Assessment of irrigation physics in a land surface modeling framework using non-traditional and human-practice datasets

Abstract. Irrigation increases soil moisture, which in turn controls water and energy fluxes from the land surface to the planetary boundary layer and determines plant stress and productivity. Therefore, developing a realistic representation of irrigation is critical to understanding land–atmosphere interactions in agricultural areas. Irrigation parameterizations are becoming more common in land surface models and are growing in sophistication, but there is difficulty in assessing the realism of these schemes, due to limited observations (e.g., soil moisture, evapotranspiration) and scant reporting of irrigation timing and quantity. This study uses the Noah land surface model run at high resolution within NASA's Land Information System to assess the physics of a sprinkler irrigation simulation scheme and model sensitivity to choice of irrigation intensity and greenness fraction datasets over a small, high-resolution domain in Nebraska. Differences between experiments are small at the interannual scale but become more apparent at seasonal and daily timescales. In addition, this study uses point and gridded soil moisture observations from fixed and roving cosmic-ray neutron probes and co-located human-practice data to evaluate the realism of irrigation amounts and soil moisture impacts simulated by the model. Results show that field-scale heterogeneity resulting from the individual actions of farmers is not captured by the model and the amount of irrigation applied by the model exceeds that applied at the two irrigated fields. However, the seasonal timing of irrigation and soil moisture contrasts between irrigated and non-irrigated areas are simulated well by the model. Overall, the results underscore the necessity of both high-quality meteorological forcing data and proper representation of irrigation for accurate simulation of water and energy states and fluxes over cropland.