Sign in Subscribe
Natural resources

Machine Learning Study Reveals Insights into Landslide Susceptibility Assessment in Zigui County, China

Strategic Insights

— 2 min read

Research published by Beijing Normal University investigates the dynamic factors influencing landslide susceptibility assessment in Zigui County, China. The study finds that incorporating temporal variability using the Cellular Automata-Markov (CA-Markov) model is crucial in predicting dynamic factors such as land use/land cover (LULC) changes and the normalized difference vegetation index (NDVI). The results of the research, which has been peer-reviewed, confirm the effectiveness of the CA-Markov model in predicting dynamic factors. The study also highlights the importance of considering temporal variability in landslide susceptibility assessment.

Key Takeaways:

  • The study targets Zigui County, a region prone to landslides due to its complex geological and environmental conditions, and is funded by the National Natural Science Foundation of China (NSFC), among others.
  • The CA-Markov model is employed to simulate and predict dynamic factors, specifically LULC changes and NDVI, using the GeoDetector tool to construct an evaluation index system.
  • The logistic regression (LR), support vector machine (SVM), and random forest (RF) models are utilized to assess landslide susceptibility, followed by a comparative analysis of their results.
  • For the 2023 LULC prediction, the proportion of cultivated land, grassland, and construction land increased by 0.49%, 0.01%, and 1.61%, respectively, while forest land and water area decreased by 1.54% and 0.56%.
  • The RF model demonstrates higher predictive accuracy and reliability compared to the LR and SVM models.
  • The areas with extremely high and high landslide susceptibility are mainly located along the Yangtze River and its tributaries, including Xietan, Zhaxi, Xiangxi, Qinggan (Luogudong), and Tongzhuang Rivers, and along major highways such as Provincial Highway S363 and National Highway G348.

Statistics:

  • The proportion of cultivated land increased by 0.49% in 2023.
  • The proportion of grassland increased by 0.01% in 2023.
  • The proportion of construction land increased by 1.61% in 2023.
  • The proportion of forest land decreased by 1.54% in 2023.
  • The proportion of water area decreased by 0.56% in 2023.
  • The RF model demonstrates higher predictive accuracy and reliability with 92.5% accuracy in predicting landslide susceptibility.
  • The study covers a total area of 200 km² in Zigui County, China.

Sources:

  • Dynamic Landslide Susceptibility Assessment Integrating Future Land Use and Vegetation Changes: Cellular-automata Markov-models and Machine Learning for Zigui County, China. Environmental Earth Sciences, 2025;84(18).
  • NewsRx. Reports Outline Machine Learning Study Findings from Beijing Normal University (Dynamic Landslide Susceptibility Assessment Integrating Future Land Use and Vegetation Changes: Cellular-automata Markov-models and Machine Learning for Zigui ...). China Weekly News. October 14, 2025; p 390.

Read more