The Hydrology team from the YukonU Research Centre – Climate Change Research Group – is leading a project that will improve understanding and the ability to forecast the timing and intensity of river ice breakup in communities of the Yukon where the risk of ice jam floods is significant. This project is supported by Environment and Climate Change Canada and the Government of Yukon and represents a collaboration between YukonU and the University of Alberta.
We were in the Tr’ondëk (Klondike River) valley from April 25 to May 2 to gather information about the river ice breakup. This blog post emphasizes the breakup of the Chu kon’ dëk (Yukon River in Hän) near Dawson.
How can we describe the 2025 breakup of the Chu kon’ dëk?
In a few technical words, it was a thermal breakup with a rather dynamic ending.
This means that the ice cover melted over an extended period (starting on April 1, which is earlier than usual) while the flow of the Yukon River (including the contribution from the Tr’ondëk) barely rose. This scenario was imposed by alternating cool and warm conditions, overcast skies, and occasional snowfalls during most of April.
The famous tripod moved on the morning of April 30 because of an ice run and associated wave coming from upstream, possibly as far as the Stewart River. It carried enough power to push, break, and mobilize the solid ice sheet in front of Dawson. An ice jam formed some 65 km downstream of Dawson later that day, and it stayed in place for several days before essentially melting in place. This breakup sequence indicates that:
- The flow generally remained below average (despite the occurrence of a significant wave associated with the breakup ice run), partly because of the low snowpack in Chu kon’ dëk tributaries.
- The ice cover was relatively weak upstream of Dawson but remained fairly strong several 10s of kilometres downstream of the town.
Can we forecast the timing of river ice breakup?
Yes, to some degree. Our team has already developed models and reports for this (link), but the model only becomes accurate a week or so ahead of local breakup (and our team does not participate in the also-famous annual breakup lottery). The following figure shows how breakup forces evolved at Dawson prior to April 30. For breakup to occur, the driving force (in blue), controlled by the rising flow and by ice runs, needs to rise above the resisting force (in grey), controlled by the ice thickness and structural strength (which declines under warm and sunny conditions). For readers who wonder, the forces are expressed in Mega Newtons (MN), and 1 MN equals 1,000,000 kg (or 1,000 tons) multiplied by the gravity.
The model struggled to simulate the initial, thermal phase of breakup and needed to be recalibrated. At some point, it appeared inevitable that local breakup would happen between April 29 and May 2. What caused the driving force to rise above the resisting force was ultimately the ice run that came from upstream. The intensity of breakup was accurately simulated: the yellow zone indicates a normal breakup peak water level.
Do we know the flow associated with the mobilization of the ice cover?
The river flow (or discharge) in the presence of an ice cover is generally measured indirectly through under-ice water depth and velocity measurements. For instance, the Water Survey of Canada (WSC) calculated a flow of about 600 m3/s at the end of March in the Chu kon’ dëk at station 09EB001 in Dawson.
Estimating the flow becomes extremely difficult when ice conditions are unstable or changing rapidly. The Chu kon’ dëk hydrograph (flow over time) was reconstructed for April 29-30 to the best of our knowledge following a technique explained here. The top graph below shows the water level (corrected to sea level elevation) obtained from the WSC station 09EB001 whereas the bottom graph presents the estimated flow and ice-induced backwater.
This analysis suggests that the flow was essentially multiplied by 3 when the ice run came to Dawson. This additional flow included:
- Ice from upstream river segments (and from the Stewart River) that was set in motion (becoming part of the flow).
- Water that had been slowed down by the presence of (or stored under) stationary ice at the very beginning of winter.
Is the ice bridge causing some form of resistance to breakup?
In theory, a thicker ice cover offers more resistance at breakup and can cause ice jams. In Dawson, this mainly applies if the ice run coming from upstream attempts to “plow” its way through the ice bridge. In most years, the ice sheet in front of Dawson, on which the ice bridge is located, is gradually “lifted” by the rising flow and “carried” downstream with only gradual fragmentation.
In 2025, our research team, with colleague Stephanie Saal as the certified drone pilot, in collaboration with Yukon Government hydrologists, could document a typical “plowing type” of breakup process at the Dawson ice bridge. The video footage prepared by Stephanie (including accelerated segments) confirms that the ice bridge, with its artificially thickened cover, did offer more resistance than the ice cover immediately upstream and downstream.
Does this mean that Dawson could be threatened or flooded by the presence of the ice bridge through the formation of an ice jam in future breakups? This scenario is possible but unlikely. Indeed, the process observed in 2025 was due to an average ice run taking place at relatively low flow, and the water level remained 5 metres below the dike crest.
Since it is not impossible that the intact ice bridge, or thick ice bridge fragments, could be involved in a significant ice jam at or below Dawson, drilling holes in the bridge after its closure every spring would be recommended. The presence of holes, whose diameter would increase over time through thermal erosion, would reduce the resistance of the ice sheet and prevent the plowing process shown in the above video.
Can a laser be used to measure ice surface or water level variations during breakup?
This year, under the leadership of colleague Tyler de Jong and with the support of Environment and Climate Change Canada, our team has begun testing a distance laser water-level sensor at Dawson. This approach is interesting because a laser represents a no-contact (or non-invasive) means of measuring water level fluctuations. The laser was installed downstream of Dawson prior to breakup.
The following graph compares the results derived from the laser (measuring a direct distance from which a vertical distance is calculated using a measured angle) and the water level directly measured by Water Survey of Canada (WSC) equipment. The approach seems promising and could represent the only reliable continuous water level data source if the primary WSC instrument failed (which has occurred in the past). Given that the WSC station in Dawson is critical to foreseeing and forecasting breakup, testing a complementary instrument is probably justified.
The laser signal was high prior to breakup, and it remained high after breakup compared with the expected signal. This is probably because the laser beam was pointing at snow-covered ice (either grounded or floating) before breakup and at stranded ice blocks (called “shear wall”) left along the riverbank after breakup. Therefore, a setup adjustment will be required in future tests. The 0.3 m-spike that occurred soon after noon on April 30 could be the laser beam hitting a high elevation floating ice block. We could also see that precipitation (rain in this case) does interfere with the laser readings.
Are we done with breakup in the Yukon?
No. At the time of this blog post, the Ch’oodeenjìk (Porcupine River) in Old Crow does not feel spring conditions yet.
Our research team is currently testing a breakup model similar to what is presented above but adapted to the different context of the Ch’oodeenjìk.
A freeze-up jam at the beginning of winter combined with recent snowfalls and cool conditions in recent weeks could lead to an intense breakup if air temperatures were to warm up quickly by mid-May.