Near-infrared spectroscopy as a rapid and simultaneous assessment of agricultural groundwater quality parameters

Abstract

Groundwater quality in agricultural area is highly affected by human activities. To determine groundwater quality, several methods are widely applied. Yet, most of them are based on standard laboratory analysis which is normally time consuming, expensive, and involve chemical materials from which may cause another environmental pollution. Thus, a rapid, effective and simple alternative method is required to assess groundwater quality. Fourier transform near-infrared spectroscopy (FT-NIRS) is considered to be employed due to its advantages. The main purpose of the present study, is to evaluate the feasibility of FT-NIRS technology in assessing groundwater quality parameters: total dissolved solids (TDS) and Sulfate concentration (SC). Transmission spectra data were acquired for groundwater samples from 8 different wells in wavelength range from 1000 to 2500 nm. Spectra data were corrected by multiplicative signal correction (MSC), while TDS and SC prediction models were established by using partial least squares regression (PLSR) and validated by full cross validation method. Obtained results showed that FTIR is able to detect and predict TDS and SC rapidly. Achieved maximum correlation coefficient (r) and RPD index were 0.86; 1.82 for TDS and 0.83; 1.76 for SC prediction respectively. It may be concluded that FT-NIRS combined with proper multivariate approach, can be used to assess groundwater quality parameters rapidly and simultaneously.