Modelling of leaching of chloride and nitrogen species in an experimental citrus grove

A numerical model solved by the finite element method, MELEF, is applied to simulate the one-dimensional water flow as well as the chloride and nitrogen transport through the unsaturated zone of the soil in an experimental site located on the Castellón Plain aquifer. s respect, we have considered a set of data measured in an experimental citrus grove during the year 1991. The agricultural soil was studied in order to determine some of the main biophysical parameters that were used in the simulations. Water uptake by plant roots was evaluated in the soil profile from regional evapotranspiration data and root densities. The model assumes that chloride is excluded from water uptake by plant roots and therefore it would be mainly concentrated in the remaining water of the soil. On the other hand, ammonium and nitrate are considered to be absorbed in the roots only by the water advection produced during the transpiration process of the plant. In general, the calculated chloride contents present some similarities with respect to the observed data. Concerning the NH+4-N contents, water uptake by roots, nitrification, immobilization and probably adsorption-desorption are the causes of the low concentration found in the soil water. The NO−3-N not involved in the immobilization process is leached towards the deepest levels; however, denitrification could be responsible for the loss of mass. The highest concentration of organic-N is found in the superficial level on the 180th day, just after the last fertilization was applied. Then, a quick decrease of the concentration is observed, two peaks appear coinciding with clayey levels where the soil moisture is larger and the mineralization of organic-N is more difficult. In general, there is some agreement between the measured data and the simulation results, especially in the deepest parts of the unsaturated zone, where the transpiration effect is not present. In superficial measurements, the differences between calculated and measured data can be explained by the possible overevaluation of irrigation rates as well as the lack of sufficient data. The lack of knowledge about the high lithological heterogeneity of the soil could also have some influence on the results.