A Random Forest-Based Approach to Map Soil Erosion Risk Distribution in Hickory Plantations in Western Zhejiang Province, China
AbstractIncreasing agroforestry areas with improper management has produced serious environmental problems, such as soil erosion. It is necessary to rapidly predict the spatial distribution of such erosion risks in a large area, but there is a lack of approaches that are suitable for mountainous regions. The objective of this research was to develop an approach that can effectively employ remotely-sensed and ancillary data, to map soil erosion risks in an agroforestry ecosystem in a mountainous region. This research employed field survey data, soil-type maps, digital elevation model data, weather station data, and Landsat imagery, for extraction of potential variables. It used the random forest approach to identify eight key variables—slope, slope of slope, normalized difference greenness index at leaf-on season, soil organic matter, fractional vegetation at leaf-on season, fractional soil at leaf-off season, precipitation in June, and percent of soil clay—for mapping soil erosion risk distribution in hickory plantations in Western Zhejiang Province, China. The results showed that an overall accuracy of 89.8% was obtained for three levels of soil erosion risk. Approximately one-fourth of hickory plantations were at high-risk, requiring the owners or decision makers to take proper measures to reduce the soil erosion problem. This research provides a new approach to predict soil erosion risk, based on the primary variables that can be extracted directly from remotely-sensed data and ancillary data. This proposed approach will be valuable for other agroforestry and plantations, such as Torreya grandis, eucalyptus, and the rubber tree, that are playing important roles in improving economic conditions for the local farmers but face soil erosion problems. View Full-Text
- Supplementary File 1:
PDF-Document (PDF, 166 KB)
Share & Cite This Article
Cheng, Z.; Lu, D.; Li, G.; Huang, J.; Sinha, N.; Zhi, J.; Li, S. A Random Forest-Based Approach to Map Soil Erosion Risk Distribution in Hickory Plantations in Western Zhejiang Province, China. Remote Sens. 2018, 10, 1899.
Cheng Z, Lu D, Li G, Huang J, Sinha N, Zhi J, Li S. A Random Forest-Based Approach to Map Soil Erosion Risk Distribution in Hickory Plantations in Western Zhejiang Province, China. Remote Sensing. 2018; 10(12):1899.Chicago/Turabian Style
Cheng, Zhenlong; Lu, Dengsheng; Li, Guiying; Huang, Jianqin; Sinha, Nibedita; Zhi, Junjun; Li, Shaojin. 2018. "A Random Forest-Based Approach to Map Soil Erosion Risk Distribution in Hickory Plantations in Western Zhejiang Province, China." Remote Sens. 10, no. 12: 1899.
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.