Aerosol_cci ECV
Aerosol_cci > Thomas Holzer-Popp > ESA Living Planet Symposium, Bergen, 1 July 2010
slide 2
Major improvements over precursor AOD datasets
Regional scoring differences between both versions (scoring grows with increased spatial + temporal
correlations and decreased bias; based on daily data)
Precursor dataset 2008 Aerosol Optical Depth maps
Aerosol_cci final dataset
Aerosol_cci > Thomas Holzer-Popp > ESA Living Planet Symposium, Bergen, 1 July 2010
slide 3
Comprehensive validation
• Validation of pixels and gridded data; global, regional, land/ocean • Ground-based references, fully independent
• AOD / AE: AERONET / MAN • Stratospheric extinction: NDAAC lidar
• Intercomparison to other satellite datasets • AOD/AE: MODIS, MISR, SeaWIFS, POLDER • Stratospheric extinction + AOD: SAGE (II, III), OSIRIS, CALIOP
AOD, pixels AOD, gridded profiles, gridded
Aerosol_cci > Thomas Holzer-Popp > ESA Living Planet Symposium, Bergen, 1 July 2010
slide 4
Pixel level uncertainties
Aerosol_cci > Thomas Holzer-Popp > ESA Living Planet Symposium, Bergen, 1 July 2010
slide 5 12,4 % 21,1 % 0,5 % 66,0 %
Cloud mask consistency Aerosol_cci / Cloud_cci
5 selected days Sep 2008 – safety zone included by Aerosol_cci
Aerosol_cci > Thomas Holzer-Popp > ESA Living Planet Symposium, Bergen, 1 July 2010
slide 6
Phase 1 products
1985 1995 2005 2015
tropospheric - AOD (amount) - Angstrom (size) … with local errors
stratospheric - extinction/AOD - effective radius
- absorbing aerosol index (qualitative)
2008
ATSR-2 SU
3* MERIS
AATSR SU
GOMOS
OMI
3* ATSR
focus on the best ATSR product
Aerosol_cci > Thomas Holzer-Popp > ESA Living Planet Symposium, Bergen, 1 July 2010
slide 7 1985 1995 2005 2015
AOD Angstrom fine-mode AOD aerosol „type“
stratospheric ext (!AOD)
absorbing aerosol index (use in case study)
2008
ATSR-2 AATSR (3 algorithms)
TOMS / GOME / SCIAMACHY / GOME-2 / OMI
SCIAMACHY/AATSR
GOMOS dust AOD IASI
more reference AOD (3 selected. regions) POLDER
Phase 2 plans
Aerosol_cci > Thomas Holzer-Popp > ESA Living Planet Symposium, Bergen, 1 July 2010
slide 8
Comparison to GCOS planned upgrades P2
ECV Parameter Requirement
Source
Performance CCI Phase 1
Explanatory Text (bullets)
Aerosol optical depth + Fine mode AOD, aerosol type Horizontal Resolution 5-10km
1°1000km GCOS-154 User survey
10km, 1°
Temporal Resolution 4h
500 obs./hour
GCOS-154
MACC
1 to 6 days
Typical >1000 / orbit
- depends on sensor swath; data availability limited by presence of clouds - MACC requirement for data assimilation (1 orbit = 1.7 hrs.)
Accuracy (rmse)
Max (0.03, 10%) Max (0.05, 20%) @10km
Max (0.02, 10%) @1°
0.02 @ 1000 km
GCOS-154 User survey
≥ 0.04 (ocean) ≥ 0.06 (land) ≥ 0.07 / 50% ≥ 0.02
- 12 months ACCI data compared to AERONET
- Precursor dataset (GlobAER): 0.18 - Phase 1: 0.06 (ocean), 0.09 (land)
Stability 0.01 GCOS-154 Mean bias stable 2002-2010
Temporal extent Minimum 3-10 years User survey 1995-2012, 2015 Phase 1: 2002-2012
uncertainty on single pixels User survey provided per pixel Qualitative information so far, but appreciated by users; needs harmonization
Spectral coverage Up to 4 AODs (VIS/NIR) User survey for several sensors - to allow derivation of size information (e.g. Ångström coefficient) - sensor and algorithm dependent
Absorption: Absorbing aerosol index Horizontal Resolution 5-10km GCOS-154 0,25° - OMI pixel resolution
Temporal Resolution 4h GCOS-154 daily - coverage limited by sensor swath
Accuracy 0.03 GCOS-154 N/A - for single scattering albedo (SSA) Stability 0.01 GCOS N/A - for single scattering albedo (SSA)
Temporal extent Minimum 10 years User survey 2008 - potential time series 1979-2012
Best suited absorption parameter
Absorption Aerosol Optical Depth (AAOD)
User survey Round robin exercise for AAOD
- AAOD is under development over ocean - GCOS-154 requires SSA (difficult for low AOD); under development as a research product
(land & ocean)
Stratospheric extinction profile Horizontal Resolution 200-500km GCOS-154 2.5° x 10°
Vertical Resolution 1-2km GCOS-154 1-3km - additional: integrated stratospheric column AOD for discrimination stratospheric and
tropospheric AOD Temporal Resolution weekly GCOS-154 Monthly - determined by occultation principles and available data points
Accuracy 10% GCOS-154 10 % - comparison to SAGE-II, SAGE-III, OSIRIS – Phase 1: 10-25% Stability 20% GCOS Not yet assessed - baseline product only 1 year
Temporal extent Minimum 10 years User survey 1985-2012 - potential GOMOS time series 2003-2012 – Phase 1: 2008
All aerosol products Format NetCDF user survey NetCDF - metadata and filenames following CCI guidelines / CF1.4 convention
Updating frequency annual user survey two years - according to system specification document (phase 2) Geographic Coverage Global GCOS-154 global - except polar latitudes
Aerosol_cci > Thomas Holzer-Popp > ESA Living Planet Symposium, Bergen, 1 July 2010
slide 9
system
Expertise for validationand intercomparison
Developmentpriorities
Expertise foralgorithms
Validation / QCIntercomparisonStandard tools
Satelliteinput
Ancilliarydata
ArchiveDisseminationDocumentation
Expertise forECV use
Algorithmdevelopment
The Aerosol_cci system
Operational ECV Production system
External sciencecommunities
Spaceagencies
instrumentteams
Science Team
ExternalECV users
Referencedata
AerosolECV productsLv2 and lv3
Expertise for validationand intercomparison
Developmentpriorities
Expertise foralgorithms
Validation / QCIntercomparisonStandard tools
Satelliteinput
Ancilliarydata
ArchiveDisseminationDocumentation
Expertise forECV use
Algorithmdevelopment
The Aerosol_cci system
Operational ECV Production system
External sciencecommunities
Spaceagencies
instrumentteams
Science Team
ExternalECV users
Referencedata
AerosolECV productsLv2 and lv3
Aerosol_cci > Thomas Holzer-Popp > ESA Living Planet Symposium, Bergen, 1 July 2010
slide 10
system
Aerosol_cci > Thomas Holzer-Popp > ESA Living Planet Symposium, Bergen, 1 July 2010
slide 11
" Network of aerosol remote sensing scientists (retrieval & validation experts, data centers, users): unfunded, open, independent, international
" Close collaboration with related initiatives: AEROCOM (models), ICAP (forecasts), GALION (lidar) and AERONET (sun photometers)
" Supported by scientists working with: NASA, ESA, NOAA, EUMETSAT, KNMI, DLR, FMI, ECMWF, MPI, JAXA, HUJ, METNO, LOA, CAS
" Goals " Promote the use of satellite aerosol data " Achieve open and active exchange of information
" Working groups invite your active collaboration
" Next meeting: 27+28/09/2014, Steamboat / Colorado (with AEROCOM) " Contacts: R. Kahn / NASA & T. Holzer-Popp / DLR
AEROSAT International Satellite Aerosol Science Network
http://aero-sat.org
Aerosol_cci > Thomas Holzer-Popp > ESA Living Planet Symposium, Bergen, 1 July 2010
slide 12
Continued from P1 • Extend Climate Research group (stratosphere, aerosol cloud
interaction; data assimialtion through CMUG; GoeMIP indirect)
• Valdiation tools and concepts (add assessing stability and aerosol type)
• Round robin exercises where appropriate (IASI, several options)
• Consistency analysis (extended, with other ECV projects)
• Uncertainties (harmonized, with metrologists, new WG)
• WGs: clouds (extend with international inter-comparison exercise, surface, components
Aerosol_cci > Thomas Holzer-Popp > ESA Living Planet Symposium, Bergen, 1 July 2010
slide 13
New in P2 • Long time series production
• Fill mission time series for all datasets • Add IASI • POLDER reference sites
• User case studies
• Sustainable but distributed system • System engineering working group • Common standards and tools (e.g. Versioning) • Obs4MIPS
• More • Sentinel perparation • Joint aerosol-cloud retrieval
Aerosol_cci > Thomas Holzer-Popp > ESA Living Planet Symposium, Bergen, 1 July 2010
slide 14
New in P2: case studies
AAI record for West-Africa (Tilstra et al., 2011). Red GOME-1, Brown
SCIAMACHY blue GOME-2A
EMAC moldel radiative forcing of stratospheric aerosol (Brühl et al., 2012/3
simulated regional aerosol direct radiative forcing (Kinne et al., 2013)
Changes in liquid water path due to anthropogenic aerosol (ECHAM6-
HAM2; Lohmann, et al. 2010)
Aerosol_cci > Thomas Holzer-Popp > ESA Living Planet Symposium, Bergen, 1 July 2010
slide 15
WP plan
Aerosol_cci > Thomas Holzer-Popp > ESA Living Planet Symposium, Bergen, 1 July 2010
slide 16
Recommendations • Data standards
• Need to get frozen to allow implementing them
• System engineering • Need to assure approval by science leaders for major
decisions
• Involve ECV project in defining requirements for dissemination
Aerosol_cci > Thomas Holzer-Popp > ESA Living Planet Symposium, Bergen, 1 July 2010
slide 17
Consistency analysis