Streamflow measurement consists of measuring water velocity and water depth across a river channel. In a small creek, as few as 10 velocities might be measured, while more than 100 may be required in a very large river. These depth and velocity measurements are then used to calculate the total volume of water that flows through the river every second. Site visits to directly measure streamflow are usually carried out every 1 to 2 months.
For each site visit, one calculated flow rate (discharge) and one water level (stage) are measured. After several site visits it is possible to generate a general relationship between river stage and river discharge. The resulting “stage-discharge curve”, or “rating curve”, is unique to a particular river at a particular location. Because rivers transport sand and gravel downstream, over time the rating curve can slowly change. Occasionally, large flooding events can completely reshape a river channel. Although hydrometric stations are generally placed at locations where the channel is expected to remain fairly stable, rating curves can still change over time. In many river reaches the river bed may be so mobile that it is not possible to establish a reliable rating curve relationship. The key “Athabasca River Below Ft. McMurray” station is located in a stable, rocky area.
In practice, a hydrometric station will continuously measure and record only the river’s water level. The flow rate is then estimated from the established rating curve. Because many hydrometric stations transmit their recorded water levels several times a day, it is possible to generate “near-real-time” discharge estimates. Official daily discharge rates are published by the Water Survey of Canada once a year, after several months of careful review.
During the winter months, most rivers in Alberta become covered in ice. The presence of ice changes the rating curve of a river. How much a rating curve changes depends on the thickness and roughness of the ice cover. Because ice conditions change from day to day, and year to year it is usually not possible to develop a single winter rating curve for a river. Winter discharge estimates are therefore inherently less accurate than “open water” discharge estimates. Accurate near-real-time winter flow estimates of river discharge remains an active area of academic research.