1) Climate change will impact Canada's water resources, but these impacts have not been fully incorporated into water management efforts. 2) To project climate change impacts, water managers need future climate datasets and modeling approaches to represent hydrologic processes. 3) Global climate models are used to estimate future climate but lack local detail, requiring downscaling methods to develop local climate projections. 4) The document presents a case study applying modeling to project climate change hydrologic impacts on a river watershed in Ontario.

1. ClimaTe Change
The headwaters of
the Credit River near
Orangeville, Ontario.
How-To
Strategies
Projecting the impact
of climate change.
By BOB WAlkeR ANd dAvid vAN vlieT
Canada’s water resourCes Michael Garraway is the water budget over several decades, the interactions of
sCientists and enGineers generally coordinator with the Ontario Ministry processes that determine weather and
agree that climate change will have of Natural Resources. According to climate. Numerous modelling centres
an impact on our water; however, we Garraway, the Province recognizes that a around the world have developed GCMs
have not incorporated climate change thorough understanding of the potential that are used for long-term (250-year)
assessments into many of our water impacts of climate change on hydrology simulations to characterize the evolution
management efforts. Will the water levels is important to develop climate change of temperature, precipitation, solar
in the Great lakes decline? Will the rivers adaptation strategies. “To our knowledge, radiation, winds, and other parameters
of the Prairies overflow their banks more the Guide is the first of its kind developed well into the future.
often and will they sustain our needs in and is equally relevant for all Canadian GCMs produce global scale output at
the summer? How long will the glaciers water resource managers, planners and relatively coarse grid point spacings of
of the Rockies continue to support icy practitioners,” he says.
rivers? While the state of climate change As described in the Water resource managers are recognizing
science and modelling technology Guide, water resource
has not advanced sufficiently to make practitioners have two the need to develop adaptation and
singular and reliable predictions, it is broad considerations mitigation strategies to guard against
practical and prudent for our industry when setting out to
to use available resources and tools to project climate change anticipated changes at the local scale.
evaluate the expected range of climate impacts: they need to
change impacts. develop or obtain future climate datasets, 250 to 400 kilometres. Simulations are
Water resource managers are and they need to select a modelling or designed to characterize future climate
recognizing the need to develop assessment approach that suitably on an annual, seasonal, and monthly
adaptation and mitigation strategies to represents the hydrologic processes that basis. At present, there are more than
guard against anticipated changes at may be influenced by a changing climate. 60 future climate simulations available
the local scale. To that end, the Ontario from the collaborative research of the
Ministry of Natural Resources and Global climate models members of the intergovernmental Panel
Ministry of the environment sponsored The primary tool used to estimate on Climate Change (iPCC).
the development of the Guide for future climate are global climate models This large set of simulations is the
Assessment of Hydrologic effects of (GCMs). GCMs are complex physically result of applying more than 20 GCMs
Climate Change in Ontario, produced based three-dimensional models that to four greenhouse gas (GHG) emission
by eBNFlO and AquaResource inc. in represent the earth’s atmosphere, scenarios. The four emission scenarios
2010. oceans, and land surfaces and simulate, have been developed by the iPCC and
18 WATER CANADA july/AuGuST 2011 WAT e R C A N A d A . N e T

2. ClimaTe Change
reflect atmospheric levels of GHGs
under four storylines, each representing Figure 1: Mean annual precipitation and
a different future trend in terms of global temperature changes for 57 climate scenarios
economy, demographics, governance,
at a southern ontario location.
technology, and GHG emission levels.
While the iPCC considers each of these
outcomes equally plausible, the climate
outcomes differ significantly in terms
of annual and monthly temperature
and precipitation change from existing
conditions at all model grid points. This
disparity is illustrated in Figure 1 and is
due to differences in the GCMs and the
emission scenarios.
downscaling simulations
When projecting possible hydrologic
impacts from climate change, water
resource practitioners are faced with
the challenge of developing or obtaining
climate datasets that reflect the temporal
and spatial level of detail needed at the
local scale. GCMs lack the local scale
parameterization and feedback from
locally significant features (i.e., topography
WAT e R C A N A d A . N e T WATER CANADA july/AuGuST 2011 19

3. ClimaTe Change
Figure 2: study area location.
at a climate station in southern Ontario.
This figure displays the significant level
of disparity among GCM models and
emission scenarios as mean annual
temperatures range from +1.7 Celsius to
+4.9 Celsius , while annual precipitation
changes range from -6 per cent to +17 per
cent. The percentile method selects 10
scenarios from this set. Five scenarios are
selected based upon the simulated changes
in temperature and five are selected based
upon changes in precipitation. For both
parameters, scenarios are selected that
best represent the 5th, 25th, 50th, 75th and
95th percentiles. This subset represents the
groups mean, variability and extremes.
an ontario case study
The Guide provides a case study
demonstrating the application of the
percentile method and change field
approach to project the impact of
climate change on the headwaters of the
Credit River (Figure 2), near the Town
of Orangeville, Ontario. The case study
includes the application of a continuous
streamflow generation model (HSP-F) and
groundwater flow model (MOdFlOW) to
and surface water) to reflect local scale of local observations, the local climate. estimate the hydrologic effects of a series
climate directly in model output. The change field method is a simple of future climate scenarios.
Research institutions continue to approach to develop future local climates The water resource quantity issues
develop various methods of downscaling that reflect large scale average features associated with the study area and the
GCM simulations. They include and allows the use of all GCM and GHG headwaters of the Credit River include
statistical downscaling, weather emission scenarios. the following:
generators, and regional climate models.
• Future availability and reliability of
The most established methodology for the percentile method
municipal groundwater supplies;
estimating future local climates uses the The iPCC recommends that water
GCM simulations to estimate annual, resource practitioners use as many • Baseflow/low flow concerns from a
seasonal, or monthly changes for each future climate simulations as possible fisheries and wastewater assimilation
climate variable for a future time period when conducting a climate change perspective; and,
relative to a baseline climate period. impact assessment. However, in most
• High flow and peak flow concerns
These relative changes, called change assessments it is impractical to conduct
from a flood risk and instream erosion
fields, are used to adjust the observed an evaluation with the full set of over
perspective.
60 future climate
When assessing the impacts of simulations. The Guide The hydrology of the study area
describes a method, is relatively complex. in addition to
climate change, understanding the referred to as the natural processes and sources of water
distribution of streamflow throughout percentile method, for in the system there are several human
selecting a smaller subset influences with significant consequences.
a particular year is critical. of climate scenarios These influences include groundwater
from the full list that is pumping for municipal supply,
climate station data time series to reflect statistically representative of the full set. wastewater discharges into the Credit
future conditions. This approach results Figure 1 illustrates a scatter plot of River, storage in a lake with controlled
in an altered input climate time series GCM simulated annual mean change in outflow and the many incremental effects
that reflects the average relative change temperature and precipitation for the of agricultural development, aggregate
in each parameter and, through the use period of 2041 to 2070 for 57 scenarios extraction, and urbanization.
20 WATER CANADA july/AuGuST 2011 WAT e R C A N A d A . N e T

4. ClimaTe Change
Figure 3: Projected range of future mean monthly streamflow.
the year. This seasonal shift is due to
warmer air temperatures which lead to
an earlier spring snowmelt. in addition,
the majority of future climates will
cause winter streamflow to experience
increases of up to 50 per cent. increases
in winter streamflow are primarily
due to winter precipitation occurring
predominantly as rainfall, rather than
snowfall. Several scenarios estimate
that summer low flows will range below
current levels due to changes in summer
precipitation and higher summer
evapotranspiration in the future.
Canadian water resource practitioners
are going to be increasingly relied upon
to make predictions of the impacts of
climate change on our water resources.
The Guide for Assessment of Hydrologic
effects of Climate Change in Ontario
describes an impact assessment
To assess potential climate change basis over a 30-year period. To evaluate methodology that is suitable across all of
impacts to streamflow and the water changes in streamflow, it is necessary Canada. While we cannot make definite
budget, the following tasks were to summarize long term hydrographs predictions of our future climate, the
completed as set forth in the Guide: into statistical metric(s). examples state of climate change and hydrologic
1 Selection of future climate datasets.
of such statistical
The selected climate change datasets metrics are mean
included nine GCM scenarios selected annual and monthly Water resource practitioners are going
using the percentiles method as well flow, median monthly
as the baseline scenario. flow, maximum daily to be increasingly relied upon to make
streamflow, and low
2 development and calibration of a
flows statistics, such as
predictions of the impacts of climate
physically based hydrologic model.
Simulated parameters included
7Q streamflow. change on our water resources.
When assessing the
streamflow and water budget
impacts of climate
parameters (precipitation, runoff,
change, understanding the distribution science is sufficient for our industry to
recharge, and evapotranspiration)
of streamflow throughout a particular estimate the likely range of impacts that
3 development and calibration of a year is critical for a variety of water might be expected across Canada. With
physically based MOdFlOW. The management aspects, including peak this information in hand we can begin to
groundwater flow model utilized and low flow management, and fisheries identify potential impacts and suitable
monthly groundwater recharge rates management. For this reason, changes mitigation and adaptation measures. WC
as simulated by the hydrologic model. to monthly streamflow must also be
4 Simulation of climate change impact investigated.
scenarios. The hydrologic and Average monthly simulated streamflow
hydrogeologic models were used to for the baseline climate and the nine
simulate current conditions (1961 future climates are shown in Figure 3. Bob Walker is president of eBNlFO
All nine future climates have streamflow environmental in Waterloo,
to 1990) and nine separate future
Ontario and is a specialist in water
climate datasets for 2041 to 2070. distributions that follow a similar trend
resources modelling and climate
wherein flows peak in early spring, change impacts.
5 evaluation of results.
decline to annual lows through the
The model simulations provide summer and rise through the fall and david van vliet is vice
president of AquaResource
predictions of runoff, evapotranspiration, early winter. However, the timing of
inc., a specialized water
streamflow, groundwater discharge, the decline in flow from the spring to resources consulting firm
and water levels on an hourly or daily the summer is shifted to earlier in located in Waterloo, Ontario.
22 WATER CANADA july/AuGuST 2011 WAT e R C A N A d A . N e T