Relationship between soil moisture and vegetation in the Kairouan plain region of Tunisia using low spatial resolution satellite data

Abstract

The present paper proposes an empirical approach for the modeling of vegetation development, using moisture measurements only. The study is based simply on the use of two databases: one containing soil moisture products derived from ERS scatterometer data over the period 1991–2006 and the other containing normalized difference vegetation indices (NDVI) derived from advanced very high resolution radiometer over the period 1991–2000. The study is applied over the Kairouan plain, the central semiarid region of Tunisia (North Africa). Soil moisture products were first validated on the basis of comparisons with Global Soil Wetness Project, Phase 2 Data, outputs and rainfall events. The soil moisture distribution during the rainy period between October and May is described and is found to be correlated with the vegetation dynamics estimated using the NDVI products. Finally, a semiempirical model is proposed, based on satellite moisture and NDVI products, which allows the NDVI value to be estimated for a period of 1 month during the rainy season as a function of the moisture profile estimations obtained during the previous months. This approach could prove very useful and provide a simple tool for the modeling of vegetation dynamics during rainy seasons in semiarid regions.