Miniaturised visible and near-infrared spectrometers for assessing soil health indicators in mine site rehabilitation

Abstract. Mining can cause severe disturbances to the soil, which underpins viability of terrestrial ecosystems. Post-mining rehabilitation relies on measuring soil properties that are critical indicators health. Soil visible–near-infrared (vis–NIR) spectroscopy is rapid, accurate, and cost-effective for estimating a range properties. Recent advances in infrared detectors microelectromechanical systems (MEMSs) have produced miniaturised, relatively inexpensive spectrometers. Here, we evaluate spectra from four miniaturised visible NIR spectrometers, some combinations, full-range vis–NIR spectrometer modelling 29 physical, chemical, biological used assess health at mine sites. We collected topsoil samples reference, undisturbed native vegetation, stockpiles seven mines Western Australia. evaluated spectrometers' repeatability accuracy spectroscopic models built with statistical machine learning algorithms. The could estimate sand, silt, clay similar or better than However, spectrometers estimates chemical By combining improved their property estimates, were those spectrometer. combinations predicted 24 moderate greater (Lin's concordance correlation, ?c?0.65). was full-range, portable comparably accurate system an order magnitude more expensive, particularly when combined sensor. Thus, form basis diagnostic capacity support site deliver significant positive economic environmental outcomes.