Streamflow estimates form the foundation of water management and inform myriad management activities. From water resource allocation to flood planning and management, data collected at stream gauge stations drive research conclusions and management decisions for watersheds around the world.
Because measuring flow directly is a complex and time-consuming process, managers and researchers rely on stage-discharge rating curves to estimate flow on the basis of water levels. Uncertainty haunts these estimates, however: Measurement error, model imperfection, and physical and biotic factors all impede accurate streamflow estimates. Combined, these obscuring agents can result in uncertainty of 50% or higher in some circumstances; in other words, an estimated flow of 100 cubic meters per second (m3/s) could actually range from 50 to 150 m3/s.
With so many entities relying on accurate streamflow estimates, errors of that size can result in significant economic losses, and it is necessary to calculate uncertainty when estimating flow. Multiple methods exist to derive uncertainty, but no single approach is recognized as the best option.
To better understand these different approaches and the assumptions that underpin them, Kiang et al. conducted the first experimental review of streamflow uncertainty estimation methods. The study recruited research teams from the United States and Europe to evaluate seven uncertainty estimation methods. Each method was applied to data from three stream gauges spanning a range of conditions. The data were selected from waters in France, New Zealand, and the United Kingdom that had sufficient minimum information to conduct the review.
The study found that uncertainty estimates of streamflow vary widely—up to a factor of 4. The variation between methods was especially pronounced for stream reaches with low flows. Streamflow estimates that relied on extrapolation to estimate flow along the stage-discharge rating curve also proved problematic and resulted in wider and more variable uncertainty estimates.
On the basis of the results of the study, the authors recommended that discharge uncertainty estimates be assessed wherever possible. The choice of method is particularly important when dealing with low-flow streams or streams with minimal flow information. The decision should be driven by the user, the application requirements, and the stream conditions.
To aid in decision-making, the authors included a flowchart to help determine the appropriate uncertainty estimation method for the circumstances and make the techniques more accessible for both managers and researchers using streamflow data. (Water Resources Research, https://doi.org/10.1029/2018WR022708, 2018)
—Aaron Sidder, Freelance Writer
The post Exploring Uncertainty in Streamflow Estimates appeared first on Eos.
from Eos https://eos.org/research-spotlights/exploring-uncertainty-in-streamflow-estimates?utm_source=rss&utm_medium=rss&utm_content=exploring-uncertainty-in-streamflow-estimates
via IFTTT