Open AccessRapid Prediction of Soil Quality Indices Using Near Infrared Spectroscopy
Author(s) -
Yuswar Yunus,
Devianti DEVİANTİ,
Purwana Satriyo,
Agus Arip Munawar
Publication year - 2019
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/365/1/012043
Subject(s) - principal component analysis , leverage (statistics) , correlation coefficient , principal component regression , macro , linear regression , soil quality , environmental science , near infrared spectroscopy , soil test , mathematics , soil science , statistics , computer science , soil water , optics , physics , programming language
To determine soil macro nutrients and other quality indices, conventional and laborious procedures were employed. However, this method is time consuming, involve chemical materials and laborious. Thus, alternative fast and environmental friendly method is required to determine several quality indices in agricultural soil. This present study is aimed to apply near infrared spectroscopy (NIRS) in determining soil macro nutrients namely N, P and K. Diffuse reflectance spectrum of soil samples were acquired and recorded in wavelength range from 1000 to 2500 nm. Near infrared spectrum were enhanced using de-trending (DT) method. Prediction models, used to predict N, P and K, were established using principal component regression (PCR) algorithm followed by leverage validation. The results showed that NIRS method can determine all three quality indices with good accuracy and robustness. Maximum correlation coefficient (r) for N, P, K prediction were achieved using DT correction method with r = 0.86 for N prediction, r = 0.90 for both P and K prediction. Based on obtained results, it may conclude that NIRS can applied as an alternative rapid and simultaneous method in predicting soil quality indices.