Tuesday, November 3, 2009
Convention Center, Exhibit Hall BC, Second Floor
Abstract:
The development of a more effective nitrogen (N) decision tool using remote sensor is an important step in establishing a more profitable and environmentally-sound sugarcane industry in Louisiana . The use of remote sensor-based nitrogen decision tools in wheat and corn production has resulted in improved fertilizer N use efficiency. To develop a similar tool for sugarcane, the first and major step is to determine the feasibility of using early- to mid-season canopy reflectance to predict sugarcane yield potential and responsiveness to N fertilization. Collection of normalized difference vegetation index (NDVI) readings using the GreenSeekerTM handheld sensor from multiple soil fertility trials located at the LSU AgCenter Sugar Research Station, St. Gabriel , LA was initiated in 2008. A total of 12 sensing dates were conducted from mid-April to first week of July from sugarcane plots planted with different varieties (LCP85-384, HoCP96-540, Ho95-988, L97-128, L99-233, and L99-226) of different crop age. Regression analysis was performed to determine the relationships of NDVI with cane and sugar yield, and the association between the NDVI-estimated N response index and the actual sugarcane response to spring N fertilization. The NDVI readings collected in early May showed the highest association with actual cane (r2=0.46) and sugar (r2 = 0.50) yields. Early season estimates of sugarcane response to N fertilization using NDVI were also correlated with increases in cane (r2 = 0.74) and sugar (r2 = 0.62) yield attributed to spring N fertilization. The initial findings of this study showed that optimum sensing dates fell within the timeframe where spring N fertilization is commonly done, and the use of NDVI readings that were collected early in the season to predict in-season cane and sugar yield potential, and the actual sugarcane response to N fertilization is feasible.