Search

Abstract Past water resource developments in dryland regions rarely estimated the full suite of environmental and social impacts arising from damming and diverting water for food production. Nowadays there is a greater focus on sustainable resource development which considers the economic, social and environmental costs and benefits. A challenge is to apply tools and methods which can capture the often disparate knowledge and data describing many costs and benefits. This paper describes a proof of concept application of a large Bayesian Decision Net to estimate the total utility of water resource development according to criteria of social, economic and environmental sustainability. We focus on two water-scarce catchments in remote northern Australia which are under investigation for development. The study catchments contain a diverse set of ecosystem services and socio-cultural values, including important Indigenous values and high value freshwater ecosystems. The Bayesian Decision Net was shown to have many properties that made it useful for performing a social, economic and environmental sustainability assessment, in particular its ease of construction; its ability to handle quantitative and qualitative data types; its preservation of system knowledge and; its ease of use in aiding decision making. From the perspective of the sustainability assessment in our case study, the total utility of water resource development for new irrigation is negative in both the studied catchments. The overall utility of water resource development could be positive if irrigation development is highly sensitive to the environment and there are very low environmental impacts, and much higher net economic returns to irrigators eventuate, possibly through higher commodity prices, lower capital costs of irrigation development or some combination of both. Highlights • Development of water resources involves trade-offs between economy and environment. • Integrated social-environment-economy sustainability (SEES) assessments support trade-offs analyses. • Bayesian Decision Networks (BDNs) support SEES assessments with diverse data types. • BDN applied to irrigation development scenarios of two remote catchments in north Australia. • The economic and ecosystem service benefits of the case study irrigation developments are questionable. [ABSTRACT FROM AUTHOR]