Cognitive Based Decision Support for Water Management and Catchment Regulation

Abstract

The effect of climate change on water ecosystems include increased winter precipitation, severe floods, leading to fluctuations in stream flow in areas and affecting both fish survival and water supplies. Several methods exist for establishing projections of changes in precipitation with regards to river flows and water levels at the river-basin scale, but hydrological characteristics change remain difficult to predict. Ensuring optimization techniques for water systems becomes significantly important especially with the degradation of water ecosystems and increased risks for fish population. On the other hand, water demand has increased in the recent periods with the population growth. Further changes in the irrigation water system demand are determined by climate change precluding the reliability of current water management systems and affecting on the water-related ecosystems. To address these challenges real time water management and optimization strategies are required to facilitate a more autonomous management process that can address requirements for water demand, supply and ecosystem preservation. We present a cognitive based decision system that performs river level prediction for water optimization and catchment regulation for preserving Usk reservoir ecosystem in South Wales. The research is conducted on the Usk reservoir in South Wales reservation that is seeking to preserve the ecosystem and for which we propose a more informed decision system for catchment regulation and water management. Our system provides five days river level prediction to regulate river levels by pumping from/to reservoirs and to create artificial spates during the salmon migration season and to coincide with periods of low river flow.