Near real time mapping of soil moisture using SAR images
Soil moisture plays a crucial role in the continental water cycle, more specifically on the distribution of precipitation between surface runoff and infiltration, which is the main driver behind most hydrological and geomorphologic processes. A better knowledge of the initial conditions of soil moisture could reinforce the early warning capacities of flooding. The main goal is to assess the soil moisture with a spatial accuracy and a temporal repetivity in adequacy with the requirement of the hydrological modelling. The French pilot basin of the Touch was selected because it is representative of Mediterranean type flash floods.
Works conducted in 2004-2005 have allowed assessing the behaviour of radar signal as a function of soil moisture. Results have shown that for a better estimate of soil moisture when using SAR data, it is necessary to use a radar configuration that minimizes the effects of surface roughness. The estimation of soil moisture is optimal at low and medium incidence angles (<35°). SAR images acquired at both low and high incidence angles (20° and 40°) makes it possible to eliminate the effects of roughness and thus to improve the accuracy on the estimation of soil moisture. The root mean square error on the estimation of soil moisture decreases by about 6% with one incidence (current SAR sensors) to 3.5% with two incidences.
The optimum configuration for soil moisture mapping with only one incidence can be attained by uniquely using the European ASAR sensor, 5 to 7 times per month on a study site in Europe. The arrival of new sensors (ALOS, RADARSAT-2 and Terra SAR-X, SENTINEL-1) should improve this periodicity. These results are tested face to the requirements of hydrological model and appear promising for the development of simplified operational algorithms for retrieving soil moisture from SAR data, and for monitoring multi-temporal moisture changes.
HyMeX – Hydrological cycle in the Mediterranean Experiment 2010-2020