Regolith modeling and its relation to earthquake induced building damage : a remote sensing approach

Regolith thickness is known as a major factor in influencing the intensity of earthquake induced ground shaking and consequently building damages. It is, however, often simplified or ignored due to its variable and complex nature. To evaluate the role of regolith thickness on earthquake induced building damage, a remote sensing based methodology is developed to model the spatial variation of regolith thickness, based on DEM derived topographic attributes and geology. Regolith thickness samples were evenly collected in geological formations at representative sites of topographic attributes. Topographic attributes (elevation, slope, TWI, distance from stream) computed from the ASTER derived DEM and a geology map were used to explore their role in spatial variation of regolith thickness. Stepwise regression was used to model the spatial variation of regolith thickness in erosional landscape of the study area. Topographic attributes and geology, explain 60% of regolith thickness variation in the study area. To test, if the modeled regolith can be used for prediction of seismic induced building damages, it is compared with the 2005 Kashmir earthquake induced building damages derived from high resolution remote sensing images and field data. The comparison shows that the structural damages increase with increasing regolith thickness. The predicted regolith thickness can be used for demarcating site prone to amplified seismic response.