Public defence

2020-04-29, digitally via conference (Zoom), public link, Stockholm, 13:00 (English)


Xu, Chong-Yu, Professor


Lyon, Steve W., Professor

Dissertation (pdf)

Klick here


Water is a vital resource for survival on the Earth. Sustainable management of water resources is therefore required for the wellbeing of present and future generations. A cornerstone of water resources management is scientific guidance supported by relevant data (in terms of quantity and quality). Most developing regions, where such guidance is crucial due to the intimate connection between natural resources and livelihoods, unfortunately face data limitations. This thesis aims to develop systematic approaches for informing water resources management in data limited regions. Specifically, this work targets Tanzania’s Kilombero Valley (KV) basin as an exemplar of a data limited region undergoing social-economic development through expansion and intensification of agriculture and other water-related interventions. Through a synthesis of lessons learned from the ongoing evolution of hydrological modelling development for water resources management in the Eastern Africa, several promising approaches were identified that could potentially be robust despite data limitations across the region. Putting these approaches into practice, recession analysis based on non-continuous discharge data in conjunction with estimations of the actual evapotranspiration (ET) using remote sensing techniques provided a basis to improve process understanding and help characterize the hydrological systems in the KV basin. This understanding translated into more-informed parameter estimation and improved accuracy when integrated into the development of a hydrological modelling framework using the Soil and Water Assessment Tool (SWAT) model. The modeling framework established for KV has potential to be used as tool for estimating impacts of water resources management strategies relative to future anthropogenic pressures and climatic changes. What is even more promising, is the possibility to derive scientific guidance to assist water resources management in a data limited region through implementation of an integrated workflow which employs state-of-the-science approaches. The methodological framework for model development adopted in this thesis could be applied in any data limited region facing similar challenges as those of the KV basin.