Evaluation of a hybrid remote sensing evapotranspiration model for variable rate irrigation management

Authors

J. Burdette Barker, University of Nebraska-LincolnFollow
Christopher M. U. Neale, University of Nebraska-LincolnFollow
Derek M. Heeren, University of Nebraska-LincolnFollow

Date of this Version

11-2015

Citation

An ASABE Meeting Presentation, Revised Nov. 9, 2015, Paper Number: 152142641.

Comments

Copyright by the authors. Used by permission.

Abstract

Accurate generation of spatial irrigation prescriptions is essential for implementation and evaluation of variable rate irrigation (VRI) technology. A hybrid remote sensing evapotranspiration (ET) model was evaluated for use in developing irrigation prescriptions for a VRI center pivot. The model is a combination of a two-source energy balance model and a reflectance based crop coefficient water balance model. Spatial ET and soil water depletion were modeled for a 10 km² area consisting of rainfed and irrigated maize fields in eastern Nebraska for 2013. Multispectral images from Landsat 8 Operational Land Imager and Thermal Infrared Sensor were used as model input. Modeled net radiation and soil heat fluxes compared well with measurements from eddy covariance systems located within three fields in the study area. Modeled sensible heat flux did not compare well. Latent heat flux compared well for the only mid-summer image, but poorly for the one spring and two fall images. The water balance ET compared well with the two-source energy balance ET for irrigated maize, but not for dryland maize. Image frequency is thought to be a contributing factor in the poor performance of the water balance. In 2015 the hybrid model will be used to generate irrigation prescription maps for a VRI system located in the study area based on modeled soil moisture depletion. Future research will focus on model parameterization and utilize aerial imagery and satellite imagery from other sensors for improved image frequency. Note: this is a revision of the original paper correcting erroneous data where one of the flux sites was mistakenly analyzed as soybeans, when it was actually maize. Mean biased error signs have also been corrected.