Final Report Summary - SENSORVEG (Staff Exchanges to estimate vegetation structure and biochemistry from remote sensing in connection to carbon and water fluxes)
Remote sensing can provide systematic global estimations of vegetation structure and biochemistry imparting data such as leaf area index (LAI), water content or fractional light interception by green vegetation (fPAR). However a global network of standardize validation sites is required to reduce the uncertainty of these products. As part of that effort, SpecNet (Spectral Network), an independent international collaborative research iniciative, promotes the acquisition of field optical remote sensing and vegetation properties. These sites are also generally part of the flux tower network (FLUXNET) to better understand carbon and water vapor fluxes. SENSORVEG promoted SpecNet by linking joint research activities to develop methodologies to estimate vegetation biochemistry and structure from remote sensing, including the design, validation and assessment of bi-dimensional radiative transfer models, comparing spectral information at leaf, canopy and ecosystem level using laboratory and field spectroscopy, airborne hyperspectral image, Lidar data and multispectral satellite data.
Standard protocols for field sampling on vegetation canopy water content in Mediterranean natural ecosystems and crops were developed and established consensus on the metadata to be used. Data acquired in the field was correlated with airborne hyperspectral and satellite data.
The potential limitation of the PROSPECT Radiative Transfer model was analyzed when applied to Mediterranean drought adapted species and an updated version of the model was made publicly available. Vegetation biochemical and biophysical parameters related to water are pivotal to understanding the water cycle and its interactions with carbon and energy balance. Estimations from hyperspectral AVIRIS PROSAIL radiative transfer model inversion were compared to in-situ measurements stratified by cover type (i.e. grasses, shrubs and forest) made at Stanford University’s Jasper Ridge Biological Preserve, California, USA.In addition, It was demonstrated that spectral indexes derived from airborne optical multispectral remote sensing data is sensitive to diurnal changes in canopy water content over well irrigated almond and pistachio orchards in the southern San Joaquin Valley of California. These results improve our understanding of the water cycle and can help managing irrigation of agricultural crops.
Accurate spatial information of clumping index contributes to better understand light regime within the canopy and the physiological processes associated with it. Terrestrial and airborne laser scanning estimated the clumping index based on the analysis of the gaps in the canopy and the distribution of returns after converting the point cloud into a 3D voxel-based model, so they have the potential to be operationally applied to other areas since it is not an empirical fit.
Airborne hyperspectral images acquired by the CASI sensor (Compact Airborne Spectrographic Imager) estimated biophysical vegetation parameters related with plan water content, vigour and structure were also performed at the Majadas del Tietar Fluxnet site (Spain) using Normalized Difference Indices. To scale-up to satellite image analysis, a program was written for automatic atmospheric and topographic conversion of Landsat 8, with which satellite time series can be built. Earlier Landsat sensors time series were used to estimate canopy water content and grass biomass at the Majadas Fluxnet site (Spain) from 1985 to 2010 that were sensitive to extreme changes in time series of precipitation and temperature. From the results of the estimation of biomass, we analyzed the evolution of carbon stocks. A temporal series of synthetic images was generated for a full phenological year (May 2010-April 2011). Gross Primary Production (GPP) was estimated and the results showed high agreement witth GPP measured in the flux tower. In relation also to fluxes, mean midday gross ecosystem production in a subalpine grassland of the Italian Alps equipped with an eddy covariance flux tower was estimated successfully through chlorophyll-related indices from a field spectrometer.
Finally, the impacts of tropospheric ozone on conifer health in California and Catalonia were investigated using hyperspectral imagery and GIS variables related to plant water relations and ozone uptake. The Ozone Injury Index (OII) field metric applied to Pinus ponderosa and Pinus jeffreyi in the USA and adapted to Pinus uncinata in Spain included chlorotic mottling, needle retention, needle length, and crown depth. Species-level classifications of AVIRIS and CASI hyperspectral imagery were all near 80% overall accuracy for the target bioindicator species.
Standard protocols for field sampling on vegetation canopy water content in Mediterranean natural ecosystems and crops were developed and established consensus on the metadata to be used. Data acquired in the field was correlated with airborne hyperspectral and satellite data.
The potential limitation of the PROSPECT Radiative Transfer model was analyzed when applied to Mediterranean drought adapted species and an updated version of the model was made publicly available. Vegetation biochemical and biophysical parameters related to water are pivotal to understanding the water cycle and its interactions with carbon and energy balance. Estimations from hyperspectral AVIRIS PROSAIL radiative transfer model inversion were compared to in-situ measurements stratified by cover type (i.e. grasses, shrubs and forest) made at Stanford University’s Jasper Ridge Biological Preserve, California, USA.In addition, It was demonstrated that spectral indexes derived from airborne optical multispectral remote sensing data is sensitive to diurnal changes in canopy water content over well irrigated almond and pistachio orchards in the southern San Joaquin Valley of California. These results improve our understanding of the water cycle and can help managing irrigation of agricultural crops.
Accurate spatial information of clumping index contributes to better understand light regime within the canopy and the physiological processes associated with it. Terrestrial and airborne laser scanning estimated the clumping index based on the analysis of the gaps in the canopy and the distribution of returns after converting the point cloud into a 3D voxel-based model, so they have the potential to be operationally applied to other areas since it is not an empirical fit.
Airborne hyperspectral images acquired by the CASI sensor (Compact Airborne Spectrographic Imager) estimated biophysical vegetation parameters related with plan water content, vigour and structure were also performed at the Majadas del Tietar Fluxnet site (Spain) using Normalized Difference Indices. To scale-up to satellite image analysis, a program was written for automatic atmospheric and topographic conversion of Landsat 8, with which satellite time series can be built. Earlier Landsat sensors time series were used to estimate canopy water content and grass biomass at the Majadas Fluxnet site (Spain) from 1985 to 2010 that were sensitive to extreme changes in time series of precipitation and temperature. From the results of the estimation of biomass, we analyzed the evolution of carbon stocks. A temporal series of synthetic images was generated for a full phenological year (May 2010-April 2011). Gross Primary Production (GPP) was estimated and the results showed high agreement witth GPP measured in the flux tower. In relation also to fluxes, mean midday gross ecosystem production in a subalpine grassland of the Italian Alps equipped with an eddy covariance flux tower was estimated successfully through chlorophyll-related indices from a field spectrometer.
Finally, the impacts of tropospheric ozone on conifer health in California and Catalonia were investigated using hyperspectral imagery and GIS variables related to plant water relations and ozone uptake. The Ozone Injury Index (OII) field metric applied to Pinus ponderosa and Pinus jeffreyi in the USA and adapted to Pinus uncinata in Spain included chlorotic mottling, needle retention, needle length, and crown depth. Species-level classifications of AVIRIS and CASI hyperspectral imagery were all near 80% overall accuracy for the target bioindicator species.