|
Fire Research » Fire & Climate Change » National & International Activities
Development of Daily Weather and Fire Danger Scenarios Using Two General
Circulation Models
Daily data for North America were obtained from two general circulation
models (GCMs), namely the Canadian Climate Centre (CCC) and Hadley Centre
models. CCC data consisted of modeled surface temperatures, relative humidities
(RH), wind speeds, and precipitation variables from the Canadian Global
Circulation Model (CGCM1) for the years 1975–1995, 2020–2040,
and 2080–2100. Similar weather variables for the periods 1975–1990
and 2080–2099 were obtained for the Hadley 3 (Had3GGa1) model. The
period 1975–1995 roughly corresponds to the 1 x CO2 scenario
and the period 2080–2100 to the 3 x CO2 scenario.
The grid cell size for the CGCM1 is 3.75° x 3.75°, whereas the
grid for the Hadley models has slightly better resolution, at 3.75°
x 2.5°. Within the Hadley model, the wind is modeled on a different
grid than the other variables. All variables from the Hadley model were
converted to a common grid structure for the analysis.
Model Calibrations
The modeled variables were compared with actual, observed data during
the fire season for the 1975–1995 time period, and adjustments were
made to RH and precipitation. These adjustments were based on rainfall
and humidity frequency distributions. Several different correction methods
for RH and rainfall were tested on both models. The rainfall correction
was a subtraction of 2.0 mm daily for the CCC data and 1.5 mm daily for
the Hadley data. For the CGCM1, the RH correction involved first correcting
the modeled specific humidity (SQ) values on the basis of observed values.
Then RH was calculated from the adjusted SQ and the saturation mixing
ratio. Had3GGa1 RH was calculated from the noon vapor pressure and the
saturation vapor pressure. The final correction factors were then applied
to the future scenarios.
The following maps illustrate this process for RH with CCC model data,
with carbon dioxide (CO2) levels unchanged from the present.
|
|
|
|
CCC mean RH, 1 x CO2 (unadjusted)
|
CCC mean RH, 1 x CO2 (adjusted)
|
|
|
Temperature Difference |
Daily temperature difference
(°C) |
Temperature differences between the 3 x CO2 and 1 x
CO2 scenarios were calculated daily for each grid cell.
Only May through August data were used in the analysis. Both models
showed an increase in temperature over the next 100 years.
|
|
CCC temperature difference
(3x - 1x)
|
Had3GGa1 temperature difference
(3x - 1x)
|
|
|
Mean Seasonal Rainfall |
Rainfall (mm) |
Mean seasonal rainfall (May to August) was calculated for each
model and time period using the adjusted daily rainfall amounts.
Both models had most of the precipitation falling in the eastern
portion of the continent. The future scenarios appeared to be drier
in the central United States but became wetter further north.
|
|
0–100 |
|
101–200 |
|
201–300 |
|
301–400 |
|
401–500 |
|
501–750 |
|
75–11000 |
|
>1000 |
|
CCC mean seasonal rainfall,
1 x CO2
|
CCC mean seasonal rainfall,
3 x CO2
|
|
|
Had3GGa1 mean seasonal rainfall,
1 x CO2
|
Had3GGa1 mean seasonal rainfall,
3 x CO2
|
|
|
FWI Calculations
The daily weather output from the GCMs was used to create noon fire weather
values for use in the Canadian Forest Fire Weather Index (FWI)
System. Output from the FWI System gives a measure of possible fire danger
across the country. The results for each model and time period were compared
to identify changes in future fire danger and also differences between
the models.
The Seasonal Severity Rating (SSR) is a seasonal (May to August) average
of the Daily Severity Rating (DSR), which is a weighted version of FWI
that gives a measure of the daily fire control difficulty. SSR ratios
of 3 x CO2 to 1 x CO2 were calculated for the fire
season for both models. This is a good overall measure of fire danger
over the next century. In most regions of North America the SSR appears
to be increasing greatly.This increase may correspond to an increase in
fire activity, which will affect forest biodiversity, productivity, and
carbon dynamics. Fire management agencies may need to adjust their strategies
to deal with a changing fire regime.
|
Had3GGa1 SSR ratio (3x/1x)
|
|
|
|
|
0–100
|
|
101–200
|
|
201–300
|
|
301–400
|
|
401–500
|
|
501–750
|
|
751–1000
|
|
>1000
|
|
Acknowledgments
We thank Walter Skinner (Climatologist, Meteorological Service of Canada)
for his contributions of weather station data, David Viner (Climate Change
LINKS Project) for providing the daily Hadley data sets, the Canadian
Climate Centre Modeling Group for allowing access to their daily GCM data
sets, and the Climate Change Action Fund (CCAF) for providing funding
in support of this research.
For more information, contact:
K.A. (Kim) Logan |
|
Canadian Forest Service
Great Lakes Forestry Centre
1219 Queen Street East
Sault Ste. Marie, ON P6A 2E5 |
Phone: (705) 541-5760
Fax: (705) 541-5701
e-mail: Kim.Logan@NRCan.gc.ca
|
B.M. (Mike) Wotton |
|
Canadian Forest Service
Great Lakes Forestry Centre
1219 Queen Street East
Sault Ste. Marie, ON P6A 2E5 |
Phone: (705) 541-5700
Fax: (705) 541-5701
e-mail: Mike.Wotton@NRCan.gc.ca |
M.D. (Mike) Flannigan |
|
Canadian Forest Service
Great Lakes Forestry Centre
1219 Queen Street East
Sault Ste. Marie, ON P6A 2E5 |
Phone: (705) 541-5541
Fax: (705) 541-5701
e-mail: Mike.Flannigan@nrcan.gc.ca |
|
|
|
|
|