AUTHOR=Bois Benjamin , Pauthier Basile , Brillante Luca , Mathieu Olivier , Leveque Jean , Van Leeuwen Cornelis , Castel Thierry , Richard Yves TITLE=Sensitivity of Grapevine Soil–Water Balance to Rainfall Spatial Variability at Local Scale Level JOURNAL=Frontiers in Environmental Science VOLUME=8 YEAR=2020 URL=https://www.frontiersin.org/journals/environmental-science/articles/10.3389/fenvs.2020.00110 DOI=10.3389/fenvs.2020.00110 ISSN=2296-665X ABSTRACT=

In rainfed agriculture systems, rainfall water management (harvesting, storage, and efficient use) is a key issue. At local scale (i.e., from 100 m to 50 km), the impact of rainfall spatial and temporal variability on crop water availability is seldom addressed. In order to accurately depict the space and time variations of rainfall at local scale, a dense rain-gauges network composed of 45 rain-gauges has been deployed over 28-km2 area, in Burgundy vineyards (North-East France). Rainfall data collected by each rain gauge from 2014 to 2016 were used as input variables in the Lebon et al. (2003) grapevine water balance model. All other climate variables, vineyard, and soil parameters were kept the same for each simulation in order to capture the impact of the sole spatial variability of rainfall on vineyard water status. As rainfall dynamics impact on the vineyard depends on the soil water content, water balance was modeled considering soils with low (50 mm) and medium (150 mm) soil water-holding capacities, representative of the soils of the area. The impact of modeled soil water availability for grapevine was assessed using the water deficit stress index (WDSI), i.e., the relative stomatal conductance. Local rainfall variability throughout the vine vegetative period leads to large variations in WDSI; it varied up to 0.3 within the study area due to because of rainfall spatial variability. Using a set of 34 weather stations at mesoscale level over Burgundy (186 km from North to South), we showed that local rainfall might contribute to change in grapevine water status as large as 50% of the simulated regional water balance spatial variability. Our results indicate that local rainfall and its impacts on agricultural production are probably not sufficiently considered in farming systems, potentially leading to inaccurate water management (cover-crop, irrigation) due to sparse rainfall network.