We mapped landslides (> 100 m²) in the Nepal Himalayas due to 8 monsoons and large landslides (> 0.1 km²) since 1970. Adding inventories of Holocene, giant landslides (> 1 km³), and of landslides from the 2015 Gorkha earthquake, we constrain the size-frequency distribution of landslides induced by monsoons and earthquakes. They both contribute ~ 50 % to a long-term (> 10 kyr) total erosion of ~ 2 mm/yr, matching long-term exhumation rate. Large landslides, rarer than ¹⁰Be sampling time, drive erosion.

In Bangladesh, we measure how water and sediment move through tidal channels in an area heavily impacted by both sea level rise and the strategies that humans use to adapt. We found that the amount of water moving through the system is largely controlled by the strength of the tides, while the mass of sediment being transported is strongly connected to the season, and to a lesser extent the tides. These results are important for developing strategies to protect both human and natural landscapes.

Using data obtained from multiple UAV flights over a debris-covered glacier in the Himalaya between 2013 and 2018 we show that the adjacent moraines erode at rates of up to 30 cm per year, contributing to this debris cover. With retreating ice and resulting instability of moraines this causes the glacier to cover itself in ever thicker layers of rocks, resulting in a possible future decrease of local melt.

Our study reveals that worldwide mean erosion rates on the million year scale are very similar to present-day erosion rates in contrast to the majority of the previously published results. Concerning the dependence of erosion on climate we found that the long-term erosion efficacy of the tropical zone has been about five times higher than that of the cold zones, while the erosional efficacy of the present-day arid zone has been as high as that of the temperate zone.

In many coastal, riverine, deltaic, marine, and subglacial systems, plentiful fine bed sediment is produced and moved to change the surface features and surrounding environments along its pathway. Our numerical research shows that, in these systems, weakened bed level changes can reduce bed friction and produce higher sediment yield, whose effects can be amplified under greater water depth. This finding can help predict sediment movement and bed level changes of water channels.

Mountain-building processes are often investigated by assuming a steady-state, meaning the balance between opposing forces, like e.g. mass influx and mass outflux. In this study we show that the assumption of steady-state for the Olympic Mountains is depending on the timescale and the spatial geometry, and particularly affected by the Quaternary glaciation. This indicates that mountain ranges can be out of steady-state during parts of their evolution.

We present in this mostly theoretical contribution a systematic treatment of tracer particle mixing in soils. We elaborate the consequences of rarefied (non-continuum) conditions of transport and mixing, and we augment this with numerical analyses that reveal important information not readily apparent in the analytical formulations, including an illustration of the variability in ¹⁰Be concentrations and OSL ages of individual particles in soils, with implications for interpreting field data.

We evaluated the performance of the low-cost seismic sensors Raspberry Shake (RS) to identify and monitor rock fall activity in alpine environments. The sensors have been tested for 1-year period in high alpine environment, recording numerous rock failure events as well as local and distant earthquakes. This study demonstrates that the RS instruments provide a good option to build low seismic monitoring networks to monitor different kind of geophysical phenomena.

High resolution digital elevation models (DEM) can be now frequently acquired using airborne Laser scanner. This allows to analyse in great details the geometry of landslides. Several large landslides were triggered by the Typhoon Talas in Japan in 2011. The comparison of pre- and post-DEM allowed to test a method defining the landslide failure surfaces before the catastrophic movements. It provides new results about the curvature of the failure surface and the volume expansion of the deposit.

Quantitative analysis of digital topography is an important part of many modern geologic investigations. We have constructed a series of open source tools, the Topographic Analysis Kit (TAK), built on top of the existing TopoToolbox library, to allow for efficient and easy analysis of topography that limits the required computing and domain specific knowledge necessary to successfully produce robust analyses of topographic data.

This research article provides a comparison between the traditional method of measuring erosion on rock shore platforms using a Traversing/Micro Erosion Meter (T/MEM) and a new approach using Structure from Motion (SfM) Photogrammetry. Our results indicate that SfM Photogrammetry offers several advantages over the T/MEM allowing measurement of erosion at different scales on rock surfaces with low roughness while also providing a means for identifying different processes and styles of erosion.

This work demonstrates how seismic networks installed in the Alps can be used for country-wide real-time monitoring of rockslide activity. We suggest several simple methods which allow to detect, locate and characterize rockslides using the seismic signals they generate. We developed an automatic procedure that allows to locate rockslides with kilometer accuracy over hundreds of kilometers distance. Our findings highlight how seismic networks can help to understand the triggering of rockslides.

The uppermost part of the permafrost thaws during summer and freezes again during winter. We use a passive seismic method to monitor these changes continuously between April and August 2016 close to Ny Ålesund, Svalbard. Our results reveal some potential pitfalls when interpreting temporal variations. However, we show that after a careful analysis this method has the potential to become a very useful tool for long-term permafrost monitoring, being of special importance in a changing climate.

We show a new and novel use for the SRTM-C digital elevation model from February 2000 and the newer TanDEM-X dataset from ~ 2015. We difference the datasets over hillslopes and gravel-bed channels to extract vertical land-level changes. These signals are associated with incision, aggradation, and landsliding. This requires careful correction of the SRTM-C biases using the TanDEM-X, with the possibility to extend the method to moderate relief areas with SRTM-C and TanDEM-X coverage.

Landscape evolution models (LEMs) aim to capture an aggregation of the processes of erosion and deposition, and predict evolving topography. A key aspect of any LEM is how water is chosen to be routed down the surface, which can impact the model results and importantly the numerical accuracy. I find that by treating rivers as lines within the model domain, and by distributing water down all slopes, the results are independent of resolution, pointing to a new method to model landscape evolution.