Hydrobal: An eco-hydrological modelling approach for assessing water balances in different vegetation types in semi-arid areas

Abstract

In semiarid areas, water is a limited resource and its management is a challenge. Water-balance models can improve the management of water resources by determining the effect of vegetation type on the soil–water balance and aquifer recharge. Here, we present HYDROBAL, an eco-hydrological modelling approach for assessing the water balance with a daily resolution. HYDROBAL is suitable for investigating the temporal variability in soil–water content determined by vegetation water uptake as a function of climatic conditions. The processes, mechanisms, and water flows involved in soil moisture changes are modelled based on daily rainfall and micrometeorological variables and used to predict changes in daily soil–water content. The model outputs include actual evapotranspiration, runoff, and aquifer recharge (deep percolation). The model was applied in a semi-arid area of south-eastern Spain, with six vegetation cover types: bare soil (B), open Stipa tenacissima steppe (St), thorn shrubland (S), dry grassland (G), and Aleppo pine (Pinus halepensis) afforestation of S and G (AS and AG, respectively). A dynamic evaporative coefficient (k) was calibrated for each vegetation type to estimate the soil–water consumption. The model was verified in base on its ability to predict the daily measured soil moisture content in plots with different vegetation types. Comparison between the estimated and measured soil moisture contents (θmodel vs. θTDR) indicated good model performance for all vegetation cover types in both wet and dry periods. High value of the coefficient of determination in the linear regressions for θmodel = ƒ (θTDR) demonstrate the accuracy of the hydrological model. All correlations between measured and predicted soil–water content were strong and significant (R2 > 0.69, p < 0.001)