Spatial and temporal variation in historical fire regimes of the Blue Mountains, Oregon and Washington: the influence of climate
To identify the influence of climate on spatial and temporal variation in fire, low- to high-severity fire regimes were reconstructed from tree-rings in the Blue Mountains. Fire recurrence, extent, severity and seasonality (low-severity fires only) were determined from fire scars and ages of 1426 trees sampled on 2914-8585 ha grids in 4 watersheds. Before 1900, fire regimes varied at regional (among watersheds) and local (within watersheds) spatial scales, although not all parameters of fire varied at both scales. Regionally, fires in ponderosa pine-dominated forests burned more frequently and earlier in the growing season in southern than northern watersheds, consistent with the occurrence of longer and drier fire seasons in the southern Blue Mountains. Fire extent did not vary regionally. Locally, fire recurrence varied with topography (aspect or elevation) in steep terrain but not in gentle terrain, while local variation in fire extent was unrelated to topography in any watershed. Temporal variation in the extent of low-severity fires was compared to existing tree-ring reconstructions of regional precipitation and an index of the Southern Oscillation (SOI). In southern watersheds, fire extent varied inversely with precipitation on annual and longer time scales. In northern watersheds, SOI tended to be low (El Nino conditions) during fire years, consistent with shorter snow-cover duration to the north during El Nino years. Prior year's climate (regional precipitation or SOI) did not influence fire extent in any watershed. Despite some regional synchrony in precipitation, fires rarely burned in more than one of the sampled watersheds during a given year, probably because processes that influence the ignition and/or spread of fire operate at sub-regional spatial scales (e.g., lightning strikes, precipitation from convective storms). After about 1900, few fires occurred in any of the watersheds. These results suggest that to predict spatial variation in fire regimes within a given forest type, the spatial scales at which the controls of fire (e.g., climate and topography) operate must be considered. These results also imply that future fire regimes could be affected by changes in the duration of snow cover or by changes in the amount or timing of ignition.
- Forestry