Bare soil (BS) identification through satellite remote sensing can potentially play a critical role in understanding and managing soil properties essential for climate regulation and ecosystem services. From 191 papers, this review synthesises advancements in BS detection methodologies, such as threshold masking and classification algorithms, while highlighting persistent challenges such as spectral confusion and inconsistent validation practices. The analysis reveals an increasing reliance on satellite data for applications such as digital soil mapping, land use monitoring, and environmental impact mapping. While multispectral sensors like Landsat and Sentinel dominate current methodologies, limitations remain in distinguishing BS from spectrally similar surfaces, such as crop residues and urban areas. This review emphasises the critical need for robust validation practices to ensure reliable estimates. By integrating technological advancements with improved methodologies, the potential for accurate, large-scale BS detection can significantly contribute to combating land degradation and supporting global food security and climate resilience efforts.
Funding
This research was supported by the UKRI BBSRC (BB/W009439/1) as part of the Collaborative Training Program for Sustainable Agricultural Innovation (CTP-SAI).
History
Author affiliation
College of Science & Engineering
Geography, Geology & Environment