Ice crystals, a scientific itinerary into the heart of waterfall ice

How are waterfalls formed? How does this very specific type of ice react to a climber’s progression and temperature variations? All climbers have asked themselves these questions at one time or another before attempting a tricky waterfall climb. Only their experience can guide them. But beyond pre-conceived ideas, what is the physical or mechanical reality of these frozen structures? In 2006, to answer these questions, the Petzl Foundation launched a fundamental research program, in partnership with the Laboratory of Glaciology and Environmental Geophysics of Grenoble.


© Monica Dalmasso


  • Project partner: Laboratory of Glaciology and Environmental Geophysics of Grenoble (LGGE),, and University of Fairbanks
  • Country: France and Alaska, United States
  • Project type: Gaining knowledge
  • Budget: €286,000 since 2006

An original research project

Many of the questions addressed to the glaciologist Luc Moreau, during his ENSA training sessions for novice mountain guides, are behind this project. Although glacier snow and ice have been the topic of many scientific studies in the past, until now there has not been a serious scientific publication on the issue of waterfall ice.A world première, this approach aims to provide a better understanding of the formation and behavior of waterfall ice, specifically for climbers but also for the scientific community.


A highly-skilled team

Under the banner of the Petzl Foundation, a working group was created around two LGGE researchers: Maurine Montagnat for the study of ice microstructures, which is essential to the understanding of the waterfall ice formation; Jérôme Weiss for the mechanical aspect of the study, which aims to understand how and why waterfall ice breaks.
In addition, the glaciologist Luc Moreau brought his know-how of “moving ice”. He filmed the movement of waterfall ice throughout the study.

The team © Monica Dalmasso

From the guides point of vue:

François Damilano and Didier Lavigne. These two mountain guides who are experts in waterfall ice-climbing took an active role in the study. 

Francois Damilano © Monica DalmassoDidier Lavigne © Didier Lavigne"As guides, this approach has been particularly enriching for us. Our role as climbers was to contribute as much of our knowledge gained through experience as we could. This study has enabled us to back up what we already know through experience."

Fundamental or applied research?

Maurine Montagnat © Maurine Montagnat"For the scientists, this field of study is totally unexplored, Their only starting point was the observations given to them by experienced ice-climbers. This meant that as researchers, we approached this study from the “fundamental” angle. In other words, they had no specific financial or practical goal in mind, which is generally the aim of “applied" research. At this stage, our goal is of a fundamental nature rather than to evaluate objective risks. However, we hope that our observations will ultimately help climbers and professional guides to improve their evaluation of field conditions », said Maurine Montagnat.

Video of the project

What conclusions can we reach?

The data collected can help us reach a number of conclusions relating to the influence of environmental conditions on waterfall ice:

Cascade de glace

  • However, a succession of fairly mild days:
    these conditions can lead to detachment of the ice from its bedrock.
  • However, a succession of fairly mild days, up to about 5 ° C maximum, interspersed with relatively cool nights (a few degrees below 0° C) to prevent the detachment seems to be a rather favorable situation: "hot" ice will behave in a “plastic” way which is not conducive to the propagation of fractures: it is the "sorbet ice".
  • If intense cold is conducive to the formation of waterfall ice, it is not necessarily conducive to climbing.
    Brutal chills generate stress in the structures (thermal contraction of ice) that can trigger their collapse, especially free-standing.
    In addition, cold ice will be brittle, which leads to the propagation of fractures, with the possibility of a collapse triggered by climbers themselves.
    Apparently, stalactite - or cigar - like structures are less sensitive to the conditions described above.
  • Slower cooling, spread over several days, and / or more stable cold spells are less critical in theory. However, under these conditions, cold ice becomes brittle under the blows climbers’ ice axes.

These conclusions remain theoretical, and can never replace the experience gained in practice.

The life story of a frozen waterfall

How are waterfalls formed? What is the structure of the ice in them? How do the temperature variations affect them?
What can we learn about the growth mechanism from an analysis of ice microstructure?

Every winter, from 2006 to 2010, experiments were carried out on waterfall ice to understand the formation of these ephemeral ice structures. Essential to this understanding is the study of microstructure, and the temperature variations, which affect it.

For this very vertical study, it was necessary to develop a new research protocol and design specific tools. A new portable drilling system was developed specifically for vertical ice sampling, in collaboration with the Petzl design office. This innovative tool, made at the LGGE was successfully used to extract samples at depths of 60 cm from the ice walls.

Carottage dans le free-standing de la cascade "Nuit-Blanche" © Monica DalmassoExtraction d'une carotte prélevée dans le free-standing de la cascade "Nuit-Blanche" © Monica Dalmasso
Stalactites and ice samples were extracted were extracted on site.

These samples are then sliced to obtain thin sections, 0.2 to 0.3 mm thick, fixed on glass plates.

Découpage de la glace au laboratoireLame mince de glace sur plaque de verre, au laboratoire © Monica DalmassoObservation d'une lame mince en lumiere polarisée © Monica DalmassoObservation d'une lame mince au microscope © Monica Dalmasso
When observed under polarized light, these thin sections reveal the microstructure of the ice (grain sizes and shapes).

What is ice microstructure?

Analyse de la structure de glace © Monica Dalmasso
Ice is a crystalline material with a hexagonal crystallographic structure.

The orientation of a single ice crystal, also called ‘grain’, is given by the longer axis of the hexagon. The microstructure represents the shape and size of the grains. The entire grain or orientation of sample as well as its microstructure is closely linked to the solidification mechanisms or the deformation and recrystallization history of the ice structure.

What can we learn from ice microstructure?

  • If the ice is white when the grains are small, it means that it has formed suddenly, after a sudden cooling, which has caused a large number of small air bubbles to be trapped inside it.
  • If the ice is black, it means that it has formed slowly, under stable temperature conditions, so it has little or no trapped air bubbles, and therefore it is very transparent.
  • Vertical ice structures grow rapidly at first by aggregation of stalactites. After this initial phase, the volume of the ice structure reaches an asymptotic value. Water continues to flow inside the structure. The ice insulates this water from the outside temperature, thereby preventing it from freezing. 
    Cracking due to the passage of a climber may cause a new flow of water outward and resulting in a new growth phase of the waterfall ice.

Growth development of the Nuit Blanche free-standing:

To monitor the growth of frozen waterfalls throughout the winter season, digital cameras set up near the waterfalls took pictures at regular intervals (six photographs per day).

From the successive images, a simple image analysis procedure was used to estimate the area occupied by the frozen waterfalls within the camera’s field of vision. In addition, they enabled the researchers to analyze their growth throughout the season.

Evolution of the free-standing
One of the automatic cameras.
Evolution of the free-standing © Luc MoreauEvolution of the free-standing © Luc MoreauEvolution of the free-standing © Luc Moreau
28th November 2007 - 17 December 2007 - 1st January 2008

A sudden end: mechanical stability and instability of the waterfall ice

Is there a link between waterfall ice color and strength? How does this ice react to climber’s progression and temperature variations?What are the best and worst climate conditions?

If the birth of a waterfall is progressive, its death, especially for vertical structures such as stalactites or free-standing, is often sudden, which raises obvious safety problems!

tests d'impacts sur la glace © François Damilano
Ignoring their vertigo, the researchers were suspended at height, to extract samples or to make impact tests on the ice.

embedded sensor at the heart of the ice © François Damilano
An embedded sensor, of the size of an ice screw, was placed at the heart of the ice, at the bottom of the Nuit Blanche free-standing. It measured internal pressure data and recorded the temperature within the ice. The data allowed the mechanical stresses supported by the structure during the season to be analyzed.

How do waterfall collapse? What are the triggering mechanisms?

The initiation and propagation of fractures are the result of mechanical stress which pulls the ice to its breaking point. Several mechanisms generate such stress in the waterfalls, but what are the causes of collapse?

  • The weight of the structure: 
    the stresses generated are too weak to reach the breaking point of the ice, and the weight of a climber will be even more negligible.
  • The freeze-thaw cycles: 
    high stress may occur in specific areas, but it does not seem to trigger general failure.
  • The mechanism of contraction / expansion due to temperature variations within the ice, however, is the most important and  most dangerous factor!
    The ice expands when it warms, and contracts when it cools. A free-standing column, under the effect of a substantial cooling, will tend to become shorter, which is impossible because of the attachment points at the top and bottom.
    The resulting vertical stress of the entire structure may be sufficient to trigger the sudden collapse of the waterfall!

Such "sudden death" scenario was observed on Shiva Lingam in 2008: after a sharp temperature drop of about fifteen degrees °C in the night, the next morning the free-standing structure had disappeared!

The end of the Shiva lingam waterfall stalactite:

Photographic monitoring highlights the initiation and propagation of horizontal fractures which are often very rapid at the upper anchorage point of the structure. In the case of structures protruding from the rock (stalactite and free-standing), this often leads to their collapse, as observed at Shiva Lingam in 2008 and 2009:

The end of the Shiva lingam waterfall stalactite - 1 © Luc Moreau The end of the Shiva lingam waterfall stalactite - 2 © Luc Moreau The end of the Shiva lingam waterfall stalactite - 3 © Luc Moreau The end of the Shiva lingam waterfall stalactite - 4 © Luc Moreau

Hot and cold waterfall ice

From 2006 to 2010, initial research carried out on the waterfalls of Argentière led to the first scientific publications. A new phase of research started in 2011 on "hot" waterfall ice in Samoens (France) and "cold" waterfall ice in Fairbanks (Alaska).

The objective was to confirm the initial research findings on other types of waterfall ice, termed "hot" or "cold" by the scientists. In February 2012, the Nant d’Ant waterfall in Samoëns (a "hot" ice fall) was chosen for the installation of various monitoring sensors. As the winter season started late, this season was short. However, we saw alternate periods of very cold and very mild temperatures, which are of great interest in scientific analysis.

Carottage dans la cascade Dragonfly, en Alaska © Maurine Montagnat During winter 2012-2013, a partnership was set up between the Laboratory of Glaciology of Grenoble and the University of Fairbanks in Alaska. In February, researchers from the Laboratory of Glaciology, Maurine Montagnat and Jerome Weiss, went to Alaska to meet Erin Pettit and her team from the University of Fairbanks where they organized a seminar to present the "Ice Crystal" project to the research group.

The French researchers used a mini-core drill to extract four columns of ice from the waterfall core, as well as inserting pressure and temperature sensors. In addition to these instruments which had already been used in France, an infrared camera was used to measure different temperature zones in the waterfall ice of Alaska. Erin Petit also used radar to measure ice density.

Find out more

Updated in september 2014



Since 2012, the Petzl Foundation has funded research into mountain sport accidents. Our long-term aim is to improve prevention. (...)

Find out more