Context and objectives
The objective of the PROBA-VET project is to assess the benefits and possible improvement of using biophysical parameters derived from moderate to high resolution remote sensing (e.g. PROBA-V) for the monitoring of the evapotranspiration at the kilometre scale over large areas (continental scale). More specifically, this project aims at 1) the improvement of the LSA-SAF land evapotranspiration product over large continental areas, and also 2) studying the interest of biophysical variables derived from the upcoming satellite for other land surface modelling applications.
The project is decomposed into two parts: 1) assess the benefits of using biophysical parameters from PROBA-V like data to improve land evapotranspiration (ET) monitoring at MSG/SEVIRI scale (~3-5 km) over large areas, i.e. continents, 2) study the possibilities of using PROBA-V like data to monitor ET at the sub-kilometer scale.
In the first part, we develop four different methods for using high resolution derived biophysical parameters and land covers, jointly or not with low resolution derived parameters, to improve ET monitoring at MSG/SEVIRI scale. SPOT-VGT and MODIS (PROBA-V proxy) imagery is used. The methods are evaluated by a careful validation against local in-situ measurements (e.g. FLUXNET) or model intercomparison.
In the second part, we explore the feasibility to monitor ET at sub-kilometre scale using MODIS data as proxy for the future PROBA-V satellite. High resolution vegetation indices and biophysical parameters are used to downscale lower resolution data and provide input for an ET model at higher resolution.
Prior to the PROBA-V launch, we have explored the benefits of several products that could later be derived from PROBA-V satellite and that could be ingested in evapotranspiration models: vegetation parameters (leaf area index), surface albedo, water bodies, land cover map. We have developed a test methodology for use of such products for ET monitoring over large areas, i.e. continents, at a few kilometres scale, and especially for the LSA-SAF ET model.
From this study, we recommend the use of PROBA-V like products for improved results expected in vegetation transition zones (thanks to an improved spatial resolution of the land cover map), in zones with clear hydrological contrasts (eg irrigation, lowlands, thanks to a better surface albedo assessment).
First tests towards monitoring land evapotranspiration at the sub-kilometre scale over large areas, by studying new synergies between remote sensing data across different spatial scales, have been obtained satisfactorily at the irrigated district of Barrax-Albacete in Southern Spain.
Period: January 1, 2011 – May 31, 2012
Funding source: Proba-V preparatory programme (contract NR CB/34/18).
Ghilain, N. and Gellens-Meulenberghs, F. 2014: Impact of land cover map resolution and geolocation accuracy on evapotranspiration simulations by a land surface model. Rem. Sens. Letters, 5(5), 491-499.
Ghilain, N., De Roo, F. and Gellens-Meulenberghs, F., 2014: Evapotranspiration monitoring with Meteosat Second Generation Satellites: improvement opportunities from moderate spatial resolution satellites for vegetation. International Journal of Remote Sensing, 35(7), 2654-2670, DOI:10.1080/01431161.2014.883093
Ghilain N., Arboleda A., Gellens-Meulenberghs F., 2012: Monitoring evapotranspiration at sub-kilometer scale: downscaling MSG/SEVIRI images using moderate resolution remote sensing derived data, Advances in Geosciences, 466-474, Proceedings of the 32nd EARSEL symposium, Mykonos, Greece.
Ghilain, N., Arboleda, A., Sepulcre-Cantò, G., Batelaan, O., Ardö, J. and Gellens-Meulenberghs, F., 2012: Improving evapotranspiration in a land surface model using biophysical variables derived from MSG/SEVIRI satellite. Hydrol. Earth Syst. Sci., 16, 2567-2583.