Climate Change and Early Warning Systems for Wildland Fire - Reducing Disaster: Early Warning Systems for Climate Change - page 125

Fig. 7.1 Cumulative Severity Rating anomalies for the HadCM3 A2 scenario for 2041-2050 rela-
tive to the 1971-2000 base period (See Flannigan et al. 2013 for study design)
The models were selected to provide a range of expected future warming conditions.
There are four emission scenario storylines (A1, A2, B1, and B2) that set out distinct
global development direction to the end of this century (IPCC 2000 ). The Flannigan
et al. ( 2013 ) study used the following three scenarios: A1B, representing a world of
very rapid economic growth with global population peaking by mid-century, rapid
development of effi cient technology, and a balanced use of fossil fuel and nonfossil
fuel sources; A2, representing a world of increased population growth, slow eco-
nomic development, and slow technological change (business-as-usual scenario);
and B1, representing the same population as A1, but more rapid change in economic
structure, and moving towards service and information technology.
The GCM-emission scenarios were used to calculate fi re weather conditions dur-
ing the next century. Fire weather data (temperature, relative humidity, wind speed,
24-h precipitation) were used to calculate daily component values of the Canadian
Forest Fire Weather Index (FWI) System (Van Wagner 1987 ). Fire season length
was calculated using a temperature approach, with the start of the fi re season defi ned
as three consecutive days of 9 °C or greater, and the end of the fi re season by three
consecutive days of 2 °C or lower. Fire severity was calculated using the Daily
Severity Rating (DSR), which represents the increasing diffi culty of control as a fi re
grows (Van Wagner 1970 ) and is a simple power function of the Fire Weather Index
component of the FWI System. Changes in fi re severity were measured using the
Cumulative Fire Severity (CSR), which was the sum of DSR values during the fi re
season divided by the fi re season length. In this way, the CSR was a seasonal length-
scaled version of the DSR. Changes in future fi re season length and CSR were sum-
marized by decade as anomalies from the 1971-2000 period (results were only
presented for mid-century and end of century).
Figures 7.1 and 7.2 show CSR for the HadCM3 model and the A2 scenario for
2041-2050 and for 2091-2100. These examples are representative of all the GCMs
and scenarios maps that show a signifi cant worldwide increase in CSR especially for