IWRA Proceedings

AbstractIn the sudano-sahelian zone, the struggle for food self-satisfaction since 60 years based on a green revolution has been a failure. The concept of rational innovative management requires the understanding of the variability of the agronomical spatio-temporal potentials of ecosystems. The study deals with the identification and hierarchisation of dominants factors that control the hydrological functioning and the fertility of soils and their modification under the impact of climatic change. The variability of millet yield is almost of the same order of magnitude at the scale of the hillslope (0 to 1136kg.ha-1) than at that of the field (21 to 1273 kg.ha-1) during a year of strong hydrological stress. These high values of extreme yields cannot result from the fertility of the different morphopedological units or from the agricultural practices. The analysis of the results show that the median yield on the slope is 228 kg.ha-1 against 448 kg.ha-1 on the terrace with coefficient of variations (CV) of 91% against 58%. On the terrace the hydrological report increases the median yield from 116 (CV 50%) to 446 kg.ha-1(CV 58%) and from 145 (CV 25%) to 1045 kg.ha-1 (CV 32%) on cheeselled and ploughed parcels. Ploughing with mineral fertilisation diminishes the variability of yield from 40 to 50% approximately. The runon which is a part of the run-off and is defined as “a natural, complementary and simultaneous irrigation to the rainfall that has generated it in according to natural (topographical, morphological, permeability) anthropogenic (soil work) conditions” must be taken into account in the hydrological balance. The increase of the hydrological and soil nutriments stock due to the runon is correlated to the slope, to the microtopography and to the soil work. These results demonstrate that this hydrological report completed by the deposit of nutriments and sequestration of organic carbon is the dominant factor of this variability. Peasants strategies to face the variability of soils potential have been poorly investigated and badly taken into account by agronomists; thus agronomical research must not ignore anymore the empiric peasant knowledge. With the continuous accentuation of the rain deficit and the erosion, the existence of the runon must be integrated in sustainable models of agricultural management. The multiscale simulation of expected yields on the morphopedological units of the hillslope will help to estimate the productivity gains due to the hydrological report. Then the innovative and traditional biophysical techniques to control the run-off and maintain the runon on the units with a good potential to secure an ecological sustainable development must be tested with the peasants.