Global estimates of water used for irrigation are so inconsistent that they cannot be trusted—and no amount of finer computer modelling has fixed the problem. This matters because irrigation accounts for the largest share of human freshwater use, yet scientists cannot agree on how much water is actually being withdrawn. The disagreement points to deep, unexamined assumptions buried in the models themselves. The DAWN project will assemble a team of hydrologists, statisticians, philosophers, and anthropologists to expose those assumptions. They will also interview traditional irrigators—farmers who manage water by experience rather than equations—and compare their practical knowledge with the scientific rules embedded in global models. If successful, the research will not produce a single “correct” number. Instead, it will reveal where the uncertainties come from and how much they matter. This could transform how governments design irrigation policies—shifting from false precision to decisions that work even when the underlying numbers are uncertain. The work is fundamental science: it questions the conceptual foundations of a field that has chased ever-finer algorithms without first checking whether the algorithms rest on solid ground.
View original technical description
The volume of water globally withdrawn for irrigation agriculture is a measure of the human impact on freshwater resources. Although scientists have attempted to produce an accurate quantification of irrigation water withdrawals through increasingly detailed mathematical models, their estimates do not seem to converge. This discrepancy points towards the existence of deep uncertainties in our conceptualization of irrigation water withdrawals that evade the development of finer-grained algorithms. Without thoroughly exposing and understanding the relevance of these uncertainties, our knowledge of the influence that humans have on the hydrological cycle will remain on fragile grounds. DAWN will assemble a five-member team to unfold, examine and assimilate the deep uncertainties that condition our understanding of irrigation water withdrawals. Firstly, DAWN will unravel the underlying assumptions of all global irrigation water withdrawal models, ponder their effect on the estimations and assess the solidity of the main belief systems grounding the simulations. Secondly, DAWN will retrieve insights on irrigation withdrawals from traditional irrigators, and compare their understandings of the premises that govern irrigation water use with the scientific knowledge of irrigation embedded in global models. Thirdly, DAWN will combine the knowledge systems of scientists and traditional irrigators and develop cost-effective uncertainty/sensitivity analysis methods to explore how their ambiguities impact the modeling of global irrigation water withdrawals. By merging approaches from hydrology, statistics, philosophy and anthropology, DAWN proposes ground-breaking research to dramatically robustify our comprehension of irrigation withdrawals. This will ultimately enhance our capacity to design model-based irrigation policies that deliver under irreducible ambiguities.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know