Speaker
Description
Vegetation phenology is strongly impacted by ongoing climate change. With the increase of drought periods, it also becomes crucial to improve methods aiming a better management of irrigation. Remote sensing provides increasing available data at high spatial and temporal resolution which can help the monitoring of vegetation growth and the water resources. A brief overview of the main surface characteristics which can be assessed from remote sensing is presented. Some operational tools (BVNET model, can-eye software developed at INRAE) allow to compute spatial indicators (or biophysical variables) describing the crop development. Soil moisture products (SMP) developed from Sentinel1 (S1) and 2 (S2) images using neural network techniques [1] are delivered at plot scale every 6 days via the Theia French public platform [2]. Up to now, these products were computed and validated mainly on cereals and grasslands. Mediterranean plots are often small and present a wide variability of agricultural practices. Among them, orchards, which require high quantify of water for irrigation are not represented in SMP because of their structure heterogeneity. Our team has evaluated different models based on spectral indices computed from S1 and S2 to assess soil moisture of ochardsof a small mediterranean watershed. The main results will be presented compared to ground measurements. These different variables derived from remote sensing can be used in crop models to predict yields and computed water balance from field scale to large regions. Some study cases will be illustrated which show that it is possible to monitor the vegetation development of various crops from Sentinel images. A crop model (STICS) can assimilate remote sensing data to assess the spatial and temporal variability of yieds at the territory scale. An example of integrated modeling is presented where remote sensing is combined with different models to assess grassland production and ground table recharge due to irrigation. With such approaches, various scenarios can be evaluated to quantify the impact of landuse modifications or changes in agricultural practices on water resources
