2009 Aqua Senior Review

Claire L. ParkinsonAqua Project Scientist

Presented at the AMSR-E Science Team meeting, Arlington, VA, June 25, 2009

2009 Senior Review

• Purpose: Determine which of the reviewed missions warrant continued operation and what are the appropriate funding levels.

• Missions involved: ACRIMSAT, Aqua, Aura, CALIPSO, CloudSat, EO-1, GRACE, ICESat, Jason-1, QuikSCAT, SORCE, Terra, and TRMM (i.e., 13 missions all beyond their prime mission lifetimes).

• Period covered: FY2010 – FY2013.• Proposal requirements: science section, technical/budget section,

appendices providing a data product inventory, budget spreadsheets, references, and a list of acronyms.

• Key Dates:– March 24, 2009, proposals due to NASA Headquarters.– May 12-14, 2009, meeting of the Senior Review Panel (Aqua

presentation May 13).– July 2009, anticipated receipt of the Panel results.

2009 Aqua Senior Review Cover Page and Table of

Contents

Compelling Aspects of the Aqua Proposal• Strong science

– Four productive science teams (AIRS/AMSU/HSB, AMSR-E, CERES, and MODIS).

• Over 600 publications using Aqua data in each of the last two years.• Over 7,000 citations to publications using Aqua data in each of the last

two years.– Significant contributions to each of NASA’s six interdisciplinary Earth

science focus areas (especially the first five). – Role in extending many key climate data records.

• Many contributions to National Objectives/Applied Science– Many Aqua data products being used in weather forecasting.– Contributions to each of NASA’s 12 Applied Science Program Elements.– Value of the Aqua data to the U.S. Forest Service, USDA, EPA, FAA,

NOAA, DOD, USAID, USGS, U.S. Coast Guard, Dartmouth Flood Observatory, Washington Volcanic Ash Advisory Center, Alaska Volcano Observatory.

• Healthy spacecraft and instruments• Enough fuel to last at least through 2018• Cheap to continue relative to the initial investment

May 13, 2009 Meeting with the Senior Review Panel

• Very nice opening and closing compliments from the Panel regarding the Aqua mission and the proposal.

• Presentation centered on 8 sets of questions that the Panel provided ahead of time.

Question Set #1

“ Please present your optimal proposal budget; describe the additional work which will be funded by the optimal budget, the benefits of the added work, and justify the additional budget requested. ”

‘Optimal’ Additions: Mission Operations (ESMO) and AIRS/AMSU

• ESMO: Additional funds are requested for FY10 (only) to complete key security upgrades to the ground systems. (Much of the current ESMO ground systems hardware and operating systems are not compliant with many NASA Information Technology (IT) security policies.)

• AIRS/AMSU: 2.1 FTE/yr are requested to enhance the AIRS CO2 product, to produce retrievals in the stratosphere and lower troposphere.

364 370 376 382

AIRS mid-troposphere CO2 concentrations (ppm) for July 2003 (from Mous Chahine and the AIRS Science Team).

AIRS/AMSU Mid-Tropospheric CO2 ProductSeptember 2002 – July 2008

(animation from Mous Chahine, the AIRS Science Team, and the NASA GSFC Scientific Visualization Studio)

‘Optimal’ Additions: AMSR-E, CERES, and MODIS

• AMSR-E: 0.6 FTE/yr are requested to take pressure off an over-extended Team Leader Science Computing Facility (TLSCF), currently comprised of only 1 full-time and 1 part-time person. The additional 0.6 FTE would allow 2 full-time people.

• CERES: 1.75 FTE/yr are requested to bring the CERES Data Management Team back to its level of staffing prior to funding cuts begun in FY08, increasing the efficiency of CERES data processing.

• MODIS: 0.75 FTE/yr are requested from Aqua and 0.75 FTE/yr from Terra optimal budget for the MODIS Rapid Response effort, to bring to full funding the sole Rapid Response scientist and to add 1.0 FTE for software development (website redesign, vector imager processing, etc.).

ARCTAS image over Alaska, 5/08

Fires in Nepal, 3/12/09

Ash plume, Chaiten volcano, Chile, 5/28/08Dust storm in Iraq, 9/16/08Hurricane Gustav over Jamaica, 8/29/08

Sample Aqua MODIS Rapid Response Imagery

Arctic composite image, 5/5/09

Question Set #2

“ … The AMSU instrument has experienced anomalies … it is unclear how AIRS product quality and accuracy would be affected by the loss of AMSU … Have potential synergies between AMSU (23-90 GHz) and AMSR-E (6.9-89 GHz) been investigated? Could AMSR-E provide a similar role as AMSU in the production of AIRS retrievals? ”

Extent to Which the Loss of AMSU Would Affect AIRS Product Quality

• Several test studies have compared results using the current operational AIRS/AMSU algorithm (Version 5.0) and the AIRS-Only version. Results show only minor degradation with the AIRS-Only version.

Susskind, Blaisdell, Iredell, Molnar 23

GlobalSeptember 6, 2002 January 25, 2003 September 29, 2004

Percent of All Cases Included Layer Mean RMS Temperature ( °C) Layer Mean BIAS

Comparisons for global averages (plots courtesy of Susskind, Blaisdell, Iredell, and Molnar)

Investigation of Synergies Between AMSU and AMSR-E

• AMSR-E would not be able to replace AMSU in the production of AIRS retrievals.– The current AIRS/AMSU retrievals require that all channels (IR and

microwave) observe the same air mass at the same time and viewing angle. AIRS and AMSU have closely coupled scan patterns that allow this.

– AMSR-E has a conical scan (versus the cross-track scanning of AIRS and AMSU) and does not view the same air mass as AIRS.

• AMSR-E data can help the AIRS retrieval in other ways. E.g., SSTs and water vapor from AMSR-E are used for cross-comparisons with AIRS/AMSU.

AIRS/AMSU integrated scan patterns

IR = infraredSST = sea surface temperature

Question Set #3

“ Aqua has two CERES instruments (FM-3, FM-4), as does Terra (FM-1 and FM-2). … Although all other channels continue to work well, the shortwave channel on CERES FM-4 ceased to operate properly in 2005 ... Does this failure affect continued product generation, product accuracy and the full range of variable retrievals? Aqua’s FM-3 is still operational allowing either cross-track or rotating azimuth plane modes, but not both; which of these two modes (cross-track or rotating azimuth) is currently used and does this activity perform synergistically with CERES on Terra? ”

Consequences of the CERES FM-4 SW Channel Failure

• Answers to the Panel’s specific questions:– The failure of the FM-4 shortwave (SW) channel has no effect on

continued product generation, product accuracy, or the full range of variable retrievals.

– Aqua’s FM-3 is operating in cross-track mode.– There is no effect on the synergistic Terra and Aqua measurements.

• Background:– The FM-3 is kept in cross-track mode to ensure continued daily, global

product generation.– The reason for two CERES on Aqua was to allow cross-track

measurements throughout the mission plus to allow angular information from the rotating azimuth plane (RAP) scan mode in order to create new Angular Distribution Models (ADMs) for converting radiances to radiative fluxes. By the time the FM-4 SW channel failed in March 2005, sufficient RAP data had been collected to produce the new ADMs.

Question Set #4

“ There are currently 36 core MODIS products covering a wide range of maturity levels. Most of these products are listed at stage II maturity; there are also two provisional products and eight other products at stage I maturity. Are there specific plans to bring these products to stage III maturity during the proposed mission continuation? ”

Data Product Validation Stages

• Provisional: Product quality is sufficient for use by the general research community, but users are urged to contact the MODIS science team before using the data in publications.

• Stage 1 Validation: Product accuracy has been estimated using a small number of independent measurements obtained from selected locations and time periods and ground-truth/field program efforts.

• Stage 2 Validation: Product accuracy has been assessed over a widely distributed set of locations and time periods via several ground-truth and validation efforts.

• Stage 3 Validation: Product accuracy has been assessed and the uncertainties in the product well established via independent measurements in a systematic and statistically robust way representing global conditions.

MODIS Data Product Maturity, p. 1

MODIS Data Products

Current Maturit

yExpecte

dWhe

n CommentsMODIS Level-1 (radiance and reflectance) products: (see also http://www.mcst.ssai.biz/mcstweb/index.html)MODIS Level-1A Radiance Counts (MYD 01) 3 3 NowMODIS Level-1B Calibrated Geolocation Data Set (MYD 02) 3 3 NowMODIS Geolocation Data Set (MYD 03) 3 3 Now

MODIS Atmosphere Products (see also http://modis-atmos.gsfc.nasa.gov/products.html )

MODIS Aerosol Product (MYD 04) 2 2 NowFunded via

ROSES

MODIS Total Precipitable Water (MYD 05) 2 2 NowFunded via

ROSES

MODIS Cloud Product (MYD 06) 2 2 NowFunded via

ROSES

MODIS Atmospheric Profiles (MYD 07) 2 2 NowFunded via

ROSES

MODIS Level-3 Atmosphere Product (MYD 08) 2 2 NowFunded via

ROSES

MODIS Level-2 Atmosphere Joint Product (MYD ATML2) 2 2 NowFunded via

ROSES

MODIS Cloud Mask (MYD 35) 2 2 NowFunded via

ROSES

MODIS Land Products (see also http://modis-land.gsfc.nasa.gov/)MODIS Surface Reflectance; Atmospheric Correction Algorithm Products (MYD 09) 2 2 Now

Funded via ROSES

MODIS Land Surface Temperature and Emissivity (MYD 11) 2 2 NowFunded via

ROSESMODIS Land Cover Product (MCD 12): Land Cover/Land Cover Dynamics 2/P 2 2011

Funded via ROSES

MODIS Vegetation Index Products (NDVI and EVI) (MYD 13) 2 2 NowFunded via

ROSES

MODIS Thermal Anomalies - Fires and Biomass Burning (MYD 14) 2 2 NowFunded via

ROSES

MODIS FPAR/LAI (MYD 15) 2 2 NowFunded via

ROSES

MODIS Evapotranspiration (MYD 16) P 1 2011Funded via

ROSESMODIS Gross Primary Productivity (GPP)/Net Primary Productivity (NPP) (MYD 17) 2 2 Now

Funded via ROSES

MODIS BRDF/Albedo Parameter (MCD 43) 1 2 2011Funded via

ROSESMODIS Burned Area Products (MCD 45) (new product added after 2007 Senior Review) P 2 2011

Funded via ROSES

MODIS Data Product Maturity, p. 2

MODIS Cryosphere Products (see also http://modis-land.gsfc.nasa.gov/)MODIS Snow Cover (MYD 10): Fractional Snow Cover/Snow Albedo 2/1 2/1 Now Funded via ROSESMODIS Sea Ice (MYD 29): Extent/Ice Temperature 2/1 2/1 Now Funded via ROSES

MODIS Ocean Products (see also http://oceancolor.gsfc.nasa.gov/)MODIS Sea Surface Temperature (MYD 28) 3 3 Now Funded via ROSESMODIS Normalized Water-leaving radiances 2 2 Now Funded via ROSESMODIS Sub-surface Chlorophyll-a concentration 2 2 Now Funded via ROSESMODIS K490 Diffuse attenuation 2 2 Now Funded via ROSESMODIS Aerosol Optical Thickness (for ocean color atmospheric correction) 2 2 Now Funded via ROSESMODIS Aerosol Angstrom exponent 2 2 Now Funded via ROSESMODIS Aerosol Single-Scattering Epsilon 2 2 Now Funded via ROSESMODIS Particulate Organic Matter (POC) (new product added after 2007 Senior Review) P 1 2011 Funded via ROSESMODIS Calcite (new product added after 2007 Senior Review) P 1 2011 Funded via ROSESMODIS Fluorescence Line Height (FLH) (new product added after 2007 Senior Review) P 1 2011 Funded via ROSES

Other MODIS Products either not part of the "core product" list or under development.

Polar Winds 3 3 NowNot funded by

NASAMODIS Disturbance Index (DI) (not a standard product) 1 2 2012 Funded via ROSES

MODIS Data ProductsCurrent Maturity

Expected When Comments

Question Set #5

“ The AMSR-E and CERES core product algorithms are listed in the proposal as mature; does this mean that the resulting products have reached a level of maturity that requires algorithm maintenance only? Can more detailed information be provided regarding the level of product (not algorithm) maturity, including the specific level of maturity (stage I-III) for each of the core products, similar to what is presented for MODIS and AIRS/AMSU/HSB? ”

AMSR-E Data Product Maturity

• Maturity levels for each of the AMSR-E data products are listed in the table to the right.

• Even with Stage 3 maturity, algorithm refinements continue to be made (i.e., we are not at a stage of algorithm-maintenance only).

CERES Data Product Maturity

• All CERES Edition 2 data products are Stage 3 validated. Validation activities are documented in Data Quality Summaries, which are updated as appropriate.

• No algorithm changes are made within a particular edition of the data (e.g., Edition 2). Hence Edition 2 data products require algorithm maintenance only.

• However, algorithm development continues for Edition 3.– Edition 2 validation activities have provided information to help

guide algorithm improvements for Edition 3.– Edition 3 reprocessing will start in 2009 and will be used for

processing CERES Terra, Aqua, NPP, and NPOESS data.

NPP = NPOESS Preparatory ProjectNPOESS = National Polar-orbiting Operational Environmental Satellite System

Question Set #6

“ CERES is a PI instrument ... The other Aqua instruments are Facility instruments, where most members of the AIRS/AMSU, AMSR-E and MODIS teams are selected and funded through ROSES. However, the minimum staffing request (60 FTE/yr) for CERES is roughly twice the level requested for MODIS. Why is there such a difference? ”

Why CERES Requires More Staff from Senior Review Funding than MODIS Does

• With CERES being a PI instrument, the CERES Senior Review budget includes support for an integrated instrument-algorithm-validation science team that provides all data products (Levels 1-3) with the highest level of data fusion of any EOS instrument team. The CERES effort combines 11 instruments on 7 spacecraft, all integrated to obtain climate accuracy. The effort includes ~ 1.7 million lines of QC and validation code and 0.75 million lines of production code.

• In contrast, MODIS is a Facility instrument, and the MODIS Senior Review budget includes funding only for the MODIS Characterization Support Team, Science Data Support Team, and MODIS Administrative Support Team. MODIS Level 2 and 3 algorithm improvement and validation are funded separately through ROSES.

QC = Quality control

CERES Workforce Breakdown and History

• Workforce breakdown– 25% for source code and optimization efforts, completing code

migration from an old SGI to a new IBM Linux cluster environment.– 45% for algorithm and code maintenance and improvement and

support of data product deliveries.– 30% for validation, QC, and documentation of the data products.

• Workforce history, actual and projected, 2003-2012

Question Set #7

“ The budget request for AIRS/AMSU for science algorithm development and processing is significantly higher (9.8 FTE/yr) than other Aqua instruments. Why isn’t this covered through ROSES like the other instruments (except CERES)? The AIRS/AMSU team is requesting an additional 2.1 FTE/yr to enhance the utility and achieve quality retrievals of CO2 in the stratosphere and boundary layer; how is this request is different from the proposed science algorithm development activities? ”

ROSES vs. Senior Review Funding for AIRS/AMSU Science Algorithm Development and Processing

• Despite the terminology used, the AIRS Project is in general not developing data product algorithms. As with MODIS and AMSR-E, the algorithm development is being done through ROSES.

• The relevant funding for the AIRS Project is for:– Science integration, testing, and validation. Total: 4.6 FTE/yr.– Software and data operations. JPL integrates software received

by the science team into a single processing system, etc. Total: 5.2 FTE/yr.

• AIRS staffing levels have decreased by 50% since launch.

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

FY02 FY03 FY04 FY05 FY06 FY07 FY08 FY09 FY10 FY11 FY12 FY13

AIRS FTE's Actuals and Planned Since Launch

Further reduction in 2010 planned

AIRS CO2 Optimal vs. In-Guide Efforts

• Maintenance of the mid-tropospheric AIRS CO2 product is part of the in-guide budget.

• The Optimal budget includes funding for the development of corresponding CO2 products for the stratosphere and near surface. Proposed activities include:– Channel selections.– Retrieval optimization.– QC algorithms.– Software development.– Testing and diagnostics.– Validation.– Product generation.– Documentation (GES DISC, journal publications).– User support.

Question Set #8

“ There are large differences in the number of publications, data users, and data volume downloads between MODIS and other Aqua sensors. There are many potential reasons for these differences, including the broad range of products and user communities addressed by MODIS. However, are there other reasons for these differences, such as other sensor products being intrinsically more difficult to analyze than MODIS and reduced software support for these other products? ”

Reasons Why MODIS Has More Publications, Data Users, and Downloads than the Other Aqua Sensors

• The MODIS science team is by far the largest.• The MODIS publication numbers cover Terra as well as Aqua.• The Terra MODIS has been operating since February 2000,

whereas no AIRS or AMSR-E was in space until May 2002.• The CERES science team is focused on long-term climate

information, whereas the MODIS team also has a considerable focus on much shorter-term events.

• MODIS has the highest resolution of the Aqua instruments, allowing local details that have generated great interest and usage, e.g., for imagery of forest fires, dust storms, and phytoplankton blooms.