Predicting Soil Organic Carbon Content Using Hyperspectral Remote Sensing in a Degraded Mountain Landscape in Lesotho

Abstract

Soil organic carbon constitutes an important indicator of soil fertility. The purpose of this study was to predict soil organic carbon content in the mountainous terrain of eastern Lesotho, southern Africa, which is an area of high endemic biodiversity as well as an area extensively used for small-scale agriculture. An integrated field and laboratory approach was undertaken, through measurements of reflectance spectra of soil using an Analytical Spectral Device (ASD) FieldSpec® 4 optical sensor. Soil spectra were collected on the land surface under field conditions and then on soil in the laboratory, in order to assess the accuracy of field spectroscopy-based models. The predictive performance of two different statistical models (random forest and partial least square regression) was compared. Results show that random forest regression can most accurately predict the soil organic carbon contents on an independent dataset using the field spectroscopy data. In contrast, the partial least square regression model overfits the calibration dataset. Important wavelengths to predict soil organic contents were localised around the visible range (400–700 nm). This study shows that soil organic carbon can be most accurately estimated using derivative field spectroscopy measurements and random forest regression.

Soil Organic Matter Inversion Based on Imaging Spectral Data in Straw‐Covered Noncultivated Land

“…The maximum value of RPD in the PLSR model was 1.68, with an average value of 1.55, which was less than the DOAR model. Therefore, it is concluded that based on the bow curvature, the DOAR model can predict the content of SOM, so that the accuracy of the DOAR model to predict the soil organic matter content was better than the PLSR model [25,26].…”

Dynamic UAV Phenotyping for Rice Disease Resistance Analysis Based on Multisource Data

“…A good performance model has a value for R 2 close to 1, with a lower RMSE and a higher RPD. RPD > 2.5 means that the model has excellent predictive ability, 2 < RPD < 2.5 means a good model, 1.5 < RPD < 2 means a relatively fair model, and RPD < 1.5 means that the model is incapable of prediction [ 48 ].…”

A deep spectral prediction network to quantitatively determine heavy metal elements in soil by X-ray fluorescence

“…In other studies, Kumar 22 applied CARS and PLSR to analyze toluene and pyrene mixtures based on excitation–emission matrix fluorescence spectroscopy. F. Bangelesa et al 23 utilized a portable spectrometer to capture the reflectance spectra of the soil and combined it with RFR to predict the soil organic carbon concentration.…”