Use of Visible and Near-Infrared Reflectance Spectroscopy Models to Determine Soil Erodibility Factor (K) in an Ecologically Restored Watershed

Qinghu Jiang,Yiyun Chen,Feng Liu,Jialiang Hu

doi:10.3390/rs12183103

Abstract

This study aimed to assess the ability of using visible and near-infrared reflectance (Vis–NIR) spectroscopy to quantify soil erodibility factor (K) rapidly in an ecologically restored watershed. To achieve this goal, we explored the performance and transferability of the developed spectral models in multiple land-use types: woodland, shrubland, terrace, and slope farmland (the first two types are natural land and the latter two are cultivated land). Subsequently, we developed an improved approach by combining spectral data with related topographic variables (i.e., elevation, watershed location, slope height, and normalized height) to estimate K. The results indicate that the calibrated spectral model using total samples could estimate K factor effectively (R2CV = 0.71, RMSECV = 0.0030 Mg h Mj−1 mm−1, and RPDCV = 1.84). When predicting K in the new samples, models performed well in natural land soils (R2P = 0.74, RPDP = 1.93) but failed in cultivated land soils (R2P = 0.24, RPDP = 0.99). Furthermore, the developed models showed low transferability between the natural and cultivated land datasets. The results also indicate that the combination of spectral data with topographic variables could slightly increase the accuracies of K estimation in total and natural land datasets but did not work for cultivated land samples. This study demonstrated that the Vis–NIR spectroscopy could be used as an effective method in predicting K. However, the predictability and transferability of the calibrated models were land-use type dependent. Our study also revealed that the coupling of spectrum and environmental variable is an effective improvement of K estimation in natural landscape region.

Highlights

Soil erosion is a severe eco-environmental issue worldwide that threatens sustainable land utilization and ecosystem stability [1,2,3]
soil organic carbon (SOC) had the highest variability with the coefficient of variation (CV) of 49.67%, whereas the CVs of pH, bulk density, and K factor were 7.27%, 12.02%, and 13.10%, respectively (Table 1)
Our study found that the incorporation of topographic variables into the partial least-squares regression (PLSR) model could conditionally increase the prediction accuracy of K factor in the total dataset

Summary

Introduction

Soil erosion is a severe eco-environmental issue worldwide that threatens sustainable land utilization and ecosystem stability [1,2,3]. The Universal Soil Loss Equation (USLE) and revised USLE (RUSLE) were the two most widely used empirical equations for erosion prediction [11] In these models, soil erodibility (introduced as the K factor) is calculated by using several soil properties, including particle size composition, soil organic carbon (SOC) content, structural stability, soil permeability, and clay minerals [12]. Soil erodibility (introduced as the K factor) is calculated by using several soil properties, including particle size composition, soil organic carbon (SOC) content, structural stability, soil permeability, and clay minerals [12] To accurately measure these properties, field soil samplings and laboratory analyses are necessary, which are labor intensive, time-consuming, and expensive. A reliable, fast, and economic measuring method is needed to assess K factor

Objectives

Methods

Results

Discussion

Conclusion