Towards improved understanding of land use effect on soil moisture variability: analysis and modeling at the plot scale

Understanding and characterizing the soil moisture spatial variability and its relevant physical controls is a main challenge in hydrological sciences. In this work we examine the spatial variability characteristics of soil moisture data at 0-30 cm depth collected over three years (2006-2008) on a plot (about 200 m(2)) in Grugliasco (Po Plain, Northern Italy) by means of 21 Time Domain Reflectometry probes. The plot is divided into two subplots: one covered by grapevine plants, the other covered homogeneously by grass. The site is almost flat and the soil is sandy. The characteristics of the site allow to isolate the contribution of soil hydraulic properties and land use to soil moisture variability. Examination of the data shows higher soil moisture values in the vineyard than in the meadow, implying the influence of vegetation cover during the growing season; correspondingly, the spatial soil moisture variability is systematically lower in the vineyard than in the meadow. Evaluation of the main physical controls on the spatial mean and the variability of soil moisture is carried out by using a simple bucket model, forced by using local rainfall and evapotranspiration data. The model is calibrated by using mean soil moisture daily time series over one year for the two sites. The model accuracy is verified for the other two years, showing a relatively good prediction capability. The model is also shown to be able to capture the main differences between the two sites in terms of spatial variability. by Elsevier B.V