ISSI March 2019 Platnick - International Space Science Institute

NASA Progress in Generating Multi-instrument Imager Cloud Data Records for Climate Analysis and Aerosol Interaction Studies

International Space Science Institute (ISSI)Bern, Switzerland4-8 March 2019

Steven Platnick1, Kerry Meyer1, Robert Holz2, Steven A. Ackerman2, Andy Heidinger3,2, Nandana Amarasinghe4,1, Galina Wind4,1, Chenxi Wang5,1, Benjamin Marchant6,1, Richard Frey2, et al.

1NASA GSFC, 2U. Wisc./CIMSS/SSEC, 3STAR/NOAA, 4SSAI, 5U. MD/ESSIC, 6USRA/GESTAR

▶ NASA* imager cloud data records• Product/algorithm status

o MODIS standard product: Collection 6.1o MODIS/VIIRS continuity products: Version 1o GEO efforts

▶ Dataset uses for ACI studies • Derived datasets• Example ACI analysis/model evaluation

Outline

ISSI, Platnick et al., March 2019

* Independent product generated by NASA CERES team for use in radiation budget algorithm not discussed.

▶ NASA imager cloud data records• Product/algorithm status



Outline


MODIS Atmosphere Team Collection History

Collection Start of Reprocessing MODIS Terra Start of Reprocessing MODIS Aqua6.1 Sept. 2017 (completed Dec. 2017) Dec. 2017 (completed March 2018)6.0 2014 20135.1 2008 20085.0 2005 20054 2002 20023 2001 20021 2000 –


▶ L1B• C6.0: Aqua VNIR spatial “re-registration”, Terra VNIR/SWIR radiometric

corrections (RVS)• C6.1: Terra IR cross-talk corrections, further Aqua/Terra RVS corrections

▶ C6 L2 cloud properties• Cloud mask threshold updates, new 1km Cloud Top Product• Optical/microphysical: numerous updates

o New rad. transfer, ice particle models, phase algorithm, surface ancillary (C5 gap-filled land spectral surface albedo, variable wind speed ocean model). Additional error sources in pixel-level uncertainty calculations. Spectral cloud effective radius retrievals included explicitly. Processing of lower quality “partly cloudy” pixels + failed retrieval info.

C6.0/C6.1 Highlights


0 90E 180 90W 0

-90

-45

0

45

90

0.0

15.0

30.0

0 90E 180 90W 0

-90

-45

0

45

90

0.0

15.0

30.0

[ Annual mean ]

AQUA ! (C51) LIQUID

JUL, 2002 to JUN, 2001

0 90E 180 90W 0

-90

-45

0

45

90

0.0

15.0

30.0

0 90E 180 90W 0

-90

-45

0

45

90

0.0

15.0

30.0

[ Annual mean ]

TERRA ! (C51) LIQUID

JUL, 2000 to JUN, 2001

0 90E 180 90W 0

-90

-45

0

45

90

-20.0

-10.0

0.0

10.0

20.0

0 90E 180 90W 0

-90

-45

0

45

90

-20.0

-10.0

0.0

10.0

20.0

[ %b0 (per decade) ]

AQUA ! (C51) anomaly (LIQUID) TREND (TOTAL)

Time Series, Monthly :JUL, 2002 to JUN, 2010

5% significance level

0 90E 180 90W 0

-90

-45

0

45

90

-20.0

-10.0

0.0

10.0

20.0

0 90E 180 90W 0

-90

-45

0

45

90

-20.0

-10.0

0.0

10.0

20.0


TERRA ! (C51) anomaly (LIQUID) TREND (TOTAL)



C5.1 Cloud Optical Thickness (COT) annual mean, liq. water clouds

MODIS Terra MODIS Aqua

C5.1 COT trend (%/dec) July 2002-June 2010, 5% sig. level

30

0

15

30

0

15

20

-20

-10

10

0

20

-20

-10

10

0


Climate Data Record Challenges: VNIR Calibration

0 90E 180 90W 0

-90

-45

0

45

90

0.0

15.0

30.0

0 90E 180 90W 0

-90

-45

0

45

90

0.0

15.0

30.0

[ Annual mean ]

AQUA ! (C51) LIQUID

JUL, 2002 to JUN, 2001

0 90E 180 90W 0

-90

-45

0

45

90

0.0

15.0

30.0

0 90E 180 90W 0

-90

-45

0

45

90

0.0

15.0

30.0

[ Annual mean ]

TERRA ! (C51) LIQUID

JUL, 2000 to JUN, 2001

0 90E 180 90W 0

-90

-45

0

45

90

-20.0

-10.0

0.0

10.0

20.0

0 90E 180 90W 0

-90

-45

0

45

90

-20.0

-10.0

0.0

10.0

20.0


AQUA ! (C51) anomaly (LIQUID) TREND (TOTAL)



0 90E 180 90W 0

-90

-45

0

45

90

-20.0

-10.0

0.0

10.0

20.0

0 90E 180 90W 0

-90

-45

0

45

90

-20.0

-10.0

0.0

10.0

20.0


TERRA ! (C51) anomaly (LIQUID) TREND (TOTAL)



30

0

15

30

0

15

20

-20

-10

10

0

20

-20

-10

10

0

C5.1 Cloud Optical Thickness (COT) annual mean, liq. water clouds

C6.1 COT trend (%/dec) July 2002-June 2017, 5% sig. level

MODIS Terra MODIS Aqua



18-yr time series, ±60° zonal mean

land

ocean

C5.1: -13%/dec

C5: -6%/dec

C6: -0.5%/dec

C6: -2%/decCO

TCO

T

influence of0.67 µm channelcalibration drift

0.86 µm drift

Terra MODIS Liquid water Cloud Optical Thickness (COT) Trends



AIRS & MODIS Ice COT Trend (2002-2016)

Brian Kahn (left fig. from Kahn et al., ACP, 2018)

AIRS% MODIS%AIRS (IR) MODIS MYD06 (VNIRS/SWIR)

COT Trend/14-yrs


Cloud Fraction, Terra 18-yr time series, ±25° zonal mean over ocean

ocean C6.1: ~0.1%/dec

C5.1C6.0

Clou

d Fr

actio

n influence of IR crosstalk impacting 8.5 µm channel (primarily)


Climate Data Record Challenges: IR Calibration

Cloud Top Height, Terra 18-yr time series, ±60° zonal mean

land

oceanC6.1: ~1%/dec

C6.1: ~0%/dec

C6.0: 4%/dec

C6.0: 5%/dec

influence of IR crosstalk 5-

7 km

3-5

km


Climate Data Record Challenges: IR Calibration


C6: Pixel-level Uncertainty Datasets

Error Sources: L1B calibration (scene-dependent), surface reflectance (land, ocean Cox-Munk wind vector), water cloud eff. variance, atmospheric correction (CTP, reanalysis state vector, O3), 3.7 µm emission (Tc, Tsfc) and solar irradiance, …

–1

central U.S. data granule, Aqua MODIS, 8 June 2014(Fig. 13, Platnick et al., 2017)

COT, CER_2.1 µm, CER_3.7 µm and uncertainties

MODIS CLDPROP (continuity product algorithm) prelim. retrieval,MYD021KM.A_2014033.2040.061_L1B, SE Pacific

WP derived from CER_2.1 & CER_3.7 µm retrievals

WP uncertainty


COTupper COTupperCOTupper

sepa

ratio

n di

stan

ce (k

m)

Probability of MODISIce Phase Detection

Probability of MODISLiquid Phase Detection

Probability of MODIS Multilayer Detection

[PH test off]

C6: Multilayer Detection & Phase: CloudSat/CALIPSO Evaluation

from 1-yr A-Train collocation (2008)

Ben Marchant et al., in progress


Removing cloudy pixels flagged as Multilayer by

MODIS (COT> 4)

All MODIS ice cloud CER retrievals

CC Single Layer IceCC

Multilayer

CER2.1 µm

CER2.1 µm

Most CER artifacts associated with multilayer clouds have been removed

C6: Multilayer Impact on Ice CER2.1µm: CloudSat/CALIPSO Evaluation


C6: Global MODIS Multilayer Detection Probability

B. Marchant et al., in progress


ORACLES region browse imagery: modis-images.gsfc.nasa.gov/oracles/

25 July (206) 2017

K. Meyer et al. (2015)

Research Product: Simultaneous Cloud/Above-Cloud Absorbing Aerosol

ISSI, Platnick et al., March 2019K. Meyer et al. (2015)

Can’t calculation DRE without accurate cloud reflectance

Research Product: Simultaneous Cloud/Above-Cloud Absorbing Aerosol


Fig. 3aFig. 2

Witte et al., GRL, 2018

• “Validation” efforts sometimes find positive retrieved CER biases. Explanations typically focus on retrieval algorithms and cloud heterogeneity. In situ uncertainty regarded as minor.

• CAPS suite (CDP + 2DC/CIP) biased low relative to Phase Doppler Interferometer (PDI) probe from three Sc aircraft campaigns (MACE, POST, VOCALS). No sig. bias wrt MODIS C6 (~0.2 µm).

• CAPS and PDI agree well in small CER conditions, but PDI typically measures larger CER than CAPS and PVM in the presence of broad DSDs, that are associated with larger CER.

Uncertainty: In Situ DSD/CER


Fig. 3c,d

Witte et al., GRL, 2018

CER

PDI–

CER

CAPS

• “Validation” efforts sometimes find positive retrieved CER biases. Explanations typically focus on retrieval algorithms and cloud heterogeneity. In situ uncertainty regarded as minor.

• CAPS suite (CDP + 2DC/CIP) biased low relative to Phase Doppler Interferometer (PDI) probe from three Sc aircraft campaigns (MACE, POST, VOCALS). No sig. bias wrt MODIS C6 (~0.2 µm).

• CAPS and PDI agree well in small CER conditions, but PDI typically measures larger CER than CAPS and PVM in the presence of broad DSDs, that are associated with larger CER.



Discrepancies between Holographic Detector for Clouds (HOLODEC) vs. CDP & 2DC

Fig. 3, Glienke et al., GRL, 2017

GOES-15/LaRC

deff≈ 44 µm (HOLODEC)deff≈ 26 µm (CDP)


C6 Documentation

1 | P a g e

Terra MODIS Collection 6.1 Calibration and Cloud Product Changes Version 1.0 (27 June 2017)

Authored by Chris Moeller and Richard Frey (University of Wisconsin) Calibration (MOD021KM) The Terra MODIS Photovoltaic (PVLWIR) bands 27-30 are known to experience an electronic crosstalk contamination. The influence of the crosstalk has gradually increased over the mission lifetime, causing for example, earth surface features to become prominent in atmospheric band 27, increased detector striping, and long term drift in the radiometric bias of these bands. The drift has compromised the climate quality of C6 Terra MODIS L2 products that depend significantly on these bands, including cloud mask (MOD35), cloud fraction and cloud top properties (MOD06), and total precipitable water (MOD07). A linear crosstalk correction algorithm has been developed and tested by MCST. It uses band averaged influence coefficients based upon monthly lunar views by Terra MODIS and has been adopted for implementation into C6.1 operational L1B processing (Wilson et al. 2017). In stressing test cases from selected time periods of the Terra mission lifetime, the correction algorithm maintains or restores the MODIS PVLWIR band radiances to nominal performance as indicated through comparisons to Aqua MODIS radiances. The Figure 1 example for band 29 shows that the crosstalk influence has increased over time, driving the Terra MODIS C6 brightness temperatures (red) away from the C6 Aqua MODIS brightness temperatures (blue). For crosstalk corrected C6.1 data (green), the Terra and Aqua MODIS brightness temperatures are brought much more closely into agreement.


JANUARY 2017 VOLUME 55 NUMBER 1 IGRSD2 (ISSN 0196-2892)

Mean ice cloud optical thickness and effective radius retrieved from the NASA MODIS imager on Aqua for November 2012. The latest versionof the MODIS cloud optical product, Collection 6 (C6), is shown in comparison with the earlier version (C5).

Platnick et al., 2017




Outline


▶ Direct continuity with MODIS cloud products not feasible• spectral channel differences, pixel size/sampling, relative radiometry

▶ Approach: apply common algorithms to both VIIRS and MODIS using common/near-common subset of spectral channels. • cloud mask and optical properties (MOD06/MOD35 Collection 6

heritage), cloud-top properties (NOAA GOES-R AWG heritage)▶ Status

• In production at U. Wisconsin SIPS• Public distribution from LAADS (ladsweb.modaps.eosdis.nasa.gov) imminent• Need user feedback!

MODIS/VIIRS continuity products


MODIS vs. VIIRS Radiometry

VIIRS ChannelM5

(0.67µm)M7

(0.87µm)M8

(1.24µm)M10

(1.61µm)M11

(2.25µm)

Radiometric Adjustment

(Expected VIIRS TOC/Observed)

vs C6 0.94 0.96 0.98 0.98 0.97

vs C6.1 0.95 0.97 0.99 0.98 0.97

• Large COT biases in initial VIIRS implementation suggested relative radiometric biases between MODIS & VIIRS.

• Multiyear MODIS Aqua/VIIRS match files generated to assess relative radiometry using warm maritime clouds as reflectance target (account for bandpass, atmo. correction, etc.).

• Adjustments applied to VIIRS, final values consistent with aerosol team (Sayer et al., 2017).


2013 2014 2015 2016 2017 20180.0

0.1

0.2

0.3

0.4

0.5

0.6

PF (I

CE)

VIIRS, AQUA & MYD08 (PF ICE and LIQUID) zonal avg [0, 50N][158E, 128W]DATA: APR, 2012 to OCT, 2018 :: AQUAV 1.0dev15, VIIRS 1.0dev18

2013 2014 2015 2016 2017 2018Year

0.0

0.1

0.2

0.3

0.4

0.5

0.6

PF (L

IQUI

D)

VIIRSMODIS AQUAMYD08

2013 2014 2015 2016 2017 20180.0

0.1

0.2

0.3

0.4

0.5

0.6

PF (I

CE)

VIIRS, AQUA & MYD08 (PF ICE and LIQUID) zonal avg [0, 50N][158E, 128W]DATA: APR, 2012 to OCT, 2018 :: AQUAV 1.0dev15, VIIRS 1.0dev18

2013 2014 2015 2016 2017 2018Year

0.0

0.1

0.2

0.3

0.4

0.5

0.6

PF (L

IQUI

D)

VIIRSMODIS AQUAMYD08 2013 2014 2015 2016 2017 2018

400

500

600

700

800

CTP

(hPa

)

VIIRS & AQUA (CTP and CTH) zonal avg [0, 50N][158E, 128W]

DATA: APR, 2012 to OCT, 2018 :: AQUAV 1.0dev15, VIIRS 1.0dev18

2013 2014 2015 2016 2017 2018Year

2

3

4

5

6

7

8

CTH

(km

)


2013 2014 2015 2016 2017 2018400

500

600

700

800

CTP

(hPa

)

VIIRS & AQUA (CTP and CTH) zonal avg [0, 50N][158E, 128W]


2013 2014 2015 2016 2017 2018Year

2

3

4

5

6

7

8

CTH

(km

)


6-yr Time Series, N. Pacific VIIRS continuity MODIS continuity MYD08

2013 2014 2015 2016 2017 20180.6

0.7

0.8

0.9

1.0

CF D

AY

VIIRS, AQUA & MYD08 (CF TOTAL & DAY) zonal avg [0, 50N][158E, 128W]


2013 2014 2015 2016 2017 2018Year

0.6

0.7

0.8

0.9

1.0

CF T

OTA

L

VIIRSMODIS AQUAMYD08Feb. 2014 VIIRS CLDMSK

CTP: 400-800 hPa

Phase Frac. liquid: 0-0.6

CTH: 2-8 km

Phase Frac. ice: 0-0.6

Cloud Fraction: 0.6-1.0

2013 2014 2015 2016 2017 201820

25

30

35

40

2.x µm

RE

(ICE)

(µm

)

VIIRS & AQUA ( 2.x µm RE LIQUID and ICE) zonal avg [0, 50N][158E, 128W]


2013 2014 2015 2016 2017 2018Year

14

15

16

17

18

19

20

2.x µm

RE

(LIQ

UID)

(µm

)


2013 2014 2015 2016 2017 201820

25

30

35

40

2.x µm

RE

(ICE)

(µm

)

VIIRS & AQUA ( 2.x µm RE LIQUID and ICE) zonal avg [0, 50N][158E, 128W]


2013 2014 2015 2016 2017 2018Year

14

15

16

17

18

19

20

2.x µm

RE

(LIQ

UID)

(µm

)


2013 2014 2015 2016 2017 20186

8

10

12

14

COT

(ICE)

VIIRS & AQUA (COT LIQUID and ICE) zonal avg [0, 50N][158E, 128W]


2013 2014 2015 2016 2017 2018Year

6

8

10

12

14CO

T (L

IQUI

D)


2013 2014 2015 2016 2017 20186

8

10

12

14

COT

(ICE)

VIIRS & AQUA (COT LIQUID and ICE) zonal avg [0, 50N][158E, 128W]


2013 2014 2015 2016 2017 2018Year

6

8

10

12

14

COT

(LIQ

UID)


6-yr Time Series, N. Pacific VIIRS continuity MODIS continuity MYD08

2013 2014 2015 2016 2017 20180.6

0.7

0.8

0.9

1.0

CF D

AY

VIIRS, AQUA & MYD08 (CF TOTAL & DAY) zonal avg [0, 50N][158E, 128W]


2013 2014 2015 2016 2017 2018Year

0.6

0.7

0.8

0.9

1.0

CF T

OTA

L

VIIRSMODIS AQUAMYD08Feb. 2014 VIIRS CLDMSK

COT liquid: 6-15

CER_2.x liquid: 14-20µm

COT ice: 6-15

CER_2.x ice: 20-40µm

Cloud Fraction: 0.6-1.0

CLD VIIRS

-90

-45

0

45

90CF

DAY

0.0

0.5

1.0

-90

-45

0

45

90

CTH

(km

)

0

7

14

0 90E 180 90W 0-90

-45

0

45

90

CTP

(hPa

)

202

601

1000

CLD VIIRS - CLD MODIS

CLD VIIRS - MYD08

-0.4

-0.2

0.0

0.2

0.4

-2.0

-1.0

0.0

1.0

2.0

0 90E 180 90W 0

0 90E 180 90W 0

-100

-50

0

50

100

CF-DAY MEAN , CTH, CTP (AQUA 1.0dev15,VIIRS 1.0dev18)

All Monthly Averages are Pixel Weighted Monthly MeansFeb 2014

Production version of common algorithm, includes SW Radiometric Bias CorrectionPixel-weighted multi-day aggregation over common MODIS swathDaytime only

a separate MODIS product run with the common algorithm is required for continuity

CTH

(km

)C

loud

Fra

ctio

nVIIRS VIIRS–MODIS Aqua

common algorithmVIIRS–MYD08


CLD VIIRS

-90

-45

0

45

90

CF M

EAN

TOT

0.0

0.5

1.0

0 90E 180 90W 0-90

-45

0

45

90CO

T IC

E

1 5 10 50

0 90E 180 90W 0-90

-45

0

45

90

COT

LIQ

1 5 10 50


CLD VIIRS - MYD08

-0.2

-0.1

0.0

0.1

0.2

-6.0

-3.0

0.0

3.0

6.0

0 90E 180 90W 0

0 90E 180 90W 0

-6.0

-3.0

0.0

3.0

6.0

CF MEAN TOT, COT(Ice), COT(Liq) (AQUA 1.0dev15,VIIRS 1.0dev18)

All Monthly Averages are Pixel Weighted

CLD VIIRS

-90

-45

0

45

90

RF_1

.6 µ

m

0.0

0.5

1.0

-90

-45

0

45

90RF

_2.x

µm

0.0

0.5

1.0

0 90E 180 90W 0-90

-45

0

45

90

RF_3

.7 µ

m

0.0

0.5

1.0


CLD VIIRS - MYD08

-0.2

-0.1

0.0

0.1

0.2

-0.2

-0.1

0.0

0.1

0.2

0 90E 180 90W 0

0 90E 180 90W 0

-0.2

-0.1

0.0

0.1

0.2

(RF-Liquid) (AQUA 1.0dev15,VIIRS 1.0dev18)All Monthly Averages are Pixel Weighted

(RF_LIQ_grid_npts=0 & RF_ICE_grid_npts=0) masked out from each RF_grid_nptsnum[idx]/deno[idx] where idx = where (RF_LIQ_grid_npts ge 0 & mask_day_grid_npts gt 0)

Monthly MeansFeb 2014

Production version of common algorithm, includes SW Radiometric Bias CorrectionPixel-weighted multi-day aggregation over common MODIS swathDaytime onlyHighest Quality (non-“Partly Cloudy” pixels)

CO

T liq

.R

etrie

val F

ract

ion

liq. VIIRS VIIRS–MODIS Aqua

common algorithmVIIRS–MYD08


6.9

1.9

1.2

5.36.7

1.2: MODIS mean1.4: VIIRS mean

mean:MYD06 Population

CLDPROP Population

19 Feb. 2014 (N. Pacific)

pixel FOV, pixel growth affects retrievals and CSR population


MODIS/VIIRS L2 Cloud Continuity Product Status

65 pp. w/references to C6.1 user guide sections as appropriate

CLDMSK_L2_VIIRS_SNPP VIIRS/SNPP Cloud Mask 6-Min Swath 750mCLDPROP_L2_VIIRS_SNPP VIIRS/SNPP Optical Cloud Properties 6-min Swath 750mCLDPROP_D3_VIIRS_SNPP VIIRS/SNPP Optical Cloud Properties Daily 1°CLDPROP_M3_VIIRS_SNPP VIIRS/SNPP Optical Cloud Properties Monthly 1°CLDMSK_L2_MODIS_AQUA MODIS/AQUA Cloud Mask 5-Min Swath 1kmCLDPROP_L2_MODIS_AQUA MODIS/AQUA Optical Cloud Properties 5-min Swath 1kmCLDPROP_D3_MODIS_AQUA MODIS/AQUA Optical Cloud Properties Daily 1°CLDPROP_M3_MODIS_AQUA MODIS/AQUA Optical Cloud Properties Monthly 1°

• Final version 1 code successfully delivered to SIPS Dec. 2018. Public release imminent.

• User guide draft completed• Filename convention


EOS MODIS and SNPP VIIRS Cloud Properties: User Guide for the Climate Data Record Continuity Level-2 Cloud Top and Optical Properties Product

(CLDPROP)

Version 1 February 2019

CLOUD MASKING TEAM STEVEN A. ACKERMAN6, RICHARD FREY6

CLOUD TOP PROPERTY TEAM ANDREW HEIDINGER2, YUE LI6, ANDI WALTHER6

CLOUD OPTICAL PROPERTY TEAM STEVEN PLATNICK1, KERRY G. MEYER1, GALA WIND3,1, NANDANA AMARASINGHE3,1, CHENXI WANG4,1, BENJAMIN MARCHANT5,1

PRODUCT ASSESSMENT SUPPORT ROBERT E. HOLZ6, STEVEN DUTCHER6, PAUL HUBANKS7,1

1 Earth Sciences Division, NASA Goddard Space Flight Center, Greenbelt, MD 2 NOAA NESDIS/STAR/CIMSS, Madison, WI 3 Science Systems and Applications, Inc., Lanham, MD 4 Earth System Science Interdisciplinary Center, University of Maryland, College Park, MD 5 Goddard Earth Sciences Technology and Research, Universities Space Research Association, Columbia, MD 6 Cooperative Institute for Meteorological Satellite Studies, University of Wisconsin, Madison WI 7 ADNET Systems, Inc., Bethesda, MD




Outline


GSFC GEO Cloud Continuity Research Efforts

▶ Since VIIRS and ABI/AHI have similar spectral bands, a ‘seed’ effort was initiated by NASA to demonstrate porting VIIRS atmosphere algorithms to ABI/AHI. • Leverages U. Wisconsin SIPS capabilities for a limited research

study for both cloud and aerosol algorithm teams; cloud team also leveraging SEVIRI effort (SEV06) at ICARE.

• Initial cloud effort: 2km retrievals, cloud optical retrievals (AWG cloud mask & cloud-top), focused on AHI

• Includes discussions with NOAA STAR and how to work collaboratively on tasks that best leverage agency activities

ICWG-2, Oct. 2018

AHI preliminary NASA cloud

research products15-16 May 2016

ICWG-2, Oct. 2018

COT-liquid CER-liquid 2.24 µm

For overlap region defined by VZA ≤15°, ≤ 5 min (16 May 2016)

AHI

AHI

VIIRS VIIRS

AHI vs VIIRS Retrievals: contiguous cloudy pixels only

ICWG-2, Oct. 2018

J. Riedi, M. Patou, et al. AOGS Annual Mtg., 2017


2019-22-02, North Pacific, GOES-17 2019-15-01, Hawaii, GOES-17




Outline


Droplet Concentration

Bennartz and Rausch, “Global and regional estimates of warm cloud droplet number concentration based on

13 years of AQUA-MODIS observations”, ACP, 2017

!" ~ $%

&'()*+,-.

Droplet Concentration

S. Zeng et al., “Study of global cloud droplet number concentration with A-Train satellites”, ACP, 2014

Bennartz-like algorithm

Yong Hu algorithm (lidar cloud-top depolarization ratio + passive CER).

Doesn’t rely on adiabatic assumption.

Cloud Susceptibility from MODIS C5

Oreopoulos and Platnick (2008)




Outline


0 1 2 3 4 5 60

1

2

3

4

5

6

7

0 1.3 3.6 9.4 23 60 1501100800

680

560

440

310

1800

CF: 84.09 RFO: 3.46CR1

0 1 2 3 4 5 60

1

2

3

4

5

6

7

0 1.3 3.6 9.4 23 60 1501100800

680

560

440

310

1800

CF: 96.69 RFO: 2.99CR2

0 1 2 3 4 5 60

1

2

3

4

5

6

7

0 1.3 3.6 9.4 23 60 1501100800

680

560

440

310

1800

CF: 89.30 RFO: 5.02CR3

0 1 2 3 4 5 60

1

2

3

4

5

6

7

0 1.3 3.6 9.4 23 60 1501100800

680

560

440

310

1800

CF: 92.49 RFO: 3.77CR4

0 1 2 3 4 5 60

1

2

3

4

5

6

7

0 1.3 3.6 9.4 23 60 1501100800

680

560

440

310

1800

Cloud optical thickness

Clo

ud to

p pr

essu

re(m

b)

CF: 86.71 RFO: 3.68CR5

0 1 2 3 4 5 60

1

2

3

4

5

6

7

0 1.3 3.6 9.4 23 60 1501100800

680

560

440

310

1800

CF: 82.24 RFO: 6.99CR6

0 1 2 3 4 5 60

1

2

3

4

5

6

7

0 1.3 3.6 9.4 23 60 1501100800

680

560

440

310

1800

CF: 95.27 RFO: 2.44CR7

0 1 2 3 4 5 60

1

2

3

4

5

6

7

0 1.3 3.6 9.4 23 60 1501100800

680

560

440

310

1800

CF: 86.19 RFO: 4.92CR8

0 1 2 3 4 5 60

1

2

3

4

5

6

7

0 1.3 3.6 9.4 23 60 1501100800

680

560

440

310

1800

CF: 88.86 RFO: 7.62CR9

0 1 2 3 4 5 60

1

2

3

4

5

6

7

0 1.3 3.6 9.4 23 60 1501100800

680

560

440

310

1800

CF: 68.49 RFO: 7.28CR10

0 1 2 3 4 5 60

1

2

3

4

5

6

7

0 1.3 3.6 9.4 23 60 1501100800

680

560

440

310

1800

CF: 51.04 RFO:10.36CR11

0 1 2 3 4 5 60

1

2

3

4

5

6

7

0 1.3 3.6 9.4 23 60 1501100800

680

560

440

310

1800

CF: 29.66 RFO:41.47CR12

0.1 0.2 0.4 0.8 1.5 3 6 10 15 20 25 35Cloud fraction (%)

Cloud Regimes (CRs) from MODIS C6

Oreopoulos et al. JGR (2017)

high

clo

uds

dom

inan

tm

id-le

vel

dom

inan

tLi

quid

/low

clo

uds

dom

inan

t

Relative Frequency of Occurrence of CRs

Variety of CRs can occur in same region =>consider AIE vs. CR

• Morning AOD: 1° grid from MODIS Terra C6 Dark Target or MERRA2, 12 years, 50°S-50°N

• Afternoon Cloud Parameters (Cx):Precip, CF, CTP, COT, CER, CRE

• dlnCx/dlnAOD at each grid cell by CR

• Only statistical sig. sensitivities used to calculate distribution

Lazaros Oreopoulos, Nayeong Cho, et al. (in progress)

ocean landCR ACI Example Regional Results

median/interquartiledistribution

yellow=no statist. sig.

Model Evaluation Example: Aerosol-Cloud Interactions

▶ Ban-Weiss et al. (2014) examined model cloud sensitivity to aerosol optical depth (!a)• MODIS !c , re mapped to N, LWP (with

assumptions)• sensitivity of N to !a with/without

relative humidity (meteorology) influence for model (COSP, fixed SST) compared with MODIS

▶ Models tend to show significantly higher sensitivity than MODIS. Suggests droplet size controlled more by RH than aerosol.

Using proxies to pick apart correlations between aerosols and clouds

−0.4

0.0

0.4Δln(N )/Δln(τa)

∂ln(N )/∂ln(τa)

MODIS AM3 CAM5 ModelE2

Sens

itivi

ty

after Ban-Weiss et al. 2014, 10.1002/2014JD021722 Ban-Weiss et al., Fig. 7

ISSI March 2019 Platnick - International Space Science Institute

Documents

Transcript of ISSI March 2019 Platnick - International Space Science Institute