Classification of Irrigated and Non-Irrigated Land Using Remote Sensing Techniques: A Case Study in Nebraska

Detailed and spatially explicit information on irrigated and non-irrigated land is critical to reliable estimation of recharge for building numerical groundwater models. Currently, such information is typically difficult and costly to compile, and its accuracy is often subject to considerable uncertainties. To develop a cost-effective land use product, a method integrating pixel-based remote sensing indexes and object-oriented classification is proposed for classifying irrigated and non-irrigated land. The method was tested in a study area in south-central Nebraska, and the results showed that the method can produce reasonable accuracy for irrigated and non-irrigated land classification. The method is expected to be applied in other parts of the state.