Calibrating Energy Cane Biomass and Nitrogen Uptake With Vegetation Indices Derived From Canopy Reflectance At The Red, Red-Edge, and Near Infrared Wavebands.

The application of remote sensing technology to manage nitrogen (N) fertilizer in energy cane production is understudied. A study was initiated in 2012 at the LSU AgCenter Sugar Research Station in St. Gabriel, LA to identify vegetation indices derived from canopy reflectance readings within the red, red-edge, and near infrared (NIR) wavebands that can be used to characterize energy cane biomass and N uptake. A 2 x 4 factorial treatment structure was superimposed on 9 m x three 1.8-m bed plots using a split-plot in randomized complete block design with four replications. The treatments included two energy cane varieties (Ho 02-113 and US 72-114) and four N rates (0, 56, 110, and 224 kg N ha^-1). Canopy reflectance readings were collected using a Jaz ® hyperspectral spectrometer (300 to 1100 nm) and 2- and 4-band handheld active sensor from a 1 m² area of each plot along with chlorophyll readings before biomass cuttings were sampled. Field data collection began three weeks after N fertilization and done once a week for three consecutive weeks. The relationships among measured parameters were evaluated using regression analysis. Chlorophyll readings, N uptake, tiller number, and biomass of US 72-114 variety responded to N fertilizer only at three weeks after application (P<0.05). The relationships of vegetation indices and agronomic parameters were different between the two varieties. The near infrared to red-edge (710 nm) reflectance ratio had a weak positive relationship with biomass of Ho 02-113 (r²=0.22) and N uptake of US 72-114 (r²=0.24) while the normalized difference vegetation index computed from NIR and red reflectance readings had a better association (r²=0.40) with N uptake of both varieties. Thus far, a few vegetation indices have been identified useful for non-destructive characterization of energy cane biomass and N uptake. Future research will focus on building the sensor database and refinement of the relationships among energy cane N-related agronomic parameters and sensor based-vegetation indices.