Irrigation benefits outweigh costs in more US croplands by mid-century

Projected increases in global food demand and rising temperatures in the United States necessitate exploring climate adaptation strategies for agriculture. Expanding irrigation is a potential solution, but its sustainability is uncertain due to future water availability. While irrigation significantly boosts crop yields (often accounting for over half of total crop sales in the US, despite being used on less than 20% of cropland), it's the largest consumer of freshwater resources globally. Past irrigation has caused environmental damage, including depleting groundwater and harming natural water flows. The feasibility of irrigation as a climate change adaptation strategy is therefore questioned due to concerns about water resource depletion and climate variability. Previous assessments have offered varying conclusions, with some suggesting a decline in total irrigated water applied due to water scarcity or reduced demand, while others highlight the need for substantial irrigation increases to meet future crop water demands. These discrepancies arise from uncertainties surrounding future precipitation changes and methodological differences in crop models and studies. This study aims to address these uncertainties by projecting the future value of water used for irrigation at a high resolution across the US, combining these projections with estimates of water availability to identify areas where irrigation could be both profitable and sustainable.

The paper reviews existing literature on the impact of climate change on agriculture and the role of irrigation as an adaptation strategy. It highlights the conflicting conclusions of previous studies regarding future irrigation requirements, which are partly due to uncertainties in precipitation projections and methodological differences in crop models. The review emphasizes the need for a high-resolution projection of the future value of irrigation water in the US, considering both benefits (marginal yield gains) and costs (capital, environmental, and social costs). Global analyses have advanced understanding of current irrigation benefits and water values, but a detailed projection for the US remains lacking.

The study uses a gridded version of the Decision Support System for Agrotechnology Transfer model (pDSSAT) to simulate maize and soybean growth at a 5-arcminute resolution across the US. Simulations were run for historical (1981-2010), mid-century (2036-2065), and end-of-century (2071-2100) periods under moderate (SSP245) and high (SSP585) emissions scenarios from five statistically downscaled CMIP6 Global Climate Models (GCMs). The GCM ensemble was selected to represent a range of future climate projections. Historical simulations were calibrated using USDA data on crop management, progress dates, and yields. Future simulations incorporated potential adaptations in planting dates and maize cultivar selection, based on a random forest model trained on historical data. Irrigation was simulated by applying 10 mm of water when plant available water fell below 40%. The study assessed irrigation feasibility using two metrics: (1) the irrigation benefit-to-cost ratio (IrB/C), which compares marginal yield gains from irrigation to pumping and equipment costs, and (2) the irrigation groundwater deficit, comparing available water (groundwater recharge minus other uses) to irrigation water applied. Costs were calculated using regional energy prices, depth to water, and irrigation equipment costs. Future groundwater availability was assumed to be at current levels. The analysis considered 30-year averages and also explored interannual variability.

The study projects significant increases in mid-century irrigated and rainfed yields throughout much of the Corn Belt and eastern US, with minor decreases in the southern High Plains under a SSP245 emissions scenario. By the end of the century, significant yield decreases are projected in the southern High Plains, but yields in the Midwest remain above current levels under SSP245. However, SSP585 results indicate substantial yield reductions across much of the US. Soybean yields show greater increases than maize, particularly under irrigation. The economic return on irrigation (IrB/C) shows that historically irrigated areas in the High Plains have the highest returns, but mid-century projections suggest increased IrB/C values in the north-central US and upper Midwest, particularly for soybeans. The frequency of positive IrB/C values also increases significantly in the future, especially for soybeans. However, many currently irrigated areas, particularly in the southern High Plains and California's Central Valley, show substantial water deficits. The study finds an expansion in areas where irrigation could be both profitable and sustainable, particularly for soybeans in the eastern US and maize in northern Michigan and Wisconsin. This suggests an increased incentive for irrigation installation and thus additional pressure on groundwater resources in regions with limited recharge rates. Notably, the analysis reveals a possible continued incentive for unsustainable irrigation practices in the High Plains despite increasing costs. In addition, changing the cultivar to a longer season type will likely improve the yields, especially for maize.

The findings show that while maize and soybean yields in the Midwest may remain near current levels without significant irrigation investment through mid-century, irrigation remains a crucial drought mitigation strategy, as the frequency of droughts is expected to increase. The increased profitability of soybean irrigation in the East and Midwest could lead to shifts in cropping patterns and increased pressure on water resources. The study highlights the complexity of irrigation decisions, balancing yearly costs, market values, weather forecasts, and risk mitigation. The authors acknowledge the limitations of their simplifying assumptions about costs and the complexities of farmer decision-making. The results emphasize the need for policymakers to consider local water availability and sustainability when promoting irrigation expansion.

This research provides valuable insights into the future of irrigation in US agriculture. While mid-century yields in the Midwest are projected to remain high with limited irrigation, expanded soybean irrigation appears feasible and profitable in multiple eastern US regions. Irrigation, particularly for soybeans, emerges as a vital adaptation strategy in many areas, but sustainability remains a critical concern requiring policy measures to manage water withdrawals and optimize resource allocation. Future work should focus on refining groundwater availability scenarios, exploring deficit irrigation strategies, and addressing the uncertainties in future crop management practices and land-atmosphere interactions.

The study's model makes several simplifying assumptions, including constant future crop and energy prices, constant future groundwater levels (despite potential changes in pumping rates), and a simplified irrigation scheme. The projection of future adaptations in crop selection and management practices is also uncertain. The model does not account for the impact of irrigation on local and regional climate. Finally, the results may be subject to inherent limitations within the pDSSAT crop model.