|Abstract or Summary
- A crossdated fire history was reconstructed for a 1562 km2 area in the southern
Willamette foothills of Oregon, using fire scars and tree origin years from twelve sites.
The purpose of this study was to determine fire frequency for each site and to quantify
temporal and spatial variability of fire frequency. Fire frequency distributions were
related to climate history and the patterns of human settlement, and compared with
other regional fire histories.
Dendrochronological methods were used to assign calendar years to fire scars
and pith dates. General Land Office maps and surveyor notes were used to determine
site and study area level changes in vegetation and Euro-American land use patterns.
Climatic influence on fire over time was determined using superposed epoch analysis.
Forty-three fire dates were reconstructed for the 290 year period from 1700 to
1990. The minimum and maximum fire intervals were two years and 191 years; the
study area mean fire return interval was 49 years. Over two-thirds of all site level fire
intervals were less than 40 years, and less than one-fifth were longer than 80 years.
Fire frequency differed between the east and west sides of the Willamette
Valley. West side sites experienced more frequent fires on average than the east side
sites: 57% of all west side site fire intervals were less than 20 years, while 68% of all
east side site fire intervals were less than 40 years. Both human and vegetative factors
were potentially influencing the fire regimes of each side of the valley. West side sites
were within, or adjacent to, woodland forest cover types and were closer to earlier and
more intensive Euro-American settlement.
Fire frequency throughout the study area did not change substantially over
time. Fire occurred every decade or so, with occasional longer fire intervals, until
after 1905, at which time fire occurred roughly every 15 to 20 years until 1979.
Unlike other regional fire history studies, this study found no statistically significant
differences in the number of fires by time periods of varying land use and climate.
When the rule set for defining fire events was modified, fire frequency could be shown
to be weakly significantly associated with these time periods. This indicates that
methodology can appreciably influence the results of fire history reconstruction.
Moreover, the sample size of trees dating from the 1 700s was small, so the estimate of
fire frequency for the 1 700s is uncertain.
Fire frequency was not related to drought over the whole study period (1700-
1990), but fire was more likely to occur three years following a drought year over the
period from 1700 to 1849, whereas after 1850, fire was significantly associated with
drought years. Fire occurrence was as expected (but not significantly so) during the
period 1700 to 1800, slightly higher than expected (but not significantly) for the
period 1800 to 1850, higher than expected (but not significantly) after 1850, and lower
than expected (but not significantly) after 1925. Several hypotheses could explain
these findings: (1) the relationship between wildfire and drought is different during
cooler, wetter periods than warm, dry periods; (2) Native Americans influenced fire
during the 1700-1849 period, obscuring the relationship between fire and drought; (3)
European settlement of the Willamette Valley in the mid to late 1 800s increased the
occurrence of fire and enhanced the effects of a warming climate, and (4) fire
suppression after 1920 resulted in a decrease in fire occurrence.
More sampling of older stumps in the Willamette Valley foothills would likely
lend credibility to the record of fire frequency before 1800 and would increase the
sample depth for use in the Superposed Epoch Analysis of the fire-drought
relationship. Sampling along transects extending from the foothills into the Cascades
and Coast Ranges would foster a better understanding of the spatial scope to which
Native American burning influenced forests adjacent to the valley floor.