Validation of FAPAR products derived from optical sensors: … · 2009. 4. 29. · Ref: Turner et al. 2005 FAPAR ≈1.-exp(-0.5) from PCA_LICOR Advanced procedure

1Davos - LPV/CEOS - 15 March 2007

N. Gobron1, B. Pinty1, O. Aussedat1, J. Chen2, W. B. Cohen3, R. Fensholt4, V. Gond5, K. F. Huemmrich6,7, T. Lavergne1, F. Mélin1, J. L. Privette6, I. Sandholt4, M. Taberner1,

D. P. Turner8, M. M. Verstraete1 and J.-L. Widlowski1

Validation of FAPAR products derived from optical sensors: method and results

(1) EC-Joint Research Centre, Institute for Environment and Sustainability, Italy(2) Department of Geography and Program in Planning, Univ. Toronto, Canada

(3) USDA Forest Service, USA (4) Institute of Geography, Univ. of Copenhagen, Denmark

(5) Forest Department of CIRAD, Guyanne Francaise(6) Code 923, NASA/Goddard Space Flight Center, USA

(7) Joint Research Center for Earth System Technology, Univ. of Maryland Baltimore County, USA(8) Department of Forest Science, Oregon State University, USA

Gobron N. et. al. (2006) ‘Evaluation of FAPAR Products for Different Canopy Radiation Transfer Regimes: Methodology and Results using JRC Products Derived from SeaWiFS and Ground-based Estimations’, JGR, 111, D13110, DOI 10.1029/2005JD006511


Validation protocol

• The performance for detecting various events (verisimilitude) is assessed against the time-series of well-known land surfaces.

• Comparisons against ground-based estimates can be achieved recognizing the following issues:– Date and time of acquisition, spatial sampling, H fluxes…– Impact of assumptions when ground-based estimates are

derived from measurements of intercepted rather than absorbed fraction of radiation.

– Categorization of land validation site in their respective radiative transfer regime at medium spatial resolution


Fire event(Pearsoll Peak, CA)

The FAPAR values suddenly decline after the 2002 summer fire events and keep low values, by comparison to those obtained for the period 1998-2001, until the end of 2003.

The 2003 FAPAR estimations suggest that the vegetation has not recovered from the 2002 fire stress.

Long Time series of 10-day

Active Fire Detection 2002

FAPAR September 2002



Rice(Pavia, IT)

Long Time series of 10-day

The FAPAR values are stable over all years.

The 2003 FAPAR estimations suggest that the growth period was in advance comparing to a normal year.

FAPAR July 2003



Validation protocol

• The performance for detecting various events (verisimilitude) is assessed against the time-series of well-known land surfaces.

• Comparisons against ground-based estimates can be achieved recognizing the following issues:– Date and time of acquisition, spatial sampling, H fluxes…– Impact of assumptions when ground-based estimates are

derived from measurements of intercepted rather than absorbed fraction of radiation.

– Categorization of land validation site in their respective radiative transfer regime at medium spatial resolution


SRF1: 09 July 2004

0.2

0.25

0.3

0.35

0.4

0.45

0.5

8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00

Time of day

Fapa

r

Source: M. Meroni (DISAFRI-Unitus)

• Three issues and caveats among others:- Date and time of acquisition

Ground-based Estimations

±0.1

+2 hours


• Three issues and caveats among others:- Date and time of acquisition- Spatial resolution: in-situ measurements at very high

resolution (1 m) to be extrapolated at the medium resolution pixel (number and spatial distribution of the samples, impact of horizontal fluxes)


Widlowski J-L. et. al.(2006) 'Horizontal Radiation Transport in 3-D Forest Canopies at Multiple Spatial Resolutions: Simulated Impact on Canopy Absorption', RSE, 103, 379-397, doi:10.1016/j.rse.2006.03.014.

FAPAR MeasurementsFAPAR Measurements


• Three issues and caveats among others:- Date and time of acquisition- Spatial resolution: in-situ measurements at very high

resolution (1 m) to be extrapolated at the medium resolution pixel (number and spatial distribution of the samples, impact of horizontal fluxes)

- Approximation of FAPAR by the intercepted radiation (impact of woody elements), i.e.


↓

↑−

↓↓

−=

+−=

Ground

TOCGroundGroundGround

TFIPAR

TTFAPAR

.1

)(.1 ρρ



FAPAR (μ0) - FIPAR (μ0)

Sparse <LAI> = 1.24

Medium<LAI> = 2.00

Dense<LAI> = 4.82

RT Model


0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

140 160 180 200 220 240 260 280 300

DOY (2004)

FiparFapar

Source: M. Meroni (DISAFRI-Unitus)

FAPAR (μ0) - FIPAR (μ0)Actual measurements


RT regime (adapted from Davis and Marshak, 2004)

“Fast” variability1-D RT theory on full domain

“Slow” variability1-D RT theory locally and Independent Pixel Approximation (IPA) on full domain “Resonant” variability3-D RT theory


Field Site Identification Land cover type


Dahra and Tessekre (Fensholt et al. 2004)Sevilleta (Turner et al. 2005)

semi-arid grass savannah

desert grassland

Ground-based estimations are categorized with respect to their anticipated radiation transfer regime to better understand sources of uncertainties.

Validation of

Leaves are randomly distributed but with preferred orientation



The FAPAR values are very small over this semi-arid grass savannah, and agree with ground estimations within the predefined daily accuracy of ±0.1

June

September

November

“Fast” variabilityValidation of

● JRC-FAPAR SeaWiFS at 2.17 km

□ Ground-estimation

±0.1

Ref: Fensholt et al. 2004FAPAR ≈ 1.-exp(-G(µ0)<LAI>/µ0) <LAI> from PCA_LICOR

● JRC-FAPAR MERIS at 1.2 km

● JRC-FAPAR MODIS at 1.0 km

MODIS Ascii SubsetsMOD15A2

NASA-MISR F06_0019


◊ 1x1 * 3x3



Gobron N. et. al. (2007) ‘Uncertainty estimates for the FAPAR operational products derived from MERIS – Impact of TOA radiance uncertainties and validation with field data’, RSE, submitted



NASA-MISR NASA-MISR

NASA-MISRNASA-MISR




NASA-MISR NASA-MISR

NASA-MISRNASA-MISR


.0

.2

.4

.6

.8

.0

◊ 1x1 * 3x3NASA-MODIS

◊ 1x1 * 3x3 NASA-MODIS





NASA-MISR

Ref: Turner et al. 2005FAPAR ≈ 1.-exp(-0.5<LAI>)<LAI> from PCA_LICOR Advanced procedure for spatio-temporal changes of local LAI








desert grassland

“Slow” variability1-D RT theory locally and Independent Pixel Approximation (IPA) on full domain

Bondville, AGRO (Turner et al. 2005) Harvard (Turner et al. 2005)De Inslaag (Gond et al. 1999)Konza Prairie (Turner et al. 2005)

corn and soybean

conifer/broad-leaf forestconifer/broad-leaf/shrub forestGrassland/shrub-land/cropland


Validation of



“Slow” variability

AGRO

Validation of


1km

3km


“Slow” variability

AGRO

● JRC-FAPAR MODIS at 1.0 km

Validation of


● JRC FAPAR MOS at 500 m

● JRC-FAPAR SeaWiFS at 2.17 km

● JRC-FAPAR MISR at 275 m

MODIS Ascii SubsetsMOD15A2

* 9x9◊ 1x1NASA-MISR








desert grassland

“Slow” variability1-D RT theory locally and Independent Pixel Approximation (IPA) on full domain

Bondville, ie AGRO (Turner et al. 2005) Harvard (Turner et al. 2005)De Inslaag (Gond et al. 1999)Konza Prairie (Turner et al. 2005)

corn and soybean

conifer/broad-leaf forestconifer/broad-leaf/shrub forestGrassland/shrub-land/cropland

“Resonant” variability3-D RT theory

Metolius (Turner et al. 2005)Mongu (Huemmrich et al. 2005)

dry needle-leaf forestshrub land/woodland


Leaves are NOT randomly distributed but clumped into crowns and woody parts contribute.

Validation of



“Resonant” variability Metolius

Validation of


1km


“Resonant” variabilityMETL

assumes random distribution of black leaves with no preferred orientation !!!

Validation of



23Davos - LPV/CEOS - 15 March 2007 Pinty B. et. al. (2006) ‘Simplifying the interaction of land surfaces with radiation for relating

remote sensing products to climate models’, JGR, 111, doi:1029/2005JD005952.

True <LAI> =2.0

3-D heterogeneous system

True <LAI> =2.0True <LAI> =2.0

3-D heterogeneous system

True <LAI> =2.0

1-D system representation

True <LAI> =2.0

1-D system representation

True <LAI> =2.0

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛ ><−=><

− μ0

1 2exp)( LAILAIT

direct

D

Direct transmission at 30 degrees Sun zenith angle,

= 0.312


0.596=><−

)(3 LAIT direct

D


0.596=><−

)(3 LAIT direct

D

))(()( 013 μξ><=>< −− LAITLAIT directD

directD

Domain-averaged structure factor

Problem related to clumping (1)

FIPAR ≈ from 0.688 to 0.404 !


Conclusions

A multi-faceted protocol for validation has been applied to the JRC-FAPAR products using various medium resolution optical sensors (MERIS, MODIS, MISR and SeaWiFS).

Assessment of the verisimilitude of the products based on multi-year time series

performance of the physics of the algorithm which gives quite similar results from various sensors.

Comparison against ground-based estimations


Suggested recommendations

1- Categorization of land validation sites with respect to their respective dominant radiative transfer regime at medium spatial resolution:

Better evaluation of the expected problems and difficulties to be faced when performing in-situ measurements

Enhanced evaluation of the accuracy, reliability and uncertainties associated with these in- situ estimations

2- Assessment of the impact of canopy internal variability for estimating domain-averaged values of the directly transmitted flux (probably a good proxy for FAPAR)


Acknowledgments• We thank for their respective ground-based estimates the ‘real world’ people!

–Senegal: Rasmus Fensholt & Inge Sandholt (Univ. Copenhagen)–Bigfoot: Warren B. Cohen (USDA, Forest Service) & David P. Turner (Oregon State Univ.) –Safari 2000: Karl Fred Huemmrich & Jeffrey L. Privette (NASA Goddard)

• The authors are grateful to the SeaWiFS Project (Code 970.2) and the Distributed Active Archive Center (Code 902) at the NASA Goddard Space Flight Center, Greenbelt, MD 20771, for the production and distribution of the SeaWiFS data, respectively.

• MODIS data were acquired as part of NASA’s Earth-Sun System Division and archived and distributed by the Goddard Earth Sciences (GES) Data and Information Services Center (DISC) Distributed Archive Center (DAAC).

• MERIS data are provided by the European Space Agency (ESA) through Brockman’s Consult.• MISR data were obtained from the NASA Langley Research Center Atmospheric Sciences Data Center.

Documents

Validation of FAPAR products derived from optical sensors: … · 2009. 4. 29. · Ref: Turner et al. 2005 FAPAR ≈1.-exp(-0.5) from PCA_LICOR Advanced procedure