Within-field wheat yield prediction from IKONOS data: a new matrix approach

Abstract

This study demonstrates a unique matrix approach to determine within-field variability in wheat yields using fine spatial resolution 4 m IKONOS data. The matrix approach involves solving a system of simultaneous equations based on IKONOS data and post-harvest yields available at entire field scale. This approach was compared with a regression-based modelling approach involving field-sensor measured yields and the corresponding IKONOS measured indices and wavebands. The IKONOS data explained 74–78% variability in wheat yield. This is a significant result since the finer spatial resolution leads to capturing greater spatial variability and detail in landscape relative to coarser spatial resolution data. A pixel-by-pixel mapping of wheat yield variability highlights the fine spatial detail provided by IKONOS data for precision farming applications.

Acknowledgments

The authors would like to thank Prof. Dr Adel El-Beltagy, Director General, and Dr Eddy De Pauw, Agroecologist of ICARDA for permission and support to conduct this research using ICARDA's facilities. Mr A. F. Tarsha is thanked for providing some of the information on ICARDA farms. The funding for the project comes from the National Aeronautics and Space Administration (NASA) Earth Science Enterprise grant number NAG5-3853.