modis cloud microphysics product (mod pr06od) data ... · test 10: improved ancillary handling test...

1
Introduction Our group has a significant number of science improvements planned for MODIS data collection 6. In this poster we introduce our proposed changes, show some of the available results for changes already implemented and discuss our overall progress. Terra, 07.30.2001 19:00UTC (California) Brief Summary of Planned Changes Improved multilayer cloud detection algorithm that includes two additional tests and stores values as a true confidence level. New forward libraries with expanded ranges of optical thickness and effective radius, based on DISORT with no asymptotic theory. These libraries have improved computationtal resolution that allow us to capture fine features of the phase functions better than before. Single scattering is now added on the fly during retrieval instead of being a part of library reflectance, thus reducing the complexity needed to capture the forward peak. The new library set also includes Cox-Munk ocean BRDF model and new ice crystal phase functions from UW-Madison. We also provide tables of g and !0 values for ice clouds as part of collection 6 MOD06 files so users can rescale our output as needed. Interpolation is now done using more reasonable limits based on scattering angle. A number of improvements to our handling of ancillary data, such as temporally interpolating model atmospheric profiles, new surface albedo libraries, high resolution cloud top properties data. Improvements to our cloud thermodynamic phase algorithm Improved clear sky restoral algorithm with color tests to better distinguish between dust and uniform cloud. Improved retrieval solution logic Improved retrieval quality settings based on our pixel-level retrieval uncertainty results Improvements to our 1.6!m and 3.7!m retrievals, such as decoupling of answers from 2.1!m results, iteration on correction for cloud emissivity, improved atmospheric emission correction and better surface temperature data. Sample Results Many of these planned changes have already been implemented. In this section we show granule test results of new forward libraries. This test does not include Cox-Munk BRDF or new ice crystal phase functions. Science Tests and Future Developments We are currently conducting a series of science tests to examine precise impact of each of our planned changes. We process data one month at a time and aggregate resulting Level-2 files all the way to monthly Level-3. We have very fine control over the test system, so we are able to study each individual change with everything else held constant. At this time we have planned 12 primary science tests, with re-test procedures included. We have completed two science tests already with the third one being prepared for execution. This test system implemented through MODAPS gives us unprecedented depth of analysis of our results. Cloud Optical Thickness Cloud Effective Radius 2.1!m Cloud Effective Radius 1.6!m Cloud Effective Radius 3.7!m Cloud Optical Thickness 1.6-2.1!m Cloud Effective Radius 1.6-2.1!m Before: Operational Collection 5 After: New forward reflectance libraries: Lambertian surface, collection 5 phase functions, new re-interpolation thresholds, different processing direction (along-track) Difference: New - Old 0 10 100 0 10 100 2 16 30 5 32 60 2 16 30 5 32 60 2 16 30 5 32 60 0 10 100 0 10 100 2 16 30 5 32 60 -4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4 -5 -2.5 0 2.5 5 -5 -2.5 0 2.5 5 -5 -2.5 0 2.5 5 -5 -2.5 0 2.5 5 -5 -2.5 0 2.5 5 -5 -2.5 0 2.5 5 -5 -2.5 0 2.5 5 -5 -2.5 0 2.5 5 Liquid Ice Liquid Ice Liquid Ice Liquid Ice Liquid Ice Liquid Ice Liquid Ice Liquid Ice Liquid Ice Liquid Ice Liquid Ice Liquid Ice Approximate Test Schedule 12.30.09 PGE06 v.6.0.2 Test 1: Baseline against PGE06 v.5.12.15 PGE06 v.6.0.5 01.20.10 Test 2: New multilayer cloud detection scheme Test 3: New forward libraries: Lambertian, C5 phase In preparation, ETA 02.10.10 ... Test 4: Test 3 + Cox-Munk Test 5: Next day NISE Test 6: CSR color test Test 7: New solution logic Test 8: New cloud phase Test 9: Improvements to 3.7!m retrieval Test 10: Improved ancillary handling Test 11: New ice libraries Test 12: New surface albedo 02.20.10 02.25.10 03.15.10 04.15.10 05.15.10 06.15.10 07.15.10 08.15.10 09.15.10 MODIS Cloud Microphysics Product (MOD_PR06OD) Data Collection 6 Updates Gala Wind*, Steven Platnick, Michael D. King NASA Goddard Space Flight Center, Greenbelt, MD 20771 *SSAI, Inc. Lanham, MD 20706

Upload: others

Post on 25-Jul-2020

2 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: MODIS Cloud Microphysics Product (MOD PR06OD) Data ... · Test 10: Improved ancillary handling Test 11: New ice libraries Test 12: New surface albedo 02.20.10 02.25.10 03.15.10 04.15.10

IntroductionOur group has a significant number of science improvements planned for MODIS data collection 6. In this poster we introduce our proposed changes, show some of the available results for changes already implemented and discuss our overall progress.

Terra, 07.30.2001 19:00UTC (California)

Brief Summary of Planned Changes– Improved multilayer cloud detection algorithm that includes two additional tests and stores values as a true confidence level.

– New forward libraries with expanded ranges of optical thickness and effective radius, based on DISORT with no asymptotic theory. These libraries have improved computationtal resolution that allow us to capture fine features of the phase functions better than before. Single scattering is now added on the fly during retrieval instead of being a part of library reflectance, thus reducing the complexity needed to capture the forward peak. The new library set also includes Cox-Munk ocean BRDF model and new ice crystal phase functions from UW-Madison. We also provide tables of g and !0 values for ice clouds as part of collection 6 MOD06 files so users can rescale our output as needed. Interpolation is now done using more reasonable limits based on scattering angle.

– A number of improvements to our handling of ancillary data, such as temporally interpolating model atmospheric profiles, new surface albedo libraries, high resolution cloud top properties data.

– Improvements to our cloud thermodynamic phase algorithm

– Improved clear sky restoral algorithm with color tests to better distinguish between dust and uniform cloud.

– Improved retrieval solution logic

– Improved retrieval quality settings based on our pixel-level retrieval uncertainty results

– Improvements to our 1.6!m and 3.7!m retrievals, such as decoupling of answers from 2.1!m results, iteration on correction for cloud emissivity, improved atmospheric emission correction and better surface temperature data.

Sample ResultsMany of these planned changes have already been implemented. In this section we show granule test results of new forward libraries. This test does not include Cox-Munk BRDF or new ice crystal phase functions.

Science Tests and Future DevelopmentsWe are currently conducting a series of science tests to examine precise impact of each of our planned changes. We process data one month at a time and aggregate resulting Level-2 files all the way to monthly Level-3. We have very fine control over the test system, so we are able to study each individual change with everything else held constant. At this time we have planned 12 primary science tests, with re-test procedures included. We have completed two science tests already with the third one being prepared for execution. This test system implemented through MODAPS gives us unprecedented depth of analysis of our results.

Cloud Optical Thickness Cloud Effective Radius 2.1!m Cloud Effective Radius 1.6!m Cloud Effective Radius 3.7!m Cloud Optical Thickness 1.6-2.1!m Cloud Effective Radius 1.6-2.1!m

Before: Operational Collection 5

After: New forward reflectance libraries: Lambertian surface, collection 5 phase functions, new re-interpolation thresholds, different processing direction (along-track)

Difference: New - Old

0 10 100 0 10 100 2 16 30 5 32 60 2 16 30 5 32 60 2 16 30 5 32 60 0 10 100 0 10 100 2 16 30 5 32 60

-4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4-5 -2.5 0 2.5 5 -5 -2.5 0 2.5 5 -5 -2.5 0 2.5 5 -5 -2.5 0 2.5 5 -5 -2.5 0 2.5 5 -5 -2.5 0 2.5 5 -5 -2.5 0 2.5 5 -5 -2.5 0 2.5 5

Liquid Ice Liquid Ice Liquid Ice Liquid Ice Liquid Ice Liquid Ice

Liquid IceLiquid IceLiquid IceLiquid IceLiquid IceLiquid Ice

Approximate Test Schedule

12.30.09

PGE06 v.6.0.2

Test 1: Baseline against PGE06 v.5.12.15

PGE06 v.6.0.5

01.20.10

Test 2: New multilayer cloud detection scheme

Test 3: New forward libraries: Lambertian, C5 phase

In preparation, ETA 02.10.10

...Test 4: Test 3 + Cox-Munk

Test 5: Next day NISE

Test 6: CSR color test

Test 7: New solution logic

Test 8: New cloud phase

Test 9: Improvements to 3.7!m retrieval

Test 10: Improved ancillary handling

Test 11: New ice libraries

Test 12: New surface albedo

02.20.10

02.25.1003.15.10

04.15.10

05.15.10

06.15.10

07.15.1008.15.10

09.15.10

MODIS Cloud Microphysics Product (MOD_PR06OD) Data Collection 6 Updates

Gala Wind*, Steven Platnick, Michael D. KingNASA Goddard Space Flight Center, Greenbelt, MD 20771

*SSAI, Inc. Lanham, MD 20706