Water Requirement Modelling for Wheat under Arid Climatic Conditions

Abstract

Algeria is an agriculture country where wheat is staple food of this region. Wheat is a winter crop and sown almost throughout the country. Winter season starts from October and normally ends in April. Water requirement of wheat followed variations in terms of location and time of the season. The northern of the country receives good rains during growth stages of wheat. But the southern of the country always requires irrigation and the groundwater is used extensively. Thousands of hectares are irrigated by pivot centres in extreme weather conditions. Better manage of limited water resources is the main objective of this study. Estimation of wheat maximum evapotranspiration with a Semi empirical approach used directly by the farmers, in lack of meteorological parameters, allows reaching thisobjective. The water balance of wheat was obtained by using non-weighing lysimeters. It was found that the Maximum daily crop Evapotranspiration (MET) for each wheat growth stage is linked to mean of daily temperature (Tm °C) and Leaf Area Index (LAI). Multivariate analysis shows that LAI does not contribute significantly for realizing a model in case of linear correlation. Using mean temperature of each growth stage, the obtained result shows a strong correlation between the calculated values of maximum evapotranspiration and its simulated values. In case of non linear correlation, both LAI and mean temperature contribute significantly for construction of the model. In this case the performance criteria are satisfactory. The determination coefficient (R²), Nash-Sutcliffe efficiency and the Relative Mean-Squared Error (RMSE) are respectively: 0.92, 0. 92 and 0.75 mm/day.