| Abstract/Description or Keywords |
According to the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report, it is now “very likely” that observed widespread warming of the atmosphere and oceans are due to historical anthropogenic greenhouse gas emissions, predominantly from fossil fuel use. Further, climate change trends will persist with continued emissions of greenhouse gases, such that we can expect further changes in global, regional and local temperature and precipitation patterns. Continued warming and changing precipitation patterns will have a large effect on the hydrology of western North America, with the possibility for subsequent impacts to various water-related resources and activities, including hydroelectric generation, municipal water supply, flood management, in-stream flow needs and fish habitat, irrigated agriculture, recreation and navigation. Although these water-related issues are germane to British Columbia, sustainable and self-sufficient generation of electricity in British Columbia is a significant concern and a major policy objective. Hydroelectricity is BC’s largest source of electric power generation and much of this hydroelectric power is generated by BC Hydro, the third largest electrical utility in Canada. Approximately 85% of BC Hydro’s generation is produced by hydroelectric means from large heritage assets in the Peace and Columbia River systems which may be susceptible to the hydrologic impacts of climate change. A high-resolution, physically-based macro-scale hydrologic model has been applied to quantify the hydrologic impacts of projected climate change within the Peace, Campbell and Upper Columbia watersheds in British Columbia. The three study watersheds contain numerous important BC Hydro heritage assets for hydroelectric generation and represent a range of hydro-climatic regimes and scales. Streamflow projections were made for several project sites within the study areas, corresponding to current BC Hydro heritage asset sites, potential sites of future hydroelectric development (i.e., Site C), as well as several natural drainages. This study utilized a suite of eight global climate models (GCMs) driven by three emissions scenarios, intended to capture a range of high, medium and low projected greenhouse gas emissions and to project a wide range of potential climate responses for the 2050s time period (2041-2070). Climate projections were statistically downscaled and used to drive the hydrology model at high spatial resolution. This methodology of selecting multiple GCMs coupled to three emissions scenarios covers a large range of potential future climates for BC and explicitly addresses both emissions and GCM uncertainty in the final hydrologic projections. The general conclusions of this work can be summarized as follows: Climate response: _ All projections indicate higher temperatures in all seasons and all study areas by the 2050s, with strong agreement between GCMs and scenarios. The highest temperature increase is projected for the winter season in all three study areas for all three emissions scenarios. _ Precipitation projections are less robust for the 2050s (i.e., the range of individual GCM projections includes both positive and negative changes), but suggest increased precipitation in the winter, spring and fall for all study areas and all emissions scenarios. Increased annual precipitation is projected for the interior study areas (Peace and Upper Columbia) but negligible changes are projected for the coastal Campbell River study area. Regional differences are also apparent for summer precipitation trends, with decreased precipitation projected for the Campbell and Upper Columbia study areas (southern BC) versus negligible changes in summer precipitation projected for the Peace study area (northern BC). Annual discharge: _ Annual discharge is projected to increase in the Peace River study area, a response that is generally consistent between project sites, although local inflow to the Peace River above Pine River shows a weaker response. Annual discharge in the Upper Columbia study area is projected to increase at the majority of project sites for all emissions scenarios. Annual discharge changes for the Campbell River study area are projected to be negligible. _ At all three study areas, increases in annual discharge are attributed to projected changes in annual precipitation for the 2050s. The variation in annual discharge response between study areas is due primarily to regional variation in projected precipitation trends, where increased annual precipitation is projected for the interior study areas (Peace and Upper Columbia) but negligible changes are projected for the coastal Campbell River study area. Monthly discharge: _ Monthly streamflow projections for the Peace River project sites show a consistent response of higher future discharge during fall and winter, an earlier onset of spring freshet, higher peak monthly discharge, and reduced discharge during late summer and early fall. Changes in the timing and duration of the spring freshet result in the largest absolute changes in monthly discharge. Differences in the monthly discharge response between the three emissions scenarios are negligible. Monthly streamflow projections for the Upper Columbia are similar, although between sites, projections are less consistent regarding changes in the month of peak discharge as well as changes in the magnitude of peak monthly discharge. _ Monthly streamflow projections for the Campbell River study area show a strong shift in seasonality due to a transition from a hybrid nival- pluvial regime to an almost exclusively pluvial regime. This transition results in large increases in fall and winter discharge, and decreases in spring, summer, and early fall discharge, resulting in a longer and more severe low flow period, although remnant freshet runoff is still projected to occur in the 2050s. _ Changes in monthly streamflow timing, seasonality and magnitude are largely attributed to projected changes in the dynamics of natural snow storage. These changes include 1) changes in the proportion of winter precipitation received as rainfall versus snowfall, 2) changes in seasonal snow accumulation, and 3) changes in the timing and magnitude of snowmelt. The most prominent regional variation is apparent in the degree to which snow storage dynamics in the three study areas respond to projected climate response. _ The coastal Campbell River site is projected to undergo the most dramatic change, shifting from what is already a transitional hybrid regime to a predominantly pluvial regime. Although the Peace River in northeastern BC shows signs of shifting to a more hybrid regime in the 2050s, it will still retain sufficient snow that the monthly hydrograph will maintain the characteristic signal of a nival regime, albeit with a freshet that will be advanced in time. The Upper Columbia arguably shows the least sensitivity to climate change, although it is still responsive to changes in temperature and precipitation. This is largely attributed to a hypsometry that places much of the study area at high enough elevation to avoid significant changes in snow storage dynamics, despite rising temperatures. In fact, in contrast to the Peace and Campbell, snow storage throughout much of the Upper Columbia reflects winter precipitation trends more so than temperature trends. Glacier mass balance: _ Glacier mass balance in the Upper Columbia study area is projected to vary with elevation, being predominantly negative at elevations below 2400 m and increasingly more positive at elevation greater than 2400 m. Total cumulative mass balance between 1995 and 2070 for the entire study area, based on the ensemble medians for the A1B, A2, and B1 scenarios, is negative for the A1B and A2 scenarios, but slightly positive for the B1 scenario. Differences in overall cumulative mass balance are mainly attributed to differences in projected temperature changes, which become progressively less pronounced for the A1B, A2 and B1 scenarios, respectively. Glacier area is projected to shrink by roughly 50% for all three emissions scenarios. Nevertheless, the projected trends in mass balance and glacier area may be under- and over-estimated, respectively, due to the absence of glacier dynamics in the hydrology model. Global climate model and emissions sensitivity: _ For annual and monthly discharge at the Peace and Upper Columbia project sites, differences between GCM runs combined with inter-annual variability for any given GCM-driven run are larger than streamflow differences between emissions scenarios, suggesting that projections for the 2050s are largely insensitive to the chosen emissions trajectories. _ Only projections for Campbell River at Strathcona Dam indicate some potential sensitivity to differences between the three emissions scenarios for the 2050s period. For this small coastal watershed, the hydro-climatic response to the mid-21st century emissions projected by the A1B scenario may be sufficiently stronger than that from either A2 or B1 scenarios that annual and monthly discharge displays a detectably different response. Only the A1B ensemble indicates statistically significant increases in annual discharge (although only a 4% difference in median historic and future discharge is projected in this case). The monthly streamflow projections from the A1B-prescribed emissions also tend to exhibit larger changes than those derived from either A2 or B1. Nevertheless, as the climate response for this small study area is likely derived from a very limited number of GCM model grid cells, such differences should be interpreted cautiously. Several next steps are recommended for future work, including: _ Incorporating a coupled dynamic glacier response within the hydrologic modelling process. This would incorporate a hydrologic response due to more realistic changes in glacier mass, volume, area and hypsometry as a response to transient climate warming. Coupled modelling would also incorporate important feedbacks based on differential mass balance response by elevation and changes in surface albedo. _ Updating the current hydrologic projections by incorporating new climate projections. Climate projections from the IPCC’s upcoming Fifth Assessment Report, based on the latest GCM technology and new emission projections scenarios, will be available in the very near future. _ Extension of the current work into the investigation of sub-monthly hydrologic and streamflow phenomenon, such as changes in the magnitude and frequency of extreme, or threshold design events. This will require refinement of current downscaling approaches to explicitly capture the daily transient climate response. _ Investigation of climate and hydrologic impacts projected to the end of the 21st century. Although the A1B emissions scenario displays the largest impacts for the mid-21st century, differences between emissions scenarios are small. Scenario differences are anticipated to become substantial by the end of the 21st century, when the A2 scenario will result in the largest impacts. Hydrologic impacts projected for the 2050s based on the chosen emissions scenarios are expected to become even more pronounced by the end of the 21st century. |
