Evaluación de los modelos numéricos de flujo de agua en el suelo hydrus-2d y simdas en riego localizado

Knowledge about water distribution in the soil profile under drip irrigation conditions is very important to prevent percolation losses and to obtain optimum yield. Therefore, during the initial step of drip irrigation design, to carry out previous field tests is very recommendable. These tests consist in applying a volume of water according to the crop needs by means of an isolated point emitter of a definite discharge rate and then to measure maximum depth and horizontal extension at 30 cm depth of the wetting front. Nevertheless, due to the effort that field tests endure, in many cases they are not carried out. An alternative to field test is to apply simulation models, that is; allowing to predict soil water distribution under drip irrigation conditions. Previous knowledge of irrigation time, discharge rate, and soil hydraulic properties or alternatively soil texture, are necessary to run these models. In this work to carry out the simulations two numerical models has been used: HYDRUS-2D and SIMDAS. The second one has been modified to take into account the ponded area under the emitter and van Genuchten-Mualem soil hydraulic functions has been implemented. Resultant soil water distribution with both models has been very similar. To validate the results of both models an experimental test in a Typic Calcixerepts soil (SSS, 1998) was done. The experimental soil water distribution, after to apply a total amount of 25 l with different discharge rates: 2, 4, 8.5 y 25 l·h -1 , has been compared with the results of the simulations. In the test, measures of ponded area under the emitter and soil water content in the soil profile have been done. The soil water content measures was taken with a TDR probe (IMKO TRIME-T) immediately before, during and after irrigation, the measures were taken at different horizontal distances of the emitter and until 140 cm deep. Simulated results of soil water content were in agreement with experimental data and are good enough for drip irrigation design. 1. Introduccion