Water conservation in US irrigation – the RHEO tool is now available

Last year, Future Farming readers were introduced to a tool addressing these issues called RHEO (regional hydro-economic optimization). It was created by Dr Hemant Kumar, now at the Indian Institute of Technology, Dr Sankar Arumugam at North Carolina State University and Dr Tingju Zhu of Zhejiang University.

RHEO works at the county level, gathering long-term and seasonal rainfall forecasts, groundwater level data from the US Geological Survey and soil characteristics. The model combines all that with data on the water consumption of each crop, the cost of irrigation on a county basis, crop price data and crop production budget data.

The current tool is for regional water managers. It is open-source and free of charge

RHEO then predicts the yield per unit area for each crop on a county-level basis, but also forecasts the irrigation costs for each crop for various irrigation strategies and recommends the crop and irrigation strategy that’s most profitable and environmentally-sustainable.

The current tool is for regional water managers. It is open-source and free of charge, explains Kumar. “It can be immediately used by the regional water planners in the Flint River Basin and the immediately surrounding regions. We can expand it to other crops based on the feedback from the community.”

_{Text continues below picture}

Completion of tool

Development of the RHEO model took about 8 months. Kumar points out that the goal was not merely reproduce the historical observations, but also incorporate the information on basin-related characteristics and crop production practices. It also needed to account for spatial variability in soil, weather, water availability and so on.

The first challenge facing the team was to merge the datasets from different sources, which had varying lengths and precision in terms of their records. “For example the annual crop yields and harvested acreage were available for the entire 30 years of study period but the percentage of acreage under irrigation was available once every five years,” Kumar explains.

The second challenge was to merge the modules with each other. The first module is a crop yield simulation module which runs at a daily scale but the third module (optimization) runs at seasonal time scale. The scientists therefore had to develop a hierarchical model based on Bayesian inference to link these two modules. “This hierarchical model also helped us reduce the runtime of the optimization module,” adds Kumar.

First water, biggest impact

In setting up RHEO to determine the irrigation strategy that’s the most profitable and environmentally sustainable, Kumar says he and his colleagues made use of the phenomenon that’s long been noted in crop watering in times of water scarcity. That is, the first ‘drink’ of water supplied to a crop at these times results in the highest gain in crop yield; each subsequent unit of irrigation water results in smaller and smaller increments in eventual yield.

The changing climate (greater risk of drought) is prompting more and more farmers to turn to irrigation in the US and beyond

“Thus, our tool found that deficit irrigation is most profitable and environmentally sustainable,” says Kumar.” (In deficit irrigation, water is only given when needed and in the minimal amount.)

Irrigation expansion

The changing climate (greater risk of drought) is prompting more and more farmers to turn to irrigation in the US and beyond. Tools like RHEO can help support this expansion because they promote maximum water conservation – the same water for more land.

Indeed, Kumar notes that there are two ways RHEO can relieve pressure on over-exploited groundwater aquifers. One is through providing data to support the use of more-efficient irrigation strategies for existing crops and the other, in areas of particular water scarcity, is to explore the cultivation of less water-intensive crops.