Scholarlink

<p>Landslides can be defined as sudden downslope movement of earth mass and debris under the action of gravitational forces. While rainfall intensity is a well-known trigger, soil moisture plays a critical yet often overlooked role in landslide initiation. However, the lack of a dense network of sensor/gauges limits the ability to capture the large heterogeneity of these hydro-meteorological variables in the mountainous region. Remote sensing technologies offer the advantage of monitoring both precipitation and soil moisture over large spatial and temporal scales. This study evaluates the gridded precipitation and soil moisture products in comparison to ground-based data and develops a dynamic landslide susceptibility framework which integrate precipitation, soil moisture, topography, and proximity to roads and streams using fuzzy logic. The analysis shows that the ERA5 precipitation product is the most suitable alternative to ground-based precipitation measurements especially for landslide hazard analysis, achieving the median RMSE of 10.73&#xa0;mm, correlation of 0.40 and relative bias of 0.21 across 21 rain gauge stations, with 30-day cumulative precipitation correlation reaching 0.69. For soil moisture, GLDAS 2.1 (0–10&#xa0;cm layer) showed correlations of 0.92 and RMSE of 0.04, while the FLDAS (0–10&#xa0;cm layer) performed marginally better with correlation and RMSE of 0.95 and 0.05 respectively. These validated products allowed derivation of hydro-meteorological thresholds for landslide initiation under different environmental conditions. The result of susceptibility framework shows strong seasonal susceptibility pattern with November emerging as the safest month with minimal regional risk, while June to September represents the most hazardous period with major susceptibility which declines in October. However, the Jammu region exhibits elevated susceptibility during December to January, with high-risk zone progressively shifting eastward through subsequent month. Arunachal Pradesh has been in a consistently high state of susceptibility throughout the year whereas the patterns of susceptibility in the Western Himalayan are seasonal. These findings demonstrate how landslide susceptibility dynamically responds to evolving precipitation and soil moisture conditions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Dynamic modelling of landslide susceptibility in response to hydro-meteorological variability in the Himalayas

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月29日

查看全文
  • Salil Sharma,
  • Siddik Barbhuiya,
  • Subhankar Das,
  • Vivek Gupta,
  • Kala Venkata Uday

摘要

Landslides can be defined as sudden downslope movement of earth mass and debris under the action of gravitational forces. While rainfall intensity is a well-known trigger, soil moisture plays a critical yet often overlooked role in landslide initiation. However, the lack of a dense network of sensor/gauges limits the ability to capture the large heterogeneity of these hydro-meteorological variables in the mountainous region. Remote sensing technologies offer the advantage of monitoring both precipitation and soil moisture over large spatial and temporal scales. This study evaluates the gridded precipitation and soil moisture products in comparison to ground-based data and develops a dynamic landslide susceptibility framework which integrate precipitation, soil moisture, topography, and proximity to roads and streams using fuzzy logic. The analysis shows that the ERA5 precipitation product is the most suitable alternative to ground-based precipitation measurements especially for landslide hazard analysis, achieving the median RMSE of 10.73 mm, correlation of 0.40 and relative bias of 0.21 across 21 rain gauge stations, with 30-day cumulative precipitation correlation reaching 0.69. For soil moisture, GLDAS 2.1 (0–10 cm layer) showed correlations of 0.92 and RMSE of 0.04, while the FLDAS (0–10 cm layer) performed marginally better with correlation and RMSE of 0.95 and 0.05 respectively. These validated products allowed derivation of hydro-meteorological thresholds for landslide initiation under different environmental conditions. The result of susceptibility framework shows strong seasonal susceptibility pattern with November emerging as the safest month with minimal regional risk, while June to September represents the most hazardous period with major susceptibility which declines in October. However, the Jammu region exhibits elevated susceptibility during December to January, with high-risk zone progressively shifting eastward through subsequent month. Arunachal Pradesh has been in a consistently high state of susceptibility throughout the year whereas the patterns of susceptibility in the Western Himalayan are seasonal. These findings demonstrate how landslide susceptibility dynamically responds to evolving precipitation and soil moisture conditions.