International audiencePredicting soil hydraulic properties and understanding their temporal variability during the irrigated cropping season are required to mitigate agro-environmental risks. This paper reports field measurements of soil hydraulic properties under two drip irrigation treatments, full (FT) and limited (LT). The objective was to identify the temporal variability of the hydraulic properties of field soil under high-frequency water application during a maize cropping season. Soil hydraulics were characterized using the Beerkan infiltration method. Seven sets of infiltration measurements were taken for each irrigation treatment during the cropping season between June and September 2007. The first set was measured two weeks before the first irrigation event. The results demonstrated that both soil porosity and hydraulic properties changed over time. These temporal changes occurred in two distinct stages. The first stage lasted from the first irrigation event until the root system was well established. During this stage, soil porosity was significantly affected by the first irrigation event, resulting in a decrease in both the saturated hydraulic conductivity Ks and the mean pore effective radius ξm and in an increase in capillary length αh. These hydraulic parameters reached their extreme values at the end of this stage. This behavior was explained by the “hydraulic” compaction of the surface soil following irrigation. During the second stage, there was a gradual increase in both Ks and ξm and a gradual decrease in αh when the effect of irrigation was overtaken by other phenomena. The latter was put down to the effects of wetting and drying cycles, soil biological activity and the effects of the root system, which could be asymmetric as a result of irrigation with only one drip line installed for every two plant rows. The processes that affected soil hydraulic properties in the two irrigation treatments were similar. No significant change in ξm and αh was observed between FT and LT. However, as a result of daily wetting and drying cycles, which were strongest in LT, the soil in this treatment was found to be more conductive than that of FT. This showed that most of the changes in pore-size distribution occurred in the larger fraction of pores. The impact of these temporal changes on the dimensions of the wetting bulb was studied using a simplified modeling approach. Our results showed that there were marked differences in the computed width and depth of wetting bulb when model input parameters measured before and after irrigation were used. A temporal increase in capillary length led to a more horizontally elongated wetting bulb. This could improve both watering and fertilization of the root zone and reduce losses due to deep percolation. As a practical result of this study, in order to mitigate agro-environmental risks we recommend applying fertilizers after the restructuration of tilled soil. Further studies using improved models accounting for temporal changes in soil hydraulic properties are needed

IntroductionIrrigation has a dominant role in agricultural production especially in Mediterranean climate, because of variant distribution of the rainfalls over the year. Improving irrigation management is important not only for saving water, but also for improving crop profitability. In addition, there is a growing competition for water by agricultural, domestic and industrial uses, hence there is a need for farmers to save water and make judicious use of it, especially during the dry season. Efficient use of water in agriculture requires proper irrigation scheduling and sowing date to obtain optimum water use and yield (Adekalu and Fapohunda, 2006). Field experiments are time consuming, expensive, and limited to the prevailing soil, climate, and crop … conditions. The experiments generally conducted in this aim, often give a partial response because the range covered (of soils types, climates, crop managements) are limited compared with agricultural conditions in which these systems can be used. Furthermore, the required time to get a response is too long considering the rapid change of varieties and all the cropping system components used and modified by farmers. Model simulations in different climates can help the farmer to better identify the best crop management towards different cropping system. Jamieson et al. (1998) believe that developing empirical models provides a good basis for decision support at the farm level by giving quick estimations of the likely costs and benefits of farm management decisions. Models that satisfactorily simulate the impacts of water stress on yield can be reliable tools in irrigation management (Cavero et al. 2000). In addition, crop models are useful tools for considering the complex interactions between a range of factors that affect crop performance, including weather, soil properties and management (Timsina and Humphreys, 2003). Where pests and diseases are controlled, and nitrogen is not a limiting factor, water management is the main factor influencing yield for a given environment. Mechanistic crop models typically require a large number of parameters and are therefore highly data-demanding to give accurate and reliable simulation results. Even