Pattern of Landslide Distribution Reflects Degree of Hillslope Adjustment in a Waipaoa River Catchment, New Zealand

Abstract:

Using landslide morphology to determine the state of hillslope transience provides a diagnostic tool to identify the extent of landscape adjustment. Here we test for a temporal and spatial progression of landslides reflecting the degree of adjustment for a Waipaoa River catchment (North Island, New Zealand). Following the shift to a warmer, wetter climate after the Last Glacial Maximum (LGM) (~18 ka), the Waipaoa River rapidly incised up to 120 m, leaving perched, low-relief hillslopes unadjusted to that base level fall. In the 16 km² study catchment—principally comprised of weak mélange—pervasive post-LGM landslides responded to >50 m of fluvial incision by sculpting and denuding >99% of the catchment, but adjustment is not yet complete. Starting ~150 years ago, European settlers deforested ~95% of the landscape, which triggered a sharp increase in hillslope erosion as widespread earthflow complexes and gully systems reactivated and/or expanded in the weak lithology with the loss of vegetation cover. Most of the remaining relict (≥18 ka) landforms are confined to the upper watershed, upholding the proposed upstream progression of hillslope adjustment. Based on previous tephrochronology and surface roughness data, present-day earthflows follow this pattern of younging upstream. Here we analyze the size, location, and distribution of ~500 earthflows (mapped from 2010 lidar and orthophotos) to determine if there is a spatial progression of earthflows reflective of hillslope adjustment and correlated to the previously identified temporal progression. Morphologic evidence from this study indicates the younger earthflows in the upper watershed have smaller areas (6000 ± 600 m² (mean ± s.e.)), a more elongate aspect ratio (A_R=3.6 ± 1.6 (mean ± s.d.)), and are generally (>50%) deposited in axial gullies, whereas downstream earthflows are larger (20000 ± 9000 m²), statistically less elongate (A_R=2.3 ± 0.9), and more frequently coupled with the main channel—in two cases, encompassing the entire sub-catchments. We suggest this difference in form is not necessarily reflective of activity, but rather a diagnostic indicator of relative age and progression of earthflow development.