Jamila Ouardi

In Brazil, the development of gullies constitutes widespread land degradation, especially in the state of South Mato Grosso, where fighting against this degradation has become a priority for policy makers. However, the environmental and anthropogenic factors that promote gully development are multiple, interact, and present a complexity that can vary by locality, making their prediction difficult. In this framework, a database was constructed for the Rio Ivinhema basin in the southern part of the state, including 400 georeferenced gullies and 13 geo-environmental descriptors. Multivariate statistical analysis was performed using principal component analysis (PCA) to identify the processes controlling the variability in gully development. Susceptibility maps were created through four machine learning models: multivariate discriminant analysis (MDA), logistic regression (LR), classification and regression tree (CART), and random forest (RF). The predictive performance of the models was analyzed by five evaluation indices: accuracy (ACC), sensitivity (SST), specificity (SPF), precision (PRC), and Receiver Operating Characteristic curve (ROC curve). The results show the existence of two major processes controlling gully erosion. The first is the surface runoff process, which is related to conditions of slightly higher relief and higher rainfall. The second also reflects high surface runoff conditions, but rather related to high drainage density and downslope, close to the river network. Human activity represented by peri-urban areas, construction of small earthen dams, and extensive rotational farming contribute significantly to gully formation. The four machine learning models yielded fairly similar results and validated susceptibility maps (ROC curve > 0.8). However, we noted a better performance of the random forest (RF) model (86% and 89.8% for training and test, respectively, with an ROC curve value of 0.931). The evaluation of the contribution of the parameters shows that susceptibility to gully erosion is not governed primarily by a single factor, but rather by the interconnection between different factors, mainly elevation, geology, precipitation, and land use.

The main landfill in the city of Rabat (Morocco) is based on sandy material containing the shallow Mio-Pliocene aquifer. The presence of a pollution plume is likely, but its extent is not known. Measurements of spontaneous potential (SP) from the soil surface were cross-referenced with direct measurements of the water table and leachates (pH, redox potential, electrical conductivity) according to the available accesses, as well as with an analysis of the landscape and the water table flows. With a few precautions during data acquisition on this resistive terrain, the results made it possible to separate the electrokinetic (~30%) and electrochemical (~70%) components responsible for the range of potentials observed (70 mV). The plume is detected in the hydrogeological downstream of the discharge, but is captured by the natural drainage network and does not extend further under the hills.

In many parts of the world, the impact of open landfills on soils, biosphere, and groundwater has become a major concern. These landfills frequently generate pollution plumes, the contours of which can be delineated by non-intrusive geophysical measurements, but in arid environments, the high soils resistivity is usually an obstacle, which results in the low number of studies that have been carried out there. In addition, such prospecting using geophysical techniques do not provide information on the intensity of the processes occurring in the water table. This study was carried out on an uncontrolled landfill in the arid Tadla plain, Morocco’s main agricultural region. A survey based on geo-referenced spontaneous potential measurements was combined with measurements of anoxic conditions (Eh-pH and O₂ equilibrating partial pressure) in the groundwater and leachates, in order to highlight a pollution plume and its geometry. The range of spontaneous potential measurement is wide, reaching 155 mV. Ponds of leachate with high electrical conductivity (20 to 40 mS cm⁻¹) form within the landfill, and present very reducing conditions down to sulphate reduction and methanisation. The plume is slowly but continuously supplied with these highly reducing and organic carbon-rich leachates from the landfill. Its direction is towards N-NW, stable throughout the season, and consistent with local knowledge of groundwater flow. The fast flow of the water table suggests pollution over long distances that should be monitored in the future. The results obtained are spatially contrasting and stable, and show that such techniques can be used on a resistive medium of arid environments.

Drainage network patterns influence the hydrological response of the watersheds and must be taken into account in the management of the water resource. In this context, it is important to identify the factors that control the configuration of drainage networks in and beyond specific climatic conditions. Here, we study 318 basins spread over three sectors (arid, semi-arid, and semi-humid) of Morocco where seven drainage network patterns have been identified. From each basin, 14 parameters were extracted, describing the relief, geology, morphometry, drainage network, land cover, precipitation, and time of concentration (Tc). Principal component analysis (PCA) and discriminant analysis (DA) processing were performed on the entire database and on each sector separately. The results show that the drainage network pattern is a feature of the landscape that contributes significantly to the variance of the basins. They suggest that the distribution of network patterns is controlled by the relationship between the different parameters, mainly those related to the relief, more than by the variations of each parameter taken individually. The network discrimination rate is 63.8%, which improves when each sector is treated separately. Confusion in discrimination are similar across all sectors and can be explained by similar conditions (active tectonic, deformation, and uplift) or transitions from one network pattern to another, due to the landscape evolution of certain sectors. A contribution of climatic variables appears locally but was attributed to a statistical coincidence, these parameters presenting a distribution close to that of the relief and geology variables.