Presentation is loading. Please wait.

Presentation is loading. Please wait.

S. Platnick 1, M. D. King 1, J. Riedi 2, T. Arnold 1,3, B. Wind 1,3, G. Wind 1,3, P. Hubanks 1,3 and S. Ackerman 4, R. Frey 4, B. Baum 4, P. Menzel 4,5,

Similar presentations


Presentation on theme: "S. Platnick 1, M. D. King 1, J. Riedi 2, T. Arnold 1,3, B. Wind 1,3, G. Wind 1,3, P. Hubanks 1,3 and S. Ackerman 4, R. Frey 4, B. Baum 4, P. Menzel 4,5,"— Presentation transcript:

1 S. Platnick 1, M. D. King 1, J. Riedi 2, T. Arnold 1,3, B. Wind 1,3, G. Wind 1,3, P. Hubanks 1,3 and S. Ackerman 4, R. Frey 4, B. Baum 4, P. Menzel 4,5, P. Yang 6, R. Pincus 7 1 NASA Goddard Space Flight Center, Greenbelt MD, 2 Laboratoire d’Optique Atmosphérique, USTL, Lille, France, 3 SSAI, Inc., Seabrook MD, 4 U. Wisconsin, Madison, WI, 5 NOAA/NESDIS, 6 Texas A&M, College Station, TX, 6 NOAA/CIRES, Boulder, CO AMS 12th Conference on Atmospheric Radiation Madison, WI 10 July 2006 The MODIS Operational Cloud Optical and Microphysical Products: Overview of the Collection 5 Processing Algorithm

2 MODIS cloud product summary Cloud optical and microphysical retrieval issues and the new Collection 5 algorithm (“collection” refers to a processing stream) Collection 5 examples Outline AMS, Platnick, 10 July 06

3 Pixel level products (Level-2) – Cloud mask. S. Ackerman, R. Frey, U. Wisconsin/CIMSS 1km, 48-bit mask/11 spectral tests, clear sky confidence in bits 1,2 – Cloud top properties: pressure, temperature, effective emissivity. P. Menzel, R. Frey, NOAA-NESDIS & U. Wisconsin/CIMSS 5 km, CO 2 slicing for high clouds, 11 µm for low clouds – Cloud optical & microphysical properties: optical thickness,  effective particle size, r e, water path, thermodynamic phase. Primary r e from 2.1 µm band. M. D. King, S. Platnick, GSFC – IR-derived thermodynamic phase. B. Baum, U. Wisconsin/SSEC – Cirrus reflectance (via 1.38 µm band). B.-C. Gao, Naval Res. Lab Gridded & time-averaged products (Level-3): statistics, histograms, contains all atmosphere products (clouds, aerosol, clear sky aggregations) MOD06, MYD06 MOD35, MYD35 Summary of MODIS Operational Cloud Products AMS, Platnick, 10 July 06

4 Optical/Microphysical Retrieval Issues Critical issues (esp. for global processing): To retrieve or not to retrieve? Cloud thermodynamic phase: liquid water or ice libraries? Ice cloud models. Multilayer/multiphase scenes: detectable? Surface spectral albedo, including ancillary information regarding snow/ice extent. Atmospheric correction: requires cloud top pressure, ancillary information regarding atmospheric moisture & temperature profiles. Cloud-top temperature, ancillary surface temperature: needed for 3.7 µm emission (band contains solar and emissive radiance). 3D cloud effects. AMS, Platnick, 10 July 06

5 Some MOD06 Optical/Microphysical Collection 5 Changes (http://modis-atmos.gsfc.nasa.gov/products_C005update.html) New “Clear Sky Restoral” algorithm implemented after cloud mask (to identify pixels incorrectly identified as cloud or partly-cloudy pixels). Updated cloud phase algorithm (still a difficult problem)! New ice cloud models (Baum et al. 2005). New research-level multilayer cloud flag. Level-3 code separately aggregates single layer and multilayer cloud fraction, as well as single layer retrievals. New MODIS-derived global snow-free land surface spectral albedo maps; snow/ice spectral albedo maps for Antarctica, Greenland; hemispheric average ecosystem-based snow/ice albedo over land and for sea ice; new IGBP ecosystem map. Available for download on Atmosphere team web site. New 1.6-2.1 µm retrievals over ocean and snow/ice surfaces. New pixel-level , r e, WP retrieval uncertainties (baseline) and estimates of uncertainty of L3 means. AMS, Platnick, 10 July 06

6 Example MODIS Data Granule Canadian Fires, MODIS Terra, 7 July 2002 AMS, Platnick, 10 July 06

7 Example MODIS Data Granule Canadian Fires, MODIS Terra, 7 July 2002 sea ice smoke ice cloud true colorSWIR composite

8 Cloud Mask overall conf.“Clear Sky Restoral” cloudy probably cloudy spatial/spectral tests edge detection 250m cloud mask probably clear clear AMS, Platnick, 10 July 06

9 SWIR compositeCloud Mask overall conf.Retrieval process. phase cloudy probably cloudy liquid water ice undetermined probably clear clear AMS, Platnick, 10 July 06

10 Optical Thickness, Effective Radius Retrievals optical thicknesseffective radius (µm) 100 101 1 icewater 402552010 30 101520 icewater AMS, Platnick, 10 July 06

11 Retrieval Uncertainty Estimates AMS, Platnick, 10 July 06 opt. thickness: effective radius: 5002550025 0 0 icewater  c  /  c (%)  r e  / r e (%) Error sources: cal./fwd. model (5%), sfc. albedo(15%), atmo. correction (20% PW c )

12 MODIS Aqua Collection 5, MODIS Aqua Collection 5,  c  /  c (%) liquid water cloud daily aggregation, 1 April 2005 (assumption: correlation between pixels =1) liquid water cloud monthly aggregation, April 2005 (assumption: daily uncertainties uncorrelated) AMS, Platnick, 10 July 06

13 MODIS Aqua Collection 5, MODIS Aqua Collection 5,  r e  / r e (%) liquid water cloud daily aggregation, 1 April 2005 liquid water cloud monthly aggregation, April 2005 AMS, Platnick, 10 July 06

14 Multilayer Flag Multilayers of different phases: disagreement between IR-phase retrieval and phase derived for optical/microphysical retrieval (SWIR bands, cloud mask tests, …). Multilayers of different phases: disagreement between IR-phase retrieval and phase derived for optical/microphysical retrieval (SWIR bands, cloud mask tests, …). General multilayer: 0.94 µm water vapor absorption band. General multilayer: 0.94 µm water vapor absorption band. AMS, Platnick, 10 July 06 liquid water ice undetermined ML, retr’d as water ML, retr’d as ice ML undetermined SWIR composite

15 Fraction of cloudy pixels (all phases) where the Multilayer Flag is set MODIS Aqua Collection 5 Multilayer Cloud Flag, monthly L3, April 2005 AMS, Platnick, 10 July 06

16 Spatially complete “white- sky” albedo in the MODIS 0.86 µm band for four 16- day periods in 2002 (after Moody et al. 2005a). Northern hemisphere multi-year average white-sky spectral snow albedo as a function of selected IGBP ecosystem classifications from 2000-2004 MOD43B3 data (after Moody et al. 2006b). Spectral Surface Albedo Examples

17 MODIS Atmosphere Team Collection 5 status: MODIS Atmosphere Team Collection 5 status: – Aqua reprocessing/forward processing began in early April 2006, reprocessing nearly complete – Terra reprocessing/forward processing will begin after completion of Aqua; reprocessing expected to be completed in Dec. 2006 All Atmosphere Collection 5 products now being archived and distributed via the MODIS production system: LAADS (disk storage archive w/ftp access, search capability, subsetting, etc.). Distribution from Goddard DAAC will be discontinued. All Atmosphere Collection 5 products now being archived and distributed via the MODIS production system: LAADS (disk storage archive w/ftp access, search capability, subsetting, etc.). Distribution from Goddard DAAC will be discontinued. MODIS operational cloud algorithms benefit from instrument’s spectral coverage, spatial resolution, and radiometric/spectral characterization. MODIS operational cloud algorithms benefit from instrument’s spectral coverage, spatial resolution, and radiometric/spectral characterization. Collection 5 enhancements include: clear sky restoral, new ice models, L2 and L3 uncertainty estimates (for a subset of error sources), new surface albedo maps, multilayer flag research product. Collection 5 enhancements include: clear sky restoral, new ice models, L2 and L3 uncertainty estimates (for a subset of error sources), new surface albedo maps, multilayer flag research product. MODIS Atmosphere Team: http://modis-atmos.gsfc.nasa.gov Collection 5 distribution through LAADS: https://ladsweb.nascom.nasa.gov/ Summary AMS, Platnick, 10 July 06


Download ppt "S. Platnick 1, M. D. King 1, J. Riedi 2, T. Arnold 1,3, B. Wind 1,3, G. Wind 1,3, P. Hubanks 1,3 and S. Ackerman 4, R. Frey 4, B. Baum 4, P. Menzel 4,5,"

Similar presentations


Ads by Google