Exploring the Implications of Modeling Choices on Prediction of Irrigation Water Savings

Abstract

AbstractWater use in irrigated agriculture is high, so reducing non‐beneficial consumption (non‐crop evaporation) of irrigation water is a potential option to free up water for other users. Yet, in practice, proper water accounting and accompanying policies (e.g., caps on irrigation expansion) are necessary to ensure that reduced non‐beneficial consumption results in extra water reallocated to other uses, rather than ending up as increased beneficial consumption (crop evaporation): paradoxically counteracting reductions. The extra water for reallocation is called water savings and several studies have questioned whether previous estimates of water savings sufficiently reflect the paradox described above. However, studies that focus on the uncertainty associated with quantified estimates of irrigation water savings are rare. This study addresses that gap and explores how the modeling choices that are necessary to represent the hydrological effects of reduced non‐beneficial consumption influence the prediction of water savings. Three such modeling choices include: (a) selecting a hydrologic model to partition irrigation water into consumed (e.g., evapotranspiration) and non‐consumed (e.g., runoff) components, (b) selecting a set of hydrologic model parameters, and (c) adopting a convention to represent non‐beneficial consumption. This study also attributes the uncertainty (i.e., variance) in predicted water savings to each of these choices. The results show that parameter selection and alternative conventions of representing non‐beneficial consumption are the largest sources of uncertainty in water savings, contributing ∼49% and ∼33%, respectively, to overall variance. These results provide a quantitative estimate for the minimum range of uncertainty one may expect when considering policy options that depend on quantified estimates of water savings.