Succession was studied in a cold-temperate forest in the Boundary Wate
rs Canoe Area Wilderness (BWCAW) of northeastern Minnesota. The 13 x 1
8 km study area comprises a complex forest mixture of jack (Pinus bank
siana) and other pines, quaking aspen (Populus tremuloides): paper bir
ch (Betula papyrifera), black spruce (Picea mariana), balsam fir (Abie
s balsamea), and white cedar (Thuja occidentalis) on thin soils over t
he Canadian Shield bedrock. The main objectives of this study were to
examine the relationships between spatial patchiness, spatial scale, a
nd canopy succession in the southern-boreal forest of the BWCAW, and t
o evaluate under what conditions successional direction may remain sta
ble, converge, or diverge. Knowledge of the successional direction of
old forests in the BWCAW that are undergoing demographic transition fr
om even-aged to uneven-aged is important because the landscape now has
many old stands as a result of reduced fire frequency. Rotation perio
ds for fires have changed from approximate to 50-100 yr in presettleme
nt times to >1000 yr since 1910. Analyses were conducted at spatial sc
ales ranging from the individual tree (0.01 ha) to the large stand (16
ha). Two permanent mapped plots (of area 0.53 and 0.56 ha) were estab
lished in stands of different age. Fine-scale age structure succession
al change, transition from one species to another, and development of
small scale age structure, succesional change, transition from one spe
cies to another, and development of small patches (of area <0.25 ha) w
ere studied by means of stand history reconstruction with increment co
res, spatial autocorrelation, and analysis of replacement trees in can
opy openings. Spatial processes at nested scales of 1, 4, and 16 ha we
re examined on 15 square 16-ha tracts of upland forest, which are dist
ributed among forests ranging from 15 to 190 yr old. Canopy species co
mposition and patch development over lime on these 15 tracts were inte
rpreted on air photos taken in 1934, 1961, and 1991. Thus, the study i
ncludes a chronosequence approach, with verification of chronosequence
validity by checking patch development processes at more than one tim
e point. This checking was done at small spatial scales by reconstruct
ion of stand history on permanent mapped plots, and at larger spatial
scales with sequential air photos of the same locations spanning a 57-
year period. Results show that the reduced fire frequency in recent ye
ars has changed the dominant successional pathways. When fire frequenc
y was high, jack pine or aspen stands usually burned while still in th
e even-aged stage of development, and the new trees after the burn wer
e the same species as before. Currently, many stands are undergoing de
mographic transition from even-aged stands of catastrophic fire origin
to uneven-aged stands. This transition parallels a change in canopy c
omposition from jack pine (occasionally red pine (Pinus resinosa)) or
aspen to an old-growth multi-aged mixture of black spruce, balsam fir,
paper birch, and white cedar. The mechanism that moves this successio
nal path forward is canopy openings, 10-30 m across on average: caused
by wind, insect, and disease, that gradually chip away at the relativ
ely uniform canopy of pines and aspen. Successional direction is indiv
idualistic in the sense that time and rate of transition from pine/asp
en to other species depends on the action of heavy windstorms, insect
infestation, and senescence of old pines that create canopy openings,
Canopy openings are often filled with one of several species, but if m
ore than one species invades an opening, monodominant patches of each
species generally result. Understory-overstory interactions are very w
eak; the dominant species within each patch is apparently independent:
of the overstory species that died when the opening was created or th
e species dominating surrounding patches. During succession, the spati
al structure of the stands at the 1-16 ha scales generally changes fro
m a matrix heavily dominated by pine or aspen to a mosaic with relativ
ely large mono-dominant patches that may be remnants of the extensive
original matrix, and finally to a mixture with small patches (mean are
a 35 m(2), maximum approximate to 0.1 ha) of black spruce, balsam fir,
white cedar, and paper birch. Thus, at 1-16 ha spatial scales, succes
sion leads to convergence on a mixture of species. At smaller spatial
scales (e.g., 0.01-0.1 ha) successional pathways appear to diverge int
o four community types. The same successional pathways can be reconstr
ucted from historical analysis of individual stands as from a chronose
quence of stands; therefore, chronosequences in this area have been st
able at least during the lifetime of the current generation of trees.