Repository logo
HARVEST

Centre for Hydrology Reports

Permanent URI for this collection

https://hdl.handle.net/10388/15190

Browse

Browsing Centre for Hydrology Reports by Author "Pomeroy, John W."

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Improving and Testing the Prairie Hydrological Model at Smith Creek Research Basin
    (Centre for Hydrology, University of Saskatchewan, Saskatoon, Saskatchewan, 2014) Pomeroy, John W.; Shook, Kevin; Fang, Xing; Dumanski, Stacey; Westbrook, Cherie; Brown, Tom
    The 2010 Prairie Hydrological Model configuration of the Cold Regions Hydrological Model was developed to include improved snowmelt and evaporation physics and a hysteretic relationship between wetland storage and runoff contributing area. The revised model was used to simulate the snow regimes on and the streamflow runoff from the five sub-basins and main basin of Smith Creek, Saskatchewan for six years (2007-2013) with good performance when compared to field observations. Smith Creek measured streamflows over this period included the highest annual flow volume on record (2011) and high flows from heavy summer rains in 2012. Smith Creek basin has undergone substantial drainage from 1958 when it contained 96 km2 of wetlands covering 24% of the basin area to the existing (2008 measurement) 43 km2 covering 11% of the basin. The Prairie Hydrological Model was run over the 2007-2013 period for various wetland extent scenarios that included the 1958 historical maximum, measured extents in 2000 and 2008, a minimum extent that excluded drainage of conservation lands and an extreme minimum extent involving complete drainage of all wetlands in Smith Creek basin. Overall, Smith Creek total flow volumes over six years increase 55% due to drainage of wetlands from the current (2008) state, and decrease 26% with restoration to the 1958 state. This sensitivity in flow volume to wetland change is crucially important for the water balance of downstream water bodies such as Lake Winnipeg. Whilst the greatest proportional impacts on the peak daily flows are for dry years, substantial impacts on the peak daily discharge of record (2011) from wetland drainage (+78%) or restoration (-32%) are notable and important for infrastructure in and downstream of Smith Creek. For the flood of record (2011), the annual flow volume and the peak daily discharge are estimated to increase from 57,317 to 81,227 dam3 and from 19.5 to 27.5 m3 /s, respectively, due to wetland drainage that has already occurred in Smith Creek. Although Smith Creek is already heavily drained and its streamflows have been impacted, the annual flow volumes and peak daily discharge for the flood of record can still be strongly increased by complete drainage from the 2008 wetland state, rising to 103,669 dam3 and 49 m3 /s respectively. This model simulation exercise shows that wetland drainage can increase annual and peak daily flows substantially, and that notable increases to estimates of the annual volume and peak daily flow of the flood of record have derived from wetland drainage and will proceed with further wetland drainage.
  • Thumbnail Image
    Item
    Yukon River Basin Streamflow Forecasting System
    (Centre for Hydrology, University Saskatchewan, Saskatoon, Saskatchewan, 2020) Elshamy, Mohamed; Loukili, Youssef; Princz, Daniel; Richard, Dominique; Tesemma, Zelalem; Pomeroy, John W.
    The Yukon River Basin is one of the main rivers 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, just across the border in Alaska. Small parts of this catchment are in Alaska. This basin has an area of 288,000 km2, from 58.8 – 65.6°N and 129.2 – 134.1°W. The southern part of the basin is characterized by large glaciers 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, 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 project developed, set up, calibrated, validated, and operationalized a streamflow discharge forecasting system for the Yukon River and several of its tributary rivers within the Yukon Territory. The Yukon River Basin streamflow forecasting system is based around the MESH (Modélisation Environmentale Communautaire - Surface and Hydrology) hydrological land surface model. MESH is a state-of-the-art semi-distributed cold regions hydrological land surface model that models both the vertical exchanges of heat and moisture between the land surface and the atmosphere as well has the horizontal transfer of water to streams that is routed hydrologically to the outlet of the basin. It includes snow, frozen soil and glacier processes as well as the full suite of warm season hydrology. MESH is driven by the Environment and Climate Change Canada GEM weather model and hindcasts are driven by GEM-CaPA which is a data assimilation product that uses local precipitation observations where they exist. The rivers forecasted includes the Yukon River Basin upstream of Eagle, AK and the Porcupine River Basin near the international boundary. MESH provides supplemental high resolution simulations and forecasts for the Klondike, Stewart, Pelly and White Rivers at their mouths. Daily river discharge and water balance forecasts are produced by the system for each river basin. Having MESH run at both 10 km and 5 km resolution provides an assessment of model resolution needed for forecasting and also of model uncertainty in the forecasts. The MESH model was driven by GEM-CaPA for hindcasts and with the GEM ECCC Regional and Global Deterministic Prediction Systems - RDPS and GDPS forecasts for forecasts of 2 and 9 days. The GEM-MESH model showed good to very good predictions in most river basins after calibration and parameter selection, with challenges for the Porcupine and White rivers due to permafrost and wetlands (Porcupine) and to extensive icefields (White) and overall to sparse to non-existent observed precipitation data to assimilate into the CaPA system. The forecast system is capable of providing reliable streamflow predictions and is run with automated scripts on Amazon Web Services. Future development of the forecasting system should focus on the very challenging permafrost hydrology of the Porcupine River Basin, and the glacier hydrology of the White River which drains the largest icefields in North America. The model does not include a river ice component, but one could be added in the future.