Yukon River Basin Streamflow Forecasting System - Vector-Based MESH Model Setup for Yukon River Basin

The Yukon River Basin the second largest river in the Arctic region of North America and is shared between Canada and the US. The Canadian part covers almost half of the Yukon Territory in addition to a small portion of the province of British Columbia, while the US part falls totally within the state of Alaska. This study is concerned with Canadian part of the Yukon River with its outlet at Eagle, Alaska - just downstream of the international boundary (288,000 km2). The southern part of the Yukon River basin is characterized by extensive icefields and snowfields at high elevations (up to 4700 m above sea level) with steep slopes, and thus generates considerable runoff. There are also mountain ranges on the eastern and northern boundaries of the basin, while the western areas are milder in slope and partially forested. Snow redistribution by wind, snowmelt, glacier melt and frozen soil processes in winter and spring along with summertime rainfall-runoff and evapotranspiration processes are thus key to the simulation of streamflow in the basin. This supplement shows the development of a vector-based MESH setup for the Canadian portion of the Yukon River Basin at Eagle. Without additional calibration, the vector-based model performance was compared to the previously generated grid-based MESH model whose development was documented in Centre for Hydrology Report #16. MESH was driven by the Environment and Climate Change Canada Global Multiscale Model (GEM) weather model forecasts with precipitation replaced with the Canadian Precipitation Analysis (CaPA) which assimilates local precipitation observations where they exist, collectively referred to as GEM-CaPA. Additionally, the models were run, without additional calibration using the newly developed Regional Deterministic Reforecast System v2 (RDRS v2) forcing. RDRS v2 forcing is being extended as a hindcast by ECCC to approx. 1980 and so will permit 40 year runs of MESH from which streamflow exceedance return periods can be calculated. Model performance was slightly inferior for the vector-based setup compared to the original grid-based one. This may be due to the full calibration applied to the grid-based model and parameter transfer to the vector-based model without recalibration. Model performance also deteriorated when the RDRS v2 was used as forcing data, as the model was originally calibrated to GEM-CaPA. It is expected that model performance will improve once it is fully calibrated using the RDRS v2 forcing data.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.