Assessment of the capability of Landsat-8 satellite imagery for predicting soil organic carbon distribution
Autor: | Abdelkader Laribi, Radjae Abdelouahad, Imane Tamine, Zahida Dehnoun |
---|---|
Jazyk: | angličtina |
Rok vydání: | 2024 |
Předmět: | |
Zdroj: | Journal of Degraded and Mining Lands Management, Vol 12, Iss 1, Pp 6781-6788 (2024) |
Druh dokumentu: | article |
ISSN: | 2339-076X 2502-2458 |
DOI: | 10.15243/jdmlm.2024.121.6781 |
Popis: | Soil organic carbon (SOC) is an important component of soil and plays a crucial role in addressing climate change. As a key component of soil organic matter, SOC directly impacts soil fertility, water retention, nutrient cycling, and overall soil health. The determination of SOC concentrations in soil often relies on costly physical sampling and chemical analysis. The aim of this research was to build a predictive model of SOC using satellite imagery of Landsat 8 OLI/TIRS over an agricultural area (Oued El Alleug) in the north of Algeria. The statistical correlations between the spectral bands (B2 and B6) and chemically measured SOC concentrations showed that it is possible to predict spatially the SOC concentrations. The results also showed that the topographic variables are not determinant in the spatial prediction of SOC concentrations. The predicted model showed an acceptable performance with a coefficient of determination (R2) = 0.7 and a root mean square error (RMSE) = 7.08 g/kg during the validation phase. The results of this study are important, as they will facilitate decision-making in soil conservation practices and enhance land management, especially in areas facing increasing agricultural and environmental pressures. |
Databáze: | Directory of Open Access Journals |
Externí odkaz: |