The prediction of heavy metals lead (Pb) and zinc (Zn) contents in soil using NIRs technology and PLSR regression method

Abstract

The case of soil contamination by heavy metals in environment shows an increasing amount along with the constructions and development period that has been done. The identification of heavy metal content in the field is extremely hard to be done without a laboratory activity. Therefore, it needs a quick and non-destructive approach or method to identify the metal content of the soil in the field. The application of Near Infrared Reflectance Spectroscopy (NIRs) technology is a method that non-destructively able to detect the heavy metal content in the soil by using Partial Least Square Regression (PLSR). Pretreatment spectrum which is done using the Multiplicative Scatter Correction (MSC) can improve the results of the prediction models of PLSR. The results of MSC pretreatment spectrum can repair and improve the accuracy of the predictions of Lead (Pb) and Zinc (Zn) in the soil. Eight samples were used for analysis of each of Pb and Zn content. The measured data were pre-treated by MSC. It was obtained that value of r = 0.98, R 2 = 0.97 and RPD = 6.46 for the Pb content measurement. Meanwhile, the measurement for Zn obtained the value of r = 0.98, R 2 = 0.97 and RPD = 6.28. Therefore, it can be inferred that the NIRs is one of technologies which is worth reckoned as the right and quick means to predict the content of heavy metals in soil in a non-destructively and environmentally friendly way.