DURHAM, N.C. – Groundwater depletion in parts of the High Plains is so extreme that peak grain production in some states has ended and production is now declining, a new Duke University-led study by a team of international scientists finds.

The scientists calculated annual groundwater use and grain production over the last 50 years in three High Plains states – Texas, Kansas and Nebraska – using U.S. Department of Agriculture and U.S. Geological Survey data. They projected future trends based on these numbers, using a new mathematical model based on equations previously used to predict peaks and declines in crude oil production.

“Our model shows that a decline in irrigated grain production will continue through at least 2050 in Texas and Kansas, which receive less rain to replenish dwindling groundwater than Nebraska does, and will grow hotter and drier under future climate change,” said Assaad Mrad, a doctoral student at Duke’s Nicholas School of the Environment, who led the study.

Hard-hit farmers in Texas and southern Kansas will increasingly need to revert to dryland agriculture, which produces lower yields because crops are watered only by rainfall. “If you irrigate corn, you get yields of 12 tons per hectare on average; if you rely only on rain, you get 8 tons,” Mrad said. “That’s a significant decrease.” Wheat, sorghum and other crops will also see decreases.

Grain production in Nebraska is predicted to continue rising through 2050 due to projected increased rainfall and milder temperature impacts from climate change in the northern Plains.

“While Texas and Nebraska had equal irrigated grain output in 1975, our analysis projects Nebraska will have almost 10 times the groundwater-based production as Texas by 2050,” said Gabriel G. Katul, Theodore S. Coile Distinguished Professor of Hydrology and Micrometeorology at Duke, who co-authored the study.

Mrad, Katul and their colleagues published their peer-reviewed study the week of October 5 in The Proceedings of the National Academy of Sciences.

The High Plains region produces more than 50 million tons of grain yearly but relies on just one groundwater source – the Ogallala aquifer, which underlies parts of eight states – for as much as 90% of its irrigation needs, especially in arid regions like the Texas panhandle.

While past methods for predicting peak production have focused on analyzing single resources such as crude oil, Mrad designed the model employed in the new study so that it can analyze groundwater use and grain production in tandem, thus yielding novel insights that might not be obtained from looking at each resource in isolation.

“Analyzing them together provides much more accurate results and allows us to more clearly identify what factors drive peak water use and peak grain production and contribute to the lag times between them and the declines that follow,” Mrad said. “This is incredibly useful information for scientists, policymakers, farmers and water managers alike.

“If groundwater withdrawals from the already overtaxed Ogallala continue at present rates and no new technologies are introduced to improve irrigation sustainability, we could see a collapse of grain production in parts of the southern Plains that could affect food security worldwide,” he said. “There is a historical precedent. In Texas, overuse of this resource over the last four decades has already twice resulted in non-sustainable peak water use and peak grain production, followed by rapid crashes.”

Funding for the study came from the National Science Foundation.

Researchers from 22 other universities or research institutions co-authored the study, which stemmed from the October 2018 Ettersburg Ecohydrology Workshop in Germany.

CITATION: “Peak Grain Forecasts for the U.S. High Plains Amid Withering Waters,” A. Mrad, G.G. Katul, D.F. Levia, A.J. Guswa, E.W. Boyer, M. Bruen, D.E. Carlyle-Moses, R. Coyte, I.F. Creed, N. van de Giesen, D. Grasso, D.M. Hannah, J.E. Hudson, V. Humphrey, S. Iida, R.B. Jackson, T. Kumagai, P. Llorens, B. Michalzik, K. Nanko, C A. Peters, J.S. Selker, D.Tetzlaff, M. Zalewski and B.R. Scanlon; Oct. 5, 2020. Proceedings of the National Academy of Sciences. DOI: www.pnas.org/cgi/doi/10.1073/pnas.2008383117