Agricultural water stress and risk assessment using integration of catastrophe theory and entropy methods

Abstract

Water is increasingly becoming a major constraint in agricultural development in many countries across the world. It is anticipated that global environmental change will deteriorate the condition further in near future. The objective of this research is to develop an evaluation approach for systematic assessment of spatial distribution of water stress and its risk to society in order to aid mitigation planning. Bangladesh is used as a case study in this research. Criteria based approach is used to select the water stress indicators. A data driven method is proposed through the integration of catastrophe theory and entropy methods for assigning weights to indicators for the estimation of water stress and its risk in a geographical information system (GIS). The novelty of the integrated method is that it can be used in multi-criteria decision making by avoiding personal judgment. Evaluation of the performance of the integrated method shows that it can predict water stress with an accuracy of 0.9, which is much higher than catastrophe (0.82) and entropy theory (0.8) based methods. Application of proposed method in Bangladesh reveals that about 5.7% area of the country is experiencing very high water stress, 32.1% area is high water stress, and the rest is moderate or no water stress for agriculture. Agriculture in western part of Bangladesh experiences more water stress compared to other parts of the country. The vulnerability map shows 25.1% area is very high vulnerable and 28.9% area is high vulnerable to water stress. The northwest region of the country is more vulnerable to water stress compared to other parts. Integration of water stress and vulnerability maps shows that northwest and southeast parts of Bangladesh have the highest risk to water stress. The analysis of the results reveal that higher agriculture water stress in northwest region resulted from water unavailability and in southwest region is closely related to poor water quality. The areas of highest water stress correspond very well with the areas that are usually thought as water scarce.