From: Landslide susceptibility assessment of South Korea using stacking ensemble machine learning
Parameter | Â | Description | Â | Formula |
---|---|---|---|---|
Topography | SRTM | Altitude data of 30-m resolution DEM | Â | Â |
 | Slope | Degree of steepness of the slope | (1) | \(\tan^{ - 1} \sqrt {\left( {\frac{dz}{{dx}}} \right)^{2} + \left( {\frac{dz}{{dy}}} \right)^{2} }\) |
 | Aspect | Direction of the slope | (2) | \(\tan^{ - 1} \sqrt {\frac{dy}{{dx}}}\) |
 | Profile curvature | Curvature in the direction perpendicular to the direction of inclination with plan curvature | (3) | \(- 2\frac{{\left( {{\text{DG}}^{2} + {\text{EH}}^{2} + {\text{FGH}}} \right)}}{{\left( {{\text{G}}^{2} + {\text{H}}^{2} } \right)}}\) |
 | Plan curvature | Horizontal curvature with respect to the direction of inclination with lateral curvature | (4) | \(2\frac{{\left( {{\text{DH}}^{2} + {\text{EG}}^{2} + {\text{FGH}}} \right)}}{{\left( {{\text{G}}^{2} + {\text{H}}^{2} } \right)}}\) |
 | TRI | Index of concave or convex conformation of the terrain | (5) | \(\frac{{\mathop \sum \nolimits_{i = 1}^{8} \left| {N - N_{i} } \right|}}{8}\) |
 | TPI | Index of degree to which a surface is soft or bumpy | (6) | \(N - \mathop \sum \limits_{i = 1}^{8} \frac{{N_{i} }}{8}\) |
 | SPI | Index of the degree of movement and erosion of sediment due to surface runoff | (7) | \(A_{s} - \tan \beta\) |
 | TWI | Index of the effects of terrain on runoff flow | (8) | \(\ln \frac{{A_{s} }}{\tan \beta }\) |
 | Valley depth | Difference of vertical distance between water channel network and DEM |  |  |
 | Soil drainage | Duration or frequency of soil being unsaturated by water |  |  |
 | Soil depth | Vertical depth of soil layer, at a depth at which the roots of the crop can extend sufficiently into the ground |  |  |
Environment | Distance to river | Distance from landslide location to water boundary | Â | Â |
 | Distance to fault | Distance from landslide location to fault |  |  |
 | Distance to road | Distance from landslide location to road |  |  |
 | FRTP | Forest type |  |  |
 | DMCLS | Forest tree diameter class |  |  |
 | AGCLS | Forest age class |  |  |
 | DNST | Forest density |  |  |
 | HEIGHT | Forest canopy height |  |  |
 | NDVI | Normalized difference vegetation index-based on Sentinel-2 data | (9) | \(\frac{NIR - Red}{{NIR + Red}}\) |
Climate | Monthly total precipitation (RN) | Total precipitation in each month, July–September |  |  |
 | RX1day | Highest precipitation amount in 1 day | (10) | \({\text{max}}\left( {{\text{RR}}_{ij} } \right)\) |
 | RX5day | Highest precipitation amount in 5 days | (11) | \({\text{max}}\left( {{\text{RR}}_{kj} } \right)\) |
 | SDII | Average precipitation on wet days | (12) | \({\text{sum}}\left( {\frac{{{\text{RR}}_{wj} }}{wet day}} \right)\) |
 | R80 | Number of days per year when precipitation ≥ 80 mm | (13) | \({\text{count}}({\text{RR}}_{ij} > 80mm)\) |
 | RD95P | Number of days when daily precipitation is greater than during the top 95% of the reference period | (14) | \({\text{sum if}}\left( {\frac{{{\text{RR}}_{wj} > 95\% }}{wet day}} \right)\) |
 | RD99P | Number of days when daily precipitation is greater than during the top 99% of the reference period | (15) | \({\text{sum if}}\left( {\frac{{{\text{RR}}_{wj} > 99\% }}{wet day}} \right)\) |