Processing spectroradiometry data for the simulation of soil physicochemical parameters
International Journal of Development Research
Processing spectroradiometry data for the simulation of soil physicochemical parameters
Received 22nd May, 2019; Received in revised form 03rd June, 2019; Accepted 11th July, 2019; Published online 28th August, 2019
Copyright © 2019, José M. Filippini Alba et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
In spite of its restricted use, due to data quantity and difficulty of interpretation, spectroradiometry is a potential application for precision agriculture. In this study, samples of lowland soils were collected according to a regular grid. Each sample was analyzed by conventional physicochemical methods (Al, Ca, K, Mg, Na, P and clay) and spectroradiometry (2151 bands, 350 to 2500 nm).The work aimed to simulate physicochemical variables using data from spectroradiometry bands. Dependence analysis, Factor analysis and Cluster analysis were applied gradually until the determination coefficients (R2) of the regression models reach100%. That occurred when at least two of the three statistical methods were applied. Two different strategies were highlighted: (1) Factor analysis with high number of factors extracted; (2) Factor Analysis with a few factors followed by cluster analysis for population segregation. Although option (1) was simpler, option (2) allowed outliers discard, in the way that the geostatistical maps were improved.
