Long-Term Application of the Crop Water Stress Index in Midwest Agro-Ecosystems.

Agricultural land in the Midwest is largely used for rainfed corn (Zea mays L.) and soybean (Glycine max (L.) Merr.) production. In addition, remnants of the native tallgrass prairie still exist, and reconstruction in on-going. Although annual rainfall typically exceeds evapotranspiration, periodic water stress can occur in the summer months which can affect carbon assimilation of those agro-ecosystems. In addition, high spring precipitation can result in very wet to water logged soil conditions. In this study, the theoretical Crop Water Stress Index (CWSI) concept was applied using eddy flux measurements coupled with infrared thermocouple sensor (IRT) data for corn and soybean canopies and a reconstructed tallgrass prairie site in Central Iowa. Therefore, the relationship between CWSI and eddy flux derived net ecosystem production (NEP) and ET as well as volumetric water content (VWC) at 10 cm soil depth was investigated. Average CWSI values varied greatly among years and agroecosystems from 0.1 to 1.1. NEP significantly decreased at all three sites with increasing CWSI (p < 0.05), showing that water stress adversely affected carbon assimilation of crop and grassland ecosystems in the Midwestern US. Prairie CWSI significantly decreased with increasing VWC (r = -0.45, p < 0.001). In addition, CWSI of corn and soybean increased with inclining VWC (p < 0.05), indicating that the CWSI concept not only applies to drought, but also to wet or water logged soil conditions. CWSI also decreased with increasing ET of soybean (r = -0.24, p < 0.001) and prairie (r = -0.29, p < 0.001), which is probably related to evaporative cooling. We conclude that the theoretical CWSI concept can be applied to measure crop water stress in rainfed corn and soybean production as well as native grasslands, induced by both dry and very wet soil moisture conditions.