Science with Local Impact: Daniela Schmidt’s research on GLOFs in El Chaltén

In this edition of the blog, we interview Daniela Schmidt, a PhD candidate at the Faculty of Exact and Natural Sciences of the University of Buenos Aires (UBA), who studies Glacial Lake Outburst Flood (GLOF) processes at Laguna Torre. Daniela shares the progress of her doctoral research, what we currently know about the mechanisms that trigger these sudden drainage events, and how they are modeled in a territory that is rapidly changing due to glacier retreat.

We also discuss the key role that citizen observations through HidroClim can play, especially in basins with scarce historical data: simple records of water levels and river conditions that can significantly improve the understanding and modeling of extreme events such as GLOFs.

Could you tell us who you are and why you chose to study GLOFs in El Chaltén?

My name is Daniela Schmidt. I am a geologist and currently pursuing a PhD at the Faculty of Exact and Natural Sciences of the University of Buenos Aires, funded by a CONICET fellowship. My research focuses on geological risk associated with GLOFs, and in particular on the potential outburst of Lake Torre and how it could affect the town of El Chaltén.

I chose to study the potential GLOF in El Chaltén because I wanted my work to serve a purpose beyond “scientific knowledge” alone. I want my research to help people, to provide information that allows them to understand nature and prepare for hazards that may affect them in their daily lives.

What is your doctoral thesis about?

My PhD thesis focuses on analyzing the geological risk associated with a potential GLOF from Lake Torre and its impact on the town of El Chaltén and nearby hiking trails. Specifically, I study how the active landslide on the slopes of Cerro Solo could release large volumes of material into the lake, triggering a sudden overflow and a downstream debris flow.

Concern about a potential GLOF at Lake Torre first arose in 2015, when Professor Winocur—my PhD advisor—and collaborators identified an active landslide on the northern slope of Cerro Solo. At that time, the research group was unable to secure funding to continue the study. However, as the slope continued to show activity, in 2021 we submitted a research proposal that allowed me to develop my doctoral work on this topic.

How does your work relate to glacier and proglacial lake risk monitoring and management?

The ultimate goal of my PhD research is to assess the risk associated with the landslide on the northern slope of Cerro Solo, which could release large volumes of water into the Fitz Roy River and cause significant downstream flooding.

To track the evolution of the slope from Buenos Aires, I use Sentinel-2A satellite imagery, which is typically updated every 3–4 days. However, this is not real-time monitoring, as any detected change has already occurred by the time it is observed.

As part of my research, two camera traps have also been installed in the area since November 2022: one monitoring the slope of Cerro Solo and another on Cerro Techado Negro. These cameras do not transmit data in real time, so the information must be retrieved every six months, either directly or with the help of the IANIGLA team. The photographs allow me to observe slope evolution, movements, and triggering factors such as snowmelt, rainfall, and water infiltration.

Additionally, I conducted three field campaigns, which made it possible to directly measure the evolution of cracks, scarps, and other signs of slope instability. Over the past three years, the Cerro Solo slope has shown significant activity, including block falls, shallow landslides involving both morainic material and forested areas, and the propagation of an “incipient scarp.” Regarding GLOF risk management, while one objective of my thesis is to propose mitigation, preparedness, and early warning measures, decision-making ultimately lies with government agencies.

What are the main results so far regarding GLOF risk?

According to the two proposed flood modeling scenarios, the southern sector of El Chaltén would be the most affected, with significant impacts on Provincial Route 41 and the bridge crossing the Fitz Roy River at the entrance to the town—two critical infrastructures for external aid in the event of flooding. Within the National Parks area, the De Agostini campground and certain sections of the trail leading to Lake Torre would also be affected. These results were communicated in a report and a presentation given to National Parks authorities in April of this year.

The northern sector of the Fitz Roy River, where most of the population, tourists, and major buildings are concentrated, would not be severely affected under these scenarios. However, both models only consider “clear water” flooding—that is, they account only for the volume of water released during a GLOF, without considering trees, blocks, or sediment.

While clear-water models are useful as a first approximation, a more realistic analysis must include sediment, trees, and large blocks transported by the flow. This is what we are currently working on. It is very likely that when these elements are included, the impacts will be much more severe than those shown in the preliminary scenarios. Even if the northern sector does not appear strongly affected in the current models, both riverbanks could be significantly impacted in more realistic scenarios.

What changes in glacier or hydrological dynamics have been observed in recent years that could increase or reduce risk?

Analysis of satellite imagery and field photographs shows that ice loss from the Torre Glacier has accelerated in recent years, doubling its retreat rate compared to earlier periods. Over the past 57 years, the glacier has lost approximately 4 km² of its surface area, and its front has retreated about 900 meters. There is also clear evidence of increased instability at the glacier front, such as the production of numerous icebergs, strong fragmentation, transverse and longitudinal crevasses, and meltwater ponds forming on the glacier surface and along its margins. This situation increases GLOF risk, as slopes that were previously supported at their base by glacier ice are now in direct contact with lake water.

At the same time, the size of Lake Torre has increased considerably, suggesting that a potential GLOF could mobilize a larger volume of water.

What are the main uncertainties or information gaps when assessing GLOF risk?

When assessing geological hazards, estimating peak discharges entering the Fitz Roy River due to a Lake Torre outburst requires the use of multiple formulas—many empirical or based on laboratory experiments—which may not fully reflect local conditions. As explained in the research, different formulas can produce very different results.

Additionally, there are no continuous data on Lake Torre water levels or Fitz Roy River discharge, which are fundamental for establishing reference water heights. There is also no active real-time monitoring system on the Cerro Solo slope.

Regarding vulnerability assessment, there is a lack of precise and up-to-date official data on population distribution, seasonal workers, tourist numbers, and housing conditions, particularly near the Fitz Roy River.

What types of data or monitoring would be needed to reduce these uncertainties?

It would be crucial to install water level gauges in both Lake Torre and the Fitz Roy River, as well as a monitoring system on the slopes of Cerro Solo and Cerro Techado Negro using stakes, inclinometers, extensometers, and piezometers to measure movement, crack opening, and pore water pressure.

Installing a real-time meteorological station and a seismic station would also be essential, as shallow earthquakes could trigger sudden slope movements. All of this should be integrated into an Early Warning System providing real-time, accurate information to authorities and the population.

Why are there differences among flood maps produced since 2019?

The first flood susceptibility map produced by SEGEMAR was an excellent starting point. While it did not show potential flood depths, it identified areas that could be affected.

Our recently published work aimed to improve upon this by modeling both flood extent and water depth, using a methodology developed by the University of Zurich and proposing two scenarios based on potential landslide volumes. However, these models only consider clear water and do not include sediment or debris, which would likely worsen impacts.

How could hydrological data like those generated by HidroClim help reduce uncertainties?

HydroClim data on water level and discharge would be fundamental, as there is currently no continuous monitoring station on the Fitz Roy River. These data are essential to establish baseline conditions and understand seasonal and event-driven variability, which greatly affects GLOF modeling outcomes.

How can hydrological information improve risk prevention and local planning?

Understanding how the Fitz Roy River’s discharge varies throughout the year and during extreme events enables more realistic flood modeling, better contingency planning, and clearer identification of evacuation zones and safe routes for both El Chaltén and nearby hiking trails.

What challenges do you face when working with hydrological data in remote glacier environments?

The main challenge is the lack of real-time data due to difficult access and harsh conditions. Transmitting data in real time is complex and expensive, often requiring satellite communication systems.

How can community participation in data collection support your scientific work and local risk management?

Community participation can be extremely valuable. For example, tour guides and residents who regularly travel local trails could record water levels and river conditions. These data would help improve GLOF flood models and support more effective local risk prevention strategies.

We would like to especially thank Daniela Schmidt for her willingness and openness in answering our questions.