Regine Morgenstern

New paper by McEwan et al. – Seismic hazard and shifting channels: Exploring coseismic river response

This is a guest post by Erin McEwan.

River systems are shaped by both gradual and sudden geological processes, and the influence of active tectonics on river behaviour is a fundamental concept in tectonic and fluvial geomorphology. Despite this, much is still unknown about how earthquake surface deformation can alter flood hazard. This is concerning as human populations are increasingly expanding onto floodplains in seismically active regions. A recent review by McEwan et al (2025) in Earth-Science Reviews addresses this knowledge gap by analysing data from 52 sites where fault deformation is known to have induced an immediate change in river behavior; otherwise referred to as a Coseismic River Response (CRR).

How Do Rivers Respond to Fault Deformation?

Figure 1: Drone aerial imagery taken 5 years following New Zealand’s 2016 Papatea fault rupture, where ~6.5 – 8.0 m of vertical offset caused the overlying braided Waiau Toa / Clarence River to partially, and then fully avulse along the Papatea fault scarp. A white truck (yellow circle) is parked atop the Papatea fault, providing a sense of scale of the change. The Waiau Toa / Clarence River has now scoured the previously productive farmland into a new ~250 – 400 m wide braid channel, while the pre-earthquake parent channel lies abandoned. Flood patterns in, and downstream of this area have been permanently altered, and river erosion continues to endanger the agricultural land and access road positioned along the true left side of the  avulsion channel. (Image credit: Regine Morgenstern 2021, GNS Science).

Case study data enabled the identification of four main classes of near-fault CRR. These range from in-channel streamflow diversion or ponding, through to overbank flooding or avulsion, wherein a channel partially or fully shifts into an enduring new course within the floodplain (Fig. 1). CRR events are most often triggered by dip-slip fault ruptures, where the elevated side of the fault partially or fully dams the river channel. More rarely, oblique motion can behead channels, or multi-meter vertical displacements can tilt riverbeds backward, reversing flow direction and promoting lake formation or avulsion upstream of the fault-river intersection (Fig. 2).

Why Do Some Rivers Respond Differently?

The study highlights key differences in how river types respond to seismic events. Braided rivers, characterized by their wide, shallow channels and mobile sediments, can distribute flow across multiple pathways, often mitigating the obstruction effect produced by fault ruptures, particularly in confined settings. In contrast, erodible substrates in unconfined environments can also facilitate rapid avulsion, as occurred during New Zealand’s 2016 Papatea fault rupture and Waiau Toa / Clarence River avulsion (Fig.1). In contrast, single thread meandering rivers are more vulnerable to rapid bank overtopping and prolonged flooding. In semi-confined settings such as intermontane basins, this can lead to the formation of a long-lived coseismic lake, as documented in Algeria’s 1980 El Asnam rupture (Fig.3). Meandering floodplains, composed of saturated fine-grained sediments, are also more susceptible to liquefaction, which can lower the ground surface and exacerbate flooding by reducing drainage capacity. In any river environment, paleochannels can enhance the potential for, and spatial extent of coseismic flooding and avulsion, highlighting the role that floodplain features play in CRR.

Figure 2A: Dip-slip dams form where vertical motion and surface offset on the fault dams an overlying channel, reducing its capacity to convey incoming streamflow and sediment (adapted from McEwan et al., 2023). Fig. 2B (top): An oblique dam, where lateral translation and attendant vertical fault offset cumulatively reduce channel capacity. Lateral translation may transfer topographic barriers (i.e., channel levees, braid bars) into the path of incoming streamflow, enhancing the damming effect produced by dip-slip motion. Fig. 2B (bottom): An oblique knickpoint, where lateral translation beheads the overlying channel, causing flooding directly onto the floodplain. Coincidently, vertical fault motion steepens the channel gradient, promoting incision across the fault. Fig. 2D: A retrograde tilt block feature, where tilt on one side of the fault reduces, or even reverses the channel slope in an overlying river system. In some cases, the orientation of fault strike relative to the river thalweg may also tilt the channel laterally, forming a retrograde-lateral tilt block.

Why Does CRR Matter?

Fault river intersections are prevalent at plate boundaries and CRR has profound implications for flood risk, especially in regions where human modification has altered natural floodplain dynamics. Globally, flooding is one of the most damaging and costly natural hazards, and models estimate that around 1.81 billion people are directly exposed to 1-in-100-year floods. Seismic activity can compound this risk by undermining flood defenses through surface rupture, liquefaction, and ground deformation. Currently, many of these impacts are overlooked when assessing flood hazard and implementing river control and flood mitigation measures.

Figure 3A: The geographic location of the 1980 El Asnam fault rupture (orange line) and Cheliff-Fodda river CRR (blue symbol). Fig. 3B: An aerial photo taken by Mustapha Meghraoui four days after the El Asnam fault rupture, illustrating the extent of the coseismic lake that formed due to the fault partially damming the Cheliff-Fodda river confluence, resulting in a Class B CRR event. Fig. 3C: A map adapted from Meghraoui and Doumaz (1996) depicting the change in seasonal flood extent within the Oued-Fodda basin following the 1980 El Asnam fault rupture. Dashed white lines represent the approximate pre-earthquake seasonal flood extent of flooding within the Cheliff and Fodda River floodplains, as interpreted by Meghraoui and Doumaz (1996). The coseismic lake that formed following the 1980 earthquake took over a decade to dissipate, reducing in extent during the dry season, and expanding again during the wetter months. In severe cases, the lake expanded to cover almost the entire basin surface.

Lessons from Past Earthquakes

New Zealand’s 2010–2011 Canterbury Earthquake Sequence (CES) provides a clear example of how seismic events can fundamentally reshape flood hazards. The CES triggered a coseismic avulsion in the Hororata River and widespread liquefaction-induced subsidence in Christchurch, permanently altering flood thresholds for thousands of residential properties. These effects rendered entire suburbs uninhabitable, leading to large-scale property buyouts and long-term shifts in urban planning. This example underscores the far-reaching consequences of earthquake-induced changes to river and floodplain systems. The pre-earthquake population of Christchurch numbered ~400,000 and the societal and economic risk posed by a large earthquake in a comparable geographic, but more densely populated and flood-prone setting, could be much higher.

Historical earthquakes in the Indian subcontinent have triggered rapid and sustained changes in river behavior, altering drainage networks and flood regimes across vast areas. However, the population density and infrastructure exposure in the region are now far greater than in past seismic events. Given the scale of settlement and dependence on floodplains for agriculture and urban development, a large earthquake today could produce severe and long-lasting disruptions to both human and natural systems. Bangladesh hosts the world’s largest lowland delta, the Ganges-Brahmaputra-Meghna (GBM) Delta, which spans ~760,000 km², half of which lies below 5 m elevation. Over 100 million people live on this delta, where seasonal flooding already affects 30–35% of the land annually, increasing to over 60% during severe monsoons. Climate change is expected to exacerbate flood hazards, while human modification of floodplains, such as stopbanks and land reclamation have reduced natural floodplain capacity and increased liquefaction susceptibility. These changes may prime river channels for coseismic flooding or avulsion following major earthquakes. Without proactive risk assessments and planning that account for coseismic river response, the consequences of such an event could be devastating for millions living near Bangladesh’s river channels.

 

Preparing for Future CRR Events

Understanding CRR is critical for seismic hazard planning. By integrating active fault databases with river centerline data, researchers can identify locations where future earthquakes may induce severe river responses. Probabilistic models can then assess the likelihood of overbank flooding due to fault rupture, while site-specific hydrodynamic modeling can refine hazard assessments. As floodplain settlement continues in earthquake-prone regions, accounting for CRR in disaster risk management is essential for mitigating future losses.

Reference

McEwan, E., Stahl, T., Langridge, R., Davies, T., Howell, A., & Wilson, M. (2025). Seismic hazard and shifting channels: Exploring coseismic river response. Earth-Science Reviews, 105042.

submit to reddit

Christoph Grützner

Christoph Grützner

works at the Institute of Geological Sciences, Jena University. He likes Central Asia and the Mediterranean and looks for ancient earthquakes.

See all posts Christoph Grützner

No Comments

No comments yet.

Leave a Comment