Global land monitoring in Europe’s Earth observation programme (GMES)

Alan BelwardJoint Research CentreIspra (VA) Italy

alan.belward@jrc.ec.europa.eu

http://www.wmo.ch/pages/prog/sat/CGMS/CGMS_home.html

“Europe has decided to develop its own operational Earth observation capacity in order to reflect the EU’s growing responsibility in European and world affairs” COM(2009) 223 final

Outline

• Europe’s Earth Observation Programme (GMES)* was adopted 9th November 2010

• This includes both spacecraft (Sentinels) and services (the GMES Initial Operations)

• A land service is underway• ‘GMES is the main space contribution of the Union to

tackle climate change’**

**COM(2010) 614; An Integrated Industrial Policy for the Globalisation Era

*COM(2009) 223; The European Earth Observation Programme (GMES) and its Initial Operations

An operational EO Programme • Overall framework

– space component (ESA coordination)– in-situ component (EEA coordination) – service component (EC coordination)

• land, marine, atmosphere, emergency response, climate change, security

• Funding leading up to the regulation– 2003 – 2006 the EU and ESA spent €100m each

on GMES projects – 2007 – 2013 the EU budgeted €430m for GMES

project work– 2007 – 2013 the EU contributed €624m to the

total ESA GMES space component budget of €2246m

• Funding for GMES Initial Operations (GIO)– 2011 – 2013 the EU has allocated €107m to the

implementation of the Regulation– 2011 – 2013 the EU has redeployed an additional

€43m from the research budget

Sentinel 1 (C-band SAR)a and b

Sentinel 2 (13 channel MSI)a and b

Sentinel 3 (21 channel OLCI)a and b

Images courtesy of ESA

€2.9bn

€150m

COM(2009) 223 final

• “The GMES services will allow policy-makers in particular to:

– prepare national, European and international legislation on environmental matters, including climate change;

– monitor implementation of this legislation; • “GMES is a tool for cooperation linked to development,

humanitarian aid and emergency situations worldwide and, more specifically, with Africa”

Land monitoring service

• Reference data: basic geographic framework

• Local component: 1 m resolution mapping of Urban Areas (2011-2012) and extension to other ‘hot spots’ e.g. Biodiversity protection sites

• Pan-European component: Land Use / Cover Area mapping at 10 m resolution

• Global component: the terrestrial essential climate variables, modelling and ‘hot spots’

Corinehttp://www.land.eu/portal/

small water bodieshttp://

bioval.jrc.ec.europa.eu/EMMA/emma/index.php

Urban atlas (Dublin) http://www.land.eu/portal/

The GCOS ECVs (2010 update)

Implementation Plan for the Global Observing System for Climate in Support of the UNFCCC (2010 Update)FINAL DRAFT (v2.0) 19 July 2010 GCOS SecretariatGCOS-138 WMO/TD-No. 1523

Main FAPAR product providers.

Projects/Institution Input data Output product Retrieval Method References

JRC-FAPARESA MERIS

Top of Atmosphere (TOA) BRFs in blue, red and near-infrared bands

Instantaneous green FAPAR based on direct incoming radiation

Optimization Formulae based on Radiative Transfer Models

Gobron et al (2000, 2006, 2008)

NASA MODIS LAI/FPAR

Surface reflectance in 7 spectral bands and land cover map.

FAPAR with direct and diffuse incoming radiation

Inversion of 3D Model versus land cover type with backup solution based on NDVI relationship)

Knyazikhin et al. (1998b)

NASA MISR LAI/FPAR

Surface products BHR, DHR & BRF in blue, green, red and near-infrared bands+ CART

FAPAR with direct and diffuse incoming radiation.

Inversion of 3D Model versus land cover type with backup solution based on NDVI relationship)

Knyazikhin et al. (1998a)

GLOBCARBON Surface reflectance red, near infrared, and shortwave infrared

Instantaneous FAPAR (Black leaves)

Parametric relation with LAI as function as Land cover type.

Plummer et al. (2006)

CYCLOPES Surface reflectance in the blue, red, NIR and SWIR bands

FAPAR at 10:00 solar local time

Neural network Baret et al (2007)

LANDSAF Visible and Near-Infrared bands FAPAR Parametric relation Roujean and Breon (1995)

JRC-TIP Broadband Surface albedo in visible and near-infrared bands.

FAPAR & Green FAPAR for direct & diffuse incoming radiation

Inversion of two-stream model using the Adjoint and Hessian codes of a cost function.

Pinty et al. (2007)

N. Gobron & M. M. Verstraete (2009) FAPAR: assessment report on available methodological standards and guides, GTOS-65

http://fapar.jrc.ec.europa.eu/Home.php

1998 onwards

http://earth.eo.esa.int/level3/meris-level3/

April 2002 onwards

Satellite Derived FAPAR AnomaliesAnomalies 1998 - 2009 - Base period 1998 - 2010

Gobron et al. 2010 GRL http://www.agu.org/journals/pip/gl/2010GL043870-pip.pdf

fAPAR Anomalies 1998 - 2009

Gobron et al. 2010 GRL http://www.agu.org/journals/pip/gl/2010GL043870-pip.pdf

Relative fraction of land surface showing fAPAR anomaly

Gobron et al. 2010 GRL http://www.agu.org/journals/pip/gl/2010GL043870-pip.pdf

May 2010 fAPAR anomalies

June 2010 fAPAR anomalies

July 2010 fAPAR anomalies

Crossing scales

20020526

0.62

ETM MERIS

ETMMERIS

Decision 4/CP.15

Source Bartholome and Belward JRC

1990

2000

2010Samples are 20km x 20km size

Systematic sampling - 4016 sample sitesTropical Latin America & Caribbean (LAC):1230 sample sites

Sub-Saharan Africa (AFR):2045 sample sites

South and Southeast Asia plus PNG and the Solomon Islands (SEA): 741 sample sites

->

Acquisition dates for satellite imagery used for the “year 2000 period”

Acquisition dates for satellite imagery used for the “year 1990 period”

Data gaps “1990 / 2000”

Data gaps “2005”

Cloud cover evaluation of TREES-3 sample sites for the “year 1990 period” (in percent)

Cloud cover evaluation of TREES-3 sample sites for the “year 2000 period” (in percent)

Tree cover loss

Tree to other wooded landand other vegetation

(red = deforestation green = aforestation)

Other wooded land loss

Other wooded land to other vegetation

(orange = loss of other wooded land green = gain of other wooded land)

Distribution of Land cover in 1990

Global monitoring of Tree Cover Changes : First results on East Africa

(Source Brink and Bodart JRC)

Photo credit championsportsradio.com/

Deforestation

Photo credit championsportsradio.com/, JRC

Deforestation; humid tropics 5.8 mha/yr

Photo credit championsportsradio.com/, JRC

Deforestation; humid tropics 5.8 mha/yr, 13 mha/yr globally Photo credit championsportsradio.com/, JRC

Deforestation; emissions ~ 1.2 Pg C yr–1

5.8 mha/yr humid tropics Achard et al. (2002), Science 297, 999-100213 mha/yr globally FAO (2010) Global Forest Resource Assessment Key Findings

http://www.fao.org/forestry/fra/fra2010/en/

Photo credit championsportsradio.com/, JRC

…less than 3 seconds to clear a football field

Emissions; van der Werf et al, 2009, Nature BiogeoSciences

Sentinel 2 bands (courtesy ESA)Sentinel 2 bands (courtesy ESA)

10 m 20 m 60 m

Sentinel 2simulations(Courtesy ESA)

Sentinel 2simulations(Courtesy ESA)

Europe’s EO satellites - 23rd November 1977…

• 19th June 1981 Meteosat 2• 22nd February 1986 SPOT 1• 15th June 1988 Meteosat 3• 6th March 1989 Meteosat 4• 22nd January 1990 SPOT 2• 2nd March 1991 Meteosat 5• 17th July 1991 ERS-1• 26th September 1993 SPOT 3• 20th November 1993 Meteosat 6• 21st April 1995 ERS-2• 2nd September 1997 Meteosat 7• 24th March 1998 SPOT 4• 1st March 2002 ENVISAT• 4th May 2002 SPOT 5• 28th August 2002 Meteosat 8• 27th October 2005 TopSat• 21st December 2005 Meteosat 9 • 19th October 2006 MetOp-A• 7th June 2007 COSMOSkyMed• 15th June 2007 TerraSAR-X• 9th December 2007 COSMOSkyMed• 29th August 2008 RapidEyes 1 to 5• 24th October 2008 COSMOSkyMed• 29th July 2009 UKDMC2, Demios 1 • 5th November 2010 COSMOSkymed

23rd November 1977… ©CNES, ©DMCii, ©ASICosmoSkyMed, ©ESA, ©EUMETSAT, ©DLR©CNES, ©DMCii, ©ASICosmoSkyMed, ©ESA, ©EUMETSAT, ©DLR

Summary

• Continuity of observation is extremely likely (virtually certain up to launch…)

• Continuity of GCOS ECV generation is very likely(GEOSS, CEOS Working Groups Climate and WGCV for characterisation / validation)

• Global cloud free optical data sets at high-resolution for key historical epochs e.g. 1990 are unlikely - but not exceptionally unlikely

• Data acquisition strategies and data policy need to (continue to) be tuned to global scales – priority areas are known

• Partnership is key; GEOSS, GCOS, CEOS

Acknowledgements

• FAPAR; Nadine Gobron

• TREES-3 optical remote sensing team; Frédéric Achard, René Beuchle, Hugh Eva, Hans-Juergen Stibig, Silvia Carboni, Rastislav Raši, François Donnay, Andreas Brink, Catherine Bodart, Philippe Mayaux, Dario Simonetti, Desirée Johansson, Ouns Kissiyar, Michael Vollmar

• FAO partners from the Forest Resource Assessment

• GMES Bureau staff

• Landsat Data Continuity Mission Science Team

• GCOS secretariat, steering committee and science panels