River bedrock erosion can occur by abrasion and by the removal of entire blocks. We observe that when there is no or few fractures most erosion occurs by abrasion, whereas with more fractures, blocks can be removed at once leading to different patterns of erosion and riverbed morphology. Fractures affect barely mean erosion rate but change the location and occurrence of block removal. Our results highlight how river bedrock properties influence erosion processes and thus landscape evolution.

Fluvial fans are a newly recognized type of river system that look like river deltas, especially when they reach lakes or oceans. This study explores how to tell them apart by measuring the size and layout of channels in these fan-shaped landforms. Understanding these differences helps to predict how these landforms respond to climate change and urbanization, and to identify them on Mars and other planetary bodies.

We analyzed ten years of satellite images of the Chikuma River in Japan to track how riparian vegetation respond to floods and relative elevation. Greenness dropped most at low elevations after floods, while higher areas stayed relatively stable. Growth followed a clear late-summer peak. These findings show when and where vegetation control can best improve flow capacity and reduce flood risk, supporting practical river management.

Bedrock strength in bedrock river is often seen as controlling incision rates and river profiles, natural changes in rock type do not always match slope changes. In the Abukuma River basin, Japan, we measured bedrock strength and despite large strength differences, slopes were nearly uniform. Numerical tests showed that the model, which includes sediment cover and erosion effects, best explained river profiles. Thus, sediment plays a greater role than bedrock strength in shaping river profiles.

This study examines how the shape of tidal dunes influences flow and turbulence under reversing currents, simulating tidal conditions in a large flume. We show that dune slope properties significantly affect flow patterns, especially the presence and size of intermittent or permanent flow separation and turbulent wake. The results highlight the key role of dune morphology in shaping flow dynamics, with implications for sediment transport and coastal morphodynamics.

Coastal areas are vulnerable to sea-level rise and earthquakes. Understanding past changes requires precise deformation estimates. Marine terraces record sea-level and tectonic histories but mapping them has relied on subjective criteria. TerraceM-3 introduces standardized workflows and a machine-learning-based approach that, combined with ICESat-2 altimetry, enhances the accuracy and reproducibility of marine terrace mapping.

Understanding and predicting the evolution of river landscapes is critical for effective river management. Traditional physics-based morphodynamic models, while accurate, are computationally intensive and often impractical for long-term applications. This study presents a robust deep learning framework, which was designed to overcome the computational limitations by enabling rapid and reliable predictions of hydrodynamic and sediment transport behaviors.