0 13.8 11.8 15.5 12.5 25.6 18.7 12.8 13.5

83 2008 Mar 23 -999.0 -999.0 2.6 -999.0 8.7 3.0 7.5 8.4 -999.0 3.6 1.4 2.7 3.5 7.2 2.8 6.3 3.1

84 2008 Mar 24 -999.0 -999.0 4.3 -999.0 4.2 5.2 6.8 7.4 -999.0 3.8 2.0 2.0 2.4 4.0 1.3 3.1 5.3

85 2008 Mar 25 -777.0 -999.0 3.6 -999.0 3.3 3.8 4.9 4.5 -999.0 3.0 1.7 1.1 1.0 2.7 0.8 1.4 3.9

86 2008 Mar 26 6.5 -999.0 6.5 -999.0 6.1 6.9 9.3 12.0 -999.0 6.3 2.3 4.6 2.0 5.7 9.8 3.3 7.0

87 2008 Mar 27 7.4 -999.0 7.5 -999.0 8.0 6.3 8.4 12.3 -999.0 7.3 5.0 4.8 4.8 9.5 18.9 5.7 6.3

88 2008 Mar 28 5.6 -999.0 6.3 -999.0 9.2 5.2 9.2 10.6 -999.0 5.1 3.6 3.8 1.9 10.3 8.0 1.8 5.3

89 2008 Mar 29 5.0 -999.0 5.8 -999.0 3.0 4.9 5.8 7.1 -999.0 4.9 4.2 4.5 3.8 4.1 5.5 4.4 5.0

90 2008 Mar 30 5.1 -999.0 6.3 -999.0 3.7 4.5 5.6 -999.0 -999.0 5.0 2.9 3.9 3.3 2.4 6.9 3.4 4.6

91 2008 Mar 31 8.1 -999.0 8.4 -999.0 7.6 7.6 9.2 -777.0 -999.0 7.4 4.7 5.7 6.4 6.2 11.3 6.9 7.6

92 2008 Apr 1 4.6 -999.0 5.9 -999.0 6.1 4.9 6.7 7.9 -999.0 3.6 4.6 4.8 2.1 6.4 6.6 3.5 5.0

93 2008 Apr 2 4.2 -999.0 3.7 -999.0 4.6 4.1 6.3 6.1 -999.0 4.1 2.4 2.3 1.0 2.6 2.1 1.7 4.3

94 2008 Apr 3 2.5 -999.0 4.1 -999.0 4.0 4.3 6.1 5.8 -999.0 3.2 2.5 1.9 2.0 2.7 1.9 3.0 4.2

95 2008 Apr 4 4.2 -999.0 4.5 -999.0 6.1 4.4 7.4 7.9 -999.0 3.2 2.9 2.2 2.5 3.3 3.9 2.8 4.5

96 2008 Apr 5 4.9 -999.0 4.1 -999.0 5.3 3.4 5.8 5.5 -999.0 3.8 2.7 2.5 3.6 2.9 5.7 5.3 3.5

97 2008 Apr 6 5.9 -999.0 4.3 -999.0 6.3 4.1 6.4 8.4 -999.0 4.7 4.4 3.5 4.6 6.1 9.7 6.6 4.1

98 2008 Apr 7 9.9 -999.0 9.1 -999.0 14.7 8.6 10.0 14.4 -999.0 9.1 7.2 7.4 7.1 10.1 10.5 5.8 8.7

99 2008 Apr 8 11.2 -999.0 7.9 -999.0 10.3 5.6 8.8 12.3 -999.0 7.6 3.7 5.4 4.0 7.8 8.7 5.3 5.7

100 2008 Apr 9 10.8 -999.0 10.4 -999.0 14.1 8.4 10.5 -777.0 -999.0 8.8 8.6 7.6 8.5 18.5 16.3 7.7 8.4

101 2008 Apr 10 8.8 -999.0 8.3 -999.0 13.1 9.0 12.1 -777.0 -999.0 7.6 9.4 5.4 4.4 11.6 5.7 3.6 9.0

102 2008 Apr 11 6.1 -999.0 8.2 -999.0 6.3 6.1 7.6 8.2 -999.0 5.4 3.8 3.6 3.5 6.5 4.4 4.0 6.3

103 2008 Apr 12 -777.0 -999.0 5.2 -999.0 6.0 3.9 6.7 6.3 -999.0 3.6 3.2 3.5 2.4 4.9 5.4 3.3 4.1

104 2008 Apr 13 -777.0 -999.0 5.2 -999.0 5.2 2.3 5.6 4.8 -999.0 2.2 2.8 2.4 1.8 4.3 4.3 2.6 2.3

105 2008 Apr 14 6.2 -999.0 6.4 -999.0 6.6 4.7 6.1 7.0 -999.0 4.1 5.3 4.4 3.1 5.3 5.2 3.7 4.8

106 2008 Apr 15 3.9 -999.0 4.1 -999.0 6.3 4.7 6.4 6.1 -999.0 3.0 3.5 2.7 2.6 3.5 3.5 2.9 4.7

A-20

JD Y M D EDMC TEOM4

EDMC TEOM3

EDME TEOM4

EDME TEOM3

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMC TEOM3

EDMS TEOM4

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

107 2008 Apr 16 3.5 -999.0 4.4 -999.0 5.1 3.2 6.1 4.4 -999.0 3.3 2.6 2.4 1.6 2.6 2.7 2.4 3.3

108 2008 Apr 17 5.8 -999.0 6.9 -999.0 5.8 4.9 7.1 7.5 -999.0 5.4 4.2 4.3 3.1 3.9 5.4 4.7 5.0

109 2008 Apr 18 2.9 -999.0 3.9 -999.0 2.3 2.8 4.5 3.5 -999.0 3.9 3.0 2.6 2.0 2.3 4.0 2.9 2.9

110 2008 Apr 19 2.8 -999.0 3.4 -999.0 1.4 3.0 3.6 2.4 -999.0 3.4 3.9 3.0 1.1 2.3 3.7 2.2 3.1

111 2008 Apr 20 2.4 -999.0 2.2 -999.0 1.1 1.6 3.4 -777.0 -999.0 2.6 2.5 1.9 1.0 1.5 2.4 1.7 1.6

112 2008 Apr 21 1.8 -999.0 1.7 -999.0 0.9 1.5 2.6 2.4 -999.0 1.7 1.9 1.2 0.8 1.4 1.4 1.5 1.5

113 2008 Apr 22 4.4 -999.0 3.9 -999.0 2.4 4.4 3.8 4.6 -999.0 4.1 3.6 3.4 3.2 2.1 3.7 3.8 4.4

114 2008 Apr 23 7.7 -999.0 7.6 -999.0 7.4 7.0 7.8 10.0 -999.0 7.6 5.8 6.8 4.8 7.1 11.9 5.9 7.1

115 2008 Apr 24 6.0 -999.0 5.6 -999.0 6.9 4.8 7.1 7.5 -999.0 5.7 5.8 5.5 4.3 9.3 13.2 5.7 4.8

116 2008 Apr 25 7.6 -999.0 7.4 -999.0 10.6 7.5 11.4 10.7 -999.0 7.6 5.3 5.6 4.0 8.1 -777.0 4.6 7.5

117 2008 Apr 26 7.4 -999.0 7.2 -999.0 8.9 7.6 16.6 11.0 -999.0 7.8 6.0 6.5 5.4 8.9 -999.0 5.5 7.5

118 2008 Apr 27 7.1 -999.0 6.2 -999.0 7.6 5.2 17.3 9.4 -999.0 6.5 6.1 6.1 6.3 8.6 -777.0 6.3 5.3

119 2008 Apr 28 6.5 -999.0 5.3 -999.0 6.1 4.6 13.4 8.1 -999.0 4.6 6.0 5.2 4.6 9.3 8.4 4.1 4.8

120 2008 Apr 29 8.5 -999.0 8.1 -999.0 7.8 5.7 14.9 9.9 -999.0 6.7 6.5 6.9 6.0 8.9 10.7 6.1 5.8

121 2008 Apr 30 4.2 -999.0 4.7 -999.0 -777.0 3.1 8.9 5.0 -999.0 3.1 5.0 3.0 3.1 8.6 9.6 -777.0 3.2

122 2008 May 1 -777.0 -999.0 5.4 -999.0 7.7 5.0 11.0 6.3 -999.0 4.3 5.2 4.9 3.8 -999.0 9.5 -777.0 5.1

123 2008 May 2 8.8 -999.0 7.8 -999.0 8.6 6.5 16.3 12.6 -999.0 8.8 3.6 5.3 6.0 -999.0 6.6 3.8 6.5

124 2008 May 3 13.7 -999.0 12.5 -999.0 15.5 15.1 29.8 24.3 -999.0 15.5 6.2 9.1 6.3 -999.0 9.6 5.4 15.1

125 2008 May 4 6.9 -999.0 7.4 -999.0 8.4 6.4 13.9 9.5 -999.0 6.4 5.0 4.5 5.3 -999.0 4.3 4.5 6.5

126 2008 May 5 14.5 -999.0 14.0 -999.0 9.6 13.7 23.3 16.5 -999.0 13.2 10.9 12.0 13.8 -777.0 12.7 14.0 13.8

127 2008 May 6 11.0 -999.0 10.3 -999.0 8.9 9.4 17.7 13.6 -999.0 11.4 5.8 7.8 12.0 -999.0 9.4 12.5 9.5

128 2008 May 7 7.6 -999.0 6.5 -999.0 6.6 6.1 13.3 8.9 -999.0 7.2 3.7 4.3 8.2 -777.0 15.8 7.7 6.2

129 2008 May 8 4.5 -999.0 4.7 -999.0 4.5 2.6 7.4 4.3 -999.0 3.5 3.6 3.4 -777.0 4.1 16.9 3.1 2.7

130 2008 May 9 7.2 -999.0 6.5 -999.0 5.5 5.7 10.9 8.3 -999.0 7.4 4.9 5.9 6.7 6.0 9.6 6.4 5.7

131 2008 May 10 6.2 -999.0 6.3 -999.0 6.0 5.2 10.9 7.8 -999.0 6.9 5.7 6.5 5.0 7.6 9.0 5.1 5.3

132 2008 May 11 -999.0 -999.0 8.1 -999.0 6.0 6.3 12.1 10.0 -999.0 6.3 6.7 8.2 5.2 8.8 9.0 4.8 6.4

133 2008 May 12 -777.0 -999.0 6.6 -999.0 6.7 6.2 13.0 10.6 -999.0 10.5 5.5 5.3 5.8 8.8 8.3 4.0 6.3

134 2008 May 13 5.8 -999.0 7.4 -999.0 6.9 5.7 12.0 8.5 -999.0 6.4 5.1 5.6 4.4 7.3 8.2 3.4 5.7

A-21

JD Y M D EDMC TEOM4

EDMC TEOM3

EDME TEOM4

EDME TEOM3

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMC TEOM3

EDMS TEOM4

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

135 2008 May 14 -777.0 -999.0 4.6 -999.0 6.5 4.3 9.1 7.3 -999.0 4.7 3.9 3.2 3.6 5.1 5.7 2.2 4.3

136 2008 May 15 -999.0 -999.0 -777.0 -999.0 5.9 6.1 12.0 20.1 -999.0 3.8 2.7 2.0 3.5 8.3 12.1 2.6 6.1

137 2008 May 16 -999.0 -999.0 4.2 -999.0 5.0 7.3 11.3 9.3 -999.0 5.7 5.2 3.9 6.9 7.5 5.3 6.8 7.3

138 2008 May 17 -999.0 -999.0 -777.0 -999.0 8.0 -777.0 18.6 14.9 -999.0 10.6 10.9 8.9 11.6 12.4 14.3 11.3 -777.0

139 2008 May 18 -999.0 -999.0 -999.0 -999.0 7.3 5.7 14.5 9.9 -999.0 7.8 7.9 7.4 7.2 9.9 12.0 5.4 5.9

140 2008 May 19 -999.0 -999.0 -999.0 -999.0 5.6 4.0 9.3 6.4 -999.0 4.2 6.1 4.3 5.3 8.0 8.0 4.4 4.1

141 2008 May 20 -999.0 -999.0 -999.0 -999.0 5.4 4.5 11.0 7.8 -999.0 4.7 5.4 4.8 7.1 8.8 11.2 7.2 4.5

142 2008 May 21 -777.0 -999.0 -777.0 -999.0 8.4 5.9 11.9 9.2 -999.0 6.1 5.7 5.7 2.8 6.6 10.5 2.0 6.1

143 2008 May 22 6.7 -999.0 4.4 -999.0 6.8 4.7 11.1 8.8 -999.0 5.0 5.8 4.9 4.8 10.3 8.7 4.0 4.9

144 2008 May 23 7.8 -999.0 5.8 -999.0 7.6 6.9 15.3 11.2 -999.0 6.3 3.8 4.7 5.7 9.2 10.3 5.9 7.0

145 2008 May 24 -999.0 -999.0 6.5 -999.0 6.5 5.7 10.8 7.9 -999.0 4.6 3.9 5.1 5.5 7.8 10.5 5.0 5.8

146 2008 May 25 -999.0 -999.0 3.6 -999.0 5.4 3.6 7.5 5.9 -999.0 3.0 2.0 2.5 2.8 5.8 4.2 2.6 3.8

147 2008 May 26 3.0 -999.0 3.5 -999.0 5.8 3.5 8.2 4.4 -999.0 2.7 3.3 3.0 3.2 7.5 5.9 3.8 3.5

148 2008 May 27 3.0 -999.0 2.9 -999.0 4.8 3.4 9.2 4.4 -999.0 2.3 3.1 2.3 4.0 8.0 3.8 4.0 3.4

149 2008 May 28 3.5 -999.0 3.7 -999.0 4.8 3.3 8.4 3.6 -999.0 2.5 3.1 2.6 2.6 6.3 4.1 2.9 3.4

150 2008 May 29 6.0 -999.0 6.2 -999.0 7.0 5.6 10.8 7.1 -999.0 6.2 6.8 4.2 4.8 6.7 6.2 3.3 5.5

151 2008 May 30 4.2 -999.0 5.0 -999.0 7.0 4.3 10.1 6.0 -999.0 4.5 3.7 2.7 3.8 6.0 3.6 3.3 4.4

152 2008 May 31 5.3 -999.0 6.3 -999.0 5.8 5.0 10.2 6.4 -999.0 4.5 4.4 3.8 5.0 8.1 6.3 4.5 5.0

153 2008 Jun 1 5.0 -999.0 4.8 -999.0 5.1 4.0 9.0 6.1 -999.0 4.2 3.2 3.8 4.1 5.6 5.5 4.2 4.0

154 2008 Jun 2 4.3 -999.0 5.3 -999.0 5.9 3.5 9.7 5.9 -999.0 3.0 3.5 3.7 4.2 7.4 5.7 4.5 3.6

155 2008 Jun 3 6.8 -999.0 9.7 -999.0 6.5 6.5 13.0 8.3 -999.0 6.1 5.5 5.5 5.4 8.9 8.4 5.0 6.5

156 2008 Jun 4 12.9 -999.0 13.0 -999.0 9.3 10.9 -777.0 15.4 -999.0 10.3 7.5 7.8 9.1 10.7 11.3 8.5 11.1

157 2008 Jun 5 7.6 -999.0 6.7 -999.0 5.8 4.4 9.4 9.5 -999.0 5.3 5.2 4.0 6.6 9.6 7.3 4.8 4.6

158 2008 Jun 6 10.5 -999.0 10.6 -999.0 9.9 8.5 13.9 12.8 -999.0 11.8 4.7 5.5 5.0 9.0 11.0 2.3 8.6

159 2008 Jun 7 6.8 -999.0 8.2 -999.0 5.9 7.0 10.9 10.2 -999.0 8.1 9.6 6.4 2.9 13.4 9.1 1.8 7.0

160 2008 Jun 8 3.5 -999.0 3.8 -999.0 4.4 3.0 7.4 6.6 -999.0 4.1 4.9 2.6 1.1 9.5 4.4 1.2 3.1

161 2008 Jun 9 3.5 -999.0 3.4 -999.0 4.3 2.4 6.6 5.3 -999.0 3.9 2.2 1.9 2.7 5.5 4.7 1.9 2.5

162 2008 Jun 10 5.2 -999.0 4.8 -999.0 5.9 4.0 7.6 6.1 -999.0 4.9 2.9 2.2 2.6 6.4 15.9 3.3 4.0

A-22

JD Y M D EDMC TEOM4

EDMC TEOM3

EDME TEOM4

EDME TEOM3

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMC TEOM3

EDMS TEOM4

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

163 2008 Jun 11 5.4 -999.0 7.9 -999.0 6.1 5.8 9.9 8.1 -999.0 6.6 2.8 4.6 3.5 7.2 7.9 2.9 5.8

164 2008 Jun 12 5.8 -999.0 4.9 -999.0 6.4 5.8 9.5 8.9 -999.0 5.6 3.5 4.4 3.1 8.1 5.5 1.9 5.9

165 2008 Jun 13 5.2 -999.0 5.6 -999.0 5.8 5.2 8.2 7.7 -999.0 5.7 3.6 4.4 3.2 7.7 4.9 1.8 5.3

166 2008 Jun 14 3.3 -999.0 3.9 -999.0 4.6 3.1 5.2 4.5 -999.0 3.5 2.1 3.4 2.1 6.0 3.5 1.0 3.2

167 2008 Jun 15 2.8 -999.0 4.1 -999.0 5.6 3.3 5.6 4.9 -999.0 3.2 2.8 2.5 2.1 5.2 3.9 0.7 3.4

168 2008 Jun 16 6.5 -999.0 7.4 -999.0 6.7 6.2 11.2 9.5 -999.0 6.6 4.7 5.6 3.5 7.0 6.1 2.1 6.3

169 2008 Jun 17 6.7 -999.0 4.9 -999.0 6.5 5.5 9.7 9.9 -999.0 6.0 3.8 5.0 5.2 7.4 9.8 4.3 5.6

170 2008 Jun 18 4.6 -999.0 5.9 -999.0 6.9 4.7 7.6 8.6 -999.0 4.5 6.0 4.0 3.5 10.8 4.7 1.9 4.8

171 2008 Jun 19 3.2 -999.0 3.8 -999.0 4.7 4.2 7.2 5.5 -999.0 4.6 2.7 2.5 1.9 7.5 3.5 1.3 4.2

172 2008 Jun 20 5.0 -999.0 -777.0 -999.0 5.1 4.4 8.0 4.9 -999.0 5.7 3.2 5.5 2.4 5.6 7.0 2.8 4.4

173 2008 Jun 21 4.7 -999.0 -777.0 -999.0 4.7 4.4 9.1 6.3 -999.0 4.5 -777.0 3.8 3.9 5.1 8.8 4.2 4.4

174 2008 Jun 22 3.8 -999.0 -777.0 -999.0 5.2 3.2 6.9 5.5 -999.0 4.3 2.9 4.1 2.5 6.6 4.9 2.3 3.3

175 2008 Jun 23 4.2 -999.0 -777.0 -999.0 4.5 4.4 7.9 6.7 -999.0 4.4 3.4 4.8 3.1 8.2 4.9 2.4 4.5

176 2008 Jun 24 5.7 -999.0 8.6 -999.0 6.7 6.0 10.4 11.7 -999.0 5.8 3.8 3.9 3.0 7.8 7.3 2.2 6.0

177 2008 Jun 25 4.4 -999.0 -777.0 -999.0 4.6 2.2 5.7 6.5 -999.0 2.8 1.4 1.8 1.3 6.0 2.6 1.2 2.3

178 2008 Jun 26 2.9 -999.0 -777.0 -999.0 7.6 2.7 6.6 11.6 -999.0 3.3 1.4 2.5 1.8 6.8 4.8 1.1 2.7

179 2008 Jun 27 5.0 -999.0 3.3 -999.0 6.4 4.8 7.2 6.6 -999.0 4.8 2.6 2.9 2.6 6.0 4.0 1.6 4.9

180 2008 Jun 28 5.5 -999.0 4.6 -999.0 5.5 5.7 9.5 9.6 -999.0 5.5 3.2 4.6 4.7 8.3 7.1 4.4 5.8

181 2008 Jun 29 10.1 -999.0 8.8 -999.0 7.9 10.8 15.1 15.6 -999.0 10.3 8.2 8.1 9.0 11.9 12.2 8.4 10.9

182 2008 Jun 30 10.0 -999.0 10.2 -999.0 7.8 11.5 -777.0 17.4 -999.0 10.0 6.4 8.4 8.6 13.1 9.9 6.5 11.5

183 2008 Jul 1 -999.0 4.3 -999.0 4.4 4.7 3.8 9.9 7.7 4.3 -999.0 2.2 2.7 3.4 6.7 5.9 3.0 3.9

184 2008 Jul 2 -999.0 7.1 -999.0 7.9 6.5 6.6 12.3 11.0 7.4 -999.0 5.5 6.2 7.0 11.1 10.5 7.3 6.5

185 2008 Jul 3 -999.0 10.3 -999.0 9.5 9.1 10.3 17.7 15.5 11.0 -999.0 7.4 10.1 13.5 13.6 16.4 10.7 10.3

186 2008 Jul 4 -999.0 10.7 -999.0 12.0 9.9 11.2 -777.0 17.9 13.8 -999.0 9.4 11.3 7.3 16.0 15.3 6.4 11.1

187 2008 Jul 5 -999.0 5.3 -999.0 5.9 6.3 4.7 -999.0 11.3 5.7 -999.0 3.6 5.1 3.9 10.1 6.3 3.5 4.6

188 2008 Jul 6 -999.0 2.7 -999.0 -777.0 5.6 2.4 -999.0 8.1 2.8 -999.0 1.9 2.5 1.7 9.7 3.8 2.1 2.5

189 2008 Jul 7 -999.0 3.1 -999.0 -777.0 4.2 3.6 -777.0 10.3 3.9 -999.0 1.3 2.3 2.8 6.7 2.3 2.3 3.7

190 2008 Jul 8 -999.0 4.2 -999.0 5.6 5.6 5.1 9.7 8.5 6.7 -999.0 2.5 3.4 3.2 7.1 3.5 3.0 5.3

A-23

JD Y M D EDMC TEOM4

EDMC TEOM3

EDME TEOM4

EDME TEOM3

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMC TEOM3

EDMS TEOM4

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

191 2008 Jul 9 -999.0 4.5 -999.0 4.7 5.8 2.9 7.8 13.5 3.0 -999.0 2.4 4.5 2.0 6.8 3.6 1.5 3.0

192 2008 Jul 10 -999.0 1.4 -999.0 2.7 4.3 1.5 6.3 10.6 1.5 -999.0 1.3 2.3 1.1 3.8 2.1 0.8 1.5

193 2008 Jul 11 -999.0 2.5 -999.0 2.9 5.5 2.5 5.4 9.8 3.1 -999.0 2.2 1.5 2.0 3.9 2.5 1.6 2.5

194 2008 Jul 12 -999.0 2.9 -999.0 5.3 4.7 3.1 7.3 6.6 3.0 -999.0 6.3 3.3 2.3 5.6 3.0 1.8 3.2

195 2008 Jul 13 -999.0 2.1 -999.0 4.2 4.9 2.1 6.9 6.7 2.8 -999.0 1.8 1.6 1.2 4.2 2.1 1.3 2.2

196 2008 Jul 14 -999.0 2.6 -999.0 3.3 4.7 2.8 6.3 9.8 2.5 -999.0 1.1 2.3 1.8 4.1 4.2 1.0 2.8

197 2008 Jul 15 -999.0 3.6 -999.0 6.3 4.8 3.4 6.5 15.1 3.4 -999.0 2.9 5.8 2.3 6.1 5.8 1.8 3.4

198 2008 Jul 16 -999.0 4.3 -999.0 6.5 7.0 4.4 8.5 17.7 4.7 -999.0 3.8 4.2 2.5 4.9 5.5 1.8 4.4

199 2008 Jul 17 -999.0 5.3 -999.0 5.6 5.1 3.6 9.5 19.5 5.5 -999.0 2.0 4.1 3.4 6.0 4.6 2.4 3.8

200 2008 Jul 18 -999.0 4.2 -999.0 6.5 6.6 4.9 9.7 13.6 5.0 -999.0 4.6 5.2 2.3 7.9 5.8 2.4 5.0

201 2008 Jul 19 -999.0 5.4 -999.0 6.5 6.4 5.1 10.6 11.0 6.2 -999.0 4.0 4.3 3.4 8.9 5.3 2.5 5.0

202 2008 Jul 20 -999.0 5.9 -999.0 6.3 6.2 6.4 11.3 12.1 6.2 -999.0 5.2 6.1 5.8 7.1 7.5 6.3 6.5

203 2008 Jul 21 -999.0 9.2 -999.0 11.0 7.3 9.5 -777.0 15.9 9.9 -999.0 7.2 7.9 7.5 8.2 9.0 5.5 9.5

204 2008 Jul 22 -999.0 2.3 -999.0 2.9 4.9 2.6 -777.0 13.9 3.2 -999.0 1.8 1.9 2.3 5.3 2.1 1.4 2.7

205 2008 Jul 23 -999.0 3.7 -999.0 4.1 5.6 4.6 9.2 15.6 4.7 -999.0 3.1 2.6 4.3 4.5 3.6 3.6 4.6

206 2008 Jul 24 -999.0 6.5 -999.0 7.0 5.8 5.9 9.9 9.5 7.3 -999.0 5.9 4.9 4.9 7.1 4.9 3.9 6.0

207 2008 Jul 25 -999.0 6.6 -999.0 7.3 6.3 7.0 12.4 9.8 8.2 -999.0 5.3 7.1 6.1 8.5 8.5 6.5 7.0

208 2008 Jul 26 -999.0 6.1 -999.0 7.2 6.8 6.8 12.8 9.7 7.3 -999.0 8.1 8.0 7.7 8.2 10.2 5.0 6.9

209 2008 Jul 27 -999.0 6.0 -999.0 7.0 7.8 6.3 12.7 11.6 7.6 -999.0 8.1 6.6 4.6 12.7 7.7 3.5 6.2

210 2008 Jul 28 -999.0 3.9 -999.0 4.1 6.4 4.9 9.3 10.7 4.9 -999.0 3.9 3.5 2.7 6.6 4.1 1.8 5.0

211 2008 Jul 29 -999.0 5.5 -999.0 5.3 6.6 6.3 11.3 10.4 6.4 -999.0 6.2 5.6 3.5 6.7 7.8 4.2 6.2

212 2008 Jul 30 -999.0 2.2 -999.0 2.4 4.7 2.3 7.9 8.4 2.5 -999.0 2.0 1.8 1.7 6.4 2.2 2.5 2.4

213 2008 Jul 31 -999.0 1.4 -999.0 2.5 4.9 2.2 5.3 6.8 2.8 -999.0 1.1 2.0 1.7 2.0 0.8 2.3 2.2

214 2008 Aug 1 -999.0 3.7 -999.0 3.2 6.2 3.5 6.9 8.1 4.4 -999.0 4.8 4.1 3.0 4.5 6.7 4.1 3.6

215 2008 Aug 2 -999.0 1.1 -999.0 1.4 4.8 1.5 5.1 4.0 1.7 -999.0 3.7 1.6 1.0 2.8 1.8 2.3 1.5

216 2008 Aug 3 -999.0 1.9 -999.0 2.3 5.0 2.5 5.7 6.1 3.3 -999.0 2.5 1.9 1.7 2.8 2.0 2.8 2.5

217 2008 Aug 4 -999.0 3.0 -999.0 4.0 5.9 4.2 8.2 6.4 5.0 -999.0 2.9 4.2 2.5 4.9 3.8 3.3 4.3

218 2008 Aug 5 -999.0 4.2 -999.0 4.7 6.6 5.6 10.3 11.4 7.1 -999.0 3.4 4.5 3.9 6.7 5.2 4.5 5.6

A-24

JD Y M D EDMC TEOM4

EDMC TEOM3

EDME TEOM4

EDME TEOM3

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMC TEOM3

EDMS TEOM4

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

219 2008 Aug 6 -999.0 6.0 -999.0 6.0 6.7 6.9 13.1 11.6 7.0 -999.0 3.7 5.4 5.8 -777.0 5.7 7.0 7.0

220 2008 Aug 7 -999.0 9.9 -999.0 7.7 8.2 11.7 16.2 15.6 13.7 -999.0 7.2 11.3 13.7 -777.0 9.5 9.5 11.8

221 2008 Aug 8 -999.0 14.1 -999.0 -777.0 9.6 16.4 22.5 22.5 16.5 -999.0 10.0 15.4 13.8 12.0 18.0 9.7 16.5

222 2008 Aug 9 -999.0 6.0 -999.0 6.3 5.6 7.5 12.1 12.9 6.3 -999.0 4.7 6.0 4.8 6.8 10.1 5.8 7.5

223 2008 Aug 10 -999.0 4.4 -999.0 5.6 6.8 6.6 11.8 10.6 5.7 -999.0 5.2 -999.0 5.0 8.4 5.4 2.5 6.6

224 2008 Aug 11 -999.0 2.4 -999.0 2.9 5.8 3.0 6.0 11.5 4.4 -999.0 2.5 -777.0 0.6 3.1 2.4 1.3 3.0

225 2008 Aug 12 -999.0 1.7 -999.0 1.8 5.5 3.3 6.7 6.2 2.1 -999.0 1.5 1.5 1.6 2.9 1.8 -777.0 3.4

226 2008 Aug 13 -999.0 3.1 -999.0 4.1 5.7 4.9 8.0 11.5 8.5 -999.0 2.3 2.6 2.9 2.8 -777.0 4.7 5.0

227 2008 Aug 14 -999.0 5.0 -999.0 6.8 6.4 6.2 10.3 11.4 8.5 -999.0 7.2 4.4 5.3 4.4 5.0 6.6 6.3

228 2008 Aug 15 -999.0 5.5 -999.0 -777.0 5.6 5.0 10.5 11.7 8.0 -999.0 8.2 5.6 5.1 6.8 8.4 6.5 5.1

229 2008 Aug 16 -999.0 7.1 -999.0 -777.0 6.6 7.1 12.3 12.6 9.4 -999.0 6.9 6.1 8.4 7.9 8.1 8.0 7.2

230 2008 Aug 17 -999.0 4.6 -999.0 -777.0 6.6 5.3 11.1 10.3 6.3 -999.0 7.2 4.5 7.8 6.1 5.7 8.6 5.3

231 2008 Aug 18 -999.0 8.8 -999.0 -777.0 7.1 9.0 -777.0 14.2 9.8 -999.0 7.5 9.0 10.3 7.4 15.2 12.0 9.0

232 2008 Aug 19 -999.0 4.9 -999.0 4.9 6.6 6.0 -999.0 13.6 7.1 -999.0 5.1 4.5 3.8 6.8 6.2 3.3 6.1

233 2008 Aug 20 -999.0 8.2 -999.0 9.1 11.3 12.6 -999.0 17.0 9.5 -999.0 6.6 8.4 4.6 10.4 8.1 4.5 12.8

234 2008 Aug 21 -999.0 5.6 -999.0 5.2 7.2 6.0 -999.0 10.7 6.4 -999.0 2.9 5.2 8.6 5.2 4.6 8.0 6.1

235 2008 Aug 22 -999.0 4.2 -999.0 5.5 6.5 5.8 -777.0 11.7 7.4 -999.0 4.7 4.0 1.5 5.2 4.7 1.0 6.0

236 2008 Aug 23 -999.0 3.0 -999.0 3.5 5.6 3.6 8.2 7.1 4.2 -999.0 2.9 3.6 3.0 2.5 5.2 2.4 3.7

237 2008 Aug 24 -999.0 2.6 -999.0 3.3 6.4 3.7 9.5 7.3 4.0 -999.0 2.3 3.1 3.3 3.6 4.5 3.6 3.8

238 2008 Aug 25 -999.0 2.1 -999.0 -777.0 5.9 3.2 8.3 7.3 4.5 -999.0 2.5 1.9 2.1 2.8 4.0 1.1 3.3

239 2008 Aug 26 -999.0 1.6 -999.0 3.0 5.6 2.4 7.4 5.1 2.2 -999.0 1.2 1.5 0.8 1.9 -777.0 0.6 2.5

240 2008 Aug 27 -999.0 4.2 -999.0 6.0 5.8 3.7 6.9 6.3 5.0 -999.0 2.8 -777.0 1.5 3.7 3.6 1.4 3.7

241 2008 Aug 28 -999.0 1.9 -999.0 4.6 4.5 3.0 5.0 6.7 1.7 -999.0 1.3 3.7 0.9 1.7 2.0 1.0 3.0

242 2008 Aug 29 -999.0 2.3 -999.0 3.8 4.9 2.3 6.0 4.8 1.8 -999.0 3.5 2.9 1.0 2.5 2.8 0.9 2.3

243 2008 Aug 30 -999.0 1.8 -999.0 2.3 3.8 1.1 3.3 3.9 1.4 -999.0 1.2 1.9 1.1 2.3 1.6 0.8 1.2

244 2008 Aug 31 -999.0 2.6 -999.0 5.2 4.1 2.5 5.7 2.7 3.1 -999.0 2.2 3.9 1.6 2.4 2.4 1.3 2.5

245 2008 Sep 1 -999.0 4.4 -999.0 5.3 5.3 4.5 7.0 6.0 4.5 -999.0 4.1 6.1 2.9 5.2 6.5 2.6 4.5

246 2008 Sep 2 -999.0 5.0 -999.0 7.2 7.1 5.5 9.5 8.9 10.3 -999.0 2.9 5.1 2.9 2.3 6.0 3.4 5.5

A-25

JD Y M D EDMC TEOM4

EDMC TEOM3

EDME TEOM4

EDME TEOM3

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMC TEOM3

EDMS TEOM4

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

247 2008 Sep 3 -999.0 3.5 -999.0 3.9 5.5 3.9 7.1 5.7 5.1 -999.0 2.0 3.1 1.5 2.4 3.2 0.8 3.9

248 2008 Sep 4 -999.0 3.9 -999.0 4.7 5.6 3.6 6.9 8.1 4.1 -999.0 1.7 2.8 1.7 2.0 3.4 1.4 3.6

249 2008 Sep 5 -999.0 4.6 -999.0 8.5 7.3 5.8 9.9 8.5 4.4 -999.0 2.8 3.1 2.2 2.6 2.6 1.6 5.8

250 2008 Sep 6 -999.0 4.0 -999.0 5.1 5.8 4.2 7.9 5.4 4.2 -999.0 2.1 4.0 2.0 2.9 3.5 2.4 4.2

251 2008 Sep 7 -999.0 3.6 -999.0 5.0 5.6 3.7 7.3 6.2 3.7 -999.0 3.8 3.6 1.6 2.8 2.8 1.3 3.7

252 2008 Sep 8 -999.0 2.9 -999.0 5.1 5.4 2.4 6.7 5.6 2.4 -999.0 2.8 3.8 1.2 2.4 3.3 1.9 2.4

253 2008 Sep 9 -999.0 3.4 -999.0 4.1 4.4 3.5 6.7 6.5 2.4 -999.0 4.8 3.1 1.8 3.7 2.3 1.2 3.6

254 2008 Sep 10 -999.0 2.1 -999.0 3.3 3.8 1.6 5.5 4.5 1.2 -999.0 1.9 1.8 1.0 1.6 1.5 1.1 1.7

255 2008 Sep 11 -999.0 3.0 -999.0 5.6 5.5 3.3 6.8 5.8 3.1 -999.0 4.1 4.2 1.6 2.2 3.7 2.1 3.4

256 2008 Sep 12 -999.0 2.7 -999.0 -777.0 5.2 2.4 5.3 7.2 1.8 -999.0 2.5 1.7 1.5 1.9 1.9 1.1 2.6

257 2008 Sep 13 -999.0 2.8 -999.0 4.1 5.5 4.7 7.0 5.6 3.0 -999.0 3.2 3.6 1.4 -777.0 3.0 1.2 4.7

258 2008 Sep 14 -999.0 3.7 -999.0 4.7 5.7 4.6 8.2 8.6 5.5 -999.0 4.5 5.2 2.8 -999.0 5.9 2.5 4.7

259 2008 Sep 15 -999.0 5.2 -999.0 5.9 6.6 5.4 9.5 10.3 6.8 -999.0 6.0 5.0 4.3 -999.0 6.9 3.8 5.4

260 2008 Sep 16 -999.0 5.9 -999.0 6.5 7.0 6.3 10.7 13.7 6.0 -999.0 3.6 4.6 4.9 -999.0 11.8 4.5 6.3

261 2008 Sep 17 -999.0 6.9 -999.0 -777.0 5.7 6.4 10.2 19.6 11.9 -999.0 3.2 4.9 5.1 -777.0 7.8 4.9 6.5

262 2008 Sep 18 -999.0 9.9 -999.0 -777.0 8.4 10.0 14.8 25.2 12.5 -999.0 6.2 8.5 6.9 8.2 8.2 5.8 10.0

263 2008 Sep 19 -999.0 5.8 -999.0 7.4 5.9 6.1 11.0 23.6 6.2 -999.0 4.0 5.1 5.8 2.6 7.4 4.7 6.1

264 2008 Sep 20 -999.0 5.1 -999.0 5.1 5.8 4.5 9.2 16.1 6.0 -999.0 6.4 4.1 6.6 3.8 8.2 5.5 4.6

265 2008 Sep 21 -999.0 6.0 -999.0 6.7 6.1 5.5 10.0 18.0 7.6 -999.0 7.7 5.4 5.8 4.0 14.3 6.1 5.6

266 2008 Sep 22 -999.0 1.5 -999.0 1.8 3.4 1.6 3.8 9.2 0.7 -999.0 3.5 1.1 0.8 3.2 2.5 0.8 1.6

267 2008 Sep 23 -999.0 1.9 -999.0 2.6 3.8 2.5 5.3 5.5 1.5 -999.0 2.7 0.8 -777.0 2.4 1.9 0.9 2.6

268 2008 Sep 24 -999.0 2.8 -999.0 4.6 4.3 2.5 5.5 6.5 2.1 -999.0 1.9 1.8 -777.0 1.7 5.5 1.1 2.6

269 2008 Sep 25 -999.0 3.4 -999.0 3.7 3.4 3.6 5.6 13.5 4.0 -999.0 1.8 3.5 3.3 1.6 7.5 3.3 3.7

270 2008 Sep 26 -999.0 2.9 -999.0 2.4 4.5 2.1 5.2 8.4 2.4 -999.0 1.7 2.0 2.5 2.7 8.0 2.8 2.2

271 2008 Sep 27 -999.0 2.6 -999.0 2.7 4.6 2.3 4.6 5.9 2.5 -999.0 2.7 2.9 1.5 3.3 3.1 1.5 2.3

272 2008 Sep 28 -999.0 1.7 -999.0 2.0 4.4 2.6 5.3 5.3 4.0 -999.0 1.1 1.2 1.8 2.3 2.0 1.3 2.7

273 2008 Sep 29 -999.0 5.1 -999.0 7.4 5.9 7.4 10.0 18.2 10.2 -999.0 3.0 4.6 3.5 2.5 5.7 2.9 7.3

274 2008 Sep 30 -999.0 7.1 -999.0 13.8 8.1 6.5 11.9 18.0 7.2 -999.0 3.3 6.3 4.3 3.0 9.3 3.7 6.6

A-26

JD Y M D EDMC TEOM4

EDMC TEOM3

EDME TEOM4

EDME TEOM3

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMC TEOM3

EDMS TEOM4

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

275 2008 Oct 1 -999.0 10.5 -999.0 12.4 9.3 10.5 15.8 23.7 18.0 -999.0 6.9 9.3 10.1 6.0 16.3 7.0 10.6

276 2008 Oct 2 -999.0 11.5 -999.0 13.7 8.9 9.0 15.4 26.8 10.2 -999.0 8.1 8.8 10.8 6.5 12.8 9.5 9.0

277 2008 Oct 3 -999.0 12.1 -999.0 13.0 10.7 11.7 17.0 24.9 14.6 -999.0 10.3 11.4 9.5 7.9 13.3 6.9 11.7

278 2008 Oct 4 -999.0 10.7 -999.0 11.0 10.6 8.3 13.9 20.4 10.3 -999.0 7.7 8.3 7.2 7.2 11.7 5.4 8.5

279 2008 Oct 5 -999.0 5.2 -999.0 5.3 10.8 5.2 11.5 9.3 4.1 -999.0 5.7 5.8 3.8 8.3 6.8 2.5 5.1

280 2008 Oct 6 -999.0 1.8 -999.0 1.8 4.1 1.5 4.9 8.7 1.7 -999.0 3.9 0.8 0.8 2.7 1.0 0.5 1.5

281 2008 Oct 7 -999.0 2.5 -999.0 2.1 4.6 1.8 4.6 7.0 2.4 -999.0 2.3 1.8 1.9 1.5 3.8 1.4 1.8

282 2008 Oct 8 -999.0 2.0 -999.0 1.9 3.9 2.1 4.9 -777.0 3.5 -999.0 2.5 1.3 0.9 5.7 1.8 0.5 2.0

283 2008 Oct 9 -999.0 2.9 -999.0 2.8 4.1 2.8 4.7 -777.0 3.0 -999.0 1.4 1.4 2.7 1.9 1.9 2.1 2.9

284 2008 Oct 10 -999.0 2.9 -999.0 3.4 4.6 4.0 6.4 6.5 3.8 -999.0 1.2 1.8 1.8 1.1 5.0 1.4 4.0

285 2008 Oct 11 -999.0 3.2 -999.0 2.3 4.9 2.1 7.8 4.9 2.5 -999.0 1.6 2.1 1.4 2.2 3.3 1.9 2.2

286 2008 Oct 12 -999.0 3.6 -999.0 4.0 4.4 4.3 8.0 5.4 4.6 -999.0 2.4 4.3 2.7 1.9 5.0 2.4 4.4

287 2008 Oct 13 -999.0 1.7 -999.0 3.3 4.0 2.0 8.0 3.1 1.6 -999.0 2.9 3.1 1.3 4.6 3.4 1.2 2.1

288 2008 Oct 14 -999.0 1.2 -999.0 1.2 3.1 1.2 3.8 17.5 2.0 -999.0 0.8 0.4 0.6 4.1 1.0 0.3 1.3

289 2008 Oct 15 -999.0 3.1 -999.0 6.5 3.6 2.8 7.3 11.9 4.0 -999.0 2.6 2.8 1.6 2.9 2.8 1.4 2.9

290 2008 Oct 16 -999.0 5.4 -999.0 8.9 5.7 4.9 10.2 11.3 5.9 -999.0 4.0 5.4 3.1 3.0 6.3 3.3 5.0

291 2008 Oct 17 -999.0 5.2 -999.0 7.1 5.6 4.9 8.7 12.1 8.6 -999.0 2.3 5.3 2.8 1.2 5.9 3.3 5.0

292 2008 Oct 18 -999.0 3.1 -999.0 3.8 4.9 2.5 4.9 3.6 5.3 -999.0 2.6 1.9 2.0 1.3 3.1 4.2 2.5

293 2008 Oct 19 -999.0 0.8 -999.0 1.6 3.2 0.9 3.6 2.4 1.5 -999.0 1.1 0.7 1.2 1.0 2.1 0.8 1.0

294 2008 Oct 20 -999.0 3.3 -999.0 4.1 5.1 3.9 7.0 6.3 6.1 -999.0 2.4 2.9 2.7 1.9 4.5 3.0 3.9

295 2008 Oct 21 -999.0 3.5 -999.0 3.3 3.8 4.6 5.3 8.6 7.1 -999.0 4.9 2.3 1.9 4.5 4.2 1.3 4.6

296 2008 Oct 22 -999.0 2.3 -999.0 0.8 3.9 1.6 5.1 6.9 2.2 -999.0 0.7 0.6 1.2 0.3 2.6 0.4 1.6

297 2008 Oct 23 -999.0 4.2 -999.0 3.7 11.6 5.6 7.0 18.9 16.3 -999.0 1.6 2.7 2.1 4.6 4.8 1.4 5.7

298 2008 Oct 24 -999.0 3.9 -999.0 6.6 6.1 4.1 6.7 8.1 7.4 -999.0 2.4 2.6 2.0 1.6 5.0 2.2 4.1

299 2008 Oct 25 -999.0 3.5 -999.0 4.0 5.8 4.9 5.4 13.3 8.2 -999.0 2.1 1.7 5.3 8.9 6.3 2.5 5.0

300 2008 Oct 26 -999.0 1.7 -999.0 1.6 1.9 1.6 2.5 2.1 2.8 -999.0 1.7 0.4 1.2 0.8 1.3 1.0 1.6

301 2008 Oct 27 -999.0 9.3 -999.0 14.1 9.5 11.5 14.5 13.7 12.8 -999.0 7.0 9.2 6.0 3.0 11.3 7.3 11.5

A-27

JD Y M D EDMC TEOM4

EDMC TEOM3

EDME TEOM4

EDME TEOM3

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMC TEOM3

EDMS TEOM4

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

302 2008 Oct 28 -999.0 15.8 -999.0 25.7 17.4 20.9 26.7 25.0 38.0 -999.0 10.1 17.1 10.3 10.0 24.2 11.3 21.0

303 2008 Oct 29 -999.0 16.0 -999.0 21.4 14.0 18.0 24.7 26.3 35.7 -999.0 15.5 14.7 6.6 6.2 21.2 8.1 18.0

304 2008 Oct 30 -999.0 3.5 -999.0 4.2 3.7 3.2 5.9 4.4 5.6 -999.0 3.6 1.1 3.7 1.3 3.3 3.0 3.3

305 2008 Oct 31 -999.0 4.9 -999.0 4.4 4.6 3.5 8.1 6.4 6.4 -999.0 4.3 2.7 4.7 1.1 5.2 4.0 3.6

306 2008 Nov 1 -999.0 7.6 -999.0 8.7 7.7 6.0 12.4 6.6 7.6 -999.0 6.5 7.0 4.3 6.1 7.8 3.6 6.1

307 2008 Nov 2 -999.0 4.7 -999.0 5.0 6.1 4.5 9.9 5.3 12.9 -999.0 3.4 2.6 3.3 -999.0 4.8 3.0 4.6

308 2008 Nov 3 -999.0 3.0 -999.0 4.6 4.7 3.5 8.5 8.2 8.3 -999.0 3.1 2.1 2.0 -777.0 2.6 1.1 3.6

309 2008 Nov 4 -999.0 6.8 -999.0 10.0 7.3 7.8 12.3 7.3 17.0 -999.0 3.2 5.0 3.6 2.1 4.6 3.6 7.9

310 2008 Nov 5 -999.0 5.6 -999.0 5.4 6.0 3.6 8.3 3.5 4.8 -999.0 2.5 2.4 5.5 1.5 5.5 5.1 3.5

311 2008 Nov 6 -999.0 4.5 -999.0 4.9 7.0 3.8 12.3 5.2 4.5 -999.0 2.6 2.6 4.6 4.1 4.9 3.6 3.9

312 2008 Nov 7 -999.0 7.1 -999.0 6.6 6.5 5.0 12.8 8.1 5.5 -999.0 3.3 4.3 3.9 3.6 4.9 3.9 5.1

313 2008 Nov 8 -999.0 2.7 -999.0 2.4 5.2 1.5 9.1 2.7 2.8 -999.0 1.6 1.7 2.2 2.2 7.3 3.0 1.5

314 2008 Nov 9 -999.0 5.1 -999.0 4.9 7.6 4.0 10.0 3.3 4.5 -999.0 3.9 5.1 4.4 2.9 4.9 5.3 4.0

315 2008 Nov 10 -999.0 7.5 -999.0 9.2 12.0 6.9 14.6 9.1 7.0 -999.0 4.8 8.7 7.3 5.3 8.7 4.8 6.9

316 2008 Nov 11 -999.0 6.1 -999.0 6.0 11.3 5.0 13.8 5.2 5.7 -999.0 2.4 5.1 6.5 -777.0 6.8 3.9 5.0

317 2008 Nov 12 -999.0 3.6 -999.0 3.7 6.7 2.6 10.3 4.6 2.4 -999.0 2.9 3.8 1.4 -777.0 5.0 2.0 2.7

318 2008 Nov 13 -999.0 2.8 -999.0 2.2 2.6 1.8 2.6 13.7 2.2 -999.0 1.8 1.8 2.5 1.0 2.0 1.6 1.8

319 2008 Nov 14 -999.0 2.4 -999.0 2.3 2.6 1.8 3.9 4.7 2.0 -999.0 1.0 1.4 1.1 0.5 1.3 0.9 1.8

320 2008 Nov 15 -999.0 2.4 -999.0 2.6 3.0 2.1 3.8 3.8 2.2 -999.0 1.8 2.0 2.3 1.2 2.1 2.5 2.2

321 2008 Nov 16 -999.0 8.2 -999.0 8.6 6.3 6.3 10.3 4.2 9.0 -999.0 2.7 4.2 3.5 1.6 4.8 2.0 6.3

322 2008 Nov 17 -999.0 2.3 -999.0 1.9 5.5 1.8 8.1 3.9 2.1 -999.0 1.0 1.7 3.0 2.5 4.2 1.9 1.9

323 2008 Nov 18 -999.0 6.0 -999.0 5.8 7.2 5.2 9.0 9.1 5.0 -999.0 5.5 5.8 1.5 6.7 7.5 5.0 5.3

324 2008 Nov 19 -999.0 3.6 -999.0 3.5 1.4 3.1 5.0 3.8 3.6 -999.0 1.7 2.7 0.7 0.4 4.0 1.9 3.1

325 2008 Nov 20 -999.0 1.2 -999.0 1.0 1.6 0.8 6.3 4.9 0.9 -999.0 0.6 0.7 0.8 1.3 1.6 1.6 0.8

326 2008 Nov 21 -999.0 5.0 -999.0 4.7 4.1 3.7 8.3 5.3 3.7 -999.0 2.8 3.5 2.2 0.8 4.2 2.6 3.7

327 2008 Nov 22 -999.0 2.0 -999.0 1.5 1.9 1.2 2.2 3.4 1.7 -999.0 1.6 1.3 0.7 0.6 2.4 1.5 1.1

328 2008 Nov 23 -999.0 1.1 -999.0 1.1 2.5 1.3 2.7 2.2 1.9 -999.0 0.9 1.1 0.7 0.4 1.6 0.8 1.4

A-28

JD Y M D EDMC TEOM4

EDMC TEOM3

EDME TEOM4

EDME TEOM3

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMC TEOM3

EDMS TEOM4

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

329 2008 Nov 24 -999.0 6.1 -999.0 14.5 7.3 8.6 10.6 10.6 12.2 -999.0 3.7 7.1 3.9 2.4 7.0 3.1 8.7

330 2008 Nov 25 -999.0 7.7 -999.0 7.4 6.1 6.0 9.6 11.5 9.9 -999.0 5.0 8.1 4.3 1.3 10.9 3.5 6.1

331 2008 Nov 26 -999.0 2.7 -999.0 4.0 3.8 3.8 6.4 9.7 3.2 -999.0 1.7 2.5 1.0 0.6 2.3 1.5 3.9

332 2008 Nov 27 -999.0 5.4 -999.0 6.1 3.9 4.5 7.8 8.5 6.2 -999.0 1.9 4.6 2.8 1.0 9.9 2.8 4.5

333 2008 Nov 28 -999.0 7.1 -999.0 12.6 7.0 8.1 12.6 11.5 7.6 -999.0 4.5 8.2 4.3 1.0 8.6 3.9 8.1

334 2008 Nov 29 -999.0 1.4 -999.0 1.1 3.0 1.4 2.8 5.5 1.0 -999.0 1.0 0.8 1.1 0.6 0.7 0.8 1.3

335 2008 Nov 30 -999.0 1.2 -999.0 2.3 3.6 1.3 4.3 3.9 1.0 -999.0 1.5 1.1 0.5 0.5 1.5 0.6 1.3

336 2008 Dec 1 -999.0 1.7 -999.0 2.3 4.0 2.4 4.0 6.4 1.6 -999.0 0.9 1.4 1.8 0.3 9.4 1.0 2.5

337 2008 Dec 2 -999.0 2.2 -999.0 1.7 1.4 2.1 4.4 7.3 1.6 -999.0 3.0 1.2 2.1 0.1 2.0 1.0 2.1

338 2008 Dec 3 -999.0 2.1 -999.0 2.2 0.5 2.1 4.6 4.0 1.7 -999.0 1.5 1.5 1.1 0.2 2.1 0.9 2.2

339 2008 Dec 4 -999.0 2.5 -999.0 4.6 -777.0 2.5 6.5 5.6 2.4 -999.0 1.1 2.5 1.0 0.3 4.3 2.2 2.5

340 2008 Dec 5 -999.0 3.0 -999.0 2.8 2.7 2.8 7.0 4.0 3.2 -999.0 2.1 2.5 2.5 1.2 3.2 1.9 2.8

341 2008 Dec 6 -999.0 4.2 -999.0 4.2 8.5 3.5 10.1 4.3 3.8 -999.0 -777.0 4.1 3.1 -777.0 4.8 3.5 3.7

342 2008 Dec 7 -999.0 9.3 -999.0 10.0 5.5 7.5 13.0 6.9 10.1 -999.0 3.7 6.8 5.4 -999.0 4.5 4.7 7.5

343 2008 Dec 8 -999.0 5.1 -999.0 5.5 3.0 5.2 9.2 6.6 6.2 -999.0 -777.0 3.1 5.7 -777.0 3.5 5.0 5.3

344 2008 Dec 9 -999.0 5.2 -999.0 4.6 5.6 4.6 8.7 9.7 5.6 -999.0 3.7 3.7 4.5 2.1 3.7 4.1 4.6

345 2008 Dec 10 -999.0 4.3 -999.0 5.6 6.1 4.6 9.7 6.6 4.8 -999.0 4.1 5.3 0.9 4.9 5.2 0.1 4.6

346 2008 Dec 11 -999.0 1.1 -999.0 1.7 3.1 1.3 5.5 5.4 0.6 -999.0 1.1 1.0 0.5 0.1 0.9 0.5 1.3

347 2008 Dec 12 -999.0 1.4 -999.0 4.3 1.0 0.5 4.6 6.0 0.7 -999.0 0.9 1.3 0.9 0.3 1.5 0.5 0.5

348 2008 Dec 13 -999.0 4.4 -999.0 5.8 -999.0 3.2 3.0 4.1 4.7 -999.0 4.3 4.4 3.6 1.5 4.4 3.2 3.3

349 2008 Dec 14 -999.0 5.2 -999.0 6.2 -999.0 4.6 6.0 4.1 6.0 -999.0 6.1 4.7 4.2 5.8 5.4 4.3 4.7

350 2008 Dec 15 -999.0 2.0 -999.0 3.8 -999.0 1.9 3.8 4.6 1.8 -999.0 2.7 1.1 1.1 1.5 2.1 1.1 1.9

351 2008 Dec 16 -999.0 6.4 -999.0 7.2 -777.0 4.8 9.6 7.7 5.4 -999.0 2.9 4.7 4.1 0.7 5.6 3.5 4.9

352 2008 Dec 17 -999.0 9.4 -999.0 14.5 -777.0 11.0 18.2 14.0 10.7 -999.0 4.9 6.9 5.0 3.3 5.6 3.5 11.0

353 2008 Dec 18 -999.0 10.9 -999.0 15.9 -777.0 15.9 21.7 15.2 12.4 -999.0 5.2 9.1 6.0 4.2 8.0 4.4 16.0

354 2008 Dec 19 -999.0 7.9 -999.0 8.2 -999.0 7.1 10.7 10.0 8.6 -999.0 6.9 6.3 7.1 4.6 7.0 6.2 7.2

355 2008 Dec 20 -999.0 11.2 -999.0 12.4 -999.0 12.3 16.1 9.3 14.2 -999.0 9.9 14.9 6.5 20.0 11.7 5.7 12.4

356 2008 Dec 21 -999.0 11.1 -999.0 16.9 -999.0 6.3 10.7 4.8 9.2 -999.0 5.7 12.5 5.1 8.4 14.6 5.7 6.4

A-29

JD Y M D EDMC TEOM4

EDMC TEOM3

EDME TEOM4

EDME TEOM3

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMC TEOM3

EDMS TEOM4

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

357 2008 Dec 22 -999.0 12.4 -999.0 14.7 -999.0 14.5 -777.0 9.9 13.6 -999.0 11.5 12.7 8.3 22.9 11.5 7.6 14.5

358 2008 Dec 23 -999.0 9.0 -999.0 8.6 -777.0 8.1 -777.0 6.3 8.6 -999.0 3.0 18.6 7.1 7.1 14.0 3.9 8.2

359 2008 Dec 24 -999.0 9.3 -999.0 10.8 -777.0 8.7 15.9 5.3 8.4 -999.0 3.3 16.4 2.6 7.3 15.2 1.7 8.9

360 2008 Dec 25 -999.0 15.3 -999.0 15.0 -777.0 14.8 21.3 9.5 18.5 -999.0 10.2 10.7 3.6 15.9 7.5 2.9 14.8

361 2008 Dec 26 -999.0 5.0 -999.0 4.4 -777.0 4.6 13.2 3.9 5.1 -999.0 3.6 4.6 1.0 7.9 9.3 0.8 4.6

362 2008 Dec 27 -999.0 4.4 -999.0 4.2 5.7 5.7 14.5 4.8 5.1 -999.0 3.0 5.3 1.6 3.0 5.9 1.8 5.9

363 2008 Dec 28 -999.0 12.2 -999.0 7.3 3.8 7.8 12.7 6.1 9.4 -999.0 -777.0 4.8 4.3 1.1 2.1 3.3 7.9

364 2008 Dec 29 -999.0 4.2 -999.0 3.7 2.0 2.5 7.9 4.2 3.0 -999.0 -777.0 2.0 2.4 0.3 8.8 4.0 2.5

365 2008 Dec 30 -999.0 4.7 -999.0 3.5 -777.0 5.2 7.4 3.5 5.9 -999.0 2.3 3.9 4.2 1.8 5.6 3.6 5.2

366 2008 Dec 31 -999.0 5.0 -999.0 3.9 -999.0 3.8 6.8 3.9 4.5 -999.0 3.4 3.1 3.8 0.8 6.6 4.0 3.9

A-30

Table A-13. Measured 24-Hour PM2.5 at monitoring sites in the Capital Region during 2009 with values above the CWS (> 30 µg/m3) highlighted in yellow, values above 20 µg/m3 highlighted in blue and PM2.5 modelling episodes for the 1.33 km fine-scale Capital Region modelling domain shown by the red boxes.

JD Y M D EDMC TEOM3

EDMC FDMS

EDME TEOM3

EDME FDMS

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS TEOM3

EDMS TEOM4

EDMS FDMS

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

1 2009 Jan 1 5.9 -999.0 5.3 -999.0 -999.0 6.1 -999.0 3.4 6.5 -999.0 -999.0 3.5 7.4 5.0 0.7 4.0 4.2 6.2

2 2009 Jan 2 4.4 -999.0 3.5 -999.0 -999.0 3.5 -999.0 5.6 4.6 -999.0 -999.0 2.5 3.2 2.9 1.0 3.7 1.8 3.6

3 2009 Jan 3 7.8 -999.0 13.1 -999.0 -777.0 6.8 -999.0 5.2 7.5 -999.0 -999.0 3.8 10.1 6.2 4.6 13.3 3.9 6.9

4 2009 Jan 4 6.8 -999.0 9.7 -999.0 -777.0 5.4 -999.0 3.4 5.9 -999.0 -999.0 4.3 10.4 2.1 5.7 10.9 1.8 5.5

5 2009 Jan 5 15.6 -999.0 13.0 -999.0 11.6 14.9 -999.0 12.4 11.9 -999.0 -999.0 2.3 5.8 5.6 2.6 7.5 4.9 15.0

6 2009 Jan 6 5.6 -999.0 5.7 -999.0 -777.0 5.0 -999.0 7.8 4.5 -999.0 -999.0 3.7 3.9 2.3 4.7 4.2 1.8 5.0

7 2009 Jan 7 4.2 -999.0 1.9 -999.0 -999.0 2.5 -999.0 5.6 3.8 -999.0 -999.0 2.1 2.0 3.3 0.0 6.3 3.4 2.5

8 2009 Jan 8 4.6 -999.0 4.6 -999.0 -999.0 4.3 -999.0 10.6 4.5 -999.0 -999.0 3.0 2.8 4.9 0.8 4.6 4.6 4.3

9 2009 Jan 9 8.6 -999.0 9.9 -999.0 -777.0 8.1 -999.0 6.8 7.9 -999.0 -999.0 5.1 8.9 6.4 4.9 11.2 5.7 8.0

10 2009 Jan 10 1.9 -999.0 4.0 -999.0 2.9 3.0 -999.0 4.4 1.9 -999.0 -999.0 1.7 2.1 1.1 1.0 1.0 1.0 3.0

11 2009 Jan 11 2.8 -999.0 4.2 -999.0 1.9 2.2 -999.0 3.0 3.0 -999.0 -999.0 3.8 4.4 2.4 5.2 8.3 1.4 2.1

12 2009 Jan 12 6.3 -999.0 5.5 -999.0 0.7 4.2 -999.0 4.9 5.5 -999.0 -999.0 1.4 1.9 4.0 0.3 4.8 1.9 4.3

13 2009 Jan 13 5.1 -999.0 7.8 -999.0 -777.0 5.5 -999.0 9.0 4.9 -999.0 -999.0 4.0 3.8 3.3 1.1 4.9 1.5 5.5

14 2009 Jan 14 3.2 -999.0 2.8 -999.0 -999.0 2.1 -999.0 4.3 2.9 -999.0 -999.0 1.4 1.4 2.3 0.1 2.8 1.1 2.0

15 2009 Jan 15 3.3 -999.0 10.2 -999.0 5.6 5.1 -999.0 5.5 2.8 -999.0 -999.0 3.3 3.9 1.4 -777.0 3.5 2.6 5.2

16 2009 Jan 16 2.9 -999.0 4.7 -999.0 3.2 3.4 -999.0 4.3 2.6 -999.0 -999.0 2.7 2.3 1.9 0.1 1.3 -777.0 3.5

17 2009 Jan 17 2.8 -999.0 6.7 -999.0 3.3 2.7 -999.0 3.5 2.6 -999.0 -999.0 3.2 3.2 1.8 1.7 2.5 1.0 2.9

18 2009 Jan 18 2.3 -999.0 4.8 -999.0 4.4 3.4 -999.0 5.6 2.6 -999.0 -999.0 2.3 3.9 1.5 1.6 2.6 2.9 3.4

19 2009 Jan 19 3.7 -999.0 10.2 -999.0 6.0 7.3 -999.0 13.0 4.7 -999.0 -999.0 3.6 6.7 1.5 2.1 5.6 3.9 7.5

20 2009 Jan 20 7.7 -999.0 11.3 -999.0 7.7 9.1 -999.0 14.7 6.9 -999.0 -999.0 7.4 10.3 -777.0 6.1 8.4 8.0 9.2

21 2009 Jan 21 10.3 -999.0 14.0 -999.0 8.9 10.8 -999.0 14.4 9.0 -999.0 -999.0 8.1 12.8 3.3 -777.0 10.7 7.6 10.9

22 2009 Jan 22 2.8 -999.0 2.5 -999.0 -777.0 1.6 -999.0 6.4 3.0 -999.0 -999.0 2.8 1.0 2.9 0.5 2.8 4.5 1.6

23 2009 Jan 23 5.1 -999.0 6.1 -999.0 -999.0 5.1 -999.0 5.4 5.2 -999.0 -999.0 9.9 4.3 3.5 2.0 4.7 2.6 5.1

24 2009 Jan 24 4.7 -999.0 7.1 -999.0 -999.0 5.0 -999.0 4.5 4.9 -999.0 -999.0 5.6 4.2 3.9 -777.0 4.5 3.8 5.1

25 2009 Jan 25 3.3 -999.0 4.7 -999.0 -999.0 4.0 -999.0 5.9 4.4 -999.0 -999.0 1.8 3.2 1.3 0.3 2.0 0.7 4.1

A-31

JD Y M D EDMC TEOM3

EDMC FDMS

EDME TEOM3

EDME FDMS

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS TEOM3

EDMS TEOM4

EDMS FDMS

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

26 2009 Jan 26 4.5 -999.0 7.0 -999.0 -999.0 5.4 -999.0 7.4 4.5 -999.0 -999.0 2.4 6.9 1.0 1.9 7.9 15.4 5.5

27 2009 Jan 27 4.3 -999.0 6.3 -999.0 5.9 4.7 -999.0 6.0 2.7 -999.0 -999.0 3.4 4.8 2.8 2.5 4.4 7.0 4.8

28 2009 Jan 28 1.8 -999.0 2.4 -999.0 2.4 1.7 -999.0 4.3 1.2 -999.0 -999.0 1.8 2.2 1.3 1.7 2.3 0.8 1.8

29 2009 Jan 29 1.0 -999.0 3.1 -999.0 3.3 1.8 -999.0 6.6 1.0 -999.0 -999.0 1.3 1.4 0.9 1.3 1.3 1.6 1.8

30 2009 Jan 30 1.9 -999.0 3.0 -999.0 3.9 2.5 -999.0 5.4 1.7 -999.0 -999.0 1.3 2.6 0.7 1.2 4.2 0.7 2.5

31 2009 Jan 31 0.9 -999.0 1.2 -999.0 1.4 0.4 -999.0 1.2 0.9 -999.0 -999.0 1.6 0.8 1.9 0.5 1.2 0.5 0.4

32 2009 Feb 1 2.2 -999.0 1.8 -999.0 2.0 1.3 -999.0 1.9 2.1 -999.0 -999.0 1.3 1.8 0.9 0.4 4.4 0.9 1.3

33 2009 Feb 2 4.7 -999.0 5.0 -999.0 6.0 -777.0 -999.0 5.4 4.3 -999.0 -999.0 0.9 3.8 2.2 3.0 3.0 2.8 -

777.0

34 2009 Feb 3 2.2 -999.0 4.0 -999.0 4.6 -999.0 -999.0 -777.0 2.4 -999.0 -999.0 0.7 4.0 0.4 2.0 6.4 1.6 -

999.0

35 2009 Feb 4 5.5 -999.0 7.9 -999.0 6.4 8.1 -999.0 9.4 5.7 -999.0 -999.0 2.4 5.2 1.5 1.8 5.8 2.3 8.1

36 2009 Feb 5 5.7 -999.0 8.7 -999.0 12.0 9.6 -999.0 7.8 7.1 -999.0 -999.0 2.9 5.3 4.9 4.0 6.9 3.8 9.7

37 2009 Feb 6 9.8 -999.0 11.7 -999.0 18.3 13.6 -999.0 12.1 10.5 -999.0 -999.0 2.6 7.2 5.0 3.8 6.4 3.2 13.5

38 2009 Feb 7 1.4 -999.0 3.0 -999.0 4.7 3.2 -999.0 3.4 2.4 -999.0 -999.0 1.8 3.2 0.7 2.6 4.2 1.0 3.2

39 2009 Feb 8 4.3 -999.0 4.5 -999.0 -777.0 4.4 -999.0 4.9 4.8 -999.0 -999.0 1.9 5.8 1.5 1.9 5.3 1.0 4.5

40 2009 Feb 9 14.3 -999.0 14.1 -999.0 -999.0 15.6 -999.0 12.7 11.1 -999.0 -999.0 4.3 5.7 5.2 12.9 4.2 4.6 15.7

41 2009 Feb 10 3.4 -999.0 3.7 -999.0 -999.0 5.5 -999.0 5.7 4.1 -999.0 -999.0 3.3 4.0 2.8 6.6 4.1 1.4 5.6

42 2009 Feb 11 9.4 -999.0 15.6 -999.0 -777.0 12.0 -999.0 12.8 8.6 -999.0 -999.0 10.1 19.8 6.1 10.0 12.5 6.0 12.0

43 2009 Feb 12 5.8 -999.0 7.4 -999.0 9.6 8.0 -999.0 8.6 7.5 -999.0 -999.0 7.2 6.2 2.9 12.0 5.2 3.2 8.1

44 2009 Feb 13 9.3 -999.0 14.3 -999.0 12.2 12.1 -999.0 12.8 9.9 -999.0 -999.0 22.7 4.9 4.8 23.8 3.5 4.0 12.1

45 2009 Feb 14 5.8 -999.0 6.2 -999.0 -777.0 5.0 -999.0 5.3 5.4 -999.0 -999.0 3.5 7.4 4.6 8.4 11.0 5.9 5.0

46 2009 Feb 15 0.9 -999.0 3.3 -999.0 -999.0 0.7 -999.0 2.1 1.6 -999.0 -999.0 0.4 2.8 1.7 1.9 4.8 3.2 0.8

47 2009 Feb 16 8.6 -999.0 9.0 -999.0 -777.0 10.1 -999.0 8.2 10.5 -999.0 -999.0 4.7 10.8 5.9 11.2 14.3 6.8 10.1

48 2009 Feb 17 22.1 -999.0 24.5 -999.0 21.2 16.4 -999.0 20.4 16.8 -999.0 -999.0 11.2 36.9 7.0 44.5 44.6 5.4 16.5

49 2009 Feb 18 8.8 -999.0 8.8 -999.0 9.5 8.7 -999.0 11.5 6.6 -999.0 -999.0 6.6 13.8 3.9 9.9 -777.0 3.6 8.8

50 2009 Feb 19 4.5 -999.0 6.0 -999.0 10.0 5.0 -999.0 5.0 4.8 -999.0 -999.0 -777.0 3.4 2.7 4.5 -777.0 2.9 5.0

51 2009 Feb 20 7.7 -999.0 5.7 -999.0 12.7 5.9 -999.0 7.0 5.6 -999.0 -999.0 -777.0 3.9 3.4 3.0 5.1 4.1 6.0

52 2009 Feb 21 11.1 -999.0 9.3 -999.0 19.3 8.0 -999.0 8.2 9.5 -999.0 -999.0 7.4 6.2 6.6 12.9 10.4 7.8 8.1

A-32

JD Y M D EDMC TEOM3

EDMC FDMS

EDME TEOM3

EDME FDMS

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS TEOM3

EDMS TEOM4

EDMS FDMS

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

53 2009 Feb 22 3.7 -999.0 3.1 -999.0 9.0 3.8 -999.0 3.7 3.0 -999.0 -999.0 4.5 2.0 4.2 1.5 2.8 6.1 3.8

54 2009 Feb 23 2.8 -999.0 4.1 -999.0 -777.0 2.8 -999.0 5.6 2.5 -999.0 -999.0 5.3 1.6 1.9 1.5 1.1 2.6 2.7

55 2009 Feb 24 4.8 -999.0 3.4 -999.0 -999.0 6.0 -999.0 6.1 4.2 -999.0 -999.0 4.3 4.4 2.2 0.4 1.4 2.2 6.0

56 2009 Feb 25 4.0 -999.0 3.3 -999.0 -999.0 4.5 -999.0 3.8 3.9 -999.0 -999.0 4.4 4.1 2.0 0.5 2.8 3.8 4.5

57 2009 Feb 26 2.7 -999.0 2.4 -999.0 -999.0 2.2 -999.0 6.6 2.5 -999.0 -999.0 4.3 4.0 1.4 0.5 4.0 2.6 2.2

58 2009 Feb 27 6.9 -999.0 7.8 -999.0 -999.0 7.2 -999.0 12.1 6.2 -999.0 -999.0 10.3 6.1 4.4 5.2 5.6 3.9 7.2

59 2009 Feb 28 4.8 -999.0 6.4 -999.0 -999.0 5.0 -999.0 5.2 4.7 -999.0 -999.0 6.3 -777.0 3.7 0.7 3.5 3.5 5.0

60 2009 Mar 1 10.0 -999.0 10.7 -999.0 -999.0 6.5 -999.0 5.4 7.2 -999.0 -999.0 11.9 -999.0 5.1 3.4 22.9 5.4 6.5

61 2009 Mar 2 8.5 -999.0 9.6 -999.0 -999.0 7.4 -999.0 9.2 -777.0 -999.0 -999.0 9.1 -999.0 6.1 4.1 7.9 8.1 7.5

62 2009 Mar 3 15.3 -999.0 15.9 -999.0 -999.0 15.3 -999.0 12.2 -777.0 -999.0 -999.0 20.5 -777.0 7.7 31.0 11.2 6.4 15.2

63 2009 Mar 4 2.7 -999.0 4.0 -999.0 -999.0 2.5 -999.0 6.3 2.6 -999.0 -999.0 11.7 6.5 1.1 -777.0 8.9 1.0 2.5

64 2009 Mar 5 2.0 -999.0 1.9 -999.0 -999.0 1.2 -999.0 4.6 3.0 -999.0 -999.0 5.9 1.9 2.2 1.1 0.5 1.5 1.3

65 2009 Mar 6 4.6 -999.0 3.4 -999.0 -999.0 2.8 -999.0 3.9 4.1 -999.0 -999.0 6.1 2.9 2.1 5.8 2.5 1.8 2.9

66 2009 Mar 7 3.2 -999.0 3.7 -999.0 -999.0 2.4 -999.0 2.1 2.6 -999.0 -999.0 7.5 2.6 3.3 6.5 5.0 3.4 2.4

67 2009 Mar 8 2.7 -999.0 2.6 -999.0 -999.0 1.5 -999.0 1.7 2.8 -999.0 -999.0 5.5 2.3 2.7 1.0 2.2 2.0 1.4

68 2009 Mar 9 4.4 -999.0 6.6 -999.0 -999.0 3.4 -999.0 4.9 4.7 -999.0 -999.0 7.4 4.1 3.1 5.0 4.5 2.7 3.6

69 2009 Mar 10 5.8 -999.0 9.7 -999.0 -999.0 6.7 -999.0 9.0 6.4 -999.0 -999.0 7.3 6.5 4.3 16.4 6.8 3.8 6.8

70 2009 Mar 11 4.6 -999.0 7.0 -999.0 -999.0 5.0 -999.0 8.0 4.9 -999.0 -999.0 5.9 6.2 3.2 9.9 6.8 3.1 5.0

71 2009 Mar 12 4.0 -999.0 6.8 -999.0 -777.0 4.9 -999.0 5.4 4.1 -999.0 -999.0 6.7 4.8 2.2 6.8 4.8 2.9 5.0

72 2009 Mar 13 1.6 -999.0 6.1 -999.0 4.7 3.0 -999.0 9.5 1.2 -999.0 -999.0 3.8 1.9 0.5 2.2 1.2 0.9 3.0

73 2009 Mar 14 5.7 -999.0 5.6 -999.0 15.7 4.2 -999.0 9.1 5.5 -999.0 -999.0 5.3 3.3 2.8 3.1 4.9 3.5 4.2

74 2009 Mar 15 4.4 -999.0 5.0 -999.0 6.4 2.8 -999.0 4.2 4.6 -999.0 -999.0 6.2 3.7 5.0 6.8 7.9 4.3 2.8

75 2009 Mar 16 8.0 -999.0 7.9 -999.0 5.6 7.0 -999.0 12.3 7.3 -999.0 -999.0 8.1 5.5 7.0 5.0 5.0 5.5 7.0

76 2009 Mar 17 8.6 -999.0 6.6 -999.0 2.8 5.5 -999.0 5.9 7.2 -999.0 -999.0 6.8 4.5 5.3 5.9 4.3 6.0 5.5

77 2009 Mar 18 10.2 -999.0 8.4 -999.0 11.9 7.8 -999.0 10.1 7.7 -999.0 -999.0 8.0 7.0 5.6 12.1 8.1 5.1 7.9

78 2009 Mar 19 12.4 -999.0 22.1 -999.0 21.8 15.5 -999.0 21.3 8.6 -999.0 -999.0 7.9 14.2 4.9 15.2 9.1 3.3 15.5

79 2009 Mar 20 8.9 -999.0 7.0 -999.0 21.8 6.9 -999.0 14.2 7.0 -999.0 -999.0 9.4 5.6 6.8 8.6 9.7 6.5 6.9

A-33

JD Y M D EDMC TEOM3

EDMC FDMS

EDME TEOM3

EDME FDMS

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS TEOM3

EDMS TEOM4

EDMS FDMS

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

80 2009 Mar 21 8.4 -999.0 8.0 -999.0 -777.0 7.1 -999.0 10.1 6.6 -999.0 -999.0 12.7 6.5 6.4 9.8 8.2 6.2 7.1

81 2009 Mar 22 4.6 -999.0 4.3 -999.0 8.0 4.9 -999.0 6.4 4.4 -999.0 -999.0 8.4 4.7 4.3 4.6 3.3 3.7 4.9

82 2009 Mar 23 4.0 -999.0 3.8 -999.0 4.3 4.7 -999.0 12.7 3.8 -999.0 -999.0 7.0 2.7 2.7 4.2 2.5 2.5 4.7

83 2009 Mar 24 3.2 -999.0 2.1 -999.0 -777.0 2.4 -999.0 8.3 2.4 -999.0 -999.0 3.9 2.5 1.4 4.7 3.9 2.5 2.5

84 2009 Mar 25 7.5 -999.0 8.0 -999.0 3.1 5.9 -999.0 13.0 4.7 -999.0 -999.0 8.4 3.7 2.4 4.7 7.4 2.6 5.9

85 2009 Mar 26 5.3 -999.0 3.4 -999.0 3.8 3.2 -999.0 9.8 -777.0 -999.0 -999.0 4.9 3.0 1.3 6.2 6.7 1.2 3.3

86 2009 Mar 27 3.2 -999.0 2.6 -999.0 2.2 2.3 -999.0 15.9 1.1 -999.0 -999.0 3.5 1.7 1.5 2.0 1.7 1.1 2.3

87 2009 Mar 28 3.0 -999.0 3.0 -999.0 2.0 2.8 -999.0 9.3 2.3 -999.0 -999.0 4.0 2.0 1.0 1.7 1.1 0.7 2.8

88 2009 Mar 29 4.4 -999.0 3.9 -999.0 4.4 3.8 -999.0 7.5 4.3 -999.0 -999.0 5.9 2.3 2.1 4.3 3.1 1.5 3.9

89 2009 Mar 30 6.8 -999.0 4.8 -999.0 2.7 3.7 -999.0 9.5 3.4 -999.0 -999.0 4.9 3.6 2.1 3.5 4.9 1.3 3.7

90 2009 Mar 31 5.5 -999.0 5.9 -999.0 7.7 5.0 -999.0 14.2 3.5 -999.0 -999.0 6.8 3.6 1.4 3.6 5.5 1.5 5.0

91 2009 Apr 1 3.6 -999.0 -777.0 -999.0 5.8 3.8 -999.0 11.6 3.4 -999.0 -999.0 6.1 2.1 1.5 3.8 3.0 1.0 3.9

92 2009 Apr 2 9.6 -999.0 -999.0 -999.0 16.4 8.1 -999.0 15.1 8.6 -999.0 -999.0 5.5 7.8 6.5 5.8 6.9 5.6 8.2

93 2009 Apr 3 18.3 -999.0 -999.0 -999.0 30.6 19.3 -999.0 21.7 18.4 -999.0 -999.0 10.8 20.1 9.1 14.0 13.0 9.6 19.3

94 2009 Apr 4 11.3 -999.0 -999.0 -999.0 15.3 7.1 -999.0 10.2 8.3 -999.0 -999.0 10.1 9.0 7.0 25.2 12.1 6.7 7.2

95 2009 Apr 5 4.4 -999.0 -999.0 -999.0 5.0 3.1 -999.0 7.2 3.7 -999.0 -999.0 6.4 3.3 3.6 9.4 2.5 3.8 3.0

96 2009 Apr 6 8.0 -999.0 -999.0 -999.0 4.2 6.1 -999.0 15.7 6.5 -999.0 -999.0 6.3 4.3 3.0 6.5 4.2 2.5 6.1

97 2009 Apr 7 8.8 -999.0 -999.0 11.4 6.7 7.3 -999.0 10.5 6.6 -999.0 -999.0 6.5 5.5 4.4 6.8 4.4 4.0 7.4

98 2009 Apr 8 6.5 -999.0 -999.0 6.2 3.7 3.6 -999.0 4.7 5.0 -999.0 -999.0 5.1 2.8 3.4 3.9 4.5 3.7 3.7

99 2009 Apr 9 8.0 -999.0 -999.0 -777.0 4.4 7.9 -999.0 10.2 7.3 -999.0 -999.0 5.8 4.5 4.9 4.0 4.7 5.0 7.9

100 2009 Apr 10 8.6 -999.0 -999.0 -777.0 11.6 8.5 -999.0 8.5 8.3 -999.0 -999.0 -777.0 6.1 5.6 9.1 7.0 5.4 8.5

101 2009 Apr 11 5.0 -999.0 -999.0 -777.0 -999.0 3.4 -999.0 11.8 3.7 -999.0 -999.0 5.5 4.0 4.0 7.1 7.8 3.2 3.4

102 2009 Apr 12 1.2 -999.0 -999.0 -777.0 -999.0 0.8 -999.0 10.1 0.8 -999.0 -999.0 3.7 0.8 0.7 1.6 1.6 0.7 0.9

103 2009 Apr 13 3.7 -999.0 -999.0 5.0 -999.0 4.0 -999.0 12.6 2.8 -999.0 -999.0 4.2 1.6 2.3 1.7 3.4 1.6 4.0

104 2009 Apr 14 5.9 -999.0 -999.0 5.5 -777.0 4.7 -999.0 9.4 4.8 -999.0 -999.0 6.5 3.5 4.1 -777.0 5.7 3.4 4.7

105 2009 Apr 15 7.7 -999.0 -999.0 8.6 4.9 7.0 -999.0 7.9 6.2 -999.0 -999.0 5.9 3.5 3.6 -777.0 6.8 3.5 7.0

106 2009 Apr 16 7.0 -999.0 -999.0 5.7 1.8 4.5 -999.0 7.1 5.6 -999.0 -999.0 5.3 3.2 4.0 3.4 6.5 3.9 4.5

107 2009 Apr 17 5.9 -999.0 -999.0 5.2 2.3 4.2 -999.0 9.2 5.3 -999.0 -999.0 6.2 3.2 3.8 4.9 6.7 3.7 4.3

A-34

JD Y M D EDMC TEOM3

EDMC FDMS

EDME TEOM3

EDME FDMS

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS TEOM3

EDMS TEOM4

EDMS FDMS

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

108 2009 Apr 18 3.4 -999.0 -999.0 3.9 1.5 2.9 -999.0 6.4 2.3 -999.0 -999.0 4.3 2.0 2.3 1.9 2.6 1.3 2.9

109 2009 Apr 19 4.2 -999.0 -999.0 7.2 3.5 5.7 -999.0 8.9 6.7 -999.0 -999.0 6.3 3.4 2.8 3.5 4.5 2.3 5.7

110 2009 Apr 20 4.1 -999.0 -999.0 5.0 4.1 5.2 -999.0 14.0 -777.0 -999.0 -999.0 5.5 5.1 2.6 3.4 4.5 1.3 5.3

111 2009 Apr 21 4.8 -999.0 -999.0 5.7 3.3 3.8 -999.0 13.6 -777.0 -999.0 -999.0 5.3 1.9 3.0 3.0 7.2 2.0 3.9

112 2009 Apr 22 4.4 -999.0 -999.0 5.1 3.7 3.8 -999.0 10.2 3.9 -999.0 -999.0 4.7 -777.0 2.7 2.3 2.4 1.7 3.8

113 2009 Apr 23 3.7 -999.0 -999.0 3.2 2.3 4.2 -999.0 9.4 3.7 -999.0 -999.0 5.3 -999.0 3.8 1.8 2.5 2.3 4.3

114 2009 Apr 24 3.5 -999.0 -999.0 5.0 7.2 4.3 -999.0 5.4 3.6 -999.0 -999.0 4.9 -999.0 3.4 3.0 2.4 2.5 4.3

115 2009 Apr 25 7.3 -999.0 -999.0 9.2 7.6 6.7 -999.0 5.2 7.5 -999.0 -999.0 7.8 -999.0 5.6 4.6 5.9 4.9 6.8

116 2009 Apr 26 5.8 -999.0 -999.0 8.6 10.8 5.8 -999.0 5.5 7.7 -999.0 -999.0 6.7 -999.0 4.2 3.6 4.5 3.2 5.9

117 2009 Apr 27 3.2 -999.0 -999.0 5.2 3.8 2.9 -999.0 6.0 3.3 -999.0 -999.0 3.9 -999.0 1.5 1.9 1.7 1.4 3.0

118 2009 Apr 28 6.4 -999.0 -999.0 8.9 6.5 5.4 -999.0 10.9 8.1 -999.0 -999.0 5.4 -999.0 3.8 3.5 5.5 3.2 5.5

119 2009 Apr 29 8.5 -999.0 -999.0 8.6 3.6 5.2 -999.0 10.3 9.4 -999.0 -999.0 4.9 -999.0 4.0 4.3 4.6 3.4 5.2

120 2009 Apr 30 6.7 -999.0 -999.0 10.4 3.9 6.2 -999.0 9.3 7.3 -999.0 -999.0 6.6 -999.0 5.0 6.2 6.0 4.1 6.1

121 2009 May 1 -999.0 6.5 -999.0 9.5 6.1 7.8 -999.0 13.4 8.4 -999.0 -999.0 7.6 -999.0 4.7 5.4 6.4 3.8 7.9

122 2009 May 2 -999.0 9.9 -999.0 16.0 11.5 11.0 -999.0 12.8 11.3 -999.0 -999.0 9.2 -999.0 8.0 6.0 7.2 7.0 11.1

123 2009 May 3 -999.0 14.0 -999.0 14.5 10.5 12.0 -999.0 15.4 13.4 -999.0 -999.0 8.6 -999.0 5.3 6.8 6.8 5.3 12.1

124 2009 May 4 -999.0 2.4 -999.0 4.4 2.4 2.0 -999.0 7.2 3.3 -999.0 -999.0 5.6 -999.0 1.2 4.3 4.0 0.4 2.0

125 2009 May 5 -999.0 -777.0 -999.0 15.4 9.0 12.7 -999.0 10.7 14.4 -999.0 -999.0 11.7 -999.0 8.1 8.5 39.2 3.0 12.7

126 2009 May 6 -999.0 -777.0 -999.0 14.4 10.6 8.2 -999.0 10.9 7.8 -999.0 -999.0 13.9 -777.0 1.5 25.7 9.5 0.6 8.3

127 2009 May 7 -999.0 16.0 -999.0 9.6 7.4 5.3 -999.0 7.4 3.8 -999.0 -999.0 6.9 5.0 2.1 6.8 6.3 1.2 5.2

128 2009 May 8 -999.0 10.6 -999.0 4.4 1.5 1.8 -999.0 3.9 2.3 -999.0 -999.0 5.8 2.2 2.9 5.4 3.2 2.3 1.8

129 2009 May 9 -999.0 12.1 -999.0 5.1 2.1 2.2 -999.0 4.7 2.9 -999.0 -999.0 4.7 2.4 2.8 2.4 2.8 2.0 2.3

130 2009 May 10 -999.0 10.4 -999.0 5.7 1.7 2.0 -999.0 4.1 2.8 -999.0 -999.0 4.2 2.2 2.0 2.6 6.6 1.5 2.0

131 2009 May 11 -999.0 11.2 -999.0 6.7 2.6 3.1 -999.0 8.7 3.6 -999.0 -999.0 5.0 2.9 -777.0 2.2 7.2 1.7 3.2

132 2009 May 12 -999.0 6.4 -999.0 3.1 1.4 1.9 -999.0 3.7 1.8 -999.0 -999.0 6.4 2.4 0.2 3.0 3.7 1.1 2.0

133 2009 May 13 -999.0 9.4 -999.0 4.6 1.9 1.0 -999.0 3.2 1.7 -999.0 -999.0 3.9 1.2 0.6 2.3 4.6 1.2 1.0

134 2009 May 14 -999.0 12.2 -999.0 8.1 5.6 6.6 -999.0 7.7 6.1 -999.0 -999.0 4.2 4.7 4.0 2.2 6.4 5.1 6.8

135 2009 May 15 -999.0 11.4 -999.0 5.3 2.7 3.9 -999.0 4.5 3.6 -999.0 -999.0 5.9 2.9 2.0 6.2 3.9 1.8 4.0

A-35

JD Y M D EDMC TEOM3

EDMC FDMS

EDME TEOM3

EDME FDMS

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS TEOM3

EDMS TEOM4

EDMS FDMS

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

136 2009 May 16 -999.0 11.4 -999.0 6.5 3.2 5.6 -999.0 5.8 4.9 -999.0 -999.0 6.2 4.0 3.4 3.4 4.3 4.0 5.7

137 2009 May 17 -999.0 11.1 -999.0 5.1 4.2 4.0 -999.0 5.7 5.0 -999.0 -999.0 6.6 4.0 3.3 3.1 5.1 4.0 4.0

138 2009 May 18 -999.0 12.9 -999.0 7.2 4.7 3.6 -999.0 3.8 4.7 -999.0 -999.0 6.0 3.7 3.3 5.0 5.4 4.3 3.7

139 2009 May 19 -999.0 8.9 -999.0 5.8 6.1 2.9 -999.0 3.7 3.1 -999.0 -999.0 4.8 2.7 2.8 2.7 4.0 3.1 3.0

140 2009 May 20 -999.0 6.3 -999.0 3.6 1.3 1.8 -999.0 4.3 2.1 -999.0 -999.0 4.8 1.6 2.0 0.9 1.9 2.1 1.8

141 2009 May 21 -999.0 8.1 -999.0 5.5 3.7 4.3 -999.0 6.6 3.9 -999.0 -999.0 5.9 3.2 2.5 2.8 3.7 2.4 4.4

142 2009 May 22 -999.0 12.7 -999.0 6.8 3.2 4.1 -999.0 8.9 5.0 -999.0 -999.0 5.4 3.6 2.5 4.7 4.2 2.6 4.1

143 2009 May 23 -999.0 14.8 -999.0 8.0 2.2 4.8 -999.0 12.2 8.3 -999.0 -999.0 7.1 5.6 5.0 3.8 9.3 5.1 4.9

144 2009 May 24 -999.0 16.2 -999.0 9.6 5.2 8.4 -999.0 14.9 11.4 -999.0 -999.0 9.2 8.0 5.5 5.7 11.0 5.1 8.5

145 2009 May 25 -999.0 12.8 -999.0 9.4 2.8 4.0 -999.0 14.2 5.0 -999.0 -999.0 6.7 3.4 8.0 3.4 5.0 5.8 4.1

146 2009 May 26 -999.0 12.4 -999.0 -777.0 5.4 5.9 -999.0 12.0 8.0 -999.0 -999.0 8.1 5.4 5.1 6.5 5.7 5.3 6.0

147 2009 May 27 -999.0 -777.0 -999.0 -999.0 1.6 2.2 -999.0 8.8 2.1 -999.0 -999.0 4.5 1.3 2.0 1.5 2.0 1.3 2.4

148 2009 May 28 -999.0 -999.0 -999.0 -999.0 1.8 3.3 -999.0 8.9 4.2 -999.0 -999.0 5.8 3.2 3.4 3.6 3.1 3.1 3.3

149 2009 May 29 -999.0 -999.0 -999.0 -999.0 3.0 5.5 -999.0 10.5 6.8 -999.0 -999.0 5.6 3.3 5.3 2.9 4.5 6.2 5.5

150 2009 May 30 -999.0 -999.0 -999.0 -999.0 1.7 2.9 -999.0 7.8 4.8 -999.0 -999.0 5.8 3.1 4.8 3.0 3.2 4.5 3.0

151 2009 May 31 -999.0 -999.0 -999.0 -999.0 4.6 6.8 -999.0 6.3 7.0 -999.0 -999.0 5.2 2.8 3.3 2.3 3.8 6.0 6.7

152 2009 Jun 1 -999.0 -999.0 -999.0 -999.0 2.2 3.3 -999.0 6.9 3.9 -999.0 -999.0 6.3 3.6 3.7 3.5 4.5 2.8 3.4

153 2009 Jun 2 -999.0 -999.0 -999.0 -999.0 6.1 6.4 -999.0 11.4 8.5 -999.0 -999.0 8.8 6.3 9.4 5.6 9.9 6.1 6.5

154 2009 Jun 3 -999.0 -999.0 -999.0 -999.0 9.5 10.7 -999.0 15.7 11.2 -999.0 -999.0 10.2 8.4 12.3 7.1 20.8 7.0 10.7

155 2009 Jun 4 -999.0 -999.0 -999.0 -999.0 3.5 3.5 -999.0 13.1 6.0 -999.0 -999.0 5.9 2.9 7.7 4.2 18.8 5.3 3.6

156 2009 Jun 5 -999.0 -999.0 -999.0 -999.0 2.5 1.8 -999.0 5.2 3.5 -999.0 -999.0 5.3 3.3 3.1 4.9 12.4 2.6 1.9

157 2009 Jun 6 -999.0 -999.0 -999.0 -999.0 3.5 2.7 -999.0 6.6 7.1 -999.0 -999.0 5.2 2.3 3.7 3.3 10.1 5.3 2.7

158 2009 Jun 7 -999.0 -999.0 -999.0 -999.0 2.1 2.2 -999.0 5.9 4.3 -999.0 -999.0 5.1 2.3 3.0 1.6 8.6 3.6 2.2

159 2009 Jun 8 -999.0 -999.0 -999.0 -999.0 2.1 1.5 -999.0 8.4 4.0 -999.0 -999.0 5.4 2.4 3.2 2.6 7.4 2.8 1.5

160 2009 Jun 9 -999.0 -999.0 -999.0 -777.0 1.8 1.7 -999.0 10.7 1.9 -999.0 -999.0 5.0 1.5 2.4 1.7 5.0 2.4 1.8

161 2009 Jun 10 -999.0 -999.0 -999.0 10.5 4.6 5.3 -999.0 8.7 5.0 -999.0 -999.0 8.0 4.5 5.0 5.2 9.2 4.8 5.4

162 2009 Jun 11 -999.0 -999.0 -999.0 15.8 10.4 10.1 -999.0 14.0 12.2 -999.0 -999.0 12.3 9.5 9.2 8.4 13.7 8.1 10.1

163 2009 Jun 12 -999.0 -999.0 -999.0 16.5 11.5 12.4 -999.0 14.0 11.9 -999.0 -999.0 14.6 10.7 10.5 15.2 15.3 9.2 12.5

A-36

JD Y M D EDMC TEOM3

EDMC FDMS

EDME TEOM3

EDME FDMS

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS TEOM3

EDMS TEOM4

EDMS FDMS

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

164 2009 Jun 13 -999.0 -999.0 -999.0 15.8 11.1 10.9 -999.0 13.0 10.8 -999.0 -999.0 12.2 11.6 7.8 10.9 20.2 9.8 11.1

165 2009 Jun 14 -999.0 -999.0 -999.0 16.5 10.9 11.3 -999.0 11.5 11.5 -999.0 -999.0 11.7 9.5 8.2 11.6 13.0 8.0 11.4

166 2009 Jun 15 -999.0 -999.0 -999.0 14.7 10.5 8.0 -999.0 12.3 8.5 -999.0 -999.0 8.3 8.4 6.3 13.3 10.8 7.4 8.0

167 2009 Jun 16 -999.0 -999.0 -999.0 12.6 6.0 6.2 -999.0 15.2 5.4 -999.0 -999.0 7.3 5.0 4.3 7.6 5.2 4.0 6.2

168 2009 Jun 17 -999.0 -999.0 -999.0 15.8 9.8 8.6 -999.0 11.9 9.2 -999.0 -999.0 9.3 9.5 5.1 10.6 9.2 5.1 8.5

169 2009 Jun 18 -999.0 -999.0 -999.0 8.9 5.1 3.7 -999.0 11.4 3.5 -999.0 -999.0 6.6 3.0 3.0 3.8 3.0 1.3 3.8

170 2009 Jun 19 -999.0 -999.0 -999.0 11.7 5.8 5.6 -999.0 12.1 5.7 -999.0 -999.0 5.8 3.8 3.8 4.5 2.9 3.6 5.6

171 2009 Jun 20 -999.0 -999.0 -999.0 11.4 6.5 7.0 -999.0 12.0 6.1 -999.0 -999.0 7.5 6.8 5.0 4.4 8.7 4.3 7.0

172 2009 Jun 21 -999.0 -999.0 -999.0 10.0 5.9 5.1 -999.0 11.1 4.4 -999.0 -999.0 5.9 4.1 4.1 4.0 4.2 3.0 5.1

173 2009 Jun 22 -999.0 -999.0 -999.0 7.9 6.0 4.2 -999.0 7.9 2.4 -999.0 -999.0 5.9 3.2 2.5 3.9 3.8 1.6 4.4

174 2009 Jun 23 -999.0 -999.0 -999.0 9.2 5.5 4.4 -999.0 11.0 2.5 -999.0 -999.0 5.6 2.4 1.5 1.4 3.5 0.8 4.4

175 2009 Jun 24 -999.0 -777.0 -999.0 12.8 4.5 4.8 -999.0 7.8 2.6 -999.0 -999.0 6.6 3.5 2.2 4.6 5.5 3.1 4.9

176 2009 Jun 25 -999.0 6.3 -999.0 13.4 7.3 5.4 -999.0 9.1 -777.0 -999.0 -999.0 6.4 3.0 3.4 3.8 6.6 2.2 5.4

177 2009 Jun 26 -999.0 4.1 -999.0 7.8 8.5 3.8 -999.0 11.9 2.4 -999.0 -999.0 5.8 2.3 2.3 1.6 4.3 2.3 3.9

178 2009 Jun 27 -999.0 4.9 -999.0 5.1 3.3 4.0 -999.0 6.0 3.4 -999.0 -999.0 6.8 2.7 3.9 2.2 4.5 4.4 3.9

179 2009 Jun 28 -999.0 4.3 -999.0 6.8 3.5 3.0 -999.0 7.8 3.0 -999.0 -999.0 4.8 2.3 5.1 2.3 3.7 4.3 3.0

180 2009 Jun 29 -999.0 4.4 -999.0 6.9 4.1 4.0 -999.0 9.4 3.5 -999.0 -999.0 5.8 3.3 2.6 2.9 3.6 2.5 4.1

181 2009 Jun 30 -999.0 5.7 -999.0 6.7 3.8 5.1 -999.0 9.9 3.8 -999.0 -999.0 4.7 2.7 2.3 1.9 3.7 1.8 5.2

182 2009 Jul 1 -999.0 5.0 -999.0 5.2 2.5 8.5 -999.0 5.6 -999.0 3.9 -999.0 5.5 3.3 2.4 2.0 3.2 2.6 8.5

183 2009 Jul 2 -999.0 6.4 -999.0 7.2 5.4 5.5 -999.0 10.0 -999.0 4.6 -999.0 6.3 3.5 3.1 2.2 3.8 2.9 5.5

184 2009 Jul 3 -999.0 6.7 -999.0 8.3 5.3 5.0 -999.0 -777.0 -999.0 6.1 -999.0 5.3 3.7 3.8 1.6 2.9 3.3 5.1

185 2009 Jul 4 -999.0 10.2 -999.0 9.5 6.4 5.8 -999.0 -999.0 -999.0 7.7 -999.0 6.9 4.3 5.8 3.2 5.0 5.3 5.8

186 2009 Jul 5 -999.0 8.6 -999.0 8.2 8.5 4.4 -999.0 -999.0 -999.0 6.2 -999.0 6.5 3.8 3.9 2.8 6.8 5.9 4.5

187 2009 Jul 6 -999.0 6.4 -999.0 6.7 4.7 2.9 -999.0 -999.0 -999.0 3.5 -999.0 4.9 2.8 3.0 2.1 7.3 4.2 2.9

188 2009 Jul 7 -999.0 6.8 -999.0 6.8 5.3 2.8 -999.0 -999.0 -999.0 3.3 -999.0 4.9 3.0 1.0 1.8 12.1 1.4 2.8

189 2009 Jul 8 -999.0 5.2 -999.0 6.0 5.9 3.1 -999.0 -999.0 -999.0 2.5 -999.0 5.0 2.7 0.6 5.7 7.6 2.1 3.1

190 2009 Jul 9 -999.0 4.4 -999.0 4.8 5.1 3.4 -999.0 -999.0 -999.0 2.9 -999.0 4.0 2.2 2.0 1.9 2.3 1.7 3.5

191 2009 Jul 10 -999.0 7.0 -999.0 7.0 3.3 3.5 -999.0 -999.0 -999.0 4.1 -999.0 8.6 4.2 2.0 3.5 5.5 3.0 3.5

A-37

JD Y M D EDMC TEOM3

EDMC FDMS

EDME TEOM3

EDME FDMS

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS TEOM3

EDMS TEOM4

EDMS FDMS

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

192 2009 Jul 11 -999.0 9.7 -999.0 9.1 6.7 6.3 -999.0 -999.0 -999.0 7.3 -999.0 5.9 3.8 6.0 1.9 6.9 6.3 6.3

193 2009 Jul 12 -999.0 6.7 -999.0 7.4 2.6 3.6 -999.0 -999.0 -999.0 4.3 -999.0 7.1 4.1 5.2 2.3 5.4 6.5 3.7

194 2009 Jul 13 -999.0 7.8 -999.0 8.5 2.9 5.5 -999.0 -999.0 -999.0 5.5 -999.0 7.6 3.7 3.6 3.5 5.1 4.3 5.6

195 2009 Jul 14 -999.0 2.7 -999.0 3.0 0.1 1.1 -999.0 -999.0 -999.0 0.8 -999.0 2.7 0.7 0.6 0.5 1.3 0.3 1.1

196 2009 Jul 15 -999.0 6.5 -999.0 6.5 1.7 3.7 -999.0 -999.0 -999.0 4.1 -999.0 6.9 3.7 2.0 1.2 4.6 2.6 3.8

197 2009 Jul 16 -999.0 8.7 -999.0 8.7 -777.0 5.8 -999.0 -999.0 -999.0 6.1 -999.0 8.8 6.4 4.6 3.6 7.4 4.4 5.9

198 2009 Jul 17 -999.0 9.0 -999.0 9.3 4.0 6.5 -999.0 -999.0 -999.0 6.3 -999.0 8.6 4.5 3.9 4.3 6.0 4.4 6.5

199 2009 Jul 18 -999.0 12.4 -999.0 14.3 10.7 10.7 -999.0 -999.0 -999.0 13.2 -999.0 12.0 7.2 6.4 6.1 8.7 7.9 10.8

200 2009 Jul 19 -999.0 2.9 -999.0 -999.0 2.6 1.5 -999.0 -999.0 -999.0 2.2 -999.0 -777.0 1.3 0.9 0.6 1.5 0.7 1.6

201 2009 Jul 20 -999.0 5.6 -999.0 -777.0 2.7 4.2 -999.0 -999.0 -999.0 4.1 -999.0 5.4 2.4 2.8 2.0 2.5 2.5 4.3

202 2009 Jul 21 -999.0 10.7 -999.0 12.4 5.2 7.4 -999.0 -999.0 -999.0 7.3 -999.0 10.2 6.6 5.5 5.9 5.5 6.4 7.5

203 2009 Jul 22 -999.0 9.8 -999.0 11.6 7.6 8.9 -999.0 -999.0 -999.0 8.6 -999.0 11.2 7.9 8.1 5.2 7.8 6.3 9.1

204 2009 Jul 23 -999.0 7.8 -999.0 -777.0 5.7 7.7 -999.0 -999.0 -999.0 7.1 -999.0 7.7 5.0 6.1 4.4 5.7 5.2 7.8

205 2009 Jul 24 -999.0 11.7 -999.0 15.5 9.0 11.7 -999.0 -999.0 -999.0 13.8 -999.0 12.0 8.9 10.4 7.9 9.5 9.2 11.7

206 2009 Jul 25 -999.0 12.8 -999.0 15.8 9.5 14.3 -999.0 -999.0 -999.0 16.3 -999.0 12.1 10.8 12.0 6.6 11.5 8.1 14.4

207 2009 Jul 26 -999.0 7.5 -999.0 8.1 5.9 5.4 -999.0 -999.0 -999.0 3.9 -999.0 6.4 5.2 3.5 3.7 3.5 2.4 5.3

208 2009 Jul 27 -999.0 10.7 -999.0 13.6 6.6 8.0 -999.0 -999.0 -999.0 9.0 -999.0 7.6 7.0 6.5 5.7 7.6 4.8 8.2

209 2009 Jul 28 -999.0 10.3 -999.0 10.3 9.0 9.8 -999.0 -999.0 -999.0 9.1 -999.0 10.6 7.4 5.8 7.0 8.3 5.5 9.8

210 2009 Jul 29 -999.0 15.8 -999.0 -777.0 12.9 15.5 -999.0 -999.0 -999.0 13.5 -999.0 9.7 9.4 12.0 9.7 9.9 13.5 15.5

211 2009 Jul 30 -999.0 20.5 -999.0 19.5 17.2 16.7 -999.0 -999.0 -999.0 18.6 -999.0 16.5 15.6 18.2 14.0 14.6 17.0 16.8

212 2009 Jul 31 -999.0 14.1 -999.0 14.0 10.8 11.8 -999.0 -999.0 -999.0 11.5 -999.0 8.7 7.5 7.9 7.2 6.3 8.0 11.9

213 2009 Aug 1 -999.0 15.0 -999.0 16.2 15.2 13.3 -999.0 -999.0 -999.0 13.8 -999.0 14.1 12.5 12.8 11.8 13.3 11.5 13.3

214 2009 Aug 2 -999.0 15.2 -999.0 18.5 13.7 12.9 -999.0 -999.0 -999.0 12.8 -999.0 11.3 12.0 14.6 11.9 10.9 10.9 13.0

215 2009 Aug 3 -999.0 5.9 -999.0 7.3 7.5 4.9 -999.0 -999.0 -999.0 5.2 -999.0 4.2 5.0 8.3 1.7 3.4 3.5 4.9

216 2009 Aug 4 -999.0 4.3 -999.0 5.3 3.9 3.1 -999.0 -999.0 -999.0 3.6 -999.0 4.2 3.6 1.7 1.9 3.1 1.8 3.0

217 2009 Aug 5 -999.0 7.0 -999.0 6.3 4.9 2.7 -999.0 -999.0 -999.0 5.7 -999.0 5.1 3.5 3.3 1.5 3.2 4.1 2.7

218 2009 Aug 6 -999.0 11.0 -999.0 10.8 9.1 7.4 -999.0 -999.0 -999.0 8.9 -999.0 7.6 6.6 5.9 3.5 7.4 5.4 7.5

219 2009 Aug 7 -999.0 14.0 -999.0 14.0 10.3 11.1 -999.0 -999.0 -999.0 12.0 -999.0 9.0 7.5 6.6 5.7 7.0 8.6 11.2

A-38

JD Y M D EDMC TEOM3

EDMC FDMS

EDME TEOM3

EDME FDMS

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS TEOM3

EDMS TEOM4

EDMS FDMS

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

220 2009 Aug 8 -999.0 13.0 -999.0 10.1 7.6 8.8 -999.0 -999.0 -999.0 9.8 -999.0 8.8 7.9 8.7 5.4 7.0 11.1 8.8

221 2009 Aug 9 -999.0 10.0 -999.0 11.2 7.2 9.3 -999.0 -999.0 -999.0 9.5 -999.0 11.5 8.7 8.8 7.3 8.0 6.4 9.3

222 2009 Aug 10 -999.0 17.5 -999.0 18.8 14.5 15.6 -999.0 -999.0 -999.0 16.3 -999.0 16.2 14.7 12.4 12.1 15.5 9.8 15.7

223 2009 Aug 11 -999.0 9.6 -999.0 11.8 7.7 7.0 -999.0 -999.0 -999.0 7.2 -999.0 9.4 6.0 3.8 7.0 5.9 2.8 7.1

224 2009 Aug 12 -999.0 5.4 -999.0 8.7 4.4 3.8 -999.0 -999.0 -999.0 -777.0 -999.0 5.6 2.8 3.0 2.2 3.0 1.9 3.8

225 2009 Aug 13 -999.0 3.0 -999.0 4.2 2.6 2.2 -999.0 -999.0 -999.0 1.7 -999.0 4.3 1.7 2.5 1.6 1.7 0.9 2.3

226 2009 Aug 14 -999.0 5.6 -999.0 5.7 4.3 2.5 -999.0 -999.0 -999.0 3.9 -999.0 3.3 2.2 2.1 0.4 3.6 2.0 2.5

227 2009 Aug 15 -999.0 9.4 -999.0 10.8 8.5 6.6 -999.0 -999.0 -999.0 5.8 -999.0 4.6 6.3 4.6 1.4 4.2 3.7 6.6

228 2009 Aug 16 -999.0 7.1 -999.0 8.2 5.3 4.8 -999.0 -999.0 -999.0 5.4 -999.0 6.5 4.1 6.0 4.0 4.3 2.2 5.0

229 2009 Aug 17 -999.0 6.6 -999.0 9.8 3.4 3.9 -999.0 -999.0 -999.0 4.3 -999.0 6.8 5.1 3.3 3.0 5.4 2.9 4.0

230 2009 Aug 18 -999.0 5.5 -999.0 6.8 3.9 -777.0 -999.0 -999.0 -999.0 3.3 -999.0 5.6 2.4 3.2 1.5 2.4 2.0 -

777.0

231 2009 Aug 19 -999.0 4.6 -999.0 5.6 2.8 3.0 -999.0 -999.0 -999.0 2.9 -999.0 4.5 2.2 1.8 0.3 2.4 -777.0 3.1

232 2009 Aug 20 -999.0 7.0 -999.0 6.2 3.7 3.8 -999.0 -999.0 -999.0 4.3 -999.0 5.3 2.8 4.2 1.0 5.3 -999.0 3.8

233 2009 Aug 21 -999.0 7.5 -999.0 8.4 4.0 5.1 -999.0 -999.0 -999.0 6.1 -999.0 8.2 5.0 4.8 2.8 6.8 -777.0 5.1

234 2009 Aug 22 -999.0 14.0 -999.0 14.9 9.1 11.3 -999.0 -999.0 -999.0 12.7 -999.0 11.6 9.5 11.0 8.2 9.0 9.7 11.4

235 2009 Aug 23 -999.0 11.1 -999.0 12.8 9.3 8.7 -999.0 -999.0 -999.0 9.9 -999.0 8.0 6.2 8.3 4.5 7.1 6.9 8.7

236 2009 Aug 24 -999.0 -777.0 -999.0 7.7 4.5 4.2 -999.0 -999.0 -999.0 4.3 -999.0 6.3 3.2 3.7 3.6 3.0 3.8 4.2

237 2009 Aug 25 -999.0 11.1 -999.0 9.4 6.3 7.1 -999.0 -999.0 -999.0 7.3 -999.0 8.4 6.4 7.6 3.4 8.5 8.1 7.2

238 2009 Aug 26 -999.0 -777.0 -999.0 10.8 8.9 7.7 -999.0 -999.0 -999.0 6.7 -999.0 7.9 5.1 6.6 2.7 5.4 7.7 7.7

239 2009 Aug 27 -999.0 16.9 -999.0 16.3 9.2 11.1 -999.0 -999.0 -999.0 11.4 -999.0 9.3 7.8 9.9 -777.0 10.3 9.4 11.3

240 2009 Aug 28 -999.0 17.9 -999.0 13.7 8.9 10.1 -999.0 -999.0 -999.0 11.2 -999.0 11.0 9.0 15.9 -999.0 10.0 12.9 10.1

241 2009 Aug 29 -999.0 13.9 -999.0 9.8 6.1 7.8 -999.0 -999.0 -999.0 8.5 -999.0 10.6 6.3 9.5 -999.0 6.9 7.7 7.9

242 2009 Aug 30 -999.0 13.4 -999.0 8.1 5.6 6.3 -999.0 -999.0 -999.0 7.2 -999.0 7.8 5.5 8.5 -999.0 5.5 9.3 6.4

243 2009 Aug 31 -999.0 17.9 -999.0 12.6 7.7 11.4 -999.0 -999.0 -999.0 11.7 -999.0 10.1 9.3 11.2 -999.0 9.3 11.5 11.5

244 2009 Sep 1 -999.0 18.9 -999.0 21.7 13.2 14.9 -999.0 -999.0 -999.0 14.7 -999.0 12.9 11.2 15.1 -999.0 11.8 12.8 15.0

245 2009 Sep 2 -999.0 20.1 -999.0 18.4 11.9 -777.0 -777.0 -999.0 -999.0 14.7 -999.0 16.1 12.8 14.1 -999.0 14.4 12.5 -

777.0

246 2009 Sep 3 -999.0 17.7 -999.0 -777.0 11.9 13.1 17.2 -999.0 -999.0 13.5 -999.0 11.3 9.4 10.7 -999.0 10.5 8.2 13.4

A-39

JD Y M D EDMC TEOM3

EDMC FDMS

EDME TEOM3

EDME FDMS

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS TEOM3

EDMS TEOM4

EDMS FDMS

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

247 2009 Sep 4 -999.0 9.2 -999.0 5.6 4.9 4.4 5.4 -999.0 -999.0 3.3 -999.0 5.1 3.3 2.0 -999.0 3.5 1.6 4.4

248 2009 Sep 5 -999.0 8.6 -999.0 5.8 2.8 2.4 5.3 -999.0 -999.0 2.0 -999.0 5.0 2.0 1.8 -999.0 7.8 2.4 2.4

249 2009 Sep 6 -999.0 9.3 -999.0 6.7 4.8 3.4 6.8 -999.0 -999.0 4.0 -999.0 7.8 5.0 4.0 -999.0 7.8 3.8 3.3

250 2009 Sep 7 -999.0 7.5 -999.0 3.1 2.9 1.4 3.5 -999.0 -999.0 0.8 -999.0 2.9 1.1 0.9 -999.0 1.8 0.7 1.4

251 2009 Sep 8 -999.0 11.2 -999.0 6.0 2.4 2.7 4.7 -999.0 -999.0 1.5 -999.0 4.2 1.7 1.1 -999.0 1.9 0.6 2.7

252 2009 Sep 9 -999.0 9.4 -999.0 7.5 3.5 4.8 7.0 -999.0 -999.0 4.3 -999.0 5.9 4.1 4.6 -999.0 3.0 2.5 4.8

253 2009 Sep 10 -999.0 5.4 -999.0 5.7 4.0 4.8 6.8 -999.0 -999.0 4.2 -999.0 5.6 2.7 4.7 -999.0 2.5 1.8 4.9

254 2009 Sep 11 -999.0 9.4 -999.0 6.7 4.2 4.4 7.8 -999.0 -999.0 4.3 -999.0 5.8 5.0 4.6 -999.0 3.1 2.8 4.5

255 2009 Sep 12 -999.0 6.6 -999.0 5.4 -777.0 3.4 6.9 -999.0 -999.0 3.3 -999.0 6.5 2.0 5.0 -999.0 3.1 5.3 3.3

256 2009 Sep 13 -999.0 6.5 -999.0 5.4 -999.0 4.3 7.2 -999.0 -999.0 4.3 -999.0 7.7 3.2 5.5 -999.0 3.9 5.9 4.3

257 2009 Sep 14 -999.0 7.4 -999.0 6.2 -999.0 5.2 7.1 -999.0 -999.0 5.5 -999.0 7.9 8.4 7.2 -999.0 8.6 6.1 5.3

258 2009 Sep 15 -999.0 15.0 -999.0 15.0 -999.0 12.7 14.9 -999.0 -999.0 13.0 -999.0 12.6 10.0 9.3 -999.0 12.2 7.3 12.7

259 2009 Sep 16 -999.0 15.5 -999.0 -777.0 -777.0 11.6 15.5 -999.0 -999.0 11.5 -999.0 11.8 8.1 14.9 -999.0 13.4 11.4 11.7

260 2009 Sep 17 -999.0 8.6 -999.0 8.5 7.6 6.8 9.1 -999.0 -999.0 5.1 -999.0 8.1 5.2 6.3 -999.0 6.6 4.5 6.7

261 2009 Sep 18 -999.0 6.9 -999.0 5.6 3.4 2.5 5.9 -999.0 -999.0 2.4 -999.0 6.0 1.6 2.0 -999.0 1.7 1.3 2.5

262 2009 Sep 19 -999.0 6.4 -999.0 5.9 4.3 4.2 7.2 -999.0 -999.0 3.7 -999.0 6.3 2.9 4.6 -999.0 4.9 1.4 4.3

263 2009 Sep 20 -999.0 3.7 -999.0 3.2 2.3 2.2 4.1 -999.0 -999.0 2.6 -999.0 4.7 1.2 1.3 -999.0 1.0 0.5 2.3

264 2009 Sep 21 -999.0 4.9 -999.0 6.6 2.5 3.5 6.1 -999.0 -999.0 3.4 -999.0 6.0 3.3 2.4 -999.0 3.9 1.6 3.5

265 2009 Sep 22 -999.0 9.3 -999.0 -777.0 4.8 7.2 10.8 -999.0 -999.0 7.8 -999.0 6.5 4.8 4.9 -999.0 6.5 2.3 7.3

266 2009 Sep 23 -999.0 14.8 -999.0 -777.0 7.6 12.1 14.3 -999.0 -999.0 11.5 -999.0 10.3 5.9 17.3 -999.0 9.8 4.8 12.2

267 2009 Sep 24 -999.0 28.2 -999.0 -777.0 14.9 20.8 21.6 -999.0 -999.0 20.4 -999.0 24.1 22.2 17.4 -999.0 23.4 11.4 20.8

268 2009 Sep 25 -999.0 12.5 -999.0 14.1 6.6 9.4 13.6 -999.0 -999.0 10.4 -999.0 7.5 6.9 7.8 -999.0 7.4 4.9 9.3

269 2009 Sep 26 -999.0 6.8 -999.0 6.0 4.3 4.4 7.0 -999.0 -999.0 3.5 -999.0 5.0 2.6 2.9 -999.0 2.1 2.7 4.4

270 2009 Sep 27 -999.0 9.4 -999.0 3.6 2.0 1.6 3.8 -999.0 -999.0 2.5 -999.0 3.7 0.8 1.7 -999.0 1.9 0.6 1.6

271 2009 Sep 28 -999.0 15.2 -999.0 5.8 2.0 0.9 4.6 -999.0 -999.0 2.2 -999.0 4.3 1.3 2.0 -999.0 1.8 1.7 0.9

272 2009 Sep 29 -999.0 8.6 -999.0 6.7 3.4 4.8 7.0 -999.0 -999.0 5.1 -999.0 7.2 3.3 3.6 -999.0 3.3 2.9 4.7

273 2009 Sep 30 -999.0 6.2 -999.0 3.9 2.3 2.6 5.8 -999.0 -999.0 2.6 -999.0 4.9 1.4 2.5 -999.0 1.6 1.7 2.7

274 2009 Oct 1 -999.0 8.7 -999.0 7.5 3.6 4.0 7.8 -999.0 -999.0 -999.0 4.4 8.4 3.2 3.8 -999.0 4.9 2.7 3.8

A-40

JD Y M D EDMC TEOM3

EDMC FDMS

EDME TEOM3

EDME FDMS

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS TEOM3

EDMS TEOM4

EDMS FDMS

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

275 2009 Oct 2 -999.0 9.7 -999.0 7.7 2.3 3.9 6.7 -999.0 -999.0 -999.0 6.1 8.1 5.4 11.4 -999.0 11.2 4.7 3.9

276 2009 Oct 3 -999.0 9.1 -999.0 9.3 7.4 6.9 9.1 -999.0 -999.0 -999.0 6.3 5.5 5.0 5.1 -999.0 3.7 3.8 6.9

277 2009 Oct 4 -999.0 5.3 -999.0 4.5 2.7 2.0 5.1 -999.0 -999.0 -999.0 2.8 4.0 0.9 2.0 -999.0 1.4 0.6 2.0

278 2009 Oct 5 -999.0 8.8 -999.0 7.4 2.6 2.9 7.9 -999.0 -999.0 -999.0 3.3 4.2 1.6 1.2 -999.0 1.9 0.4 3.0

279 2009 Oct 6 -999.0 4.6 -999.0 6.4 2.6 2.8 5.5 -999.0 -999.0 -999.0 2.7 5.0 2.4 2.0 -999.0 1.8 1.5 2.8

280 2009 Oct 7 -999.0 2.9 -999.0 2.6 0.5 1.0 3.0 -999.0 -999.0 -999.0 1.1 2.2 0.1 0.7 -999.0 0.1 0.3 1.0

281 2009 Oct 8 -999.0 3.0 -999.0 3.0 -777.0 1.1 3.0 -999.0 -999.0 -999.0 1.2 -777.0 0.5 0.9 -999.0 0.5 0.6 1.3

282 2009 Oct 9 -999.0 2.9 -999.0 2.5 -999.0 1.2 2.9 -999.0 -999.0 -999.0 0.7 0.0 0.4 0.6 -999.0 0.1 0.1 1.3

283 2009 Oct 10 -999.0 2.0 -999.0 1.7 -999.0 0.4 2.2 -999.0 -999.0 -999.0 0.3 0.1 0.1 0.4 -999.0 0.0 0.2 0.4

284 2009 Oct 11 -999.0 5.2 -999.0 5.9 -999.0 2.3 4.4 -999.0 -999.0 -999.0 2.7 0.7 2.8 1.2 -999.0 1.3 0.9 2.3

285 2009 Oct 12 -999.0 7.9 -999.0 7.0 -999.0 2.8 7.0 -999.0 -999.0 -999.0 3.5 1.9 1.6 2.4 -999.0 3.0 2.0 2.8

286 2009 Oct 13 -999.0 6.1 -999.0 5.0 -999.0 1.5 5.3 -999.0 -999.0 -999.0 2.1 1.1 0.8 2.1 -999.0 1.6 2.2 1.5

287 2009 Oct 14 -999.0 7.6 -999.0 6.0 -999.0 3.3 6.3 -999.0 -999.0 -999.0 4.0 2.3 2.2 3.6 -999.0 2.3 3.4 3.4

288 2009 Oct 15 -999.0 13.5 -999.0 15.1 -999.0 7.7 13.9 -999.0 -999.0 -999.0 8.1 6.5 7.8 6.5 -999.0 7.7 3.3 7.8

289 2009 Oct 16 -999.0 13.6 -999.0 10.9 -999.0 4.3 12.3 -999.0 -999.0 -999.0 5.1 2.5 3.9 2.7 -777.0 4.2 1.3 4.4

290 2009 Oct 17 -999.0 6.0 -999.0 9.2 -999.0 5.0 8.4 -999.0 -999.0 -999.0 3.1 1.8 9.8 1.4 1.2 2.9 2.1 5.1

291 2009 Oct 18 -999.0 6.8 -999.0 7.4 -999.0 5.2 9.2 -999.0 -999.0 -999.0 5.2 2.4 3.2 1.9 1.1 4.5 2.0 5.3

292 2009 Oct 19 -999.0 15.6 -999.0 15.6 -999.0 6.4 16.5 -999.0 -999.0 -999.0 6.1 3.3 5.0 3.1 6.0 6.7 3.2 6.5

293 2009 Oct 20 -999.0 14.3 -999.0 16.2 -999.0 6.2 12.4 -999.0 -999.0 -999.0 8.2 3.5 9.5 2.7 7.9 5.1 2.5 6.3

294 2009 Oct 21 -999.0 13.2 -999.0 10.6 -999.0 4.5 11.8 -999.0 -999.0 -999.0 8.2 1.8 3.0 3.3 4.1 5.1 4.0 4.6

295 2009 Oct 22 -999.0 7.3 -999.0 8.6 -999.0 4.2 10.6 -999.0 -999.0 -999.0 3.9 2.5 2.7 2.1 2.6 3.0 2.0 4.2

296 2009 Oct 23 -999.0 8.1 -999.0 10.3 -999.0 3.2 8.4 -999.0 -999.0 -999.0 3.3 1.2 5.9 1.0 1.6 6.4 1.0 3.2

297 2009 Oct 24 -999.0 3.2 -999.0 3.7 -999.0 1.2 4.1 -999.0 -999.0 -999.0 1.6 2.0 1.4 1.5 1.3 1.8 1.5 1.3

298 2009 Oct 25 -999.0 4.0 -999.0 4.3 -999.0 0.5 3.5 -999.0 -999.0 -999.0 0.6 0.9 0.8 0.3 1.5 1.1 0.3 0.5

299 2009 Oct 26 -999.0 8.3 -999.0 9.8 -999.0 4.4 7.2 -999.0 -999.0 -999.0 3.2 2.4 3.0 1.5 2.4 4.0 1.6 4.3

300 2009 Oct 27 -999.0 3.5 -999.0 3.1 -999.0 1.9 4.3 -999.0 -999.0 -999.0 1.6 1.8 1.7 1.8 0.5 1.8 1.4 2.0

301 2009 Oct 28 -999.0 5.2 -999.0 3.8 -999.0 2.0 5.0 -999.0 -999.0 -999.0 1.8 2.0 0.8 1.7 0.4 1.3 2.1 2.0

302 2009 Oct 29 -999.0 9.8 -999.0 7.5 -999.0 1.6 7.3 -999.0 -999.0 -999.0 2.4 1.4 1.7 2.3 4.2 2.1 2.7 1.6

A-41

JD Y M D EDMC TEOM3

EDMC FDMS

EDME TEOM3

EDME FDMS

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS TEOM3

EDMS TEOM4

EDMS FDMS

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

303 2009 Oct 30 -999.0 11.7 -999.0 13.5 -999.0 5.5 11.7 -999.0 -999.0 -999.0 6.1 5.7 6.2 -777.0 9.7 6.5 2.8 5.5

304 2009 Oct 31 -999.0 8.4 -999.0 8.4 -999.0 3.2 9.1 -999.0 -999.0 -999.0 3.6 3.7 3.5 2.2 -999.0 4.1 2.1 3.3

305 2009 Nov 1 -999.0 5.0 -999.0 4.0 -999.0 3.3 5.9 -999.0 -999.0 -999.0 3.6 0.9 1.6 1.8 -999.0 1.5 1.5 3.4

306 2009 Nov 2 -999.0 11.0 -999.0 10.7 -777.0 6.3 10.1 -999.0 -999.0 -999.0 5.8 1.0 3.7 1.4 -999.0 4.0 1.8 6.3

307 2009 Nov 3 -999.0 9.7 -999.0 9.6 8.2 4.7 10.4 -999.0 -999.0 -999.0 3.8 2.0 3.5 2.0 -777.0 2.9 2.9 4.8

308 2009 Nov 4 -999.0 8.1 -999.0 6.8 4.0 1.8 7.3 -999.0 -999.0 -999.0 2.5 0.8 2.7 1.3 2.0 3.3 1.5 1.8

309 2009 Nov 5 -999.0 8.8 -999.0 12.1 4.0 2.5 8.1 -999.0 -999.0 -999.0 3.7 1.7 5.0 1.8 3.0 7.1 1.6 2.5

310 2009 Nov 6 -999.0 4.4 -999.0 6.5 2.6 3.4 4.9 -999.0 -999.0 -999.0 1.8 2.0 2.5 1.5 1.1 2.6 1.1 3.4

311 2009 Nov 7 -999.0 3.8 -999.0 5.7 3.6 3.8 6.4 -999.0 -999.0 -999.0 2.4 1.2 3.5 0.6 1.8 3.1 0.8 3.8

312 2009 Nov 8 -999.0 3.6 -999.0 5.0 3.5 3.6 5.2 -999.0 -999.0 -999.0 3.1 2.5 2.7 1.6 1.3 3.0 1.9 3.7

313 2009 Nov 9 -999.0 5.4 -999.0 4.6 1.8 1.1 3.9 -999.0 -999.0 -999.0 2.1 0.7 1.2 0.4 1.7 2.7 0.5 1.2

314 2009 Nov 10 -999.0 7.6 -999.0 11.2 4.8 6.7 9.1 -777.0 -999.0 -999.0 6.0 3.6 5.2 2.4 2.9 4.8 2.2 6.7

315 2009 Nov 11 -999.0 8.1 -999.0 9.7 5.7 5.3 9.1 4.3 -999.0 -999.0 5.0 2.9 4.9 2.1 4.5 6.5 2.0 5.3

316 2009 Nov 12 -999.0 3.8 -999.0 9.6 3.0 3.8 6.5 2.4 -999.0 -999.0 2.3 2.0 2.4 1.1 3.1 2.4 1.0 3.9

317 2009 Nov 13 -999.0 7.3 -999.0 11.8 3.9 7.2 7.8 3.7 -999.0 -999.0 5.5 5.1 6.4 3.3 3.8 7.4 3.2 7.3

318 2009 Nov 14 -999.0 3.9 -999.0 4.3 2.3 3.4 4.5 1.2 -999.0 -999.0 2.4 2.0 1.6 2.6 3.0 2.2 1.8 3.5

319 2009 Nov 15 -999.0 4.5 -999.0 5.5 3.1 4.5 5.7 1.4 -999.0 -999.0 3.7 2.6 3.1 3.4 2.8 4.3 2.5 4.5

320 2009 Nov 16 -999.0 3.9 -999.0 6.0 2.2 3.4 4.1 1.2 -999.0 -999.0 2.8 1.0 1.7 1.4 0.7 3.5 0.5 3.5

321 2009 Nov 17 -999.0 2.8 -999.0 7.1 1.9 3.5 4.1 1.0 -999.0 -999.0 -777.0 1.5 2.1 1.4 1.3 4.0 1.0 3.5

322 2009 Nov 18 -999.0 5.0 -999.0 6.7 2.5 3.9 5.7 1.7 -999.0 -999.0 -777.0 1.3 2.8 1.3 1.7 4.5 0.8 4.0

323 2009 Nov 19 -999.0 5.3 -999.0 8.1 3.2 4.7 6.4 2.8 -999.0 -999.0 9.1 1.4 3.7 3.3 3.2 4.1 1.0 4.8

324 2009 Nov 20 -999.0 7.0 -999.0 9.1 4.4 4.6 7.4 2.8 -999.0 -999.0 10.8 1.5 2.3 4.3 2.8 9.4 3.5 4.7

325 2009 Nov 21 -999.0 5.6 -999.0 6.9 5.3 4.8 7.2 1.9 -999.0 -999.0 10.2 4.0 3.8 3.9 4.4 2.9 3.2 4.8

326 2009 Nov 22 -999.0 9.7 -999.0 9.1 4.9 3.0 7.0 2.6 -999.0 -999.0 11.8 1.4 5.7 4.9 6.5 12.0 2.9 3.0

327 2009 Nov 23 -999.0 11.4 -999.0 15.5 10.0 6.6 13.5 8.4 -999.0 -999.0 17.0 2.9 5.5 4.0 5.8 5.7 3.3 6.7

328 2009 Nov 24 -999.0 5.4 -999.0 8.0 5.4 4.0 8.0 3.4 -999.0 -999.0 9.1 1.4 2.7 1.1 3.2 2.0 0.9 4.2

329 2009 Nov 25 -999.0 4.6 -999.0 6.0 2.4 2.0 5.7 2.7 -999.0 -999.0 9.5 0.5 1.5 1.6 2.6 3.8 2.4 2.0

330 2009 Nov 26 -999.0 8.0 -999.0 13.6 8.2 6.2 10.1 5.0 -999.0 -999.0 9.8 4.4 5.9 3.3 6.6 3.8 3.1 6.3

A-42

JD Y M D EDMC TEOM3

EDMC FDMS

EDME TEOM3

EDME FDMS

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS TEOM3

EDMS TEOM4

EDMS FDMS

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

331 2009 Nov 27 -999.0 5.0 -999.0 9.1 4.9 2.8 5.4 2.4 -999.0 -999.0 8.0 1.8 2.2 0.7 2.3 1.3 1.0 2.9

332 2009 Nov 28 -999.0 4.8 -999.0 7.9 4.7 4.3 7.2 5.3 -999.0 -999.0 7.4 2.1 2.8 0.5 4.0 2.5 2.0 4.2

333 2009 Nov 29 -999.0 3.6 -999.0 5.9 3.5 2.4 5.2 2.0 -999.0 -999.0 5.1 1.6 2.3 1.3 2.6 2.8 2.8 2.4

334 2009 Nov 30 -999.0 1.7 -999.0 4.1 2.6 1.4 3.0 1.0 -999.0 -999.0 -777.0 0.6 0.5 0.8 2.7 0.5 0.8 1.4

335 2009 Dec 1 -999.0 7.6 -999.0 5.2 2.7 3.4 4.9 3.0 -999.0 -999.0 4.7 1.5 2.1 2.1 2.6 2.0 2.4 3.4

336 2009 Dec 2 -999.0 20.1 -999.0 15.2 7.5 7.7 12.7 9.7 -999.0 -999.0 12.7 7.3 8.8 3.3 13.8 8.8 2.8 7.8

337 2009 Dec 3 -999.0 22.5 -999.0 16.6 9.2 6.8 17.9 16.2 -999.0 -999.0 20.4 4.4 4.1 4.6 15.3 6.3 3.0 6.8

338 2009 Dec 4 -999.0 19.6 -999.0 12.0 12.0 5.4 11.8 8.6 -999.0 -999.0 16.0 5.1 4.9 3.6 12.0 4.8 3.6 5.4

339 2009 Dec 5 -999.0 -999.0 -999.0 2.3 2.2 0.5 1.7 1.5 -999.0 -999.0 7.9 0.2 0.8 1.0 1.5 0.4 1.2 0.6

340 2009 Dec 6 -999.0 -999.0 -999.0 5.5 2.5 2.6 5.7 4.6 -999.0 -999.0 9.9 0.1 1.6 0.8 5.1 0.7 0.8 2.6

341 2009 Dec 7 -999.0 -999.0 -999.0 11.0 4.5 4.2 11.3 8.8 -999.0 -999.0 20.1 2.9 3.6 2.0 8.7 6.3 1.2 4.3

342 2009 Dec 8 -999.0 -777.0 -999.0 16.9 9.3 7.6 14.2 12.9 -999.0 -999.0 19.3 8.0 7.4 2.1 19.2 7.2 -777.0 7.5

343 2009 Dec 9 -999.0 -777.0 -999.0 10.9 8.0 5.4 11.1 7.9 -999.0 -999.0 13.0 4.1 5.0 1.3 8.9 4.3 -999.0 5.5

344 2009 Dec 10 -999.0 7.3 -999.0 6.3 4.0 1.9 5.0 3.6 -999.0 -999.0 3.6 0.9 1.5 0.5 2.0 1.1 -777.0 2.0

345 2009 Dec 11 -999.0 8.7 -999.0 7.6 7.9 7.3 10.4 8.7 -999.0 -999.0 9.0 1.1 2.2 3.4 1.7 3.4 3.1 7.3

346 2009 Dec 12 -999.0 7.6 -999.0 7.3 -777.0 6.4 7.9 7.9 -999.0 -999.0 9.2 4.0 5.8 3.3 7.0 4.4 2.9 6.5

347 2009 Dec 13 -999.0 13.0 -999.0 15.4 -777.0 7.7 10.9 10.3 -999.0 -999.0 16.8 5.0 6.4 5.5 18.4 6.0 3.2 7.7

348 2009 Dec 14 -999.0 11.0 -999.0 14.6 -777.0 7.3 10.9 11.2 -999.0 -999.0 14.2 3.8 9.7 4.1 12.3 9.6 3.5 7.3

349 2009 Dec 15 -999.0 15.9 -999.0 14.7 -999.0 5.9 11.4 10.4 -999.0 -999.0 14.0 3.4 10.1 8.5 7.8 11.5 4.8 5.9

350 2009 Dec 16 -999.0 29.6 -999.0 31.3 16.7 14.7 24.0 20.2 -999.0 -999.0 26.7 5.8 14.2 9.3 15.7 11.3 6.0 14.8

351 2009 Dec 17 -999.0 23.9 -999.0 27.0 24.0 13.9 23.6 18.7 -999.0 -999.0 26.4 9.7 15.6 2.2 18.7 13.7 3.9 14.0

352 2009 Dec 18 -999.0 19.6 -999.0 16.2 12.6 4.6 14.6 11.9 -999.0 -999.0 15.6 2.8 11.5 2.6 6.2 15.0 1.9 4.6

353 2009 Dec 19 -999.0 14.5 -999.0 11.6 11.3 5.0 10.2 7.8 -999.0 -999.0 9.0 3.9 3.2 1.5 6.4 2.7 2.2 5.0

354 2009 Dec 20 -999.0 14.1 -999.0 11.1 8.5 2.9 8.0 6.6 -999.0 -999.0 13.0 1.8 7.2 1.1 6.2 5.0 1.4 3.0

355 2009 Dec 21 -999.0 15.7 -999.0 -777.0 7.1 2.6 7.9 5.9 -999.0 -999.0 16.1 1.8 2.9 2.0 5.4 11.5 1.1 2.6

356 2009 Dec 22 -999.0 11.2 -999.0 7.7 7.6 4.9 8.5 7.5 -999.0 -999.0 10.2 3.4 4.4 3.7 4.2 5.2 3.1 5.0

357 2009 Dec 23 -999.0 10.9 -999.0 10.9 4.2 5.3 9.8 8.4 -999.0 -999.0 9.4 4.4 3.9 2.3 10.6 4.7 1.6 5.3

A-43

JD Y M D EDMC TEOM3

EDMC FDMS

EDME TEOM3

EDME FDMS

MCIN BAM

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS TEOM3

EDMS TEOM4

EDMS FDMS

ELKI TEOM4

FTSA TEOM4

GENE TEOM4

LAMA BAM1

REDW TEOM4

TOMA TEOM4

9601 TEOM

358 2009 Dec 24 -999.0 21.4 -999.0 18.0 12.1 10.0 17.6 15.0 -999.0 -999.0 18.4 2.9 11.8 1.6 6.6 12.2 1.7 9.9

359 2009 Dec 25 -999.0 17.8 -999.0 16.2 13.0 4.7 15.3 11.8 -999.0 -999.0 17.6 3.6 21.8 3.6 8.7 15.5 2.8 4.8

360 2009 Dec 26 -999.0 23.0 -999.0 19.9 20.3 6.5 20.8 17.5 -999.0 -999.0 24.6 3.7 8.2 1.5 8.0 7.1 2.0 6.7

361 2009 Dec 27 -999.0 40.9 -999.0 54.5 47.3 30.8 46.5 41.6 -999.0 -999.0 44.0 7.0 28.9 9.0 23.5 15.0 7.9 30.9

362 2009 Dec 28 -999.0 25.4 -999.0 24.0 24.7 8.0 25.2 20.7 -999.0 -999.0 30.2 4.6 4.1 17.0 17.2 6.1 14.2 8.2

363 2009 Dec 29 -999.0 14.3 -999.0 11.7 -777.0 5.3 12.6 14.0 -999.0 -999.0 16.6 5.8 3.7 7.0 12.3 3.5 9.8 5.3

364 2009 Dec 30 -999.0 5.4 -999.0 3.0 -999.0 1.3 3.1 2.8 -999.0 -999.0 4.3 0.9 0.8 0.9 1.0 0.8 0.6 1.3

365 2009 Dec 31 -999.0 14.8 -999.0 12.5 -999.0 7.1 10.8 9.9 -999.0 -999.0 12.7 1.5 5.6 1.6 3.8 5.8 1.0 7.2

A-44

Table A-14. Measured 24-Hour PM2.5 at monitoring sites in the Capital Region during 2010 with values above the CWS (> 30 µg/m3) highlighted in yellow, values above 20 µg/m3 highlighted in blue and PM2.5 episodes shown by the red boxes.

JD Yr M D EDMC FDMS

EDME FDMS

MCIN BAM1

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS FDMS

ELKI TEOM4

FTSA TEOM40

FTSA SHARP

GENE TEOM4

LAMO BAM1

REDW TEOM4

TOMA TEOM4

90601 SES

1 2010 Jan 1 6.7 5.9 -777.0 1.5 4.6 4.2 7.4 0.6 0.8 -999.0 2.0 3.3 5.2 2.1 0.8

2 2010 Jan 2 10.5 9.9 2.5 5.8 8.1 6.4 9.8 4.0 4.3 -999.0 5.8 4.1 3.4 5.6 4.2

3 2010 Jan 3 18.8 18.3 10.4 9.2 13.7 13.3 19.5 6.3 6.6 -999.0 6.0 12.0 6.2 5.5 6.6

4 2010 Jan 4 22.4 22.2 11.0 10.1 15.6 14.9 20.6 5.6 7.6 -999.0 8.1 10.0 5.5 6.5 7.6

5 2010 Jan 5 19.0 15.7 7.6 11.6 14.8 14.5 18.7 3.8 6.5 -999.0 11.8 7.6 5.9 10.5 6.6

6 2010 Jan 6 21.6 17.4 5.5 10.0 15.9 15.3 19.5 3.6 6.1 -999.0 10.4 8.0 8.3 8.3 6.2

7 2010 Jan 7 15.0 12.2 3.3 3.8 11.3 9.1 14.6 2.2 3.2 -999.0 7.0 9.7 8.9 6.3 3.3

8 2010 Jan 8 23.7 24.0 13.7 11.6 20.7 19.2 23.3 6.4 13.4 -999.0 3.2 13.3 16.8 6.0 13.5

9 2010 Jan 9 14.6 18.6 12.7 6.0 17.3 11.0 13.3 6.1 10.1 -999.0 1.2 9.6 9.9 2.8 10.2

10 2010 Jan 10 13.1 8.9 4.9 2.7 11.5 4.7 9.2 2.1 6.4 -999.0 1.3 6.1 5.4 1.2 6.4

11 2010 Jan 11 14.5 12.1 6.0 1.1 11.5 5.9 9.8 0.4 3.4 -999.0 0.5 3.4 4.4 0.5 3.4

12 2010 Jan 12 21.4 21.4 12.4 7.2 17.4 9.8 13.2 5.1 9.2 -999.0 2.4 8.7 10.1 3.0 9.2

13 2010 Jan 13 21.6 21.2 20.2 14.7 24.9 17.3 21.9 6.0 10.9 -999.0 13.3 12.1 11.4 11.5 11.0

14 2010 Jan 14 22.2 23.5 16.3 9.8 22.6 14.1 17.6 7.0 19.0 -999.0 5.5 16.8 22.2 7.0 19.0

15 2010 Jan 15 10.5 8.6 3.6 1.5 7.2 1.9 5.7 1.2 2.3 -999.0 1.0 2.5 1.8 1.1 2.3

16 2010 Jan 16 7.6 6.6 4.3 3.5 6.8 3.9 5.7 1.6 2.1 -999.0 0.5 2.5 1.9 1.3 2.1

17 2010 Jan 17 7.5 6.5 6.9 0.8 7.1 2.8 6.4 0.2 0.8 -999.0 0.8 2.5 1.9 2.0 0.8

18 2010 Jan 18 37.5 34.2 28.4 18.0 34.4 24.3 25.1 0.9 14.2 -999.0 4.9 6.9 7.8 3.6 14.4

19 2010 Jan 19 47.2 56.9 38.0 25.1 47.0 32.6 40.3 15.8 33.7 -999.0 3.9 34.4 35.6 4.6 33.8

20 2010 Jan 20 46.5 44.0 45.0 28.9 50.6 39.2 51.0 7.4 15.6 -999.0 27.4 20.3 14.2 17.3 15.6

21 2010 Jan 21 16.3 16.1 16.9 3.7 16.5 10.5 15.9 5.8 4.7 -999.0 3.8 11.5 8.4 5.4 4.8

22 2010 Jan 22 12.8 13.4 15.0 3.9 12.4 7.5 9.1 2.9 4.2 -999.0 4.4 6.9 4.8 4.9 4.3

23 2010 Jan 23 9.0 8.7 7.5 3.4 9.1 5.9 9.0 3.0 2.7 -999.0 5.7 4.2 2.5 5.9 2.7

24 2010 Jan 24 4.5 5.2 2.9 1.4 4.9 3.1 7.9 1.5 1.0 -999.0 3.1 2.3 1.1 3.6 1.0

25 2010 Jan 25 7.8 8.5 6.0 3.9 8.5 6.6 9.7 1.7 2.0 -999.0 2.9 4.1 1.2 3.2 2.0

26 2010 Jan 26 12.8 15.1 8.9 7.5 15.2 12.9 16.1 6.1 4.8 -999.0 5.0 12.7 3.9 3.2 4.8

A-45

JD Yr M D EDMC FDMS

EDME FDMS

MCIN BAM1

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS FDMS

ELKI TEOM4

FTSA TEOM40

FTSA SHARP

GENE TEOM4

LAMO BAM1

REDW TEOM4

TOMA TEOM4

90601 SES

27 2010 Jan 27 18.5 22.0 14.0 9.7 20.5 16.5 19.3 6.8 24.6 -999.0 9.0 20.6 17.8 7.7 24.6

28 2010 Jan 28 46.2 57.9 32.7 27.4 44.3 36.7 37.2 11.1 27.7 -999.0 13.7 24.3 18.9 10.3 27.8

29 2010 Jan 29 65.4 67.9 58.7 49.7 74.3 62.3 64.3 8.8 18.1 -999.0 29.2 26.4 10.4 16.0 18.2

30 2010 Jan 30 20.9 20.0 18.1 7.0 22.0 18.1 25.2 4.1 1.9 -999.0 10.0 12.6 4.3 13.3 1.9

31 2010 Jan 31 8.5 7.6 5.2 0.4 7.9 6.2 7.4 1.1 0.3 -999.0 3.3 5.8 3.1 3.3 0.3

32 2010 Feb 1 28.6 30.2 23.9 11.1 30.7 20.5 25.6 8.1 5.2 -999.0 6.3 16.0 -999.0 6.0 5.3

33 2010 Feb 2 38.0 36.2 34.0 15.7 40.5 29.2 39.4 9.9 13.5 -999.0 16.4 14.8 -777.0 7.0 13.5

34 2010 Feb 3 37.9 34.0 33.4 14.1 36.3 26.8 36.0 10.5 11.1 -999.0 12.1 21.3 -999.0 11.3 11.2

35 2010 Feb 4 24.7 20.3 18.7 5.1 21.7 15.1 20.9 6.2 4.7 -999.0 7.8 14.1 -777.0 7.0 4.8

36 2010 Feb 5 21.9 20.7 16.1 4.7 19.3 14.3 20.3 5.7 5.9 -999.0 7.3 12.2 -777.0 7.0 5.9

37 2010 Feb 6 18.5 16.6 16.3 4.0 16.2 12.1 19.4 -777.0 8.4 -999.0 9.4 9.4 -777.0 7.9 8.4

38 2010 Feb 7 7.6 6.4 5.5 0.1 5.7 4.9 5.4 3.9 1.9 -999.0 1.8 6.7 -777.0 1.9 1.9

39 2010 Feb 8 9.6 9.1 2.9 0.7 7.7 6.0 6.0 1.3 0.6 -999.0 4.8 5.2 -999.0 4.4 0.6

40 2010 Feb 9 22.4 25.7 18.6 8.8 22.6 18.0 18.9 9.2 10.9 -999.0 5.5 22.6 -999.0 5.8 10.9

41 2010 Feb 10 15.2 21.7 16.8 5.0 17.5 11.5 14.7 5.0 7.3 -999.0 2.9 15.3 -777.0 2.6 7.4

42 2010 Feb 11 37.6 39.5 25.2 12.1 34.0 22.7 28.8 4.3 12.1 -999.0 6.3 13.1 -777.0 6.5 12.1

43 2010 Feb 12 39.3 27.7 29.8 18.5 36.1 26.8 36.4 3.7 5.1 -999.0 10.5 15.7 7.6 10.9 5.2

44 2010 Feb 13 15.7 7.0 4.6 3.0 6.9 5.1 6.7 3.1 1.8 -999.0 2.5 6.0 4.2 2.7 1.9

45 2010 Feb 14 16.0 7.9 4.3 2.8 7.6 6.2 6.7 1.8 1.4 -999.0 3.5 7.2 2.9 3.0 1.5

46 2010 Feb 15 21.6 15.1 11.6 5.5 14.7 10.1 10.4 3.4 4.2 -999.0 3.0 11.8 4.2 2.3 4.4

47 2010 Feb 16 31.5 33.4 16.7 5.8 20.1 13.6 20.6 2.1 17.0 -999.0 5.2 17.1 11.9 5.0 17.0

48 2010 Feb 17 24.6 -777.0 12.2 4.8 14.5 11.4 11.0 8.0 4.8 -999.0 4.3 19.3 4.6 3.3 4.9

49 2010 Feb 18 19.5 8.4 5.0 3.8 9.0 6.7 6.2 2.0 0.9 -999.0 2.0 5.7 2.6 1.4 1.0

50 2010 Feb 19 21.4 11.4 7.9 5.2 10.8 7.4 6.4 4.1 3.0 -999.0 3.2 8.0 4.6 1.5 3.1

51 2010 Feb 20 26.6 14.1 10.8 5.1 13.3 11.1 14.9 1.3 4.3 -999.0 8.3 6.2 5.7 4.6 4.3

52 2010 Feb 21 33.6 26.2 17.6 7.4 22.7 15.4 20.9 6.0 9.0 -999.0 6.6 16.1 8.5 4.7 9.1

53 2010 Feb 22 31.6 31.0 21.3 17.5 25.0 21.4 19.3 2.1 12.2 -999.0 6.3 5.9 4.8 6.5 12.3

A-46

JD Yr M D EDMC FDMS

EDME FDMS

MCIN BAM1

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS FDMS

ELKI TEOM4

FTSA TEOM40

FTSA SHARP

GENE TEOM4

LAMO BAM1

REDW TEOM4

TOMA TEOM4

90601 SES

54 2010 Feb 23 36.7 31.9 21.4 12.4 26.3 20.5 24.9 6.3 9.1 -999.0 9.3 15.9 10.7 8.8 9.3

55 2010 Feb 24 52.2 59.1 33.8 21.2 46.2 33.7 43.1 13.4 16.6 -999.0 9.4 32.6 18.3 5.2 16.6

56 2010 Feb 25 40.8 31.6 20.1 7.6 28.7 18.3 26.2 5.0 9.7 -999.0 4.3 15.9 11.3 2.3 9.8

57 2010 Feb 26 26.5 21.6 7.7 2.2 13.3 7.9 10.7 1.9 6.4 -999.0 0.6 8.5 10.8 1.4 6.4

58 2010 Feb 27 26.7 17.1 12.4 1.4 11.8 6.8 11.0 1.0 1.7 -999.0 2.3 13.5 2.3 2.0 1.7

59 2010 Feb 28 44.4 36.0 25.3 9.9 32.3 17.9 29.6 8.1 7.6 -999.0 7.3 24.5 7.0 7.5 7.7

60 2010 Mar 1 55.9 43.8 33.7 6.4 43.4 25.9 42.5 4.7 7.7 -999.0 5.5 23.1 8.7 4.3 7.8

61 2010 Mar 2 29.1 18.0 19.3 0.6 15.6 8.6 14.2 1.1 1.4 -999.0 1.4 9.9 5.6 3.1 1.5

62 2010 Mar 3 40.7 39.6 22.9 10.8 29.0 18.9 24.3 11.3 12.0 -999.0 4.8 29.6 11.7 4.7 12.1

63 2010 Mar 4 47.5 38.4 26.9 12.1 34.2 22.6 28.9 7.3 15.1 -999.0 7.9 23.3 15.8 6.4 15.2

64 2010 Mar 5 39.5 35.1 20.9 7.6 29.9 18.8 24.3 5.8 9.9 -999.0 4.0 22.5 11.2 5.0 10.1

65 2010 Mar 6 28.8 19.6 16.0 6.3 19.7 11.6 14.4 2.5 7.0 -999.0 2.9 10.5 7.4 2.1 7.0

66 2010 Mar 7 37.6 29.5 21.7 8.1 28.7 17.6 24.0 2.1 6.4 -999.0 4.5 14.5 5.7 5.3 6.5

67 2010 Mar 8 24.5 16.3 7.9 2.9 12.6 6.4 12.2 3.6 3.9 -999.0 2.3 15.6 5.4 2.5 3.9

68 2010 Mar 9 19.4 10.0 5.2 3.0 8.3 5.6 5.3 0.5 2.1 -999.0 0.8 3.4 3.0 1.4 2.1

69 2010 Mar 10 27.1 20.0 10.5 6.4 17.0 12.3 12.4 3.6 5.8 -999.0 2.9 9.0 8.9 2.7 5.8

70 2010 Mar 11 22.6 14.2 8.0 3.2 12.6 8.1 11.7 2.5 3.3 -999.0 2.4 9.5 8.0 2.3 3.4

71 2010 Mar 12 24.5 11.4 4.6 3.4 10.6 6.1 10.2 1.6 3.4 -999.0 1.2 5.0 3.4 1.4 3.5

72 2010 Mar 13 17.3 8.3 3.3 3.0 7.1 3.5 9.4 1.8 2.5 -999.0 2.0 3.1 2.7 2.5 2.7

73 2010 Mar 14 20.2 12.0 4.1 2.4 10.3 7.4 11.1 2.0 2.0 -999.0 2.0 7.4 3.0 2.2 2.0

74 2010 Mar 15 24.7 13.9 4.7 5.4 13.3 8.5 11.5 2.6 4.0 -999.0 2.8 7.8 5.4 2.3 4.0

75 2010 Mar 16 29.2 18.7 7.9 4.5 16.2 11.1 19.1 1.5 5.0 -999.0 2.9 7.5 6.9 2.2 5.0

76 2010 Mar 17 26.4 16.5 8.2 4.0 13.7 6.8 13.2 2.7 2.2 -999.0 2.8 13.6 4.3 2.2 2.2

77 2010 Mar 18 16.0 6.0 -777.0 1.1 4.7 2.1 8.9 1.1 0.9 -999.0 0.6 3.1 1.4 0.8 0.8

78 2010 Mar 19 18.2 7.8 -777.0 1.7 5.0 4.9 6.6 0.8 1.1 -999.0 0.2 5.9 1.3 0.7 1.1

79 2010 Mar 20 22.6 12.0 4.9 5.8 9.0 5.6 7.7 3.6 4.0 -999.0 3.4 6.8 4.8 4.0 4.0

80 2010 Mar 21 22.5 11.7 8.9 4.4 10.1 4.6 10.2 2.4 3.6 -999.0 5.4 5.3 5.8 5.4 3.7

A-47

JD Yr M D EDMC FDMS

EDME FDMS

MCIN BAM1

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS FDMS

ELKI TEOM4

FTSA TEOM40

FTSA SHARP

GENE TEOM4

LAMO BAM1

REDW TEOM4

TOMA TEOM4

90601 SES

81 2010 Mar 22 19.2 9.9 5.6 3.4 8.8 5.2 9.1 2.3 2.5 -999.0 3.3 3.5 3.3 4.2 2.6

82 2010 Mar 23 27.2 19.5 14.9 6.6 17.7 13.2 14.1 4.8 4.7 -999.0 6.0 8.6 4.6 5.0 4.7

83 2010 Mar 24 16.6 6.7 4.7 0.8 5.0 3.5 -777.0 0.7 1.9 -999.0 3.2 5.1 2.3 2.6 1.9

84 2010 Mar 25 26.1 15.6 13.4 5.7 12.9 10.4 11.0 3.8 5.6 -999.0 8.5 8.7 5.4 8.6 5.7

85 2010 Mar 26 33.3 24.3 20.9 13.1 23.6 18.2 22.6 7.4 10.4 -999.0 8.2 15.4 10.1 10.9 10.5

86 2010 Mar 27 24.4 18.3 10.9 4.9 14.6 9.4 12.0 3.6 6.4 -999.0 4.6 19.5 5.4 4.2 6.5

87 2010 Mar 28 18.7 11.6 1.2 2.5 7.8 3.4 6.7 1.8 4.1 -999.0 3.0 18.4 3.9 1.5 4.1

88 2010 Mar 29 19.0 10.0 3.2 1.2 7.0 1.0 3.8 1.4 1.6 -999.0 0.6 16.9 2.6 0.6 1.6

89 2010 Mar 30 19.0 7.6 1.6 1.1 5.3 0.9 3.7 0.9 0.7 -999.0 0.9 15.2 0.9 0.6 0.8

90 2010 Mar 31 18.0 6.4 1.7 2.1 4.7 2.0 5.3 1.7 1.2 -999.0 1.7 16.1 2.1 1.1 1.3

91 2010 Apr 1 18.1 10.0 3.9 3.3 7.6 4.7 5.7 1.9 1.7 -999.0 1.3 4.5 2.2 1.3 1.8

92 2010 Apr 2 19.1 7.0 2.2 0.8 5.1 1.7 5.6 0.6 0.9 -999.0 0.5 3.6 2.9 1.0 0.8

93 2010 Apr 3 20.3 8.1 3.0 3.1 5.2 1.3 6.1 2.0 2.0 -999.0 1.8 2.4 3.5 2.0 2.1

94 2010 Apr 4 16.4 6.8 2.2 1.8 4.1 1.0 -777.0 1.3 1.0 -999.0 1.0 1.2 1.5 0.6 1.0

95 2010 Apr 5 24.4 15.1 5.2 5.8 11.0 6.4 10.5 3.4 4.2 -999.0 7.0 4.3 5.4 8.0 4.3

96 2010 Apr 6 22.8 13.5 6.5 8.1 11.5 7.2 12.7 4.5 4.6 -999.0 8.4 4.7 5.3 7.3 4.6

97 2010 Apr 7 17.6 10.9 3.0 2.3 6.6 4.2 7.2 1.4 2.2 -999.0 0.5 3.0 2.6 0.9 2.3

98 2010 Apr 8 21.7 11.5 -777.0 5.2 9.3 5.3 7.8 2.9 3.1 -999.0 3.0 4.1 6.0 2.4 3.2

99 2010 Apr 9 15.0 4.2 -777.0 1.1 2.4 3.9 -777.0 1.2 1.2 -999.0 1.5 3.6 1.7 1.0 1.2

100 2010 Apr 10 17.3 4.5 1.7 1.6 2.3 1.8 -777.0 1.5 2.0 -999.0 2.2 1.1 2.0 1.5 2.0

101 2010 Apr 11 19.0 5.0 2.1 1.2 2.9 2.7 4.0 1.4 1.5 -999.0 2.0 2.5 2.6 1.8 1.5

102 2010 Apr 12 18.2 7.7 -777.0 1.9 5.2 5.2 5.2 2.2 2.3 -999.0 2.0 3.6 5.9 1.8 2.4

103 2010 Apr 13 22.5 10.2 -777.0 3.3 8.1 7.2 6.5 1.5 2.4 -999.0 3.5 2.8 4.1 4.5 2.4

104 2010 Apr 14 20.7 9.4 4.1 4.4 8.0 6.1 6.2 3.2 3.3 -999.0 4.4 4.6 3.5 4.2 3.4

105 2010 Apr 15 21.4 9.5 4.8 5.6 9.4 5.7 8.4 3.1 4.0 -999.0 4.2 3.8 4.0 4.0 4.0

106 2010 Apr 16 24.1 12.5 4.9 5.9 11.4 7.5 8.0 4.5 5.8 -999.0 5.5 5.2 6.1 5.5 5.8

107 2010 Apr 17 24.7 14.5 5.6 6.2 12.9 7.4 9.9 5.2 6.9 -999.0 6.3 4.5 9.4 5.8 6.9

A-48

JD Yr M D EDMC FDMS

EDME FDMS

MCIN BAM1

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS FDMS

ELKI TEOM4

FTSA TEOM40

FTSA SHARP

GENE TEOM4

LAMO BAM1

REDW TEOM4

TOMA TEOM4

90601 SES

108 2010 Apr 18 23.8 12.9 4.3 4.7 12.1 5.2 10.6 5.0 5.3 -999.0 6.8 4.4 10.6 6.2 5.5

109 2010 Apr 19 22.7 10.9 4.2 5.5 11.1 5.1 8.5 5.5 5.4 -999.0 -777.0 5.0 10.1 5.4 5.5

110 2010 Apr 20 24.5 13.1 5.6 6.4 13.8 7.4 12.4 8.2 6.7 -999.0 4.8 7.5 8.1 5.2 6.8

111 2010 Apr 21 23.1 12.8 6.0 7.5 14.1 7.3 11.9 7.3 6.0 -999.0 6.9 7.9 6.2 7.0 6.1

112 2010 Apr 22 22.8 10.8 5.3 5.7 10.0 4.7 6.6 5.0 4.7 -999.0 5.0 3.3 5.3 2.0 4.8

113 2010 Apr 23 18.8 7.9 3.8 3.6 7.9 5.4 3.6 2.0 1.7 -999.0 1.9 2.2 1.5 1.1 1.7

114 2010 Apr 24 15.5 7.3 3.5 2.9 6.7 3.0 3.9 2.6 2.3 -999.0 2.9 3.7 3.2 3.1 2.4

115 2010 Apr 25 16.5 8.5 3.0 2.8 6.2 2.8 -777.0 2.3 3.1 -999.0 2.8 3.4 3.1 1.3 3.2

116 2010 Apr 26 17.0 7.3 2.9 1.4 7.2 5.1 3.3 1.8 3.6 -999.0 1.5 2.8 4.6 1.2 3.7

117 2010 Apr 27 18.5 6.8 -777.0 1.1 5.8 1.9 3.1 1.0 6.3 -999.0 1.3 -777.0 2.5 1.5 6.3

118 2010 Apr 28 21.2 8.3 4.9 4.5 8.1 4.0 3.9 2.3 5.9 -999.0 2.9 2.6 5.2 3.3 6.0

119 2010 Apr 29 15.5 5.3 2.6 2.8 5.1 3.0 3.8 1.8 3.7 -999.0 2.0 4.2 2.3 2.3 3.6

120 2010 Apr 30 21.1 10.0 4.8 5.2 9.5 5.0 6.1 1.9 -777.0 -999.0 3.8 3.2 5.5 3.0 -777.0

121 2010 May 1 23.3 13.8 7.0 6.9 13.3 7.4 10.0 4.3 -999.0 -999.0 3.2 4.5 4.5 2.9 -999.0

122 2010 May 2 20.7 9.7 4.7 3.7 8.7 5.2 9.2 2.6 -999.0 -999.0 1.2 3.8 2.8 0.5 -999.0

123 2010 May 3 18.6 8.3 4.1 3.2 8.5 3.3 6.0 1.5 -999.0 -999.0 1.7 2.1 4.8 1.9 -999.0

124 2010 May 4 15.7 4.1 3.2 2.6 3.5 4.1 8.3 1.8 -999.0 -999.0 2.0 2.1 2.0 1.8 -999.0

125 2010 May 5 17.0 6.1 1.9 2.4 4.0 4.8 -777.0 1.3 -999.0 -999.0 2.0 1.7 2.5 2.2 -999.0

126 2010 May 6 19.0 8.5 4.0 4.8 7.1 6.3 4.8 3.3 -999.0 -999.0 2.5 4.5 3.9 2.5 -999.0

127 2010 May 7 20.1 10.4 6.0 5.1 9.3 7.6 5.9 4.5 -999.0 -999.0 2.9 4.7 6.2 3.3 -999.0

128 2010 May 8 24.2 9.4 5.4 3.8 10.2 8.4 8.0 3.9 -999.0 -999.0 3.6 5.0 5.5 3.8 -999.0

129 2010 May 9 19.7 7.3 3.3 2.8 7.6 4.4 4.6 2.4 -999.0 -999.0 2.4 2.7 3.4 2.7 -999.0

130 2010 May 10 19.0 9.2 4.0 3.7 8.7 4.2 8.5 3.0 -999.0 -999.0 3.0 3.6 5.1 3.0 -999.0

131 2010 May 11 25.1 13.9 5.6 8.4 13.8 9.8 12.0 5.0 -999.0 -777.0 5.8 5.7 7.1 4.5 -777.0

132 2010 May 12 -777.0 12.3 5.3 6.8 11.8 8.1 13.5 4.8 -999.0 3.7 4.3 5.2 7.8 4.1 3.8

133 2010 May 13 24.0 12.6 5.8 9.8 14.6 9.3 14.0 6.6 -999.0 17.2 4.3 11.5 15.2 2.9 17.1

134 2010 May 14 31.1 14.9 8.6 9.3 16.0 9.8 24.1 4.1 -999.0 12.4 5.9 3.8 10.3 5.0 12.4

135 2010 May 15 26.4 22.7 9.8 13.0 20.2 13.1 20.2 7.1 -999.0 14.9 9.1 9.4 14.6 7.4 15.0

A-49

JD Yr M D EDMC FDMS

EDME FDMS

MCIN BAM1

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS FDMS

ELKI TEOM4

FTSA TEOM40

FTSA SHARP

GENE TEOM4

LAMO BAM1

REDW TEOM4

TOMA TEOM4

90601 SES

136 2010 May 16 10.8 12.2 5.2 5.8 12.0 7.3 11.1 6.3 -999.0 8.7 6.5 7.6 10.4 5.6 8.8

137 2010 May 17 10.7 25.0 9.6 9.9 16.2 10.0 13.2 15.6 -999.0 20.5 7.4 17.1 44.3 5.2 20.5

138 2010 May 18 17.9 39.9 9.3 9.9 -777.0 7.3 16.1 7.2 -999.0 9.9 8.2 9.1 12.0 7.1 9.8

139 2010 May 19 10.6 17.1 10.6 9.2 -777.0 10.5 12.2 7.6 -999.0 18.4 9.3 10.4 12.6 4.3 18.5

140 2010 May 20 15.3 20.3 23.4 16.5 16.7 14.7 16.0 5.6 -999.0 12.1 7.3 11.0 9.2 6.0 12.3

141 2010 May 21 7.0 6.0 2.0 0.5 3.8 0.5 2.0 0.2 -999.0 0.0 0.0 1.6 0.7 0.0 0.0

142 2010 May 22 4.3 4.3 2.9 0.9 3.3 0.6 2.6 0.7 -999.0 0.0 0.5 2.1 0.7 0.3 0.0

143 2010 May 23 6.5 4.8 3.3 1.1 4.2 1.1 2.6 0.9 -999.0 0.4 1.7 2.5 0.9 0.7 0.3

144 2010 May 24 5.7 6.6 3.6 3.4 6.4 3.7 4.7 8.5 -999.0 1.8 2.9 2.7 4.4 2.8 1.8

145 2010 May 25 8.5 9.2 4.3 4.4 7.4 5.3 6.4 4.3 -999.0 4.3 3.2 3.9 7.9 3.5 4.3

146 2010 May 26 5.2 15.3 4.8 7.0 8.6 7.8 9.1 4.6 -999.0 10.2 5.0 5.6 14.8 5.1 10.4

147 2010 May 27 5.6 9.2 4.1 5.6 6.3 4.2 8.4 1.9 -999.0 6.6 3.8 2.8 22.0 4.3 6.5

148 2010 May 28 6.3 8.0 4.6 3.7 6.1 4.8 6.0 1.8 -999.0 0.9 2.3 2.6 7.2 2.4 1.0

149 2010 May 29 6.6 7.2 3.0 1.7 5.5 2.7 4.5 0.7 -999.0 6.0 1.0 3.1 2.6 1.2 6.1

150 2010 May 30 6.1 3.1 2.3 0.3 2.6 2.2 3.0 0.0 -999.0 0.0 0.1 1.9 0.1 0.5 0.0

151 2010 May 31 6.2 7.1 3.6 2.4 5.6 2.5 6.2 1.9 -999.0 1.6 1.6 3.0 1.9 1.0 1.6

152 2010 Jun 1 5.1 9.2 4.2 3.0 7.8 5.9 7.1 2.9 -999.0 6.1 2.8 3.7 4.1 3.2 6.1

153 2010 Jun 2 -777.0 8.4 4.4 2.6 7.3 3.6 7.6 2.0 -999.0 3.8 2.8 4.1 3.5 2.8 3.9

154 2010 Jun 3 17.0 9.5 4.5 5.7 8.4 4.5 6.1 4.7 -999.0 -777.0 3.1 3.2 5.1 2.6 -777.0

155 2010 Jun 4 16.0 6.9 3.3 4.2 5.6 2.6 6.5 6.4 -999.0 -777.0 2.0 2.2 2.1 1.8 -777.0

156 2010 Jun 5 21.3 11.6 14.2 7.0 11.1 7.3 9.8 3.6 -999.0 -777.0 4.0 2.0 2.7 3.3 -777.0

157 2010 Jun 6 18.8 11.5 7.8 5.0 10.3 5.8 10.0 4.5 -999.0 -777.0 6.5 5.1 3.7 4.9 -777.0

158 2010 Jun 7 17.8 12.5 10.0 5.3 11.7 5.8 10.5 3.1 -999.0 5.9 4.2 4.5 5.5 3.7 5.8

159 2010 Jun 8 18.5 8.0 3.9 1.6 7.3 4.3 5.2 1.1 -999.0 -777.0 2.6 3.6 2.3 2.7 -777.0

160 2010 Jun 9 19.2 9.1 6.6 2.9 8.9 5.3 8.6 1.5 -999.0 7.9 3.3 4.7 10.7 2.7 8.0

161 2010 Jun 10 -777.0 10.0 7.6 4.0 9.3 5.7 7.2 1.5 -999.0 4.3 3.7 3.0 3.5 2.9 4.3

162 2010 Jun 11 17.5 14.5 8.6 7.1 13.2 9.5 8.4 5.4 -999.0 12.2 5.8 7.8 8.0 5.1 12.2

163 2010 Jun 12 13.5 12.5 6.1 5.7 11.1 9.6 10.7 3.9 -999.0 6.5 4.7 8.3 6.6 5.5 6.6

A-50

JD Yr M D EDMC FDMS

EDME FDMS

MCIN BAM1

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS FDMS

ELKI TEOM4

FTSA TEOM40

FTSA SHARP

GENE TEOM4

LAMO BAM1

REDW TEOM4

TOMA TEOM4

90601 SES

164 2010 Jun 13 20.9 11.9 4.5 5.5 10.7 5.3 7.8 3.2 -999.0 6.0 5.0 6.1 5.5 4.8 6.3

165 2010 Jun 14 12.0 7.1 3.9 2.0 7.3 3.1 6.8 1.7 -999.0 -999.0 2.2 3.1 2.7 1.0 -999.0

166 2010 Jun 15 10.4 7.1 4.5 3.3 8.1 3.7 6.4 2.0 -999.0 -999.0 2.6 3.2 3.5 1.7 -999.0

167 2010 Jun 16 12.9 11.4 7.7 7.1 11.9 8.7 10.3 3.7 -999.0 -999.0 6.2 5.3 6.1 3.2 -999.0

168 2010 Jun 17 17.8 13.4 9.9 8.4 13.9 9.2 13.6 3.4 -999.0 -777.0 5.4 5.4 8.9 4.7 -777.0

169 2010 Jun 18 24.9 16.9 11.4 13.1 18.9 11.1 15.5 8.0 -999.0 13.0 10.4 11.3 16.0 5.2 13.0

170 2010 Jun 19 22.0 22.3 14.9 16.6 21.0 15.6 21.0 9.8 -999.0 18.4 13.4 15.5 21.0 8.6 18.4

171 2010 Jun 20 17.4 18.0 13.4 16.1 17.3 12.7 17.8 7.2 -999.0 14.4 10.8 11.8 20.8 6.7 14.5

172 2010 Jun 21 21.4 21.5 13.2 13.8 15.5 11.6 15.0 7.4 -999.0 9.9 9.6 11.5 10.7 8.9 10.1

173 2010 Jun 22 16.4 16.9 13.7 10.8 16.1 12.1 12.2 7.5 -999.0 12.4 6.0 12.2 11.7 6.4 12.5

174 2010 Jun 23 13.1 15.5 13.2 11.3 -777.0 15.7 18.7 4.2 -999.0 7.2 4.6 9.8 10.5 4.3 7.3

175 2010 Jun 24 8.5 10.4 6.9 4.5 -777.0 7.2 10.6 1.3 -999.0 4.7 2.7 5.2 4.1 2.5 4.8

176 2010 Jun 25 6.4 11.5 4.8 4.7 6.1 5.3 9.4 2.3 -999.0 -777.0 1.3 6.1 2.3 0.8 -777.0

177 2010 Jun 26 4.3 7.0 4.5 4.9 5.9 7.6 9.3 1.2 -999.0 -777.0 2.0 3.7 2.1 1.0 -777.0

178 2010 Jun 27 7.5 9.8 8.1 10.4 9.6 8.7 12.4 3.6 -999.0 -777.0 4.3 6.5 3.7 3.1 -777.0

179 2010 Jun 28 7.2 9.8 3.8 4.8 6.6 5.2 6.6 3.3 -999.0 -777.0 4.1 5.8 6.0 2.4 -777.0

180 2010 Jun 29 7.8 15.3 17.7 7.4 10.4 7.7 9.7 3.2 -999.0 6.9 3.3 7.1 7.0 2.3 7.0

181 2010 Jun 30 4.0 -777.0 7.3 1.8 5.0 3.9 5.2 1.0 -999.0 0.2 1.7 4.2 0.9 0.8 0.2

182 2010 Jul 1 3.5 5.7 6.6 2.4 2.7 3.8 4.6 0.9 -999.0 0.2 1.5 4.5 1.2 1.1 0.2

183 2010 Jul 2 5.2 7.6 5.8 3.6 5.6 4.0 8.3 2.2 -999.0 -777.0 2.4 5.2 2.5 1.7 -777.0

184 2010 Jul 3 4.0 7.8 5.7 4.3 4.4 4.6 5.5 2.4 -999.0 -777.0 2.2 5.9 3.6 1.6 -777.0

185 2010 Jul 4 3.7 5.8 3.7 2.4 3.3 3.1 6.1 1.3 -999.0 -777.0 1.3 3.6 1.5 1.6 -777.0

186 2010 Jul 5 5.9 7.3 5.3 4.9 4.7 5.4 4.8 4.4 -999.0 -777.0 3.0 7.2 4.6 1.8 -777.0

187 2010 Jul 6 5.9 6.9 3.9 3.7 5.8 5.5 5.0 2.5 -999.0 3.0 2.5 5.3 3.0 1.8 3.0

188 2010 Jul 7 6.5 7.9 4.9 3.8 6.0 6.1 6.5 2.8 -999.0 3.4 4.4 5.6 3.6 3.4 3.4

189 2010 Jul 8 9.2 11.6 5.2 6.0 9.0 8.5 11.2 4.8 -999.0 5.7 7.1 7.2 5.0 6.0 5.8

190 2010 Jul 9 10.0 11.8 7.7 6.6 12.4 8.2 14.0 4.3 -999.0 8.3 -777.0 6.2 5.7 6.6 8.4

191 2010 Jul 10 7.4 10.6 7.2 3.5 10.2 6.7 7.9 1.6 -999.0 5.6 3.7 6.2 2.6 2.8 5.8

A-51

JD Yr M D EDMC FDMS

EDME FDMS

MCIN BAM1

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS FDMS

ELKI TEOM4

FTSA TEOM40

FTSA SHARP

GENE TEOM4

LAMO BAM1

REDW TEOM4

TOMA TEOM4

90601 SES

192 2010 Jul 11 7.8 8.9 7.2 3.5 8.9 7.8 8.0 2.7 -999.0 4.8 2.8 7.9 4.5 1.7 4.9

193 2010 Jul 12 5.9 8.0 6.6 -777.0 7.4 5.7 8.4 0.9 -999.0 5.5 1.6 7.0 3.4 0.9 5.6

194 2010 Jul 13 7.2 8.7 10.6 10.1 8.3 8.9 8.1 7.3 -999.0 9.2 5.0 8.0 7.3 3.2 9.3

195 2010 Jul 14 8.7 10.6 8.2 7.2 10.3 11.5 8.8 5.6 -999.0 5.9 4.1 11.2 5.0 2.1 5.9

196 2010 Jul 15 6.6 9.3 4.5 4.8 7.8 8.3 8.4 5.7 -999.0 3.6 4.1 7.3 3.6 2.9 3.6

197 2010 Jul 16 7.3 10.0 8.0 5.4 7.5 8.8 7.7 3.8 -999.0 5.3 3.2 6.4 3.8 1.2 5.3

198 2010 Jul 17 6.6 6.9 4.3 4.5 5.7 7.8 8.9 2.7 -999.0 2.9 3.9 6.0 1.4 1.7 2.9

199 2010 Jul 18 6.8 8.5 4.8 4.6 6.4 8.3 6.8 4.2 -999.0 4.0 4.4 6.4 6.0 2.3 4.0

200 2010 Jul 19 6.7 10.1 8.3 4.8 6.4 7.8 8.2 3.4 -999.0 4.4 3.5 8.1 3.8 1.5 4.5

201 2010 Jul 20 7.6 10.4 5.7 4.6 6.7 8.4 6.3 4.3 -999.0 5.1 6.9 8.7 4.4 -777.0 5.2

202 2010 Jul 21 9.2 12.3 5.9 5.7 10.5 8.6 9.9 5.9 -999.0 7.8 5.0 7.9 6.3 -777.0 7.9

203 2010 Jul 22 7.7 9.3 5.7 5.3 8.3 5.6 9.8 4.5 -999.0 6.2 6.3 9.3 9.3 5.2 6.2

204 2010 Jul 23 6.0 8.6 4.9 4.5 8.8 5.4 8.3 4.8 -999.0 3.6 4.0 9.1 3.7 1.5 3.7

205 2010 Jul 24 6.1 9.1 3.7 5.2 9.4 5.1 9.8 3.2 -999.0 4.4 5.3 5.8 5.2 2.8 4.3

206 2010 Jul 25 7.7 10.1 3.5 5.3 9.5 5.5 10.1 2.5 -999.0 4.0 6.0 5.0 3.0 3.6 4.1

207 2010 Jul 26 5.1 6.2 3.8 4.6 6.9 3.6 9.1 2.2 -999.0 4.5 4.6 4.8 3.1 2.5 4.6

208 2010 Jul 27 8.7 9.2 4.1 6.8 8.7 5.2 9.2 3.5 -999.0 8.9 5.2 5.7 3.5 4.0 8.9

209 2010 Jul 28 7.7 8.0 4.7 7.0 8.7 4.4 9.4 3.6 -999.0 6.6 9.0 7.0 5.1 7.2 6.6

210 2010 Jul 29 15.4 18.5 7.7 16.4 16.0 10.2 13.1 6.0 -999.0 9.6 10.0 11.6 10.0 6.7 9.7

211 2010 Jul 30 12.0 15.9 10.6 11.9 14.2 10.3 12.8 7.9 -999.0 8.1 10.6 12.3 10.6 10.2 8.2

212 2010 Jul 31 11.9 15.2 10.1 11.0 14.5 11.2 12.2 7.9 -999.0 10.3 10.2 13.7 9.4 7.5 10.5

213 2010 Aug 1 13.5 16.8 10.2 11.3 15.8 9.8 13.3 7.7 -999.0 10.2 10.8 13.7 10.5 8.0 10.3

214 2010 Aug 2 7.7 9.1 7.2 6.5 9.0 6.2 10.5 4.2 -999.0 5.5 7.4 8.2 5.2 5.5 5.6

215 2010 Aug 3 4.4 6.0 5.7 4.2 5.7 4.2 6.3 -777.0 -999.0 3.4 4.9 5.5 3.1 2.2 3.4

216 2010 Aug 4 -777.0 13.2 9.7 10.6 12.8 11.8 14.4 -777.0 -999.0 9.3 11.8 14.2 12.6 11.7 9.5

217 2010 Aug 5 26.3 36.0 30.0 29.5 34.8 31.8 31.5 29.3 -999.0 27.1 23.8 33.0 29.9 19.1 27.1

218 2010 Aug 6 22.4 27.8 21.7 21.3 26.8 21.6 27.9 25.4 -999.0 20.9 14.5 30.7 25.7 13.0 21.0

219 2010 Aug 7 22.3 27.8 18.0 20.8 26.8 20.5 25.8 23.6 -999.0 20.9 18.3 26.2 24.7 16.6 21.0

A-52

JD Yr M D EDMC FDMS

EDME FDMS

MCIN BAM1

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS FDMS

ELKI TEOM4

FTSA TEOM40

FTSA SHARP

GENE TEOM4

LAMO BAM1

REDW TEOM4

TOMA TEOM4

90601 SES

220 2010 Aug 8 7.6 9.5 5.3 6.2 8.3 5.7 11.1 6.4 -999.0 5.1 6.8 11.3 10.1 5.8 5.1

221 2010 Aug 9 4.7 9.4 4.5 4.7 7.6 4.3 8.2 5.8 -999.0 3.8 6.0 8.6 7.0 4.2 3.9

222 2010 Aug 10 6.1 10.5 6.9 7.4 12.1 7.9 11.3 4.3 -999.0 4.2 5.7 5.3 9.7 2.7 4.3

223 2010 Aug 11 7.4 13.6 8.2 7.2 12.2 7.7 15.4 5.7 -999.0 5.2 5.9 10.1 8.2 2.7 5.3

224 2010 Aug 12 -777.0 7.5 6.0 3.5 8.0 3.3 7.7 3.3 -999.0 4.6 3.8 8.0 4.1 2.1 4.6

225 2010 Aug 13 -777.0 4.1 2.5 2.4 5.3 2.8 4.9 1.6 -999.0 3.1 2.3 4.3 2.1 1.1 3.2

226 2010 Aug 14 -777.0 6.2 2.4 3.0 6.9 3.5 7.1 1.0 -999.0 2.8 3.0 4.1 1.4 1.5 2.8

227 2010 Aug 15 -777.0 8.0 3.9 3.3 8.0 4.0 8.8 1.7 -999.0 2.3 4.5 5.2 1.7 3.5 2.3

228 2010 Aug 16 7.4 12.0 7.8 5.5 10.3 6.5 10.7 3.9 -999.0 4.7 4.9 6.4 5.2 4.1 4.7

229 2010 Aug 17 -777.0 9.8 8.7 5.1 10.3 6.6 12.4 1.5 -999.0 4.8 3.7 7.1 3.1 2.8 4.8

230 2010 Aug 18 20.0 13.2 8.1 7.4 12.0 7.5 15.3 4.6 -999.0 7.0 14.8 8.2 8.7 17.1 7.0

231 2010 Aug 19 146.8 159.6 141.7 154.7 166.4 106.6 145.0 159.7 -999.0 347.6 220.5 184.5 163.3 211.2 349.8

232 2010 Aug 20 77.1 87.5 81.9 85.1 94.5 84.8 91.6 61.0 -999.0 56.6 109.3 68.5 56.6 102.1 56.7

233 2010 Aug 21 71.8 73.2 72.0 66.8 75.4 69.2 76.0 54.7 -999.0 58.5 62.1 67.8 62.8 54.3 58.6

234 2010 Aug 22 48.9 37.1 25.6 20.0 29.8 25.3 25.8 29.2 -999.0 28.0 9.5 31.9 21.9 9.2 28.0

235 2010 Aug 23 20.0 21.4 17.4 15.4 20.1 16.4 21.0 13.7 -999.0 16.0 10.7 20.2 11.9 7.7 16.0

236 2010 Aug 24 -777.0 15.9 6.1 7.0 9.9 6.7 11.1 5.4 -999.0 6.4 5.0 8.9 5.9 3.3 6.4

237 2010 Aug 25 7.2 16.6 3.9 7.0 10.0 6.6 12.3 5.5 -999.0 3.9 4.9 7.4 8.1 3.3 4.0

238 2010 Aug 26 15.2 -777.0 5.8 6.2 9.8 4.5 15.0 2.9 -999.0 4.6 3.7 6.0 5.7 2.6 4.8

239 2010 Aug 27 -777.0 -999.0 0.9 1.6 3.5 0.4 5.1 0.7 -999.0 -777.0 0.3 5.0 0.9 0.4 -777.0

240 2010 Aug 28 -999.0 -999.0 1.3 1.0 1.6 -777.0 4.2 0.2 -999.0 -999.0 0.6 3.0 0.4 0.8 -999.0

241 2010 Aug 29 -999.0 -999.0 2.7 2.9 4.3 2.3 7.3 1.7 -999.0 -999.0 3.0 5.0 1.5 1.8 -999.0

242 2010 Aug 30 -999.0 -777.0 6.2 4.0 7.4 4.1 6.8 3.1 -999.0 -777.0 2.8 6.6 6.3 2.3 -777.0

243 2010 Aug 31 -999.0 9.7 4.2 3.0 5.9 2.7 7.6 1.9 -999.0 3.4 2.3 5.0 4.2 1.6 3.4

244 2010 Sep 1 -999.0 10.2 8.3 5.9 8.1 3.9 9.3 5.0 -999.0 5.6 3.4 6.0 3.7 2.2 5.7

245 2010 Sep 2 -777.0 9.6 3.5 3.3 5.7 3.0 9.3 2.5 -999.0 1.9 3.8 5.4 2.7 1.7 1.9

246 2010 Sep 3 7.2 7.0 1.4 2.0 4.9 3.1 10.1 1.2 -999.0 2.5 6.1 4.3 2.3 2.6 2.5

247 2010 Sep 4 5.1 7.4 3.2 3.6 5.9 1.8 10.6 1.5 -999.0 4.0 2.2 2.6 2.2 1.1 4.0

A-53

JD Yr M D EDMC FDMS

EDME FDMS

MCIN BAM1

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS FDMS

ELKI TEOM4

FTSA TEOM40

FTSA SHARP

GENE TEOM4

LAMO BAM1

REDW TEOM4

TOMA TEOM4

90601 SES

248 2010 Sep 5 4.6 6.8 3.1 1.9 3.9 1.8 9.6 1.8 -999.0 2.0 1.2 3.1 1.6 0.8 2.0

249 2010 Sep 6 4.6 5.9 2.5 2.0 3.2 2.1 8.5 1.6 -999.0 1.6 1.7 3.2 1.5 1.1 1.6

250 2010 Sep 7 6.6 7.8 4.6 4.0 6.1 3.5 10.0 1.5 -999.0 4.0 4.1 3.8 2.8 3.0 4.1

251 2010 Sep 8 6.9 7.1 3.5 2.4 5.6 3.8 8.3 0.5 -999.0 3.5 2.3 3.8 9.3 2.7 3.5

252 2010 Sep 9 7.7 7.0 6.0 2.3 4.2 3.7 6.2 0.5 -999.0 2.2 1.8 4.3 9.6 3.4 2.2

253 2010 Sep 10 6.8 8.5 7.2 4.4 6.1 5.0 6.2 2.0 -999.0 2.6 2.6 4.7 2.2 2.2 2.6

254 2010 Sep 11 6.3 8.2 2.3 3.1 4.3 3.3 5.3 3.3 -999.0 3.2 1.8 6.4 2.7 1.5 3.4

255 2010 Sep 12 6.2 9.7 7.3 4.0 5.4 3.2 6.6 1.9 -999.0 4.5 2.6 5.3 2.6 2.0 4.5

256 2010 Sep 13 6.2 6.2 5.3 2.9 4.7 3.8 8.0 1.1 -999.0 8.3 2.1 4.3 3.2 3.0 8.5

257 2010 Sep 14 5.6 5.4 2.6 1.1 4.7 2.0 8.0 0.3 -999.0 1.5 1.5 3.5 1.3 1.3 1.5

258 2010 Sep 15 9.0 11.1 8.7 3.1 7.2 4.6 7.8 2.2 -999.0 8.2 1.3 6.5 4.8 1.2 8.2

259 2010 Sep 16 3.9 5.7 3.1 2.5 3.4 1.6 4.2 0.5 -999.0 2.3 0.7 2.6 1.4 0.6 2.3

260 2010 Sep 17 4.8 6.3 3.5 3.4 4.6 2.7 5.5 1.0 -999.0 1.3 0.6 3.5 1.8 0.6 1.3

261 2010 Sep 18 6.8 6.5 3.6 2.2 4.5 3.0 7.2 1.6 -999.0 2.0 2.0 5.5 4.1 1.8 2.1

262 2010 Sep 19 4.2 5.5 2.5 0.6 1.6 0.8 5.5 0.8 -999.0 0.0 0.7 4.3 2.8 0.8 0.0

263 2010 Sep 20 3.3 5.2 2.7 1.5 2.1 2.5 3.2 0.3 -999.0 0.3 0.5 1.9 1.2 0.5 0.3

264 2010 Sep 21 7.6 8.5 5.5 3.2 5.4 4.5 6.9 2.6 -999.0 6.3 1.3 8.1 4.4 1.3 6.3

265 2010 Sep 22 9.1 9.8 3.0 3.0 6.7 4.6 8.7 1.8 -999.0 4.1 2.7 6.1 3.3 3.0 4.2

266 2010 Sep 23 9.0 8.2 6.0 4.3 6.5 3.6 9.8 1.1 -999.0 4.0 3.1 3.8 2.7 3.0 4.0

267 2010 Sep 24 7.2 10.7 2.4 3.9 6.1 4.1 6.5 2.4 -999.0 5.3 2.0 6.3 4.4 2.3 5.3

268 2010 Sep 25 6.3 7.7 0.7 2.4 4.4 1.6 6.0 1.0 -999.0 4.0 0.8 4.4 4.6 1.0 4.1

269 2010 Sep 26 6.3 7.8 2.7 3.7 5.5 1.3 7.2 2.9 -999.0 3.9 2.5 5.1 5.0 2.7 3.9

270 2010 Sep 27 7.7 7.9 2.3 2.8 4.9 2.8 6.9 1.7 -999.0 2.9 1.5 5.3 4.0 1.9 3.0

271 2010 Sep 28 6.5 3.4 2.9 2.3 3.8 1.1 5.7 1.1 -999.0 1.8 1.2 3.9 1.1 1.1 1.9

272 2010 Sep 29 6.6 4.0 3.1 3.9 5.9 4.2 6.8 1.6 -999.0 2.8 1.9 4.0 1.7 2.3 2.8

273 2010 Sep 30 8.0 6.8 5.1 5.5 7.2 5.2 8.1 2.1 -999.0 6.9 3.6 7.3 4.3 2.6 6.9

274 2010 Oct 1 5.8 4.9 2.4 2.3 3.7 4.0 6.5 2.0 -999.0 5.3 2.2 5.3 2.9 3.7 5.3

275 2010 Oct 2 13.5 12.8 9.1 11.4 13.6 11.1 17.7 6.9 -999.0 11.7 7.7 9.9 8.3 6.9 11.8

A-54

JD Yr M D EDMC FDMS

EDME FDMS

MCIN BAM1

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS FDMS

ELKI TEOM4

FTSA TEOM40

FTSA SHARP

GENE TEOM4

LAMO BAM1

REDW TEOM4

TOMA TEOM4

90601 SES

276 2010 Oct 3 13.5 12.1 8.5 7.7 12.5 8.4 12.3 4.5 -999.0 9.0 6.2 7.8 6.0 3.8 9.0

277 2010 Oct 4 15.5 12.7 12.4 10.4 14.9 9.3 -777.0 4.4 -999.0 8.2 7.9 8.0 7.9 5.3 8.3

278 2010 Oct 5 5.0 5.5 2.1 2.5 5.1 3.4 -777.0 1.9 -999.0 1.4 3.2 5.5 2.5 1.7 1.4

279 2010 Oct 6 6.2 9.0 3.1 3.9 7.7 4.7 6.0 1.2 -999.0 4.6 3.7 4.3 3.4 3.1 4.5

280 2010 Oct 7 11.7 10.9 5.7 6.5 10.7 6.3 10.6 3.4 -999.0 8.4 6.3 8.2 6.8 6.4 8.4

281 2010 Oct 8 -777.0 11.0 7.6 8.9 12.6 7.2 11.0 4.1 -999.0 7.9 7.1 6.2 9.0 5.0 8.0

282 2010 Oct 9 -999.0 6.3 4.8 5.1 11.2 6.5 10.0 2.4 -999.0 6.3 6.0 5.3 4.1 5.0 6.4

283 2010 Oct 10 -999.0 3.8 2.2 2.1 6.0 1.6 4.8 1.4 -999.0 2.3 1.6 4.9 2.5 1.2 2.4

284 2010 Oct 11 3.7 3.8 0.7 1.7 3.7 1.9 4.6 0.6 -999.0 1.6 2.0 3.5 1.1 1.2 1.6

285 2010 Oct 12 5.8 7.0 2.4 3.6 5.5 3.4 7.0 1.3 -999.0 3.0 2.4 3.3 2.3 2.0 3.0

286 2010 Oct 13 6.3 9.4 2.8 4.5 7.0 5.0 7.9 2.8 -999.0 6.6 4.8 6.1 5.2 3.5 6.6

287 2010 Oct 14 8.7 12.9 6.4 6.7 9.6 7.2 9.3 4.5 -999.0 9.1 7.7 8.0 6.4 6.4 9.2

288 2010 Oct 15 3.7 4.5 3.1 4.4 5.7 2.8 5.8 2.3 -999.0 1.8 1.7 5.0 2.5 1.0 1.8

289 2010 Oct 16 3.6 7.0 3.8 4.4 5.8 4.0 7.3 1.5 -999.0 4.0 2.1 5.0 2.7 1.2 4.1

290 2010 Oct 17 4.7 7.5 5.0 5.2 7.2 5.0 9.2 2.5 -999.0 5.2 2.6 5.3 4.3 2.3 5.3

291 2010 Oct 18 4.8 13.1 6.4 6.6 9.0 6.4 8.7 3.9 -999.0 6.9 3.5 6.7 6.0 2.2 7.0

292 2010 Oct 19 4.2 9.0 3.5 3.6 8.2 4.2 5.3 2.0 -999.0 3.1 2.3 4.6 2.4 -777.0 3.1

293 2010 Oct 20 6.8 11.0 3.6 3.7 6.3 4.0 6.2 3.8 -999.0 4.3 2.2 6.3 4.9 1.3 4.4

294 2010 Oct 21 9.5 12.5 6.7 6.5 9.3 5.2 8.8 5.0 -999.0 6.4 4.4 5.1 5.0 3.5 6.3

295 2010 Oct 22 7.7 7.4 4.7 3.5 6.5 4.2 5.8 2.1 -999.0 4.2 2.4 2.6 5.8 3.2 4.2

296 2010 Oct 23 14.5 13.2 14.6 8.4 13.4 9.2 12.3 3.2 -999.0 7.0 6.8 5.7 5.7 7.6 7.0

297 2010 Oct 24 10.9 10.3 8.0 4.6 8.8 5.8 9.4 2.1 -999.0 5.1 5.9 5.5 4.9 7.4 5.2

298 2010 Oct 25 5.0 5.5 4.2 2.6 3.9 1.6 3.8 1.9 -999.0 2.5 1.5 2.7 2.2 1.3 2.5

299 2010 Oct 26 5.2 5.7 4.5 2.8 3.5 2.9 6.5 2.5 -999.0 2.5 2.8 4.9 3.1 4.5 2.5

300 2010 Oct 27 10.8 9.5 8.4 5.8 7.9 6.9 -777.0 9.7 -999.0 9.4 6.1 13.8 7.6 7.3 9.5

301 2010 Oct 28 18.1 20.4 -999.0 8.6 15.0 14.7 17.3 5.5 -999.0 11.7 5.6 11.7 9.3 4.4 11.6

302 2010 Oct 29 24.8 28.9 -999.0 15.7 24.2 21.2 28.0 10.4 -999.0 19.4 6.6 18.1 19.3 4.6 19.5

303 2010 Oct 30 14.2 12.7 -777.0 5.6 9.8 9.5 16.3 4.5 -999.0 8.2 2.4 9.3 9.0 2.9 8.3

A-55

JD Yr M D EDMC FDMS

EDME FDMS

MCIN BAM1

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS FDMS

ELKI TEOM4

FTSA TEOM40

FTSA SHARP

GENE TEOM4

LAMO BAM1

REDW TEOM4

TOMA TEOM4

90601 SES

304 2010 Oct 31 8.5 12.1 10.3 7.2 9.2 6.9 13.6 2.4 -999.0 7.7 2.8 8.4 6.4 4.4 7.8

305 2010 Nov 1 9.7 10.4 5.1 3.9 5.6 3.4 12.6 1.6 -999.0 4.3 1.3 4.4 4.8 1.5 4.5

306 2010 Nov 2 3.3 6.3 2.2 1.6 1.8 2.4 6.8 0.9 -999.0 0.8 1.1 4.0 1.2 1.7 0.9

307 2010 Nov 3 5.3 7.9 3.2 2.7 3.7 3.4 8.7 1.7 -999.0 2.1 2.6 5.0 2.8 1.7 2.1

308 2010 Nov 4 8.1 9.0 4.3 3.3 5.2 4.0 9.9 1.6 -999.0 3.6 2.5 4.2 2.9 3.3 3.6

309 2010 Nov 5 4.9 6.7 2.5 2.6 4.4 1.7 6.3 1.2 -999.0 3.2 1.2 3.2 3.1 1.2 3.2

310 2010 Nov 6 5.2 6.1 1.1 1.9 3.6 1.9 4.6 0.8 -999.0 3.2 1.7 3.5 5.8 1.1 3.3

311 2010 Nov 7 11.0 12.5 7.1 6.5 9.6 5.8 12.3 2.6 -999.0 7.5 4.6 4.4 7.6 5.2 7.5

312 2010 Nov 8 8.5 8.1 6.8 6.2 8.3 5.2 9.7 2.9 -999.0 4.7 2.7 5.7 3.6 2.2 4.8

313 2010 Nov 9 9.1 9.0 6.3 3.3 7.0 4.7 8.2 2.0 -999.0 7.2 2.2 5.7 5.7 2.8 7.3

314 2010 Nov 10 14.7 20.0 15.3 9.8 16.2 11.3 14.4 7.5 -999.0 10.3 4.0 15.4 9.8 3.7 10.3

315 2010 Nov 11 9.1 12.5 9.0 5.3 9.0 7.3 10.6 3.2 -999.0 6.4 3.8 10.1 4.2 4.1 6.5

316 2010 Nov 12 8.8 12.8 7.1 6.7 9.5 5.9 8.6 4.2 -999.0 7.6 3.2 8.3 5.1 3.8 7.7

317 2010 Nov 13 6.5 7.9 3.9 2.6 5.3 4.2 6.8 1.3 -999.0 4.7 1.0 5.4 2.9 1.2 4.8

318 2010 Nov 14 4.5 7.3 4.0 2.0 3.5 2.5 2.6 1.7 -999.0 2.6 0.4 3.6 1.2 0.7 2.7

319 2010 Nov 15 4.6 6.7 4.7 1.2 4.2 2.2 3.4 -777.0 -999.0 1.9 0.2 5.1 0.9 0.7 1.9

320 2010 Nov 16 2.9 3.8 2.8 1.8 1.9 1.8 5.6 0.2 -999.0 1.0 1.2 2.5 1.1 1.5 1.0

321 2010 Nov 17 4.8 4.0 3.8 1.5 2.1 3.3 6.8 0.2 -999.0 7.2 1.5 3.0 6.9 1.2 7.1

322 2010 Nov 18 3.9 3.8 5.7 2.3 2.5 2.6 4.1 0.2 -999.0 3.0 0.6 1.6 3.9 0.7 3.0

323 2010 Nov 19 13.9 14.1 13.6 8.3 11.7 11.1 14.7 5.5 -999.0 19.0 3.3 11.3 11.6 2.3 19.0

324 2010 Nov 20 18.7 22.3 18.3 8.0 17.2 15.5 20.8 8.1 -999.0 15.4 4.1 28.7 8.3 3.2 15.5

325 2010 Nov 21 14.8 15.1 18.8 10.6 16.0 14.9 17.9 21.1 -999.0 13.5 4.5 18.0 5.6 3.8 13.5

326 2010 Nov 22 8.6 7.3 11.6 5.9 8.4 8.1 -777.0 2.4 -999.0 7.0 5.0 7.0 3.5 2.5 7.0

327 2010 Nov 23 11.9 12.2 11.1 6.4 9.7 11.2 -777.0 4.9 -999.0 13.8 4.4 16.0 7.8 4.5 13.9

328 2010 Nov 24 11.2 13.7 12.5 7.0 10.5 11.8 -777.0 3.7 -999.0 12.1 2.1 9.9 9.0 4.0 12.1

329 2010 Nov 25 7.5 10.2 6.1 2.4 4.3 4.3 9.5 1.3 -999.0 5.1 1.1 6.6 2.0 1.4 5.3

330 2010 Nov 26 11.2 9.7 6.7 2.9 5.6 4.9 9.7 2.8 -999.0 8.7 1.9 7.4 7.3 1.8 8.8

331 2010 Nov 27 10.0 12.1 10.3 6.5 8.6 8.0 13.9 3.8 -999.0 8.7 2.8 15.3 4.9 2.7 8.8

A-56

JD Yr M D EDMC FDMS

EDME FDMS

MCIN BAM1

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS FDMS

ELKI TEOM4

FTSA TEOM40

FTSA SHARP

GENE TEOM4

LAMO BAM1

REDW TEOM4

TOMA TEOM4

90601 SES

332 2010 Nov 28 15.9 16.2 16.5 10.2 14.0 13.5 21.8 7.9 -999.0 12.7 7.0 17.1 7.4 7.4 12.8

333 2010 Nov 29 19.4 22.5 17.3 9.4 17.8 15.8 22.7 8.3 -999.0 27.0 3.2 24.7 21.6 3.6 27.0

334 2010 Nov 30 11.9 8.8 11.2 1.9 9.5 8.1 13.4 2.0 -999.0 6.7 3.0 8.4 4.0 3.8 6.8

335 2010 Dec 1 26.9 32.0 27.9 13.6 25.5 21.9 28.5 7.2 -999.0 22.2 7.2 18.6 16.7 7.1 22.3

336 2010 Dec 2 29.4 24.7 -777.0 9.5 24.3 20.7 27.8 13.0 -999.0 34.1 14.4 32.5 18.4 10.5 34.3

337 2010 Dec 3 19.9 17.8 23.7 10.9 17.3 16.6 22.8 7.3 -999.0 13.1 9.7 -777.0 7.2 8.8 13.2

338 2010 Dec 4 18.0 18.8 18.1 10.9 17.1 17.0 23.1 7.2 -999.0 13.5 5.8 16.2 10.4 7.3 13.4

339 2010 Dec 5 16.1 18.0 20.5 12.0 18.4 18.0 19.9 3.8 -999.0 17.7 3.1 10.6 8.7 3.6 17.8

340 2010 Dec 6 35.3 47.8 38.1 24.5 35.1 31.3 32.5 13.8 -999.0 57.1 8.0 22.2 11.8 6.9 57.0

341 2010 Dec 7 43.9 46.4 45.4 24.1 40.5 39.3 45.6 35.0 -999.0 54.8 10.8 58.1 23.3 10.0 54.8

342 2010 Dec 8 38.0 33.9 35.8 14.9 31.8 30.6 38.8 7.6 -999.0 26.6 13.7 24.5 10.1 14.7 26.7

343 2010 Dec 9 12.3 10.9 12.2 4.7 9.9 8.7 12.8 3.0 -999.0 6.4 5.4 9.0 2.3 5.4 6.4

344 2010 Dec 10 12.8 7.7 11.3 7.2 8.5 9.3 11.4 2.4 -999.0 5.0 5.9 5.7 5.8 5.3 4.9

345 2010 Dec 11 8.8 6.7 7.8 2.6 5.1 5.5 9.8 0.8 -999.0 5.0 2.9 4.8 4.2 3.4 5.1

346 2010 Dec 12 9.0 7.5 7.6 2.7 5.2 6.7 9.2 1.1 -999.0 5.5 1.6 6.4 1.6 1.4 5.5

347 2010 Dec 13 21.4 20.3 22.4 9.4 17.4 16.2 18.4 3.7 -999.0 17.1 3.3 10.9 9.4 2.8 17.3

348 2010 Dec 14 17.6 15.5 12.7 3.1 11.3 8.3 12.5 1.9 -999.0 8.5 0.9 8.1 3.7 3.0 8.5

349 2010 Dec 15 9.4 10.3 8.6 5.2 7.1 7.0 9.3 5.3 -999.0 6.8 5.9 10.0 5.5 4.8 6.8

350 2010 Dec 16 8.3 7.7 6.1 4.7 5.3 5.5 8.9 2.6 -999.0 5.9 3.3 6.3 4.3 2.1 6.0

351 2010 Dec 17 14.9 10.3 10.2 4.8 9.1 8.7 12.3 2.8 -999.0 8.9 3.1 8.1 19.2 2.6 9.0

352 2010 Dec 18 13.8 12.8 14.5 7.6 11.8 12.5 12.2 1.8 -999.0 7.6 3.1 6.3 8.8 1.9 7.6

353 2010 Dec 19 9.6 7.6 9.0 5.1 8.5 7.3 8.9 2.8 -999.0 6.2 3.2 6.6 3.4 2.0 6.1

354 2010 Dec 20 11.8 9.7 11.2 5.1 9.6 7.7 12.5 4.0 -999.0 11.1 4.3 10.1 10.3 3.6 11.3

355 2010 Dec 21 13.2 13.4 11.7 5.5 10.4 7.6 12.9 2.5 -999.0 8.3 3.3 8.3 7.5 3.5 8.3

356 2010 Dec 22 25.6 22.4 27.7 16.5 25.4 22.4 33.9 1.5 -999.0 10.8 11.6 7.4 9.6 8.4 10.7

357 2010 Dec 23 33.6 30.3 33.0 20.1 31.5 28.2 37.4 2.8 -999.0 18.6 10.8 7.4 8.5 6.8 18.7

358 2010 Dec 24 14.6 14.0 16.0 1.9 12.3 10.6 16.3 2.2 -999.0 10.5 5.7 8.0 6.3 5.8 10.6

A-57

JD Yr M D EDMC FDMS

EDME FDMS

MCIN BAM1

MCIN TEOM3

MCIN FDMS

MCIN BAM3

EDMS FDMS

ELKI TEOM4

FTSA TEOM40

FTSA SHARP

GENE TEOM4

LAMO BAM1

REDW TEOM4

TOMA TEOM4

90601 SES

359 2010 Dec 25 38.0 34.8 34.0 17.9 32.9 29.1 38.5 8.2 -999.0 29.7 9.9 22.5 12.0 6.8 29.8

360 2010 Dec 26 32.9 32.7 36.4 17.1 31.7 27.8 37.5 12.1 -999.0 27.4 7.5 23.6 8.7 5.7 27.4

361 2010 Dec 27 14.6 15.2 20.0 9.5 13.5 12.4 17.1 5.2 -999.0 11.0 3.3 11.9 5.8 3.0 11.0

362 2010 Dec 28 10.8 13.0 11.4 4.9 8.2 7.2 11.8 2.7 -999.0 9.3 2.1 8.9 3.8 0.4 9.4

363 2010 Dec 29 8.3 8.2 7.0 6.0 5.7 5.3 9.3 2.1 -999.0 6.4 5.8 5.7 3.3 5.1 6.6

364 2010 Dec 30 19.9 19.8 19.9 14.1 16.9 18.9 24.5 5.3 -999.0 14.6 9.4 17.9 14.0 7.3 14.6

365 2010 Dec 31 15.8 12.5 15.0 7.1 12.9 13.8 19.5 1.8 -999.0 13.3 6.7 7.4 6.6 4.6 13.3

APPENDIX B

Province-Wide Point Source Anthropogenic Emissions Summary Table

B-1

Appendix B. Province-Wide Point Source Anthropogenic Emissions Summary Table

Table B-1. Province-wide emissions inventory categorized by inventory source and final file name.

File Name New File # of

Sources

NOx SO2 NH3 PM2.5 PM10 CO VOC

[t/yr] [t/yr] [t/yr] [t/yr] [t/yr] [t/yr] [t/yr]

2008 ESRD Industrial Survey + 2010 Updates (AAEI2008+)

AAEI2010_CapitalRegion_FIT_v2.ff10 New 419 64,166 75,029 2,200 2,358 4,532 16,301 1,502

AAEI2010_CapitalRegion_FIT_WABAMUN New 1 4,337 6,659 52 144 334 368 44

AAEI2010_CapitalRegion_noFIT.ff10 New 91 477 388 0 12 14 735 84

AAEI2010_NSRP_noCEMA_noCR_FIT.ff10 New 1,220 22,811 36,777 7 264 264 7,081 1,562

AAEI2010_NSRP_noCEMA_noCR_noFIT.ff10 New 634 13,769 11,823 330 957 1,649 12,638 2,519

AAEI2010_SSRP_FIT.ff10 New 1,038 49,640 105,932 3,175 1,769 3,045 14,285 2,808

AAEI2010_SSRP_noFIT.ff10 New 58 580 10,916 0 12 12 341 49

Total 3,460 151,444 240,867 5,712 5,372 9,517 51,381 8,524

CEMA/LAR Database

Existing_stack_emissions_MAJOR.ida Old 399 42,291 115,001 3,438 24,868 1,308

CEMA_existing_gas_edit.ff10 Modified 421 24,074 789 359 21,242 1,469

Total 820 66,365 115,790 0 3,797 0 46,110 2,777

SSRP2008 Inventory

SSRP_noSurvey_Large_confirmed.ff10 Modified 223 51,903 12,851 205 206 38,713 3,146

SSRP_noSurvey_Large_independent.ff10 Modified 80 12,836 1,080 74 75 12,534 2,998

SSRP_noSurvey_medium.ff10 Modified 1,712 18,478 2,752 756 758 25,395 318

SSRP_noSurvey_noLargeVOC_v2.ff10 Modified 76,357 41,091

SSRP_noSurvey_small.ff10 Modified 3,993 217 27 83 83 314 17

EC_noSurvey_agr.ida.IN04KD3.agr Old 13 2 0 20 76 27

EC_noSurvey_comm.ida.IN04KD3.comm Old 1 1 1

EC_noSurvey_egu.ida.IN04KD3.egu Old 8 0 110 405 14 48 35

EC_noSurvey_others.ida.IN04KD3.others Old 101 8,193 4 6 380 596 2,825

ECvoc_noSurvey_agr.ida.IN04KD3.agr Old 5 n/a n/a n/a n/a n/a n/a n/a

ECvoc_noSurvey_comm.ida.IN04KD3.comm Old 2 n/a n/a n/a n/a n/a n/a n/a

ECvoc_noSurvey_egu.ida.IN04KD3.egu Old 2 n/a n/a n/a n/a n/a n/a n/a

ECvoc_noSurvey_others.ida.IN04KD3.others Old 35 n/a n/a n/a n/a n/a n/a n/a

Total 82,532 91,629 16,825 411 1,533 1,844 79,844 47,570

B-2

File Name New File # of

Sources

NOx SO2 NH3 PM2.5 PM10 CO VOC

[t/yr] [t/yr] [t/yr] [t/yr] [t/yr] [t/yr] [t/yr]

NSRP2006 Inventory

Harmonized_CL_no_survey_source_edit.ff10 Modified 5 331 0 0 77 34

peace_river_2010_AAEIedit_v2.ida Modified 36 3,155 1,603 196 0 1,609 0

NSRP2006_noSurvey_Large_confirmedv2.ff10 Modified 180 35,090 432 151 151 31,062 5,715

NSRP2006_noSurvey_Large_independentv2.ff10 Modified 158 24,883 3,614 134 134 27,200 13,672

NSRP2006_noSurvey_mediumv2.ff10 Modified 1,844 19,227 5,937 1,196 1,200 25,672 445

NSRP2006_noSurvey_noLarge_VOCv2.ff10 Modified 41,856 40,117

NSRP2006_noSurvey_smallv2.ff10 Modified 3,473 176 46 91 92 298 18

NSRP2006_noSurvey_CapitalRegion Modified 1,242 21

ORL_alberta_noSurvey_suog Old 60,271 8,687 1,042 1,045 58,492 69,542

FAP_SIA_from_Survey_CAC_AAEIedit.ida Modified 3 101

Harmonized_FAP_SIA_no_Survey_Source_v2edit.ida Modified 31 490 109 6 6 1

EC_Survey_negu_noAAEI.ff10 Modified 53 1,635 92 36 35 364 858 93

ECvoc_noSurvey-NOSSRP-NOCEMA.ida Old 88 n/a n/a n/a n/a n/a n/a n/a

EC_noSurvey-NOSSRP-NOCEMA.ida Old 166 23,046 196 492 897 1,261 31,062

Total 49,138 168,304 20,817 528 3,747 4,253 176,332 129,657

Province-Wide Emissions Inventory Total Total 135,950 477,743 394,299 6,651 14,449 15,614 353,667 188,527

Capital Region Particulate Matter Air Modelling Assessment … · 2016-03-16 · Capital Region...

Documents

Transcript of Capital Region Particulate Matter Air Modelling Assessment … · 2016-03-16 · Capital Region...