Scientists are often tempted to classify their observations and conclusions into defined, ideally exclusive (no overlapping) categories. This structured approach of summarizing sciences facilitates knowledge transfer, supports project planning, and guides future research. For instance, over decades of observations and analyses, biologists have defined families of species, characterized their habitats, and identified their ecological roles. This organized knowledge allows us to anticipate the global impact of climate change or the effect of invasive species on ecosystems and enables proactive exploration of adaptation avenues. Geoscientists have developed similar classification models to understand physical (landscape) processes. Figure 1 presents a classification model for Canadian creeks and rivers.

Figure 1. Diagram showing the structure of the proposed watercourse classification model

This diagram divides watercourse segments in different categories based on three parameters:

  • Climate (C: Temperate, Cold, or Arctic): In Canada, streams in these different climates will present a distinct hydrological regime. For instance, in temperate climates (e.g., with mild winters summing less than 1000°C-days of freezing), maximum annual flows may occur during any season. In contrast, minimum flows are likely to occur during the summer months. In the Arctic (e.g., long winters with more than 3000°C-days of freezing), annual minimum flows occur at the end of winter, whereas maximum flows usually coincide with the snowmelt period. Climate also affects the ice period’s duration and the ice cover’s maximum thickness.
  • Size (S: Small or Large): This parameter could be based on the stream channel width, the average annual flow, the watershed size, or other comparable parameters. Turcotte et al. (2017a) proposed that small streams drain less than 2000 km2, present a maximum width of 100 m, or are affected by flooding when carrying a flow below 1000m3/s. Users may consider different thresholds based on their needs and objectives. Here in Yukon, flooding in a watershed of less than 2000 km2 occurs at a much lower discharge, generally 200 m3/s.
  • Gradient (G: Low gradient, Hybrid, and Steep): There is usually a significant morphological difference between a 5 m elevation loss or more per 1000 m of channel length (steep slope of 0.5%) and an elevation loss of less than 0.5 m per 1000 m of channel (slope of 0.05%). Steep channels present morphologies such as rapids, steps, and cascades, and their bed is usually made of coarse gravel or bedrock. In low-gradient environments, channels are often meandering, and their bed is mainly composed of small gravel and sand. In between, hybrid channels are spatially wandering, and their profile comprises relatively deep pools and riffles (short and shallow fast flowing segments). These parameters significantly impact flow conditions (i.e., velocity and turbulence), water quality, sediment transport, habitat characteristics, winter ice processes, and infrastructure design (e.g., bridges or erosion protections).

From this model, considering C, S, and G, but also two special categories (warm streams (W) and braided streams (B)) a total of 27 categories are defined. For instance, one of these categories would be C3S1G1, for a low gradient (G1), small (S1), and Arctic (C3) stream.

The following figure presents three distinct examples of Yukon watercourse segments.

Liard River near Watson Lake (C2S2G1)

Figure 2. Liard River near Upper Liard, Yukon.

This river segment flows in a cold or subarctic environment (C2, about 2700°C-days of freezing every winter), it is large (S2, draining 32,600km2 with a flood threshold in the order of 3200 m3/s), and it presents a low-gradient (G1). The combination of C2S2G1 implies several hydrological conditions and characteristics:

  • Its thalweg (main flow path) is a few meters deep most of the year, so it can be navigated with most small vessels.
  • In turn, it cannot be wadded, and its spring and summer flow can only be measured using a boat (remote-controlled or not).
  • It forms a floating (often smooth) ice cover during winter, with a thickness of up to 1 m.
  • It carries water throughout the year, including at the end of long winters.
  • It exhibits a relatively slow and predictable hydrological response to snowmelt and rain events. Its flow is not significantly influenced by isolated thunderstorms.
  • This type of river segment may be affected by dynamic ice breakup events (the single pillar of the Alaska Highway bridge presented in Figure 2 is designed accordingly).
  • It transports sediment year-round, either in suspension mode (with turbidity that is generally proportional to the discharge) or through the movement of dunes.

East Blackstone River in the Tombstone Territorial Park (C3S1G2)

Figure 3. East Blackstone River in the Tombstone Territorial Park.

This channel segment of the East Blackstone River is small (S1), it is located in an Arctic climate (C3), and its gradient is hybrid (G2). These parameters mean that:

  • Its depth rarely exceeds 1 m and can only be paddled at high flow.
  • Its discharge can be measured by wadding across the channel or using a small remote-controlled vessel equipped with a device that measures the water depth and velocity (e.g., ADCP).
  • It forms a partially grounded ice cover with possible overflow events (see previous blog post about pressurized conditions).
  • It may be affected by short-lived, zero-flow events during winter months, especially during early or mid-winter cold spells. This condition means that aquatic instruments may freeze and be damaged by static ice between October and April.
  • Its ice cover melts in place, and very limited ice pieces are carried by the increasing spring flow.
  • It may react rapidly to significant snowmelt rates and to isolated summer storms. However, its gradient (and, in this case, the presence of wetlands in the valley) tempers its hydrological answer.
  • It is only turbid (carrying suspended sediment) during high-flow events, mainly in June. Otherwise, its water is clear, and most bedload consists of the movement of small gravel*.

Note that this river includes several fairly shallow braided (B) segments (a morphological platypus, to use a biology metaphor) and a few warm (W) groundwater-fed segments where an ice cover only forms below -30°C. This spatially heterogeneous hydraulic and thermal regime implies that aquatic habitats vary significantly over short distances.

* As of July 2024, it seems that the East Blackstone River will be added to the long list of subarctic and Arctic streams and rivers that are now orange, potentially as a result of permafrost thaw in a metal-rich landscape. This new state may also be temporary, and more research is needed about this water quality transition.

Klondike River (Tr’ondëk) near Henderson Corner (C3S2G3)

Figure 4: Klondike River near Henderson Corner.

This segment of the river is composed of rapids (G3), flows in a climate that is at the margin of Arctic characteristics (C3) and can be considered a large watercourse (S2). This fluvial environment is among the most dynamic and complex:

  • Its depth is locally limited at low flow. However, its velocity and turbulence are substantial at high flow, which means that its discharge is difficult to measure under most hydrological conditions.
  • It can be navigated with a small motorboat, but its channel may be too shallow at low flow and too turbid at high flow (it is challenging to distinguish submerged obstacles that may damage vessels).
  • Freeze-up and breakup can be extremely dynamic, potentially damaging riparian vegetation, causing important ice jams, and generating unstable water levels, including floods.
  • During winter, the ice cover can be free-spanning or partial (with open water areas), which translates into hazardous travel on a snowmobile, on foot, or on skis.
  • For a river of this size, it reacts relatively rapidly (within 24 hours) to high snowmelt rates and rainstorms.
  • Aquatic instruments are susceptible to failure or loss in any season unless properly installed and protected.

***

A comparable list of hydrological conditions and characteristics could be developed for each of the 27 watercourse categories. Although the proposed classification is meant to address the needs of the private, public, and academic sectors, its simplicity represents a limit to its comprehensiveness; it can still be improved and adapted to additional regions. For instance, it currently excludes some bedrock-dominated streams of the Canadian Shield and the hydrological regime of the Canadian Pacific coast, and it does not emphasize glacier-fed rivers. Other classification models that consider watercourses from distinct angles can be found in the literature on hydrology or freshwater ecology.

Figure 1 is particularly useful as a practitioner’s guide for initial assessments for projects (e.g., pre-design, planning) along cold region watercourses, characterized by a unique hydrological regime, influenced by permafrost, and affected by ice processes during a large portion of the year. Winter processes are frequently overlooked in hydrology textbooks (e.g., Singh, 2017) and are often not adequately considered in the design of hydraulic structures in the North. This neglect represents a risk to infrastructure integrity, a potential environmental problem, and a poor taxpayer’s investment. A good example is the construction and maintenance of highways in subarctic and Arctic regions. In the early days, engineers underestimated the challenge associated with stream crossings just like they did with permafrost, and they are currently catching up using knowledge produced by researchers (e.g., Ensom et al., 2024).

Years of collaboration on various projects across different jurisdictions in Canada suggest that engineer and scientist consultants bidding on large government projects (i.e. infrastructure) involving watercourses but who are unfamiliar with cold region hydrological processes would benefit from using this classification to write their proposals (e.g., to establish a methodology, to plan activities, to identify project risks). In situations where neither contractors nor decision-makers are hydrology experts, understanding and using a detailed version of the proposed classification may yield tangible benefits to taxpayers, right holders, and the environment.

Another clear advantage of this classification is that it can support identifying and implementing adapted stream monitoring strategies (e.g., Turcotte et al., 2017b). The needs for stream monitoring are unlimited but usually include:

  • Long-term baseline water quantity or water quality data collection (e.g., hydrological statistics for land use planning and design, documentation of the evolution of water quality in a changing landscape, quantification of the impact of climate change on flows);
  • Short-term, project-specific data collection (e.g., mining site, episodic habitat assessments) or;
  • Research (e.g., hydrology, aquatic biology, limnology).

From a previous project completed by our research group, we are aware that the federal agency responsible for monitoring the quantity of water in streams and rivers year-round occasionally encounters technological and analytical issues. A similar classification model has been proposed to help inform the identification or the development of technologies and data analysis tools adapted to specific watercourse categories to produce continuous and reliable hydrological data at a reasonable cost for Canadians. Our team continues to work with other researchers in Canada to improve the performance of hydrometric stations and the quality of hydrological data.

***

Ensom, T., Kokelj, S.V., Marsh, P., Connon, R., Kamo McHugh, K., van der Sluijs, J., 2024. Hydrothermal and Terrain Effects of a Highway on Streams in Permafrost. In: 12th International Conference on Permafrost. Whitehorse, YT.

Singh, V.P., Ph.D., D.Sc., D. Eng. (Hon.), Ph.D. (Hon.), D. Sc. (Hon.), P.E., P.H., Hon. D. WRE, Academician (GFA) (editor), 2017. Handbook of Applied Hydrology, Second Edition, McGraw-Hill Education ISBN: 9780071835091

Turcotte, B., Alfredsen, K., Beltaos, S., Burrell, B.C. 2017a. Ice-Related Floods and Flood Delineation along Streams and Small Rivers, in CGU HS CRIPE Workshop on the Hydraulics of Ice Covered Rivers. Whitehorse, YT, July 10-12.

Turcotte, B., Nafziger, J., Clark, S., Beltaos, S., Jasek, M., Alfredsen, K., Lind, L., Stander, E.J. 2017b. Monitoring river ice processes: Sharing experience to improve research programs, in CGU HS CRIPE Workshop on the Hydraulics of Ice Covered Rivers. Whitehorse, YT, July 10-12.