Megafloods mostly shaped Icelandic canyon landscape
In the 1920s, geologist J Harlen Bretz proposed that cataclysmic glacial megafloods had dramatically transformed the landscape of the northwestern U.S., leaving behind distinctive landforms like the channeled scablands in eastern Washington. Though Bretz’s notion, initially controversial, has long since been accepted, resolving how much landscapes are shaped by extreme events as opposed to the slow action of everyday erosion has nonetheless remained difficult.
Now, a new study suggests that in one environment prone to catastrophic floods, the Jökulsárgljúfur Canyon in northeastern Iceland, the effects of several megafloods in the past 9,000 years have sculpted the canyon landscape far more so than “background” erosion caused by water and sediment flowing through it.
Iceland is a fertile location to study the impacts of extreme floods due to the abundance of glacier-capped volcanoes. The heat released when such volcanoes erupt can melt huge portions of the glaciers, unleashing massive torrents of floodwater. “We demonstrate that these short, extreme flood events can not only cause a great deal of damage initially, but also leave a long-lasting legacy on the landscape,” says Edwin Baynes, a graduate student in geosciences at the University of Edinburgh in Scotland and lead author of the study, published in Proceedings of the National Academy of Sciences.
The volcanic landscape around Jökulsárgljúfur, through which the Jökulsá á Fjöllum River flows, is made up of jointed basalt columns. Extreme floods along the river — when river discharge rates exceed 3.25 million liters per second, or more than 6 times the usual maximum discharge of the river — are powerful enough to dislodge large chunks of this basalt, according to the study. This rapid erosion can create knickpoints — sudden changes in elevation along the course of the river where waterfalls typically form — or rapidly shift existing knickpoints far upstream.As knickpoints move upstream and existing surfaces are washed away, previously buried rock is exposed to the atmosphere and to cosmic rays, leading to the accumulation of cosmogenic isotopes, like helium-3, in the newly-exposed surfaces. Based on known accumulation rates of different cosmogenic isotopes in different rocks, researchers can calculate how long a particular surface has been open to the atmosphere, and thus when it was first exposed. So, Baynes and his colleagues sampled surface rocks at several locations along a stretch of the Jökulsárgljúfur well known for famous knickpoints — the Selfoss, Dettifoss and Hafragilsfoss waterfalls — and analyzed the rocks’ helium-3 contents.
If the waterfalls along Jökulsárgljúfur were gradually moving upstream due to slow-and-steady erosion, then rocks left behind by the retreating waterfalls should have progressively older exposure ages the farther downstream they are from the present-day location of the falls. “But what we found is that the ages were all clustered around certain points in time, suggesting the waterfalls had moved back very rapidly,” Baynes says.
The sudden shifts of the knickpoints tell a story of violent, extreme floods having occurred about 8,500, 5,000 and 2,000 years ago, the researchers reported. Although background erosion may play a small role, in their estimation, these megafloods have caused most of the change in the Jökulsárgljúfur landscape.
The novel aspect of the study is “the helium-3 dates through which the authors can show the timing of the[se] process[es]” in detail, says Victor Baker, a paleohydrologist at the University of Arizona who was not connected with the study. The work also demonstrates the utility of field measurements in studying large-scale landscape change, Baker says. “You can't make a stream 10 meters deep flowing at 10 meters per second in a lab; it will tear all your equipment apart,” he says. “The only way we can understand [extreme flood events] is to look at field relationships, like this study does.”
Still, not everyone is sure yet of the conclusion — that a handful of extreme events have caused the bulk of the erosion in the Jökulsárgljúfur Canyon.
“It is a very sparse [cosmogenic dating] data set,” consisting of just eight dated samples, notes Kelin Whipple, a geologist at Arizona State University. “There is a good chance the authors’ interpretation is correct,” he says, adding however that he’s “unconvinced the data quantitatively demonstrate that megafloods are doing nearly all of the geomorphic work in this setting.”
Isaac Larsen, a geologist at Caltech, agrees. “The number of exposure ages is quite small, and several exposure factors, such as prior exposure and erosion, can influence the [cosmogenic isotope] concentration of any given sample,” he says. Nonetheless, the data the researchers collected “do support multiple ages of canyon carving, and are consistent with previous dating of flood deposits” in the area.
Although the effects of megafloods on landscapes can depend on many site-specific factors — including the types of rocks present, the volume and flow rate of the floodwaters, and the presence of vegetation — this study has implications for locales beyond Jökulsárgljúfur, Larsen says. “It suggests,” for example, “that terraces in other landscapes on Earth and Mars with similar jointed columnar basalt may also be recorders of the history of outburst floods.”