Hyperspectral data use for prediction of leaf nitrogen content in sugarcane

• Grant number: 13/22435-9
• Support type: Program for Research on Bioenergy (BIOEN) - Regular Program Grants
• Duration: June 01, 2014 - May 31, 2016
• Field of knowledge: Agronomical Sciences - Agronomy
• Principal Investigator: Peterson Ricardo Fiorio
• Grantee: Peterson Ricardo Fiorio
• Home Institution: Escola Superior de Agricultura Luiz de Queiroz (ESALQ). Universidade de São Paulo (USP). Piracicaba , SP, Brazil
• Assoc. researchers: Christopher Michael Usher Neale ; José Alexandre Melo Demattê

Abstract

Nitrogen is an essential nutrient for crop production, interacting in a very complex system with environment, so its monitoring is therefore important, both economically as well as environmentally. In recent years, the use of remote sensing has expanded in agricultural sciences, providing a useful tool to monitor and manage such activities, including the nitrogen fertilizer administration on crops. In this sense, the present work aims to study the application of remote sensing techniques in monitoring nitrogen at the sugarcane culture, using a field hyperspectral sensor. Evaluations shall be performed at different crop stages development in experimental areas of São Paulo Agency for Agribusiness Technology, that studies effects of different nitrogen doses in some sugarcane varieties. Spectral readings will be held of leaf and canopy; leaves will be collected and sent to laboratory for leaf nitrogen content analysis. These data will be used to generate spectral models of multiple linear regression to predict the leaf nitrogen content, thus detecting the wavelengths that are more related to the plant nitrogen status. This study aims to show the results that may guide future applications of spectral data to monitoring sugarcane nitrogen fertilization, as new research and development of dedicated sensors for the culture. (AU)