How scientists are using drones to lower the risk of catastrophic flooding from large glacial lakes

View of Llaca Lake, in Peru, taken from an un-crewed aerial vehicle. (Rodrigo Narro Pérez), Author provided

Rodrigo Narro Pérez, McMaster University

Early in the morning on Dec. 13, 1941, the citizens of Huaraz, Peru, heard a terrifying rumble echo across the valley. Within minutes, a torrent of water, ice and rocks had poured over the city, destroying a third of it and killing at least 2,000 people.

The natural dam of rocks and loose sediment that had held back Lake Palcacocha had failed. Eighty years later, its collapse remains one of Peru’s most tragic natural disasters.

This type of catastrophic event is known as a “glacial lake outburst flood.” Glacial lakes, such as those found throughout the Cordillera Blanca in the Andean mountain range, are often dammed by glacial moraines that can reach heights of over 100 metres. They are impressive, but they are often unstable.

Heavy rainfall and rock, snow or ice avalanches can raise water levels in moraine-dammed glacial lakes, generating waves that overtop the moraine dam or cause it to collapse, releasing huge amounts of water. These natural disasters are only expected to become more common in Peru — and around the world as climate warming melts glaciers at historically unprecedented rates.

Predicting future floods

This dark history has spurred international research into the stability of the moraines damming Peru’s glacial lakes. The Cordillera Blanca in northern Peru contains the highest concentration of tropical glaciers in the world. Predicting when these outburst floods will occur — and how destructive they will be — is of enormous concern to the over 320,000 people who live downstream.

Twisted steel beams near a picnic table.
The twisted remains of a steel bridge destroyed by a glacial lake outburst flood in Iceland in 1996. (Chris 73/Wikimedia Commons), CC BY-SA

Geological engineering models use variables such as the size and volume of the lake, height, width and slope of the moraine dam, and channel and valley dimensions to estimate the stability of the moraine dam and the risk of flood. Unfortunately, these models don’t include much information about the composition of the moraine dam, which can vary signifcantly depending on on its location and mode of formation.

My research, part of a collaboration between McMaster University and Peru’s National Institute for Research on Glaciers and Mountain Ecosystems (INAIGEM), focuses on establishing the origin of these moraine dams and the physical characteristics of the dams and the lakes they hold back. These features can have considerable influence on the stability of the dam and its potential for failure.

Using UAV to understand the structure of moraine dams

Glaciers create moraines by transporting, depositing and pushing boulders, sands and fine-grained silts and clays along the valley floor and adjacent valley walls, often forming a barrier. But one moraine may be much more stable than another, depending on the materials it contains and how it is formed.

Water may leak through weak points in the moraine’s stacked layers, taking sediment with it, or loose rocks may fall after a disturbance such as an earthquake. These weak points make a complete collapse of the moraine dam more likely. Locating these weak points is an important step in predicting the stability of the lake dams and can allow geoscientists and engineers to design more effective remediation strategies.

My colleagues and I are analyzing the architecture of large lateral moraines, which form along the sides of glaciers, in southern Iceland using un-crewed aerial vehicles (UAVs or drones) to collect high-resolution images. We use these images to identify and classify areas of coarse- and fine-grained sediment that may form zones of water leakage and sediment removal and cause the dam to fail. We’ve planned similar high resolution UAV surveys of moraine dams in the Cordillera Blanca for early 2022.

The research will enhance the reliability of predictive models to identify potential glacial lake flood hazards. It will also identify areas where remediation work, such as the building of additional outlet channels or armoured barriers, is most needed to strengthen the moraine.

people standing on a dam overlooking a glacial lake
Remediation work done at Llaca Lake, Peru, in 1977 included building a dam. (Rodrigo Narro Pérez), Author provided

This will be particularly important as glaciers melt more quickly, the volume of water held by these natural moraine dams builds, and the destructive power of floods also continues to increase. A recent study by researchers at the University of Calgary showed that the volume of water in glacial lakes has increased by 50 per cent globally since 1990.

Since the beginning of the 19th century, an estimated 165 moraine-dammed glacial lake outburst floods have occurred. In addition, approximately 12,000 deaths worldwide can be attributed directly to glacier floods.

Our research in Peru will provide new insights into moraine dam stability that can be applied to other regions, such as Bolivia, the Himalayas and the Canadian Rockies, which are also experiencing an increased risk of glacial lake outburst floods as climate warming continues to melt glaciers.

Rodrigo Narro Pérez, PhD Candidate, School of Earth, Environment and Society, McMaster University

This article is republished from The Conversation under a Creative Commons license. Read the original article.

3 5 votes
Article Rating
30 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
CRISP
June 12, 2021 10:52 pm

These natural disasters are only expected to become more common in Peru — and around the world as climate warming melts glaciers at historically unprecedented rates.”

Historically unprecedented? Went to the bottom of the article and found it was republished from The Conversation, an Australian ABC-run sceptic-free, rationalist-free site full of True Believers. Unworthy of WUWT.

Reply to  CRISP
June 13, 2021 12:01 am

“Expected to become more common “.
So aren’t more common yet then. I guess we’re waiting for a tipping point.

Historically unprecedented? That’s a bald-faced lie, isn’t it?

spangled drongo
Reply to  CRISP
June 13, 2021 12:12 am

Yes Crisp, if you post a rational, skeptical comment on The Con, no matter how relevant it is, you are banned for life.

saveenergy
Reply to  spangled drongo
June 13, 2021 2:45 am

Been there, done that,
got banned for life !

Scissor
Reply to  spangled drongo
June 13, 2021 6:12 am

Nice pictures though.

I’d like to know the real story behind placing a wooden picnic table underneath twisted metal remains of a bridge, and the “remediation work” photo of Llaca Lake dam appears to show some guys taking a break from their Frisbee game to warm their hands.

June 12, 2021 10:54 pm

Good plan regardless of whether climate change has any impact or not.

alastair gray
Reply to  Ralph Dave Westfall
June 13, 2021 2:23 pm

That is the bottom line . I saw no mention of AGW in the summary of the article but no-one would dispute that
1) many glaciers are shrinking
2) in many areas meltwater contained in morraine dammed lakes is increading
3) failure of such dams is catastrophoic.
4) we are quite capanle of evaluating the scale of the hazard and designing a mitigation strategy.
You dont need to be an AGW loon to see this as a hazard which can me mitigated by less extreme methods than us all deriving electric rollers skates or whatever the neo-STASI will allow us -definitely not shoe leather as that causes ruminant ructions

Rory Forbes
June 12, 2021 10:56 pm

Since the beginning of the 19th century, an estimated 165 moraine-dammed glacial lake outburst floods have occurred. In addition, approximately 12,000 deaths worldwide can be attributed directly to glacier floods.

So, what they’re saying is: this is a completely natural occurrence, as the Holocene interglacial lapses into late middle age. However modern technology could assist us to mitigate the problem before more people are killed. Breath easy, this wasn’t caused by us. We had nothing to do with it … but serendipity has led us to the problem during one of many fruitless “studies” searching for anthropogenic “climate” crises.

June 12, 2021 11:15 pm

The moraines are formed from the gravel that was contained within the glacial ice that was over a km. thick. It makes one realize the power of earlier global warming compared to today’s.

Ron Long
June 13, 2021 4:05 am

Good to study natural phenomena, but looks like obligatory CAGW comments for funding or support is necessary. Why not build a dam below the terminal moraine, with hydroelectric ability, then blow the moraine dam to fill the artificial lake? Since global warming will produce more water in the future the dam will generate a lot of electricity. “blow the moraine dam…” is an invitation to military bomber pilots to get some useful practice. Win-win! And a great video! I’m starting to like this idea.

Reply to  Ron Long
June 13, 2021 7:12 am

Good idea!
Only, the researchers above forgot to mention the part Earthquakes play in glacial moraine dam failures. A risk that any improvements, e.g., man-made dam, must be able to take in stride.

Tom Abbott
Reply to  Ron Long
June 13, 2021 8:38 am

“Why not build a dam below the terminal moraine, with hydroelectric ability, then blow the moraine dam to fill the artificial lake?”

That’s what I was thinking!

alastair gray
Reply to  Ron Long
June 13, 2021 2:24 pm

Now you are thinking like an old time doodle bugger Get a boat and lob a can of dynamite overboard

Christina Widmann
Reply to  Ron Long
June 14, 2021 3:33 am

And if we build the dam and leave the moraine in place? Two seams hold better than one.

Scott Anderson
June 13, 2021 4:16 am

The solution is for those people living in the path of the possible flood is to move out of the way; much like those living in the coast in danger of rising oceans need to move to higher ground. I live near the coast in North Carolina at about 20′ above sea level; if the oceans rise enough I will have ocean front property.

Scissor
Reply to  Scott Anderson
June 13, 2021 6:32 am

Mountain valleys are at risk of flash flooding, but they are also nice spots for cabins and recreation, etc. People knowingly or unknowingly risk their well being in the locations. It is best to know about the risks and to prepare for them, unfortunately lessons are often not taken to heart.

The following link describes a couple of deadly Colorado floods, the second of which involved the failure of a glacial moraine dam that had been “enhanced” by farmers about 80 years before it failed.

https://www.eptrail.com/2021/02/24/100-years-magazine-1976-1982-floods-wreaked-havoc-on-estes-park/

noaaprogrammer
Reply to  Scissor
June 13, 2021 10:21 am

The summer of 1976, I was programming for NOAA in Boulder, Colorado just 30 miles south of Loveland where the Big Thompson flood occurred on July 31, 1976.

The following Monday, I found all the meteorologists at work intently studying the weather maps, marveling at how all the wind vectors of the fronts involved in that thunderstorm canceled each other, preventing the storm from moving out onto the plains as it normally would have. As I recall, there were reports of 14″ of rain in the Estes Park drainage basin that night which had nowhere to go but downhill through the narrow Big Thompson Canyon.

My wife and I were personally acquainted with a woman and her two daughters who were hiking in the canyon the Saturday afternoon of the flood. They told a harrowing story of literally scrambling a steep hill, keeping just a few feet ahead of the rapidly rising water.

The next weekend I took some relatives to view the destruction, but a police barricade stopped our car. I just flashed them my NOAA card telling them I was an employee of the federal government, and they let me through. I filmed the destruction with my 8mm movie camera.

Scissor
Reply to  noaaprogrammer
June 13, 2021 11:19 am

I moved to Boulder later, but not in time for the 1982 flood of Estes Park.

A thing I find mind boggling is Boulder placing Police Headquarters and Justice Center just east of the mouth of Boulder Canyon. The main library straddles the creek!

The 2013 flood was bad but wait until another is as bad as the 1894 flood. I’m sure the meteorologists will marvel at the weather pattern setups again. Just a matter of time.

H.R.
June 13, 2021 4:38 am

The last image in the article, labeled “Remediation work done at Llaca Lake, Peru, in 1977 included building a dam. (Rodrigo Narro Pérez), Author provided” puts the lie to pretty much all of the hype and loaded CAGW language in the article.

Before anyone had even heard of CO2-based CAGW, the Peruvians had already recognized the problem of glacial moraine-dammed lakes and were taking action to remediate the problem.

Back then, people went out and actually looked at potential problem areas; no drones, no models. In this case they said, “Hey! This one is looking really sketchy. We better shore it up.”

Certainly the drones and models speed things up quite a bit. So, more moraine dams can be evaluated without sending out people to climb around for days or weeks assessing the situation. That’s great!

But the problem I’m seeing is getting the politicians to spend money on practical fixes instead of wind mills.

Politicians: “That dam is getting ready to go… I know! Let’s put in a solar farm and some wind turbines. That’ll fix it!”

Maybe they should just brace up those moraine dams with all the scrapped blades from dead wind turbines. Win-win.

(*sigh* /Sunday morning only-one-coffee grumpy comment)

Reply to  Hans Erren
June 13, 2021 7:25 am

Looks to be an excellent report.

“Changes Over Time The inconsistency between the data sources in the inventory of 2001 and the present study presents a problem in obtaining a meaningful comparison as the two inventories are not strictly equivalent. Nevertheless, an attempt was made to identify the main differences (Table 4.5).”

Unlike the drone flying kids in the article above, this looks to be a good summary without claims of future doom.

The pictures in the Nepal GLOF study, also indicate there should be some concern over what happens when the glaciers grow, again. A problem more likely to occur as our interglacial cools.
i.e., Man-made dams may not survive a glacier returning. Of course if the glacial lake is mostly frozen there isn’t an immediate worry.

One of the pictures in the Nepal study image a large slough off the sides of a mountain above a town. A significant way glacial waterways are blocked and form glacial lakes.

June 13, 2021 5:38 am

The enquiring mind would like to know who exactly is paying for the insurance for these self pleasuring children clowns…

Picture it:
The drone flies and looks, clowns give it the OK
Repeat a few times. Still OK
People then imagine it is safe to go live under the glacier – a lot more people than otherwise would if The People they’d trusted their own eyes, ears and actual experience.

You know what happens next. The drone and its clown pilots get it wrong one day…
Two three or four times as many people perish as otherwise would have.

And the drone pilots are to be found…. where?

Power without accountability – the pretty accurate definition of Climate Change.
Or a rehash of either Jevon’s Paradox or the Common’s Tragedy

Scissor
Reply to  Peta of Newark
June 13, 2021 6:35 am

Sometimes doing good doesn’t pay.

Loren C. Wilson
June 13, 2021 6:04 am

Since the temperatures in the tropics are increasing the least, the problem is that more people live below these natural dams. The solution is to allow them to farm in the flood plain but not live there. And build real dams.

June 13, 2021 6:41 am

A) Introduce a catastrophe; in this case the ‘catastrophe’ is a glacial dam failure in 1947.

B) State a portion of the current situation.

Geological engineering models use variables such as the size and volume of the lake, height, width and slope of the moraine dam, and channel and valley dimensions to estimate the stability of the moraine dam and the risk of flood. Unfortunately, these models don’t include much information about the composition of the moraine dam, which can vary signifcantly depending on on its location and mode of formation.”

Better described as ‘Their models don’t really work’

C) Make an outrageous claim.

“The research will enhance the reliability of predictive models to identify potential glacial lake flood hazards. It will also identify areas where remediation work, such as the building of additional outlet channels or armoured barriers, is most needed to strengthen the moraine.”

D) Make an outrageous claim regarding benefit of the new research.
N.B. This benefit is supposition, not documented.

“This will be particularly important as glaciers melt more quickly, the volume of water held by these natural moraine dams builds, and the destructive power of floods also continues to increase. A recent study by researchers at the University of Calgary showed that the volume of water in glacial lakes has increased by 50 per cent globally since 1990.”

Also note that the linked research is another ‘We just learned we can do this and the situation is worse than we thought’.

Models.
New model. Coupled with new gross unproven supposition.

“We use these images to identify and classify areas of coarse- and fine-grained sediment that may form zones of water leakage and sediment removal and cause the dam to fail.”

A claim that new allegedly model supports old model.
a) No references to certifying and verifying their analysis method.
b) All suppositions. No plan to survey glacial lakes and list by concern.
c) They have great field trips to Peru. Just what the Peruvians need…

June 13, 2021 7:15 am

If glaciers are melting rates how did they measure the rates of melting the last time glaciers melted so they would know this.

How fast did the glacier below melt?

4FA5C6FB-A3B7-4CD0-87C6-7F9392460E30.jpeg
Sara
June 13, 2021 11:00 am

Gee, if you look at the depth of that valley, down to the water/ice surface, in the first photo, you might understand just how deep the glacier was that filled that valley some time in the past, and the volume of water it held in the form of ice and snow.

There have probably been floods and cascades in the past that were worse than anything that could happen now. We live in interesting times, we do.

dk_
June 13, 2021 12:50 pm

The Conversation believes that better observarion is the same as risk reduction, and that risk reduction is something that scientists do . Maybe this is the yardstick we should use for any scary risks or claims for scientists made by green alarmists.

June 13, 2021 5:02 pm
June 14, 2021 2:12 am

Early in the morning on Dec. 13, 1941, the citizens of Huaraz, Peru, heard a terrifying rumble

Yes 1941 was toward the end of the previous natural warming cycle, between 1900-1940-ish.
So a glacial lake burst around then is not unusual.

In fact these people are so far separated in their thinking from the rational scientific method that they don’t realise that they have provided a useful real world test if global warming is really happening or not.

Every day it gets hotter, every day more unprecedented, every day more catastrophic, every day more idiotic.

So – where are the glacial dam bursts??

Come on – we used to hear every day how glaciers are melting worldwide. Now it’s every hour. So these dam bursts should be happening on a regular basis. All over the world. They should be in our faces in the news continually.

But they’re not, are they? Because they are not happening.

Or is this another case of “it may look good now but just you wait, our computer-prophets tell us that any day now the catastrophe and extinction will begin. Any day now…

The fact that they have to go back to 1941 for an actual Peru dam burst shows that the warming back then was probably more pronounced than it is now.

Or as Dave Middleton would call it, this whole article is just another “self inflicted gunshot wound” for the alarmists.