Search for: All records

Abstract Landslides pose a significant hazard worldwide. Despite advances in landslide monitoring, predicting their size, timing, and location remains a major challenge. We revisit the 2017 Mud Creek landslide in California using radar interferometry, pixel tracking, and elevation change measurements from satellite and airborne radar, lidar, and optical data. Our analysis shows that pixel tracking of optical imagery captured the transition from slow motion to runaway acceleration starting ~ 1 month before catastrophic failure—an acceleration undetected by satellite InSAR alone. Strain rate maps revealed a new slip surface formed within the landslide body during acceleration, likely a key weakening mechanism. Failure forecast analysis indicates the acceleration followed a hyperbolic trend, suggesting failure time could have been predicted at least 6 days in advance. We also inverted for the landslide thickness during the slow-moving phase and found variations from < 1 to 36 m. While thickness inversions provide important first-order information on landslide size, more work is needed to better understand how landslide subsurface properties and deforming volumes may evolve during the transition from slow-to-fast motion. Our findings underscore the need for integrated remote sensing techniques to improve landslide monitoring and forecasting. Future advancements in operational monitoring systems and big data analysis will be critical for tracking slope instability and improving regional-scale failure predictions. 

Phosphorus (P) is an essential nutrient for life. Deficits in soil P reduce primary production and alter biodiversity. A soil P paradigm based on studies of soils that form on flat topography, where erosion rates are minimal, indicates P is supplied to soil mainly as apatite from the underlying parent material and over time is lost via weathering or transformed into labile and less-bioavailable secondary forms. However, little is systematically known about P transformation and bioavailability on eroding hillslopes, which make up the majority of Earth's surface. By linking soil residence time to P fractions in soils and parent material, we show that the traditional concept of P transformation as a function of time has limited applicability to hillslope soils of the western Southern Alps (New Zealand) and Northern Sierra Nevada (USA). Instead, the P inventory of eroding soils at these sites is dominated by secondary P forms across a range of soil residence times, an observation consistent with previously published soil P data. The findings for hillslope soils contrast with those from minimally eroding soils used in chronosequence studies, where the soil P paradigm originated, because chronosequences are often located on landforms where parent materials are less chemically altered and therefore richer in apatite P compared to soils on hillslopes, which are generally underlain by pre-weathered parent material (e.g., saprolite). The geomorphic history of the soil parent material is the likely cause of soil P inventory differences for eroding hillslope soils versus geomorphically stable chronosequence soils. Additionally, plants and dust seem to play an important role in vertically redistributing P in hillslope soils. Given the dominance of secondary soil P in hillslope soils, limits to ecosystem development caused by an undersupply of bio-available P may be more relevant to hillslopes than previously thought.