VNIR spectrometry coupled with geostatistics for spatial prediction of soil erosion in sample area of Calabria region (southern Italy)

Soil erosion is one of the most serious types of land degradation and it has severe economic and environmental impacts in the Mediterranean belt, as well as in large areas of the Italian territory. Soil erosion can be evaluated by soil properties such as organic matter, calcium carbonate, texture, aggregation structure, infiltration capacity, etc. The aim of the study was to test laboratory reflectance spectroscopy in the visible near infrared (VNIR) range as a tool to develop an empirical model for predicting specific soil properties (organic matter and calcium carbonate content) as indicators of soil condition and erosion severity. The study was conducted in the Turbolo watershed representative of the pedo-environmental variability of the Calabria region, southern Italy. In this watershed 215 topsoil samples, representative of different soil types and various soil degradation conditions were collected and analysed for soil organic matter (SOM), calcium carbonate and spectroscopic measurements. The partial least squared regression (PLSR) analysis was applied to establish relationships between spectral reflectance and two soil properties. PLSR was performed on 161 of the 215 available samples. The optimum number of factors to retain in the calibration model was determined by cross validation. The models were independently validated and results were satisfactory. The results of this work suggest that laboratory reflectance spectroscopy in the VNIR (350-2500 nm range) coupled with geostatistics analysis can be used as tools for evaluating and mapping spatial pattern of the SOM and calcium carbonate, as well as a method for a quick screening of soil erosion conditions.