Characterizing the spatial variations and correlations of large rainstorms for landslide study
Citation
Gao, L., Zhang, L., Lu, M. (2017). Characterizing the spatial variations and correlations of large rainstorms for landslide study. Hydrology and Earth System Sciences, 21: 4573-4589. Link to paper
Abstract
This study quantifies spatial correlation structures of three large landslide-triggering rainstorms in Hong Kong to improve rainfall characterization for landslide hazard analysis. Maximum rolling rainfall fields (4, 12, 24, and 36 hours) are represented using a rotated ellipsoidal trend surface plus a random residual field. The work estimates principal directions, trend-axis lengths, and scales of fluctuation across multiple orientations, showing principal directions between 19 and 43 degrees and residual fluctuation scales from 5 to 30 km. The extracted parameters are consistent with those from ordinary rainfall events, suggesting stable spatial descriptors for stochastic rainfall modeling. The resulting model supports more robust representation of impact area, intensity distribution, and local topographic effects in landslide evaluations.