DATA PRODUCTION

Climate impact indicators
Global essential climate variables/indicators used

Precipitation and temperature from the HydroGFD2.0 dataset.
Global climate model (GCM) data (temperature and precipitation) from CMIP5 , for RCPs, 4.5 and 8.5, some available in the CDS catalogue.

Regional/local indicators used and produced

Streamflow and runoff

Global data to regional/local scale

Step 1: Bias correction of GCM data. CMIP5 GCMs data (precipitation and temperature) were bias-corrected using GFDv2 and the QQ mapping approach. Some models bias-corrected are available in the CDS catalogue.

Step 2: Downscaling of HydroGFD forcing. HydroGFD2.0 forcing data were downscaled using inverse distance weighting and IDW methodology to the resolution of the hydrologic model (i.e., HYPE, mean sub-basin size ~650 km2).Precipitation amounts were compared to those derived for HYPE from the original SMHI-derived Arctic-HYPE forcing, and other relevant Canadian datasets in the Nelson River basin (western Hudson Bay), where there is considerable disagreement among precipitation datasets.

Step 3: Calibration of HYPE hydrologic model. The hydrologic model was calibrated using HydroGFD2.0 forcing and streamflow and runoff were simulated for the entire historic period (1961-2015, model spin up from 1961-1979). Streamflow data (for calibration) from Dery et al. (2016) were observed, which presents a gap-filled, continuous dataset within the region of interest.

Step 4: Simulate streamflow and runoff. Streamflow and runoff were simulated using two future time periods (data from step1): 2030s (2021-2040) and 2050s (2041-2070). Changes in streamflow and runoff were evaluated for each time period, and for the entire period (2021-2070) relative to the simulated historic climate normal time period (1981-2010). Specific gauge locations of interest include major gauged outlets to Hudson Bay (Nelson and Churchill Rivers), and 16 control (regulation) points within the Nelson River basin.

Step 5: Compare HYPE simulations. Specific simulation metrics were compared to HYPE simulations driven by original SMHI-derived Arctic-HYPE forcing during both the historical and future projections to assess the relative impact, precipitation forcing data have on projections of future streamflow and runoff.


University of Manitoba


Global Water Futures

 

Poster displayed at the Kick–Off meeting, 7/8 September 2017, Norrköping, Sweden