Fire and climate dynamics in the boreal forest of northern Alberta, Canada, from AD 1850 to 1989

The reconstruction of climatically induced changes in the occurrence of forest fires has typically been limited to the period of historical records or to areas where long fire-scar records are available. In this study historical fire records and dendrochronological forest stand-age data are analyzed...

Full description

Bibliographic Details
Published in:The Holocene
Main Author: Larsen, C.P.S.
Format: Article in Journal/Newspaper
Language:English
Published: SAGE Publications 1996
Subjects:
Online Access:http://dx.doi.org/10.1177/095968369600600407
http://journals.sagepub.com/doi/pdf/10.1177/095968369600600407
Description
Summary:The reconstruction of climatically induced changes in the occurrence of forest fires has typically been limited to the period of historical records or to areas where long fire-scar records are available. In this study historical fire records and dendrochronological forest stand-age data are analyzed using life-table methods to reconstruct annual (AD 1950-1989) and semi-decadal (AD 1850-1989) variations in area burned in the 44 870 km 2 area of boreal forest in Wood Buffalo National Park (WBNP). Annual area burned between 1950 and 1989 was negatively correlated with mean fire-season precipitation and five annual tree-ring width indices from WBNP, and positively correlated with the fire-season means of temperature and the Seasonal Severity Rating, a Canadian Forest Fire Weather Index. Life-table estimates of semidecadal variations in mean annual per cent area burned were significantly positively correlated with historical records of annual area burned in WBNP between 1950 and 1989 and significantly negatively correlated with five tree-ring width indices from WBNP between 1850 and 1989. Peaks in semidecadal estimates of mean annual per cent area burned appear to be quasi-periodic, with an average interval of 30 to 40 years. Potential causes of these decadal scale variations in fire and climate are discussed. The results of this study indicate that forest age-structure data collected at the landscape scale can be used to infer prehistoric, decadal scale variations in the amount of area burned by forest fires.