Estimation of leaf nitrogen levels in sugarcane using hyperspectral models

ABSTRACT: Sugarcane is a good source of renewable energy and helps reduce the emission greenhouse gases. Nitrogen has critical role in plant growth; therefore,estimating nitrogen levels essential, remote sensing can improve fertilizer management. This field study selects wavelengths from hyperspectral data on sugarcane canopy to generate models for estimating leaf concentrations. The was carried out municipalities Piracicaba, Jaú, Santa Maria da Serra, state São Paulo, 2013/2014 growing season. experiments were using completely randomized block design with split plots (three varieties per plot [variety SP 81-3250 common all plots] four concentrations [0, 50, 100, 150 kgha-1] subplot) repetitions. that best correlated selected usingsparse partial least square regression. wavelength regionswere combinedby stepwise multiple linear Spectral bands visible (700-705 nm), red-edge (710-720 near-infrared (725, 925, 955, 980 short-wave infrared (1355, 1420, 1595, 1600, 1605, 1610 nm) regions identified. R² RMSE model 0.50 1.67 g.kg-1, respectively. adjusted 0.31 (unreliable) 1.30 0.53 1.96 0.54 1.46 Our results showed reflectance estimate manage application sugarcane.