Global cloud observations based on three decades of AVHRR measurements

Martin Stengel, Rainer Hollmann, Karl-Göran Karlsson, Jan Fokke Meirink

ISCCP at 30, City College of New York, NY, April 2013

2

Outline

• AVHRR-based datasets (including CM SAF CLARA-A1)

• Evaluating the accuracy of AVHRR cloud retrievals

• Sampling issue

• COSP type satellite simulator

• Aerosols and clouds

• Summary

CLARA-A1

• NWC SAF Polar Processing System (PPS) software package for CFC, CTH, CTP, CTT (Dybbroe et al., 2005a, 2005b)

• Cloud Physical Properties (CPP) software for COT, LWP, IWP, CPH (Roebeling et al., 2006)

• AVHRR-GAC of all NOAAs and MetOp, 1982 – 2009, global coverage on 0.25° lat/lon• VIS: Recalibrated visible reflectances provided by NOAA (Heidinger et al., 2010). • IR: unchanged (only onboard BB calibration)• Daily and monthly means, 1d/2d histograms

CM SAF cLoud, Albedo and RAdiation dataset, AVHRR-based, v1(Karlsson et al, 2013)

3

4

AVHRR based cloud climatologies• Advantages:

• Global coverage with similar viewing geometries (Tropics tendentially with larger SZAs)• 5 spectral channels back to 1982 (this is at the cost of spatiotemporal coverage)

• Disadvantages• Need for inter-calibration of VIS channels• Sparse temporal sampling (i.e. in the 80’s and early 90’s) • Satellite drifting

• Example datasets:• Patmos-X v6 (AVHRR Pathfinder Atmospheres - Extended version 6),

Heidinger et al. (2009, 2012)• CLARA-A1 (CMSAF cLoud, Albedo and Radiation – AVHRR based version 1),

Karlsson et al. (2013)• Cloud CCI (Climate Change Initiative – on ECV clouds, ORAC alg.)

Foster and Heidinger (2012)

5

Accuracy of AVHRR-based cloud properties

• Validating AVHRR-based cloud retrievals (of existing or planned datasets) in a common framework• CM SAF: EUMETSAT Satellite Application Facility on Climate

Monitoring; CPP (Cloud Physical Properties; Roebeling et al., 2006) algorithm developed at KNMI and PPS (Polar Platform System; Dybbroe et al., 2005a; Dybbroe et al. 2005b) developed at SMHI

• ORAC: (Oxford RAL Retrieval of Aerosol and Cloud) algorithm (Poulsen et al., 2010 and Watts et al., 1998)developed at Oxford University and Rutherford Appleton Laboratory (RAL)

• CLAVR-X: Cloud from AVHRR Extended processing scheme hosted at NOAA at University of Wisconsin (Pavolonis et al. ,2005; Walther et al., 2012)

AVHRR/NOAA18 MODIS/AQUA (1.6mic)

MODIS/AQUA (3.7mic)

Calipso CMa, CTH, CPH CMa, CTH, CPH CMa, CTH, CPH

AMSR-E LWP LWP LWPDARDAR IWP IWP IWP

Cloud detection & height assignment

6

• Sensitivity of clouds detection to cloud optical depth

• Sensitivities of cloud height assignment to cloud optical depth

We start to systematically miss clouds with COT<0.3 for passive imagers (Karlsson et al., 2013)These are global values; there is certainly a scene dependence

(vs. CALISPO)

(vs. CALISPO)

7

Detecting thermodynamic phase• Phase validation against CALIPSO, Liquid cloud occurrence

IR vs. VIS/NIR information gives different phase retrievals due to different penetration depths

CALIPSO phase vs. AVHRR CTT shows interesting features

Integrated quantities - Cloud water paths• Liquid water path against AMSR-E

• Ice water path against DARDAR (Delanoë and Hogan (2008, 2010))

also see Eliasson et al. 2013 (JGR)

All results in Stengel et al. (2013)

Recurring validation/evaluation efforts of this kind helps us to identify strengths and weakness of these datasets and helps us to understand which application are meaningful

Comparing different datasets

9

CM SAF CLARA-A1

PATMOS-x

ISCCP

10

CLARA-A1 sampling issue• Significant diurnal cycle in cloud properties create trends when satellites

drift (or jumps at transitions of different satellites)• Correcting for imperfect sampling of diurnal cycle (e.g. Devasthale et al, 2012; Foster and

Heidinger, 2012)• For potential climate model evaluation satellite simulators will solve this problem in this

contextFoster and Heidinger (2012)

11

CLARA-A1 long-term stability• Cloud fraction vs. SYNOP

(which reliably reported over the full period)

(GAC ALL = CLARA-A1)

SYNOP cloud fraction seems to be stable over this period

Decrease in CLARA-A1 also visible in afternoon satellite time series

COSP type CLARA simulator• Currently being developed in CM SAF, including specific retrieval

considerations• CTP-COT 2D histograms CLARA vs. EC-EARTH (AMIP, SST prescr.) - JAN 1999

12

CLARA-A1 ECEARTH CLARA simulated

Courtesy of Joseph Sedlar

(Temporal sampling of AVHRR not included so far)

Can we detect impact of aerosols?

13

• Is there any detectable signal of liquid cloud occurrences for different cloud top temperature?(PATMOS-x version 6, 2 years of NOAA15, NOAA18, MetOp-A data, daylight, ocean only)

-20°C

Liq

uid

clo

ud

fra

ctio

n

-25°C

-30°C

Zonal average of liquid cloud fractionfor different cloud top temperature

Global liquid cloud occurrence relative to all clouds

Very preliminary results!

more aerosol

less aerosol

14

Summary• CM SAF has recently jointed the international effort of generating AVHRR-

based long-term cloud property dataset (CLARA-A1)• Taking AVHRR-only gives the advantage of enhanced spectral information at

the cost of temporal coverage• Intercomparison study done, revisiting the accuracy of AVHRR-based retrieval

schemes• Differences among the results of the retrieval schemes are partly large

(might partly be large than the difference to products of other sensors)• Outcomes of this has been fed back to retrieval developers, already initialized further

developments (will always be a never ending process)• Cloud cover stability (Trend in CLARA-A1 CFC not confirmed by SYNOP)• CLARA-A1 simulator results• Quality of AVHRR-based retrievals have improve and enable new applications

(E.g. looking at the aerosol impact?)• Uncertainty estimates? Few schemes provide this pixel-based information (but

based on different sources. How do we include this information into higher level products (monthly means/histograms)?

15

Thank you

16

• Choi, Y.-S. et al., 2010: Space observations of cold-cloud phase change, PNAS, vol. 107, issue 25, pp. 11211-11216• Devasthale, A. et al. 2012: Correcting orbital drift signal in the time series of AVHRR derived convective cloud

fraction using rotated empirical orthogonal function, Atmos. Meas. Tech., 5, 267-273• Doutriaux-Boucher, M. and J. Quaas, 2004: Evaluation of cloud thermodynamic phase parametrizations in the

LMDZ GCM by using POLDER satellite data, Geophys. Res. Lett., 31, L06126, doi:10.1029/2003GL019095. • Dybbroe, A. et al., 2005a: NWCSAF AVHRR cloud detection and analysis using dynamic thresholds and radiative

transfer modeling - Part I: Algorithm description, J. Appl. Meteor, 44, pp. 39-54.• Dybbroe, A. et al., 2005b: NWCSAF AVHRR cloud detection and analysis using dynamic thresholds and radiative

transfer modeling - Part II: Tuning and validation, J. Appl. Meteor, 44, 55-71.• Foster M.J. and A. Heidinger, 2012: PATMOS-x: Results from a Diurnally-Corrected Thirty-Year Satellite Cloud

Climatology, J. of Climate, 26, 414-425 • Heidinger, A.K. et al., 2010: Deriving an inter-sensor consistent calibration for the AVHRR solar reflectance data

record. Int. J. Rem. Sens., 31(24), 6493-6517• Karlsson, K.-G. et al. 2012: CLARA - The CMSAF cloud and radiation dataset from 28 years of global AVHRR data

(in preparation).• Mittaz, P.D. and R. Harris, 2009: A Physical Method for the Calibration of the AVHRR/3 Thermal IR Channels 1:

The Prelaunch Calibration Data. J. Atmos. Ocean. Tech., 26, 996-1019, doi: 10.1175/2008JTECHO636.1 • Roebeling, R.A. et al., 2006, Cloud property retrievals for climate monitoring: implications of differences

between SEVIRI on METEOSAT-8 and AVHRR on NOAA-17, J. Geophys. Res., 111• SATBD1, 2009: Algorithm Theoretical Basis Document - Cloud Fraction, Cloud Type and Cloud Top Parameter

Retrieval from SEVIRI, reference no.: SAF/CM/DWD/ATBD/ CFC_CTH_CTO_SEVIRI, Version: 1.0, 10 September 2009, available at www.cmsaf.eu

• Schulz, J., et al., 2009: Operational climate monitoring from space: the EUMETSAT Satellite Application Facility on Climate Monitoring (CM-SAF), Atmos. Chem. Phys., 9, 1687-1709

References