Recent  

To study channel network development following continental glaciation, we ran small physical experiments where networks slowly expanded into flat surfaces. By changing substrate and rainfall, we altered flow pathways between the surface and subsurface. Initially, most channels grew by overland flow. As relief increased, erosion through groundwater sapping occurred, especially in runs with high infiltration and low cohesion, highlighting the importance of groundwater in channel network evolution.

Quantification of bedload transport constitutes a challenge in river dynamics due to variable transport modes. We used passive monitoring device to record acoustic signals generated by bedload particles moving with different transport modes and established relationship between signal and bedload characteristics. The findings in present study could improve the understanding of our monitoring system and bedload transport process, contributing to bedload size classification as well.

Recently, along the eastern coast of South Korea, seasonal beach erosion induced by the structures which severely block the supply of sand from the upstream side. This study proposes a coastal solution that can predict the erosion of critical points in sag curves formed downstream of longshore sediment control facilities by applying a parabolic bay shape equation (PBSE).

There is an increasing demand for tidal marsh restoration around the world. We have developed a new modeling approach to reduce the uncertainty associated with their development. Its application to a real tidal marsh restoration project in Northwestern Europe illustrates how the rate of landscape development can be steered by restoration design, with important consequences on restored tidal marsh resilience to increasing sea level rise and decreasing sediment supply.

Some types of big earthquakes create small cliffs or "fault-scarps" around a ~1–3 meters in height, where sediments can pile up and create deposits we call "Colluvial Wedges." Geologists will look at colluvial wedges and use them to understand how often big earthquakes occur. Here we made a computer simulation to find out if the way we think colluvial wedges form works with physics. We found that it does in theory, but there are conditions where it may be more complicated than we expected.

Granite residual soil landslides are widely distributed in southeastern Guangxi province, China. In this study, it conducted artificial flume model tests to investigate the failure mode of granite residual soil landslide. The research results can provide valuable references for the prevention and warning of granite residual soil landslide in southeast Guangxi.

We compiled a large image dataset containing more than 125,000 sediments and developed a model (GrainID) based on convolutional neural networks to measure individual grain size from images. The model was calibrated on flume and natural streams images covering wide range of fluvial environments. The model showed high performance in comparison to other methods. Our model showed great potential for grain size measurements from small patch of sediment in a flume to watershed-scale drone survey.

We dated 187 earthflows to determine how they impacted salmon habitat in the Teanaway basin, central Washington State, USA. We developed a new method to date earthflows that uses the texture and directionality of the landscape. Earthflows were active 3-5000 years ago, with some active in the last 1000 years. The period of earthflow activity coincides with climate change as well as large salmon populations, suggesting the wide floodplains formed by earthflows overall increased habitat.

Our work is focused on describing how and why the terrace levels of the central Amazonia were formed during the last 100'000 years. We propose to address this question through a landscape evolution numerical model. Our results show that terrace levels at lower elevation were established in response to dry-wet climate changes and the older terrace levels at higher elevations most likely formed in response to a previously higher elevation of the regional base level.

Obtaining accurate information on the volume of geomorphic features typically requires high-resolution topographic data, which is often not available. Here, we show that the globally available 12 m TanDEM-X DEM can be used to accurately estimate gully volumes after applying a correction. This allows to get a first estimate of the amount of sediment that has been mobilized by large gullies (lavaka) in central Madagascar over the past 70 years.

Himalayan rivers are full of giant boulders that almost never move except during glacial lake outburst floods (GLOFs). GLOFs therefore must be very important for driving erosion. GLOFs are rare, so little is known about how they control erosion long-term. We analyzed rivers in the Nepal Himalaya and found the slopes and widths of channels and valleys suggest GLOFs as the dominant means of erosion. This is often unaccounted for in erosion and landscape evolution studies and should be considered.

The evolution of a landscape through erosion is fundamentally a geometric problem. We use tools from classical mechanics and differential geometry to transform an erosion model into a Hamiltonian. This Hamiltonian shows how erosion acts in two directions simultaneously: normal to the surface, and along rays angled upstream. The rays follow paths of shortest erosion time, point up or down depending on how erosion scales with slope, and preserve some boundary information unchanged as they move.

We investigated the influence of sediment transport modes on the formation of bedforms using theoretical analysis. The results of the theoretical analysis were verified with published data of plane beds obtained by fieldwork and laboratory experiments. We found that suspended sand particles can promote the formation of plane beds on a fine-grained bed, which suggests that the presence of suspended particles suppresses the development of dunes under submarine sediment-laden gravity currents.

The condition of peatland largely determines the capacity for it to store carbon, but peatland condition is not accurately known. Combining the knowledge of management, vegetation and detecting differences in seasonal surface movement from satellite radar data, we map peat condition. In a blanket bog landscape we discovered the presence of wetter and dryer conditions, that could help guide restoration decisions and conclude that this approach could  transfer to peat management worldwide.