Science Service System
You are here : Home : Proposals_Summary

Summary of Proposal HYD0745

TitleRemote sensing data assimilation on SVAT model for the monitoring of evapotranspiration in a semi-arid region in North Africa
Investigator ZRIBI, MEHREZ - CNRS, CESBIO
Team Member
DR BAGHDADI, NICOLAS - CEMAGREF, UMR TETIS
SummaryClimate changes, as well as socio-economic development, are likely to put an even greater strain on the already stretched water resources in the semi-arid regions of the world in the coming decades. To face up to the increasing demand for agricultural production, in combination with a decrease in water available for irrigation, farmers in these regions will have to adopt measures that enable them to increase productivity levels, whilst using less water. In this respect, an accurate knowledge of the actual evapotransipration of these areas, particularly irrigated ones, is of paramount importance for a better irrigation management. We are interested in the context of SICMED project to one site in Tunisia. It is a project that has taken place in southern Mediterranean regions, to assess the spatial-temporal variability of water needs and consumption for irrigated crops under water shortages (Chehbouni et al., 2006). For our study, we propose to use an approach based on ICARE Soil Vegetation Atmosphere Transfer (SVAT) model (Gentine et al., 2007), coupled to Multi-configuration spatial remote sensing to estimate land surface evapotranspiration. Surface soil moisture plays a crucial role on the continental water cycle, more specifically on the partition of precipitation between surface runoff and infiltration (Beven and Fisher, 1996) and in partitioning the incoming radiation between latent and sensible heat fluxes. Therefore, the ability of measuring soil surface characteristics (particularly soil moisture) on a large scale from space with a sufficient repetitiveness and spatial precision is an attractive challenge. A large number of studies, based on radar remote sensing, have been realized to estimate surface parameters in humid regions. In arid and semi-arid regions, and particularly in our sites in North Africa (Tunisia), surface characteristics are different, with a high spatial variability of surface moisture, during rain events or irrigation, the presence of dispersed vegetation cover and small agricultural fields. In our study, we propose to develop different methodologies to estimate and monitor surface parameters and particularly soil moisture, with high spatial and temporal scales, in order to have a precise estimation of fluxes in the studied sites. A large number of radar and optical images (30 ASAR/ENVISAT and 8 SPOT/HRV) data have been acquired during vegetation cycle season. Our analysis will be based on different axes: - an estimation of soil moisture and vegetation dynamic using ASAR and SPOT data at field scale - an analysis of TERRASAR-X signal variability in the agricultural field: due to moisture variations (with irrigation) and vegetation dispersion, particularly wheat or olives. - An analysis of links between ASAR and TERRASAR measurements - An analysis of synergy between TERRASAR data and High resolution optical data (Qcuickbird). In this context, we propose a validation of our approaches using two tools: continuous instrument measurements (thetaprobe, flux station) and ground punctual campaigns over a large number of test fields with moisture, roughness, vegetation.

Back to list of proposals

DLR 2004-2016