ALBERT

Description: We investigated 335 landslides (including rockslides, rock avalanches, soil slides, and slides) in the Zagros mountain belt of southwest Iran using a digital elevation model (DEM) with 30 m resolution, Google EarthTM images, and field investigation. Individual landslides have volumes ranging from 10**4 (the lower limit of resolution) to 3 × 10**10 m**3 and cover surface areas ranging from 10**3 to 10**8 m**2. The relationship between landslide volume (VL) and area (AL) is well described by a power law of the form VL = AL**alpha, where alpha = 1.49 over five orders of magnitude of AL and seven orders of magnitude of VL. We also show that the frequency-size (i.e., volume) distribution is heavy tailed, following a power law for the largest landslides (〉10**7 m**3) with a scaling exponent beta = 1.51. Non-power-law behavior for smaller landslides is probably an artifact due to the relatively low resolution of our data, such that we are essentially missing many small landslides. Comparison of these results with the other published data sets around the globe shows that the Zagros landslides are relatively larger, and they show similar scaling behavior to those observed in other regions, especially with regard to data sets where landslides are deep-seated or relatively large and occur in relatively resistant materials (e.g., consolidated rocks, as opposed to soil). In addition, we used principal component analysis (PCA) to investigate links between the size of landslides and causative factors (e.g., geological, geomorphological, and physical factors). Our results highlight that although the size of landslides is not controlled by any single factor, their geographic distribution is strongly influenced by lithology, elevation, and slope.