[PhD defence] 29/11/2024 - Irène Kinoti: "Remote sensing for the assessment of transboundary groundwater resources - The case of the Stampriet aquifer system (STAS)" (UMR EMMAH)

Irène KINOTI will submit her thesis on 29 November 2024 on the subject of "Remote sensing for the assessment of transboundary groundwater resources - The case of the Stampriet aquifer system (STAS)".

Laboratory

UMR 1114 EMMAH - Mediterranean Environment and Modelling of Agrohydrosystems

Composition of the jury

MR ALBERT OLIOSO	INRAE	Thesis supervisor
Mr Marc LEBLANC	University of Avignon	Thesis co-director
Mr Maciek LUBCZYNSKI	University of Twente	Thesis co-supervisor
Ms Mounia TAHIRI	Mohammed V University, Faculty of Science	Rapporteur
Frédéric FRAPPART	INRAE	Rapporteur
Mr André CHANZY	INRAE	Examiner

Summary

The Stampriet Transboundary Aquifer System (STAS) is a key multi-layered aquifer shared by Namibia, Botswana and South Africa. It is a key water resource for the region, prompting UNESCO and local stakeholders to initiate studies into its sustainable management. This PhD is a continuation of previous work carried out as part of STAS, with the ultimate aim of developing an integrated hydrological model (IHM). The first objective of this PhD was to develop a conceptual hydrogeological model of the STAS. Six hydrostratigraphic units consisting of four aquifers, including a new one (the Dwyka), and two aquitards were identified. The analysis revealed that groundwater in the STAS flows in two directions, indicating the presence of a regional groundwater divide. The hydrostratigraphic and 3D geometry work also made it possible to increase the size of the STAS by 42 % compared with the limit initially delimited by UNESCO. With rainfall being the main driving force behind the HMIs, the second objective was to evaluate the performance of thirteen gridded rainfall products (GPPs) for the whole of Botswana and Namibia in terms of rainfall detection capability, quantitative accuracy and bias decomposition, in order to select the most appropriate for this study. The results showed that Multi-Source Weighted-Ensemble Precipitation (MSWEP) performed best on all the criteria used. Groundwater models and HMIs are still commonly evaluated using in situ measurements of groundwater levels and streamflow. However, these measurements are often sparse and do not represent groundwater dynamics at basin scale, which subjects the models to parametric uncertainties. The Gravity Recovery and Climate Experiment (GRACE) provides monthly anomalies in terrestrial water storage relative to a long-term average, and its potential value in HMIs lies in its ability to constrain changes in storage simulated by HMIs. The third objective of this PhD was to use GRACE terrestrial water storage (TWS) in addition to groundwater levels to constrain the STAS HMI through a multi-objective optimisation criterion. Storage change in the unsaturated zone was found to be the main component of subsurface water storage change in STAS, as most infiltrated water is stored in this zone where it eventually evapotranspires, and unsaturated zone parameters therefore play a major role in controlling subsurface fluxes. However, as most studies do not incorporate state variables for the unsaturated zone, optimisation of these parameters is quite difficult and, as a result, the values of these parameters are often obtained from the literature and are not optimised during calibration. The inclusion of GRACE TWS as an additional state variable in the STAS HMI showed that non-optimisation of the UZF parameters could lead to underestimation of gross and net recharge. It has also been observed that using GRACE TWS alone to constrain an HMI runs the risk of overestimating gross and net recharge. It is therefore important to use GRACE TWS with other state variables such as groundwater measurements or actual evapotranspiration to constrain the HMI.

Keywords Conceptual hydrological model, Rainfall products per grid cell, Groundwater, Integrated hydrological modelling, Teledetection, Stampriet transboundary aquifer system