On the ability of Sentinel-1 backscatter to detect soil moisture and vegetation changes caused by irrigation fluxes over the Po River Valley (Italy)

Worldwide, the amount of water used for agricultural purposes is rising because of an increasing food demand. In this context, the detection and quantification of irrigation is crucial, but the availability of ground observations is limited. Therefore, an increasing number of studies are focusing on the use of models and satellite data to detect and quantify irrigation. For instance, the parameterization of irrigation in large scale Land Surface Models (LSM) is improving, but it is still characterized by simplifying assumptions, such as the lack of dynamic crop information, the extent of irrigated areas, and the mostly unknown timing and amount of irrigation. Remote sensing observations offer an opportunity to fill this gap as they are directly affected by, and hence potentially able to detect, irrigation. Therefore, combining models and satellite information through data assimilation can offer a viable way to quantify the water used for irrigation. The aim of this study is to test how well modelled soil moisture and vegetation estimates from the Noah-MP LSM, with or without irrigation parameterization in the NASA Land Information System (LIS), are able to mimic in situ observations or to capture the signal of high-resolution Sentinel-1 backscatter observations in an irrigated area. The experiments were carried out over select sites in the Po river Valley, an important agricultural area in Northern Italy. To prepare for a data assimilation system, Level-1 Sentinel-1 backscatter observations, aggregated and sampled onto the 1 km EASE-v2 grid, were used to calibrate a Water Cloud Model (WCM) using simulated soil moisture and Leaf Area Index estimates. The WCM was calibrated with and without activating an irrigation scheme in Noah-MP. Results demonstrate that the use of the irrigation scheme provides the optimal calibration of the WCM, confirming the ability of Sentinel-1 to track the impact of human activities on the water cycle. Additionally, a first data assimilation experiment demonstrates the potential of Sentinel-1 backscatter observations to correct errors in Land Surface Model (LSM) simulations that are caused by unmodelled or wrongly modelled irrigation.