Climatic Inferences from Dendroecological Reconstructions - Dendroclimatology - page 282

disturbances and demography (e.g., fire, insect outbreaks, and tree mortality and
subsequent natality) can add multiple layers of complexity and non-climatic signals
in composite, ecological time series. Nevertheless, depending on the extent and
degree of spatial coherence and synchrony of disturbances or demographic pat-
terns, the relative role of climate can be assessed by the nature of the aggregated
time series, and their relations to independent climate reconstructions. The moder-
ate degree of regional synchrony in southwestern US ponderosa pine establishment
over the past several centuries, and approximate correspondence with reconstructed
drought indices (Fig. 9.10 ) , suggests that climate is at least partly controlling. We
recognize, however, that multiple other environment factors probably also determine
the nonsynchronous variations, especially at finer spatial scales of individual stands.
9.3 The Late Eighteenth-Century, Early Nineteenth-Century
Fire Gap
We close our discussion of dendroecological and climate reconstructions with a spe-
cific example of an interesting climatic episode that has been identified in multiple
fire chronologies. Fire historians in North and South America have long recognized
a very interesting 'gap,' or hiatus, in fire events during the period from about 1780 to
1840. The gap appears as an unusually long interval (relative to average intervals in
the rest of the chronologies) between fire events. This gap is evident in many forest
stands from the Southwest (e.g., Swetnam and Dieterich 1985 ; Fig. 9.11 ; see also
Fig. 9.8 ) , Baja California (Stephens et al. 2003 ; Fig. 9.12 ) , and northeastern Oregon
(Heyerdahl et al. 2002 ) . The gap seems to be detectable primarily in the intermediate
fire frequency fire regimes (i.e., mean fire return intervals between about 7 and 15
years). The timing and length of the gap varies somewhat from area to area, starting
or ending some years earlier or later, but generally within the period from the late
1700s through the middle1800s.
Kitzberger et al. ( 2001 ) noted that a similar gap of reduced fire frequency was
present during approximately the same time period (i.e., 1780s-1840s) in fire his-
tories from Austrocedrus chilensis woodlands in northern Patagonia, Argentina
(Fig. 9.13 ) . Perhaps most remarkably, the regional fire-scar time series from the
Southwest and Patagonia were statistically coherent (i.e., synchronous), especially
in the frequency range of about 2-7 years (determined in a bivariate spectral analy-
sis). Kitzberger et al. ( 2001 ) hypothesized that the common response of fire regimes
in the Southwest and Patagonia to the El Niño/Southern Oscillation (Swetnam and
Betancourt 1990 ; Kitzberger et al. 1997 ; Veblen et al. 1999 , 2000 ; Veblen and
Kitzberger 2002 ) was likely the reason for coherence of the two fire occurrence time
series from different hemispheres. Moreover, they proposed that the hiatus during
the late eighteenth and early nineteenth centuries could be due to a combination
of (1) reduced amplitude and/or frequency of ENSO events (especially La Niña
events) during this period (Fig. 9.14 ; see also Fig. 9.8 ) , and (2) a coincidence of
this period with a major global cooling phase in the early 1800s (i.e., 1810s-1820s),