Early Results from the MODIS Cloud Algorithms

1
Early Results from theMODIS Cloud Algorithms
cloud detection optical, microphysical, and cloud
top properties
S. Platnick5,2, S. A. Ackerman1, M. D. King2, W.
P. Menzel3,1, B. A. Baum4,1, et al.
1 U. Wisconsin/CIMSS, 2 NASA GSFC, 3 NOAA NESDIS,
4 NASA LaRC, 5 UMBC/JCET
AGU Fall Meeting San Francisco, CA December
19, 2000
2
Outline
MODIS (MODerate resolution Imaging
Spectroradiomter)

• MODIS - a quick introduction
• MODIS cloud products
• Algorithm descriptions and example retrievals
• Data status

S. Platnick, AGU, 19 Dec. 2000
3
MODIS instrument highlights

• Filter radiometer, 4 detector arrays, 36 spectral
  bands 250 m, 500 m, 1 km spatial resolution
• Onboard calibration via Solar Diffuser/Stability
  Monitor, Spectral Radiometric Calibration
  Assembly instruments
• First light on 24 February 2000 (Terra launch 18
  Dec 1999)
• Science teams organized into atmosphere, land,
  and ocean discipline groups

S. Platnick, AGU, 19 Dec. 2000
4
MODIS Atmosphere Global 19 April 2000 L1B True
color RGB (0.65, 0.56, 0.47 µm bands)
example data granule coverage (5 min)
S. Platnick, AGU, 19 Dec. 2000
5
MODIS cloud products, granule level

• Pixel level products (Level 2) overview
• Cloud mask for determining clear-sky
• Cloud top properties
• Cloud optical, microphysical properties
  (optical thickness, effective particle size,
  water path, thermodynamic phase, cirrus
  reflectance)
• Unique aspects
• New algorithms from greater spectral coverage
• Heritage algorithms at higher spatial
  resolution
• Includes QA (processing and assessment info)

S. Platnick, AGU, 19 Dec. 2000
6
MODIS cloud products, global composites

• Gridded time-average products (Level 3)
• Daily, 8 day, monthly composites containing
  all atmosphere products
• 1 x 1 equal angle grid
• mean, standard deviation, marginal and
• joint probability distributions
• quick look available on the web
• details at the MODIS atmosphere web page
• http//modis-atmos.gsfc.nasa.gov

S. Platnick, AGU, 19 Dec. 2000
7
Cloud mask(S. Ackerman, R. Frey, K. Strabala
U. Wisconsin/CIMSS)

• Bottom of the algorithmic food chain, input to
  all MODIS products.
• 1 km nadir spatial resolution day night, (250 m
  day)
• 17 spectral bands (0.55-13.93 µm, incl. 1.38 µm)
• 11 spectral tests (function of 5 ecosystems)
• temporal consistency test over ocean, desert
  (nighttime)
• spatial variability test over ocean
• 48 bits per pixel including individual test
  results and processing path generation of clear
  sky maps
• Bits 1,2 give combined test results as confident
  clear, probably clear, uncertain,
  obstructed/cloudy (clear sky conservative)

S. Platnick, AGU, 19 Dec. 2000
8
Cloud mask, cont.

• Spectral tests use fuzzy thresholds, examples
  include
• low cloud tests
• 3.9 µm - 11 µm BT
• high cloud tests
• 13.9 µm (CO2), 1.38 µm (H2O), 11-12 µm BT
• 1.6 µm snow/ice test
• NIR/VIS reflectance test IR tests
  (dependent on sfc emissivity, PW, aerosols) et
  al.

Ackerman, S. A. et al. 1998 JGR, 103,
32141-32157.
S. Platnick, AGU, 19 Dec. 2000
9
Cloud mask, validation activities

• Mask consistent with radar/lidar cloud boundary
  measurements at Oklahoma ARM CART site and ER-2
  observations during spring 2000 campaign
  (including correct snow identification).
• Improvements being made for sun glint, warm cloud
  in arid ecosystems, Antarctica, nighttime low
  cloud over land, nighttime snow/ice surfaces
• Regional/global validation is ongoing. Gobal
  clear sky composite images being used to identify
  problem regions.

S. Platnick, AGU, 19 Dec. 2000
10
1.6 µm image
0.86 µm image
11 µm image
3.9 µm image
cloud mask
Snow test (impacts choice of tests/thresholds)
VIS test (over non-snow covered areas)
11 - 12 BT test (primarily for high cloud)
13.9 µm high cloud test (sensitive in cold
regions)
3.9 - 11 BT test for low clouds
aa
MODIS cloud mask example (confident clear is
green, probably clear is blue, uncertain is red,
cloud is white)
11
MODIS 5-8 September 2000 Band 31 (11.0 µm)
Daytime Clear sky Brightness Temperature
12
MODIS 5-8 September 2000 Band 1, 4, 3 (R/G/B)
Daytime Clear sky Reflectance Composite
13
Cloud top properties(P. Menzel, R. Frey, K.
Strabala, L. Gumley, et al. NOAA NESDIS, U.
Wisc./CIMSS)

• Cloud top pressure, temperature, effective
  emissivity
• Retrieved for every 5x5 box of 1 km FOVs, when
  at least 5 FOVs are cloudy, day night
• CO2 Slicing technique (5 bands, 12.0-14.2 µm)
• ratio of cloud forcing in 2 nearby bands
• retrieve pc Tc from temperature profile
• most accurate for high and mid-level clouds
• Previously applied to HIRS (NOAA POES, 20 km).
  MODIS 1st satellite sensor capable of CO2 slicing
  at high spatial resolution.

S. Platnick, AGU, 19 Dec. 2000
14
Cloud top properties, validation activities

• Activities proceeding via early ER-2 effort
  (March 2000 with lidar and HIS IR
  interferometer), and NOAA HIRS intercomparisons
• Cloud top pressure compares well with HIRS and
  aircraft validation,better than 50 mb.

Frey, R. A. et al, 1999 JGR , 104, 24547-24555.
S. Platnick, AGU, 19 Dec. 2000
15
CO2 slicing

• Technique
• - ratio of cloud forcing at two near-by
  wavelengths
• - effective emissivity includes cloud fraction
  in 5x5 box
• - actual cloud emissivity assumed same for each
  band
• - radiance gradient used when clear sky not
  available
• The more absorbing bands are more sensitive to
  high clouds, weighting functions

Frey, R. A. et al, 1999 A comparison of cloud
top heights computed from airborne lidar and MAS
radiance data. J. Geophys. Res., 104, 24547-24555.
S. Platnick, AGU, 19 Dec. 2000
16
Cloud Mask MODIS 12 March 2000, 1730 UTC
snow
ARM CART site
MODIS band 31 11 µm
Cloud Mask
cleargreen cloudwhite uncertainred
17
Cloud Top Pressure MODIS 12 March 2000, 1730 UTC
ARM CART site
MODIS band 31 11 µm
Cloud top pressure
900-1000 mbpurple 500-600 mbblue 300-400 mbred
18
Comparison of ER2 lidar (nadir view), HIRS (3 hrs
later), RAOB, MODIS Cloud Properties over ARM
CART site, Oklahoma
lidar
CTP vs. HIRS
effective emissivity vs. HIRS
19
MODIS 5-8 September 2000
Cloud top Pressure
Cloud top Temperature
Cloud Fraction
Cloud Effective Emissivity
20
IR thermodynamic phase algorithm (B. Baum, S.
Ackerman, K. Strabala NASA LaRC, U.W. CIMSS)

• Tri-spectral method, 5 km resolution
• NIR, MWIR reflectance technique being developed

ice cloud April 1996 Success
water cloud Jan 1993 TOGA/ COARE
Strabala, Menzel, and Ackerman, 1994, JAM, 2,
212-229. Baum et al, 2000, JGR, 105, 11781-11792.
S. Platnick, AGU, 19 Dec. 2000
21
MODIS cloud thermodynamic phase - IR algorithm
VIS
IR window
cloud phase
Ice
Mixed Phase
Clouds over Southern India 19 April 2000
Water
Uncertain
22
MODIS 5-8 September 2000 IR retrieval - percent
liquid water
23
MODIS 5-8 September 2000 IR retrieval - percent
ice water
24
MODIS IR phase retrieval vs. Cloud Top
Temperature frequency of ice phase Tc lt 253
K statistics from 5 Sept day and night,
60ºN-60ºS, water surface only
frequency ()
25
MODIS IR phase retrieval vs. Cloud Top
Temperature frequency of ice phase 253lt Tclt
273 K statistics from 5 Sept day and night,
60ºN-60ºS, water surface only
frequency ()
26
MODIS IR phase retrieval vs. Cloud Top
Temperature frequency of ice phase Tcgt 273
K statistics from 5 Sept day and night,
60ºN-60ºS, water surface only
frequency ()
27
Cloud optical, microphysical properties(M. D.
King, S. Platnick, M. Gray, E. Moody, J. Li,
S.-C. Tsay, et al. NASA GSFC, UMBC)

• Optical thickness, particle size (effective
  radius), water path
• 1 km spatial resolution, daytime only, liquid
  water and ice clouds
• Land, ocean, snow/sea ice surfaces
• Solar reflectance technique, VIS through MWIR
  (0.65, 0.86, 1.2, 1.6, 2.1, 3.7 µm)
• MODIS 1st satellite sensor with all useful
• SWIR, MWIR bands

S. Platnick, AGU, 19 Dec. 2000
28
Cloud optical, microphysical properties, cont.

• Required input
• cloud mask (tuned for cloudy not clear using
  individual cloud mask tests)
• cloud top temperature for 3.7 µm retrieval
• cloud top pressure for atmospheric correction
  (being implemented)
• cloud phase (currently derived from individual
  cloud mask tests, not IR or solar tests)
• surface albedo (currently assigned from IGBP
• ecosystem map NISE snow/ice data set)
• Early validation effort as part of SAFARI 2000

S. Platnick, AGU, 19 Dec. 2000
29
MODIS SAFARI granule RGB composite 13 September
2000, 0925 UTC
Angola
Zambia
Botswana
Namibia
Angola
South Africa
ER-2 ground track
marine stratocumulus
Etosha Pan
Namibia
30
(No Transcript)
31
(gt99)
(gt95)
(gt66)
32
(No Transcript)
33
(No Transcript)
34
(No Transcript)
35
(No Transcript)
36
(No Transcript)
37
(No Transcript)
38
Data summary

• Atmosphere L2 products processed on P.I. system
  at GSFC (except cloud mask)
• Cloud products archived at GSFC DAAC
• series starts at 8/20/00 for L2, 10/31/00 for
  L3(1d)
• Current archived products are beta release
• early data product, useful for familiarity
  with data formats/parameters, minimal validation,
  temporary
• Consistent processing time series (instrument
  bias settings, L1B algorithm) underway
• Order through EOS Data Gateway
• - details at http//modis.gsfc.nasa.gov/DATA/

S. Platnick, AGU, 19 Dec. 2000
39
Data summary, cont.

• Data file info
• MOD35 cloud mask 48 MB/granule daytime
• MOD06 cloud product 65 MB/granule daytime,
  16 MB nighttime, 12 GB/day
• - L3 atmospheres 400-800 MB/day
• - L1B 340 MB/granule, 70 GB/day
• L2 production system limits
• currently running at approximately 1x, not
  sufficient for reprocessing needs

S. Platnick, AGU, 19 Dec. 2000
40
L3 optical thickness (liquid water) statistics,
10/2/00, from atmo web page
mean
maximum
minimum
standard deviation
S. Platnick, AGU, 19 Dec. 2000
41
Algorithms summary

• MODIS provides an unprecedented opportunity for
  cloud and other atmospheric studies
• 36 spectral channels, high spatial resolution
• Comprehensive set of cloud algorithms
• Archive of pixel level retrievals, global
  statistics
• Product intercomparison for small number of
  selected day(s) proven useful
• Validation activities ongoing (gnd. based, in
  situ, aircraft, satellite intercomparisons,
  etc.) detailed plans on atmosphere web site

S. Platnick, AGU, 19 Dec. 2000