5. CONCLUSIONS
In this study, we established the water balance of a surface and subsurface drained field and compared the runoff processes of surface and subsurface drains according to the soil hydric status. The quantification of the surface and subsurface fluxes allowed to find the characteristics of the discharges of the drained fields. In our case, the period of intense drainage extends from November to March and corresponds to a period during which the saturated zone is close to the soil surface. During this period, most of the subsurface discharge is due to a slow matrix flow of water into the soil but macroporosity seems to be active and contributes to up to 30% of the subsurface discharge. Thus, it appears that rapid flows of water in soil to subsurface drains are present throughout the year and regardless of the hydric state of the soil. When the entire soil profile is saturated, a slow matrix flow of water from the soil to the drains occurs and stays dominant as long as the soil moisture content is greater than the field capacity. Surface drainage seems mainly due to a saturation-excess runoff but when the saturated zone approaches the soil surface, water from the first few centimeters of the soil can be transferred to surface drains. Moreover, according to the water balance, 44% of the water transiting through the soil infiltrate into the S2 horizon. At the end of the intense drainage period, the decrease in water content of the LA and S1 horizons to below field capacity results in the discontinuation of base flow for subsurface drainage. From mid-march to August, subsurface tile drains, like surface drains, only operate during intense rainfall events. However, the subsurface discharge comes from a preferential flow through the macropores while surface drainage is only due to a refusal of infiltration.
The calculation of the water balance proposed in this study makes it possible to predict the hydrological functioning of the drained field for the period of intense drainage. More particularly, it constitutes a model that makes it possible to anticipate the initiation of subsurface drainage. Provided that the variability of the soils and the characteristics of the drainage networks of the other fields in the watershed are taken into account, the simplicity of setting up this model should make it possible to extrapolate it to other fields. The application of this model on all the drained fields could then lead to the understanding of the hydrological functioning of the watershed and participate in the study of dissolved and particulate transfers from the fields to the watershed.
Moreover, according to the soil hydric state, water sources variations of subsurface runoff can be one of the factors explaining the variation in dissolved and solid transfers over the year in drained systems. Concerning surface drainage by digging SDRs, it can be observed that, in addition to the runoff process usually encountered in drained fields, there is a lateral flow of water from the soil to the SDRs when the saturated zone reaches the depth of the SDR. Such a flow is therefore likely to be associated with transfers of dissolved substances stored in the first few centimeters of soil.