Implementing the dual crop coefficient approach in interactive software: 2. Model testing

Abstract

This paper is the second of a two-part series, with the first part describing the SIMDualKc model, an irrigation scheduling simulation tool that employs the dual crop coefficient approach for calculating daily crop ET and then performs a water balance for a cropped soil. The model was applied, calibrated and validated for rainfed and basin irrigated maize (Coruche, Portugal), rainfed and surface irrigated wheat (Aleppo, Syria), and furrow irrigated cotton (Fergana, Central Asia). Results show good agreement between available soil water content observed in the field and that predicted by the model. Results indicate that the calibrated model does not tend to over- or underestimate available soil water over the course of a season, and that the model, prior to calibration, and using standard values for many parameters, also performed relatively well. After calibration, the average growing season maximum estimation errors were 10 mm for maize, 8 mm for winter wheat and 9 mm for cotton, i.e., respectively 3.6, 2.9 and 5.0% of total available water. Results indicate that the separation between evaporation and transpiration and the water balance calculation procedures are accurate enough for use in operational water management. The indicators used for assessing model performance show the model to accurately simulate the water balance of several crops subjected to a variety of irrigation management practices and various climate conditions. In addition, the model was applied to alternative irrigation management scenarios and related results are discussed aiming at assessing the model’s ability to support the development of alternative active water management strategies.