Publications

Maloom, Juanito M.; Saludes, Ronaldo B.; Dorado, Moises A.; Cruz, Pompe C. Sta. (2014). Development of a GIS-Based Model for Predicting Rice Yield. PHILIPPINE JOURNAL OF CROP SCIENCE, 39(3), 8-19.

Abstract
Monitoring the growth of rice and forecasting its yield before harvest season is important for crop and food management. Remote sensing images are capable of identifying crop health as well as in predicting its yield. This study was conducted to develop a functional model for predicting rice yield from remotely sensed data. The Normalized Difference Vegetation Index (NDVI) calculated from remote sensing images has been widely-used to monitor crop growth and relate it to crop yield. This study used 16-d composite Terra MODIS NDVI data from December 2008 to April 2009 to predict rice yield in Dingras, Ilocos Norte, Philippines. Nine yield prediction models were developed through linear regression analysis (stepwise method) between NDVI and observed yield, among which Model 3 had the highest potential. It can predict yield way ahead before harvesting. The R-2 value of the models ranged 0.36-0.75, which means that about 39-75% of the yield variability could be explained by the models. The variability between predicted and actual yields are due to factors not considered in the model such as type of soil, varieties planted, weather and other cultural management practices, such as water, nutrient and pest managements used on the standing crop studied.

DOI:

ISSN:
0115-463X