HR: 17:00h
AN: B44C-05 INVITED     [Abstracts]
TI: Scale-Dependent Controls of Late-Holocene Forest Fires in British Columbia: Insights From Intra-regional Paleorecord Comparisons
AU: * Gavin, D
EM: dgavin@life.uiuc.edu
AF: Department of Plant Biology University of Illinois, 265 Morrill Hall, Urbana, IL 61801 United States
AU: Hu, F
EM: fshu@life.uiuc.edu
AF: Department of Plant Biology University of Illinois, 265 Morrill Hall, Urbana, IL 61801 United States
AB: Forest fire regimes are affected by multiple controls that operate at local (e.g. stochastic ignitions, topography, and fuel loads) to regional (e.g. climate) scales. At small spatial scales, climatic controls may be obfuscated by local factors, but the spatial scale at which climatic controls outweigh local controls is poorly understood for stand-replacing fire regimes. We addressed this hierarchy of controls by comparing Holocene fire histories (the specific times of fire events and trends in fire frequency) and fire regimes (the statistical distribution of fire intervals) based on sediment charcoal records from southern British Columbia. We compared two pairs of charcoal records. First, at sites with similar vegetation and climatic history, we expected similar fire regimes and histories. However, a comparison of two charcoal records from sites in similar vegetation only 11-km apart revealed different fire-interval distributions and the lack of any synchrony in fire dates $>$ 2500 years before present (BP). After 2500 years BP, the fire regimes converged and fire events were marginally synchronous. In addition, the composite fire-frequency record (average of both sites) showed strong coherency with regional climatic changes evidenced by late-Holocene glacial advances. This coherency probably resulted from the increased area represented by the composite record (up to ~200 ha) and from the increased regional climatic variability over the last several millennia. Secondly, at sites in different elevational forest zones, we expected different fire regimes and possibly different fire histories. In this comparison, charcoal accumulation rates suggested that prior to 3000 years BP, fire severity was high at the high elevation site (1530 m asl) and low at the low elevation site (515 m asl) site, such that charcoal peaks could not be identified at the low elevation site. After 3000 years BP, severity increased at the low elevation site, allowing distinct charcoal peaks to be identified. During this period, statistically significant synchrony was detected within 1000-year and ca. 300-year windows, suggesting climatic variability at these scales entrained fire history over the last few millennia. Overall, these results indicate 1) that stands with similar modern conditions may have experienced different past fire regimes and fire history, likely because local processes outweighed the synchronizing effect of climate over millennia, and 2) that the influence of climate varied over time, and was stronger during the past 2500 years than before because of greater climatic variability.
DE: 0400 Biogeosciences
SC: Biogeosciences [B]
MN: 2004 AGU Fall Meeting