Measurement of soil attributes using NIR spectroscopy : A review

Soil fertilizer management is important to adjust the soil fertility by supplying fertilizers at optimum level to achieve maximum yield at minimum environmental risks. Precision agriculture demands to develop better cost and time efficient methodologies for analysis of soil properties. Besides conventional techniques, sometimes spectroscopy is more accurate, fast, non-destructive and inexpensive analytical technique. Diffused reflectance measurement in the NIR region (780-2500 nm) is involved in NIRS (near infrared reflectance spectroscopy) and this is highly sensitive to both inorganic and organic phases in soil. Quantitative analysis for prediction of various soil properties and components involves NIR spectra calibration against some reference value by using some prediction model (e.g. ANN (Artificial neural network) or MPLS (Modified Partial Least Square Regression)).The goal of this paper is to review the results, modelling methods, hardware and software involved to measure various attributes of soil using NIR spectroscopy. KeywordsNIRS, Precision Agriculture, Soil properties, Spectroscopy.