Environmental assessment of soil quality indices using near infrared reflectance spectroscopy

Abstract

The main purpose of this preset study is to assess soil quality indices in form of potassium (K) and phosphorus (P) contents using a non-invasive and environmental friendly approach namely near infrared reflectance spectroscopy. Soil samples were obtained from Aceh Besar district in rice field land-use. Near infrared spectral data of soil samples were acquired and recorded as absorbance in wavelength range from 1000 to 2500 nm. On the other hand, actual P and K were measured using standard laboratory procedures by means of Kjeldahl methods. Spectral data were corrected and pre-treated using mean centering approach and applied to all dataset. Prediction models were developed using principal component regression and validated using leverage cross validation. The results showed that both soil quality indices can be predicted with maximum correlation coefficient (r) of 0.98 and ratio prediction to deviation (RPD) index of 3.47 for P, and r of 0.91, RPD of 2.68 for K respectively. It may conclude that environmental assessment, particularly for soil quality determination can be conducted rapidly and non-invasively using near infrared spectroscopy approach.