Building and Evaluating Mathematical Model to Predict Available Water in Middle Euphrates in Iraq

Abstract

    A simple linear and multiple linear regression equation was developed to predict available water in the soils of the Middle Euphrates Governorates from some easily measurable physical and chemical properties. The statistical software SAS was used. The results showed a positive correlation between soil moisture constants (field capacity, permanent wilting point, and available water) and the clay content of the soil. The clay percentage was the most influential factor compared to other properties in the amount of available water. There was a positive correlation between available water and clay and silt content, and a negative correlation between available water and apparent soil density, carbonates, and electrical conductivity. We recommend using the developed prediction model under experimental conditions using the clay fraction, as it yields acceptable results for available water with an error rate of no more than 2%.