Application of the Field Hyperspectral Sensing to Estimate the Oxidation-Reduction Potential (ORP) in Tropical Urban Waters

Abstract

Oxidation-reduction potential (ORP) is one of the critical water quality parameters used in determining the biochemical condition of water, especially in wastewater treatment. Conventional measurement of ORP requires physical contact with the water samples. A relative version of the ORP measurement that is faster and requires non-contact with the water can be helpful, especially for environmental monitoring. One way to perform this is by measuring the radiated electromagnetic wave from water bodies such as drainage, rivers, and streams that are exposed to sunlight. The reflected radiation from various water samples was obtained using a handheld hyperspectral sensor that is sensitive to 350 – 250 nm radiation and the ORP measurement was performed using the water quality probes sensor equipped with the ORP sensor. Forty-four water samples were taken from multiple locations, including rivers, tributaries, drainage, and streams. Our analysis revealed that the wave spectra at 654 nm can estimate ORP values ranging from 0 to 500 with moderate performance (R² = 0.581). The reflectance spectra need to undergo normalization, and the second derivative filtering of the Savitsky-Golay is required to obtain such desired results. This preliminary experiment demonstrated that the ORP estimation is plausible using the hyperspectral remote sensor with moderate accuracy for water that potentially experiences nitrification, denitrification, carbonaceous biochemical oxygen demand, and phosphorus waste.