133-6Improvement of Sensor-Based Yield Prediction Model in Rice Using Red Edge Reflectance.

See more from this Division: S04 Soil Fertility & Plant Nutrition
See more from this Session: Nutrient Management Using Precision Agriculture and Remote Sensing Technologies
Monday, October 22, 2012
Duke Energy Convention Center, Exhibit Hall AB, Level 1

Yumiko Kanke1, Dustin Harrell2, Marilyn Dalen3, Jasper Teboh4 and Brenda Tubana3, (1)Louisiana State University, Baton Rouge, LA
(2)LOUISIANA STATE UNIVERSITY AGCENTER, BATON ROUGE, LA
(3)School of Plant, Environmental, and Soil Sciences, Louisiana State University AgCenter, Baton Rouge, LA
(4)North Dakota State University Carrington Research Extension Center, Carrington, ND
Current yield prediction model for rice production in Louisiana was established using normalized difference vegetation index (NDVI). A known limitation of NDVI is its decreasing sensitivity when crop approaches canopy closure. In rice, it has been observed that at a high NDVI value, predicted grain yield is capped at a certain level which in most cases underestimate actual grain yields. The objective of this research was to evaluate the potential of red edge-based NDVI in improving grain yield prediction model in rice. A variety x N trial was established in 2011 in Crowley, LA using rice varieties CL152 and CL261, and five N rates at 0, 34, 101, 135 and 168 kg ha-1. Weekly collection of canopy reflectance readings were done for three consecutive weeks starting from one week after panicle initiation. Red edge position computed from the maximum point of first derivative reflectance showed strong linear relationship with yield for each sampling period (r2=0.93). Red edge-based vegetation indices were computed and regressed with grain yield. Red edge-based NDVI had higher association with grain yield (r2=0.89) compared with current red-based NDVI. Our study showed that mid-season yield prediction model can be improved by using red edge reflectance in rice.
See more from this Division: S04 Soil Fertility & Plant Nutrition
See more from this Session: Nutrient Management Using Precision Agriculture and Remote Sensing Technologies