Rapid groundwater potential mapping in data-scarce regolithic landscapes: a contribution to hydrogeology in humanitarian contexts

Abstract

The main objective of this thesis is to contribute to humanitarian efforts to provide responsive and reliable groundwater supply to displaced populations living in refugee camps or settlements across Sub-Saharan Africa. A large-scale transition is underway from rural towards peri-urban and urban settings, hence, from scattered hand pumps towards centralized motorized boreholes connected to distribution networks. However, with this shift in water supply strategy, the need to achieve yields an order of magnitude higher than those typically sought for hand pumps goes hand in hand with the need for changing the exploration strategy from ‘drilling where the people are’ to ‘drilling where the water is’. However, comprehensive hydrogeological assessments can often not be carried out as time is limited and data are scarce. It is against this backdrop that this applied research has been carried out, aiming towards developing a practical tool which can rapidly inform stakeholders in planning the siting of boreholes and on sustainable groundwater exploitation. The first section of this thesis is dedicated to the development of a rapid groundwater potential mapping methodology (RGWPM). It is based on the overlay of the two main groundwater relevant variables, i.e. the water availability (WA) reflecting hydrogeomorphological landscape units and the reservoir capacity (RC) being a proxy of the hydraulic properties, always retaining the lowest to define the groundwater potential (GWP). The RGWPM methodology was applied to the real case-study of Bidibidi refugee settlement (Northern Uganda) and used to implement eight new boreholes. The cross-validation revealed that its application significantly increased the average yield, justifying further applications in other settlements and camps. The overlay process of the RGWPM methodology revealed that the WA variable is almost always inferior and dependent on the RC, leading to a proposed revision to the RGWPM, relying only on the WA mapping. The revised version was again evaluated using the Bidibidi case-study with similar results and was subsequently applied to fourteen different refugee camps situated in similar regolithic landscapes in Sub-Saharan Africa. A cross-validation of the borehole yield with the mapped RGWPM units was carried out for all sites, again revealing a high degree of predictability. Although the revised approach was very useful and practical in visualising the spatial probability for GWP, it did so far not include any information on sustainable exploitation, as for instance how many high-yielding boreholes one unit can accommodate. The lack of information on sustainable exploitation of boreholes in the revised RGWPM methodology led to the second part of the thesis, which is dedicated to the exploration of the relationship between RGWPM units and water balance components, in particular groundwater recharge. A hydrogeomorphological analysis was carried out on twenty reference catchments, selected in similar geological (regolithic) and climatic contexts in Sub-Saharan Africa, for which all water balance components were known (i.e. including data from gauging stations) and for which the revised RGWPM units were translated into hydrogeomorphological landscape (HGM) units. The hydrogeological frameworks of these environments all fall into the topography-driven water table settings, where HGM units can be associated with surface and groundwater dynamics. The hydrogeomorphological-water balance analysis resulted in an empirical formulation of groundwater recharge based on the mapped HGM units, precipitation, and evapotranspiration. Once there was an approach to estimate groundwater recharge in any catchment with similar geological characteristics it was possible to introduce the notion of sustainable exploitation into the revised methodology. To do this on a real case-study, groundwater recharge was estimated using the empirical relationship for the Bidibidi settlement, for which the HGM units were directly obtained by analogy with the RGWPM units, while precipitation and evapotranspiration were obtained from remote sensing products. The Bidibidi RGWPM map was subsequently divided into sub-catchments, within which the cumulative extraction from motorised boreholes was compared to the sustainable groundwater potential, defined as a third of the sub-catchment groundwater recharge. This allowed mapping of the degree of sustainability of groundwater extraction within the sub-catchments. Some sub-catchments were identified to be in a state of over-exploitation, while others were identified where further groundwater development could be envisaged. In order to translate the sustainable groundwater potential into a map, the drainage sections corresponding to different sustainable cumulative sub-catchment extractions were added to the revised methodology. In the last section of the thesis, the correlation between the HGM units and the water balance components, expressed in the empirical relationship, was addressed by articulating the conceptual framework by a simplified analytical approach, and applying the obtained solution to the twenty reference catchments. Comparing the analytical results with the empirical solution indeed suggested a meta-physically based relationship between the HGM units, the water balance components, and the hydraulic properties, thereby capturing some essential complex interactions between regolithic landscapes and climate. On the one hand, this supports the easier-to-implement empirical solution, and, on the other hand, it opened a wide range of new perspectives showing the path to expanding this type of hydrogeomorphological analysis to other geological landscapes, also under diverse and changing climatic conditions.