Estimating Missing Hourly Climatic Data Using Artificial Neural Network for Energy Balance Based ET Mapping Applications.

Estimating Missing Hourly Climatic Data Using Artificial Neural Network for Energy Balance Based ET Mapping Applications Remote sensing based evapotranspiration (ET) mapping is an important improvement for water resources management. Hourly climatic data and reference ET are crucial for implementing remote sensing based ET models such as METRIC and SEBAL. In Turkey, data on all climatic variables may not be available at hourly time-step in all locations either due to cost constraints or due to equipment malfunctioning. In this study, the Artificial Neural Network (ANN) techniques was used to estimate missing hourly climatic data and reference ET for semi-humid Bafra Plains located in northern Turkey. Modeled and actual climatic and reference ET were used to derive ET maps from two Landsat Thematic Mapper data acquired on September 02, 2009 and August 04, 2010. Results indicated that that climatic data and reference ET estimated through ANN could be useful for mapping ET where hourly climatic data is missing or not available. Key Words: Evapotranspiration, METRIC, Artificial Neural Networks, Missing Climatic Data.