Skip to main content

Smart water management using data-driven hydrological model based on artificial intelligence and big data


Traditional hydrological models were developed based on model structures that heavily rely on traditional in-situ observations, such as measurements from rain gauges and hydrological stations. The breakthroughs in observations and information technologies in the past decades enable the accumulation of a variety of earth observation data (e.g. remote sensing datasets) and non-traditional observation data (e.g. photos of hazards impacts from social media). Because traditional hydrological models were not developed for digging out the insight of these newly developed big data in a variety of types, how to make use of the big data to improve water management is an emerging challenge in hydrological studies. Therefore, the proposed study aims to identify and collect different types of data (from remote sensing to non-traditional data in social media), and to develop a data-driven hydrological model on the basis of the big data and the recent development of artificial intelligence techniques for smart water management. The performance of the proposed data-driven model will be evaluated in terms of its accuracy, efficiency, and feasibility for smart water management.

Required skills

Understanding of hydrology, with skills in geographical information systems, remote sensing, and computer programing tools.

Principal Investigator


Prof. Qiming Zhou

Prof. Qiming Zhou 

Associate Dean (Research), Faculty of Social Science
Professor, Department of Geography




Dr. Jianfeng Li

Dr. Jianfeng Li

Assistant Professor,  Department of Geography