Visible and near‐infrared reflectance spectroscopy analysis of soils
Yufeng Ge, Cristine L.S. Morgan, Nuwan K. Wijewardane
Abstract
Abstract Visible and near‐infrared reflectance spectroscopy (VisNIR, 350–2500 nm) is a proximal sensing technique for rapid and nondestructive soil analysis in the laboratory or field. It can be used to estimate multiple soil properties simultaneously and is usually cheaper than conventional soil analysis methods. In addition to many soil constituents (such as clay minerals, organic matter, water, and carbonates) with which VisNIR light energy directly interacts, this method can also estimate soil properties that are related to these directly measured properties, such as cation exchange capacity (CEC), pH, and salinity. Samples are usually air dried, ground, and sieved (passing 2‐mm screens) before being scanned by a VisNIR instrument. Multivariate statistical modeling techniques (such as partial least squares regression) are commonly used to build VisNIR models from a training set. The VisNIR models estimate target soil properties from VisNIR data. The performance of the VisNIR model needs to be objectively evaluated with an independent test set. Visible and near‐infrared reflectance spectroscopy has been mainly used in the research domain, with a lot of ongoing efforts to further develop and refine this method for routine soil analysis.