|
JULY 4th BLOWDOWN in the BWCAW Lee E. Frelich Final version, February 10, 2000 SUMMARY The great derecho of July 4th 1999 in the BWCAW was one of the largest blowdowns ever recorded in North America.It is similar in size and severity of forest damage to category 3 or 4 hurricanes making landfall in forested region (e.g. Hurricane Hugo in South Carolina in 1989). There is debate at the current time as to whether the disturbance regime is changing so that these large blowdowns are becoming more common. This report examines the potential response of five forest types to blowdown alone, blowdown followed by intense fire during a significant drought, fire during a year of normal rainfall, and prescribed fire as proposed by the Fuels Risk Assessment of Blowdown in Boundary Waters Canoe Area Wilderness and Adjacent Lands (Leuschen et al. 1999).The forest types are second growth aspen-birch-red maple-conifer and primary mesic birch, aspen, spruce and fir; dry-mesic white and red pine; dry jack pine and black spruce; and lowland black spruce. In general, the blowdown alone will accelerate succession towards the shade-tolerant conifers balsam fir, black spruce and white cedar; a high-intensity fire in the blowdown slash will set succession back to aspen or paper birch; and low-to-moderate-intensity natural fires or prescribed burns will create variable mixtures of aspen and birch with conifers. Some important specific effects include the following.Some stands with shallow soil that burn may be returned to bare rock, grasses and small shrubs. There is potential for blown down jack pine-black spruce stands to regenerate to those species after prescribed fire, because the blowdown/prescribed fire combination would be equivalent to one natural crown fire in its effects on the forest.Such regeneration would not work well in older senescent jack pine, where black spruce would have an advantage. A blowdown-intense fire combination could possibly wipe out historical refuge populations in sheltered lakeshore areas for white cedar, white pine and red pine, resulting in an extreme long-lasting negative effect on recovery of these species within the blowdown. Prescribed fires might able to help the situation for white pine, red pine and white cedar if carefully done, by saving extant refuges during the burning procedure, and by reducing the chances for a landscape-scale severe fire. Many of the predictions made here are qualitative. Quantitative answers for the problem of how forests would respond to various combinations of disturbance would require an in depth study including seed dispersal, mortality functions, and growth simulators. INTRODUCTION The storm damage was caused by a derecho. Squall lines with a characteristic bow echo on weather radar produce derechos, which are a defined as a characteristic pattern of straight line wind damage over a widespread area (main axis of damage >250 miles in length) caused by repeated downbursts that emanate from thunderstorm cells near the peak of the bow echo as the storm moves across the landscape (Johns and Hirt 1987).The frequency of derechos in North America reaches a maximum in a triangular-shaped region with vertices in Minnesota, Oklahoma, and Pennsylvania (Johns and Hirt 1987, Bentley and Mote 1998). Derechos cover much larger areas than tornados and are the main cause of wind damage in the Upper Midwest. However, it should be noted that, although southern Minnesota has a very high frequency of derechos, northern Minnesota normally has a low derecho frequency, so that the July 4th storm there represents an extreme event. At this point it is not certain whether derechos are occurring further north as a result of global warming (i.e. the zone of high frequency has expanded or moved north), or whether the July 4th, 1999 event was an extreme of the historic climate. Recent data suggest the former, however, as other large blowdowns occurred in northern Minnesota in July and September 1995, and in northern Ontario in 1973 and 1988 (Flannigan et al. 1989). Several of these recent blowdowns exceed 100,000 acres in size (counting only the areas with severe tree damage where more than half of the canopy was downed), and two of them are in excess of 300,000 acres. These sizes are an order of magnitude larger than windfalls recorded by the General Land Office Surveyors in the Lake States of Michigan, Wisconsin and Minnesota prior to European settlement (Frelich unpublished data). The combination of frequency and large size of these `superblowdowns' causing forest damage beyond that previously known both suggest that such storms have only recently become an important disturbance in the Lake States northern forests. Some derechos produce F2 winds (113-157 miles per hour) on the Fujita scale (Fujita 1978). Winds of 100-110 mph or more are necessary to cause catastrophic forest damage (Frelich and Lorimer 1991), so that the most intense derechos can easily level vast areas of forest and cause the most extensive forest damage of any weather phenomenon in the interior of North America. The July 4th, 1999 derecho that hit the BWCAW and a similar one that hit northern Wisconsin on July 4th 1977, caused damage similar in scope to that of a category 3 or 4 hurricane making landfall in a forested region (Frelich unpublished data). The severity of forest damage in the BWCAW was very likely exacerbated by fire suppression and fire exclusion over the last century, which resulted in a large expanse of older forest highly susceptible to wind damage (Heinselman 1996). Under the historic fire regime (1700-1900), the rotation period for crown fire was between 50 and 100 years in jack pine, black spruce and aspen forests, which occupied the majority of the BWCAW landscape, so that most forests were young and even-aged, with relatively low susceptibility to wind damage. SOME GENERAL EFFECTS General principles of forest response to disturbance have emerged in recent years (Frelich and Reich 1995a, 1999) and these principles should hold for the BWCAW. Windthrow generally accelerates succession by releasing shade-tolerant understory seedlings and saplings, although if a stand is already at the climax stage and dominated by shade-tolerant species, then windthrow merely changes the stature of the forest with little effect on composition. Intense fire generally sets succession back to the early pioneer stage, but if the system experiences frequent fires so that the forest is already dominated by shade-intolerant pioneer species, then the fire changes only the stature of the forest. Surface fires generally have an intermediate effect between these two other disturbances, and can create mixed pioneer-climax forests or maintain mid-successional species (e.g. white pine). Also note that mixed pioneer-climax forests can result from succession of a pioneer stand in the absence of fire. The combination of windfall followed by intense fire can have extremely severe effects on the forest, beyond that of a crown fire alone. All forests will be set back at least to the pioneer stage by this disturbance combination, and some will be set back to a non-forest stage. Each disturbance event has an influence on the probability and outcome of future disturbance (Frelich 2000, in press). Windthrow increases the chance that intense fire will occur. It does not increase the chance of ignition, but it does increase the chance that a fire, once ignited, will become intense and spread more than the same ignition may have under the same weather conditions in a standing forest. Thus, the probability of burning for stands increases because of a contagion effect (i.e. each fire is larger), not because there are more ignitions. In the BWCAW, this effect probably lasts at least 8-12 years, after which downed wood is shaded by the young post-blowdown canopy, becomes punky, and does not add much to the chance of intense fire. Canopy-leveling windthrow and stand-killing fires also decrease the chance of another windthrow to nearly zero for 20-40 years, and to a low level for 40-60 years under conditions in the BWCAW. Severe fire in a standing conifer forest does not seem to influence the chances of another fire (although it is likely to reduce the intensity of fire for the next 2 decades), unless the forest happens to regenerate to a less flammable type after the fire (Johnson 1992). Under the historical natural disturbance regime in the BWCAW, this would generally not be the case (e.g. jack pine regenerating immediately to jack pine after fire). In the blowdown, however, intense fire would convert the area mostly to aspen, which is less flammable than conifers. The reduced flammability could vary anywhere from 20 years to a century or more among stands across the landscape, because the rate of reinvasion by conifers also varies, depending on the situation in each stand. The details of these differences are explained under specific effects below. Low-to-moderate intensity natural or prescribed fires will reduce the intensity of future fires for a few decades by removing some of the windfall slash, but not to the extent that a high-intensity fire would. General effects forest management practices and their interactions with natural disturbance are understood at this time. Salvage logging in windfall slash has three effects on the future forest condition. First, the logging itself necessarily kills or damages many remnant seedlings and saplings, because of the amount of physical manipulation necessary to untangle, cut, and remove interlocked fallen trees. The effect of this is to reduce the amount of acceleration towards future dominance by shade-tolerant conifers that would otherwise be caused by the windfall itself. Second, salvage cutting of paper birch and red maple often leads to stump sprouting that often does not occur when the fallen tree is left intact. Many more multiple-trunked trees then occur in the future stand, so that the average dbh is smaller (due to competition with the other sprouts from the same stump) and trees are less likely to have straight trunks. Third, salvage logging removes a substantial volume of coarse woody debris, so that any subsequent fires are apt to be less intense. Also, this coarse woody debris (30-100 years after the future canopy is established and it is well rotted), would be the future seedbed of several species of trees, including white cedar, white and black spruce, white pine, and balsam fir. Therefore, invasion of salvaged forests by these species would be relatively slow over the next century in stands that are salvage logged. This would be true whether a stand burns or not, since much of the large-diameter coarse woody debris is not consumed by fire. SPECIFIC EFFECTS--FOREST TYPES ANALYZED The forest types covered are those that were identified by Kurt Rusterholz, based on ordination of MN DNR relevŠ data from northeastern MN. The Range of Natural Variability Working Group--comprising DNR, Forest Service and University of MN people--worked out the characteristics and variability of the landscape structure for eight major forest types in the Northern Superior Uplands Section in Minnesota (Frelich 1999). Four of these are important in the blowdown area: Mesic birch, aspen, spruce and fir; Dry-mesic white and red pine; Dry jack pine and black spruce; and Lowland black spruce. Four others (Mesic white and red pine; Rich swamp with white cedar and black ash; Dry jack pine-pin oak; and northern hardwoods with sugar maple) are absent or have very little acreage within the blowdown area, and are not included. In addition, I created a second-growth category (Aspen-birch-red maple-conifer) for this analysis, since in my judgement second growth forests will in some respects respond differently to wind and fire than primary forest. No analysis of the coverage of primary and secondary forest within the blowdown has been done, but my preliminary inspection of the damage maps indicates roughly a 50:50 split. SPECIFIC EFFECTS--RESPONSE TO DISTURBANCE BY FOREST TYPE Four disturbance combinations were examined: the blowdown alone, fire following the blowdown during a significant drought year, fire following the blowdown during a year with normal rainfall, and prescribed fire following the blowdown. Fire during a significant drought generally refers to the 90th or 97th percentile simulations with flame lengths of 3-5 m as shown by Finney (1999). Similar results for tree mortality and forest succession would probably occur after the 75th percentile drought. The reason: relatively slow movement and extreme heat near the ground with the high fuel accumulation would result in a high total heat output (kw/hr X hr = Kw) that would be enough to kill live trees as effectively as the 90th or 97th percentile fires, even though soil moisture may be higher under the 75th percentile drought. Normal moisture year corresponds to the 15th percentile simulation, with flame lengths of 0-1 m. I assume that prescribed burns would be done at times when the fire weather was similar to the 15th percentile conditions. If conditions were more severe than that, the fire crew would ignite the big fire they are worried about, and if burning weather were wetter than that little burning and fuel reduction would take place. Therefore, the differences between prescribed fire and natural fire during a normal moisture year would be mainly one of size and location of fires, since locations can be chosen for prescribed fires. The following assessment of windstorm and fire impacts integrates all of my knowledge from 10 years of research in the BWCAW, everything I learned from Bud Heinselman and from the scientific literature, as well as my observations from 150 photographs of the blowdown-damaged area that were supplied to me by Brian Sandberg and Barb Soderberg of the Superior National Forest Headquarters in Duluth, MN. The following should be taken as expert opinion, not a statement of fact. There are occasional surprises when one obtains actual data from forests responding to disturbance and compares that to an expert's prior opinion. Secondary Forest Type Aspen-Birch-Red Maple-Conifer Response without occurrence of fire. The most common response would be to accelerate succession, since most stands have younger white or black spruce, balsam fir or white pine that were in the understory of the pre-windthrow stand. A few of these seedling-to-sapling-sized conifers were killed by falling trees, but most will be released from suppression. The post-wind stand will also have substantial numbers of aspen root sprouts and a few paper birch and red maple stump sprouts mixed with the released conifers. Thus, they will remain mixed conifer-hardwood stands, but the conifer proportion should be enhanced as compared to pre-windthrow conditions. A secondary response--in stands without a conifer understory--would be conversion to shrublands dominated by hazel, mountain maple, alder, willow and other shrubs. The length of time it will take for trees to regain control from the shrubs will vary from 1 decade to several decades, depending on the size of the shrubby area and wetness (wet and dry sites would retain shrubs longer than mesic sites). Response after large fire during a drought year. In older forest that originated from 1985-1930--Heinselman's (1996) `big pine logging era', succession would be set back to the same level it was at in the period 1895-1930, when the area's primary red and white pine forest was logged, often followed by burning of the slash, thus creating young aspen and paper birch dominated forests. The reason the forest regenerated without much conifer after the logging-slash fire combination, was that conifer seed trees were systematically removed by loggers. In such cases, aspen and paper birch win dominance of the post-disturbance forest by default, since they are the species with long-distance seed dispersal (Frelich and Reich 1995a). It took from one to ten decades for conifers to begin reinvading these stands, depending on how far they are from a refuge (Figure 1), and the same would be true after a large fire during a drought year in the windfall slash. Some conifers in these stands are already at seed-bearing age, and a few of those would survive the fire, but nevertheless, reinvasion by conifers would be very slow. Some red maple would survive the fire and stump sprout, and their seed dispersal distances are intermediate between that of aspen and paper birch and the conifers, so that red maple would invade the young aspen slightly ahead and/or along with the conifers. Some secondary forest originated during the pulpwood logging era from 1935-1978 (Heinselman 1996). These forest were dramatically changed by logging, resulting in removal or large reduction in proportion of conifers--mostly jack pine and spruce in this case--as compared to the virgin forests (Heinselman 1996). Because they are still quite young, they have not had significant conifer reinvasion except in limited areas, and the conifers are mainly balsam fir. A windfall slash fire during a drought year would probably kill all the conifers and consume their seeds, and return the forest to young aspen and birch as it was after the logging. Response after a large fire during a normal moisture year, and response after prescribed burns. The effect of these burns would be intermediate between those described above for no fire and intense fire during drought. Much of the conifer regeneration released by the blowdown would be killed but some would survive, and aspen and birch would fill in the rest of the forest. Mixed conifer-hardwood stands would be the result. Primary Forest Types Mesic birch-aspen, spruce and fir Response without occurrence of fire. Most of this forest was relatively old before the blowdown, due to the relatively low number of large stand-killing fires in recent years (Friedman and Frelich 1999). Therefore, most stands have copious conifer regeneration of balsam fir, black spruce, white pine, and white cedar. This understory will be released from suppression by the blowdown, accelerating succession towards a conifer-dominated forest. Many remnants of white cedar, with balsam fir and black spruce are still extant along lakeshores after the windstorm. Remnants of white and red pine also exist, but are less numerous. Conifers will be able to spread from these remnants to the surrounding forest, acting as a reinforcing factor in succession to conifers. The ultimate climax forest in much of the BWCAW uplands, in the absence of fire, is likely white cedar with a few other conifers mixed in (Grigal and Ohmann 1975, Frelich and Reich 1999). Because white cedar is a relatively short stature species and has a growth form with a low center of gravity, it was minimally affected by the blowdown. Therefore, the blowdown in the absence of fire will act to accelerate succession towards this climax in all forest types. Response after large fire during a drought year. The response of this forest type to intense fire is well known and fairly simple; it returns to young aspen and paper birch (Heinselman 1973, Frelich and Reich 1995a). Reinvasion of conifers is not usually expected until 20-100 years after fire, with the exact timing depending on distance of a given stand from conifer remnants that survived the fire (Figure 1). Remnant small groves also exist along lakeshores in indentations in the bedrock that are 5-10 m deep. Historically fires did not burn these little refuges, and many of the trees also survived the blowdown because they were protected by the surrounding topographical features and were not very tall trees to begin with. Thus these refuges serve as the seed source for recolonization following all types of natural disturbances. It remains to be seen whether these refuges can survive massive blowdown followed by high-intensity fire, since the dense slash fuels could carry fires into them in, whereas crown fires in standing timber often would not carry into them. Response after a large fire during a normal moisture year, and response after prescribed burns. Much (not all) of the conifer regeneration (mainly fir but also including some white pine, black spruce, white spruce and white cedar) would be killed by these low-to-moderate-intensity fires, so the response would be a return to dominance by aspen and paper birch with some fir, spruce and cedar mixed in. Dry-mesic white and red pine Response without occurrence of fire. Many of these forests lost most or all of their seed trees in the blowdown. The lack of surface fires in the past century has disfavored establishment of regeneration cohorts of these species underneath the canopy (Heinselman 1973, 1996). Therefore, the blowdown will favor succession towards more shade-tolerant conifers such as balsam fir, black spruce and white cedar (Figure 2). Paper birch also is an important component of late-successional stands (whether young or old) in this forest type, and should continue to be so after the blowdown (Frelich and Reich 1995b). Red pine will be reduced more than white pine by the blowdown. White pine is moderately tolerant of shade, and some white pine saplings exist, so that the species will not be extirpated to the extent that red pine will. Some stands also have red oak and red maple saplings in the understory. These species will be released, but they do not attain full stature under the climate and soil conditions of the BWCAW. Response after large fire during a drought year. High-intensity fire in this forest type will usually kill all the seedlings and most of the adult pines, as well as consume any soil seed bank that exists. The result is to set the forest back to young paper birch (Figure 2), with some aspen mixed in (Heinselman 1981, Frelich 1992). Some red pine stands are on extremely shallow rocky soil, which could burn away and/or slough off after an intense fire, and succession would go back to a bare rock or grass and small shrub stage. A major concern for this forest type, is whether significant numbers of mature white and red pine in the traditional refuges from fire along the lakeshores will survive to reseed the stands after a fire. Many of these refuge trees in locations that would normally be survivable after natural fire, were blown down. In addition, the high density and contiguity of fuel near the ground in the post-blowdown forest could propagate high-intensity fire into some of the refuge areas and kill some of those trees that did not blown down. Therefore, the future for regenerating white and red pine forests after a fire in the blowdown is in serious doubt. White and red pine will not go extinct within the blowdown area, but the young paper birch vegetational growth stage that normally occurs after wildfire during severe drought, will last longer in these forests in post-blowdown condition. Only a major study can resolve the question of whether this extended stay in the birch phase will last a few decades more than normal or will last for centuries. Data on distribution and mortality of mature pines after fire combined with seed dispersal models and growth simulators would be necessary to resolve the question. Response after a large fire during a normal moisture year, and response after prescribed burns. The combination of having most trees blown down, and then having a fairly intense surface fire, would have a more or less equal effect on red and white pine forests as a single high-intensity crown fire: namely only a few seed-producing pines would remain and paper birch along with some aspen would restock the stand. Succession back to white pine would depend on the abundance and distance from surviving seed-bearing trees. Some stands next to an area that did not blow down could be expected to have abundant pine stocking under the birch forest by age 40 and to succeed back to pine by age 100 (Figure 2). ther stands in the middle of the blowdown, and that also have nearly all seed trees killed within the stand, may not succeed back to pine for centuries. Surviving seed source after the blowdown/fire combination will be an overwhelmingly important factor for red and white pine, and it also applies to white cedar. These species do not have long-distance seed transport like birch and aspen, they do not have serotinous cones like jack pine and black spruce, and they cannot resprout from the stump or underground rootstocks. Their only mechanism for continuing the population after disturbance is for individual mature trees to survive by physically resisting the disturbance (Frelich and Reich 1995a). Therefore, extreme caution should be used during prescribed fires to make sure that the last refuges for these species are not extirpated. Refuge populations need more than a few trees if future generations of pines are to be genetically viable (Buchert et al. 1997). Groves of 10-20 trees within effective pollination distance (100 m) may be a good minimum standard, but this topic needs more discussion and research. One of the proposals for using prescribed fire would burn out windfall slash fuels between lakes to connect them into a more efficient fire break for large natural fires. This would also place most of the prescribed fires in exactly the areas where red and white pine abundance was historically the highest: between lakes on isthmuses and islands where the frequency of high-intensity fire was relatively low compared to the rest of the BWCAW landscape. The prescribed burn procedure could cut both ways for the pines. Some of the last refuge pines may be killed, but the prescribed burning plan may succeed in limiting the overall size of major natural-ignition fires, which would result in less widespread elimination of these species across the BWCAW landscape. It is possible to protect important remnant groves within a prescribed burn perimeter, and that strategy in combination with reduction of area experiencing high-intensity fire, would maximize the chance of recovery of red and white pine within the blowdown. Dry jack pine and black spruce Response without occurrence of fire. Prior to the blowdown, these forests were mostly old (100-130 years) and had dense regeneration of black spruce, balsam fir, and sometimes white cedar (Frelich and Reich 1995b, Friedman and Frelich 1999). The blowdown will shift composition towards these species and away from jack pine (Figure 3). Some younger stands existed that originated after fires in 1910, 1936 and 1976 (Heinselman 1973). The response in these younger stands will be somewhat different, because advance regeneration of shade-tolerant conifers was not very dense. Temporarily, these stands will be open sites, and invasions by hazel, aspen and paper birch and red maple are possibilities. It is also possible that some of the serotinous cones that are close to the ground will open on warm sunny days in the next few summers. Thus, jack pine will regenerate, but will be mixed with the other species mentioned above.Invasion by the late-successional conifers will probably not occur for several more decades, but the timing will depend on how close the stands are to the previously mentioned remnants of cedar/spruce/fir along the lakeshores. ˙˙˙˙˙˙˙˙˙˙˙˙˙˙ Response after large fire during a drought year.Standing jack pine and black spruce have serotinous cones, and have massive regeneration after intense fire (Heinselman 1973, 1981, Frelich and Reich 1999).˙ A blown down forest will respond differently. Because of the long duration of high heat within 2-4 m of the ground, where most of the seeds are at this time in the post-blowdown forest, the seeds that are normally released by fire could be consumed instead.˙ Also, this forest type often occurs on very severe sites with shallow soil; and many of the stands were very old due to the lack of regenerating fires over the last century; such stands have relatively little jack pine seed because most jack pines have died or become senescent.Sometimes the soil consists mostly of an organic moss layer. Black spruce would be impacted similarly with most seed and forest floor seedlings consumed by fire. All of these factors (low seed numbers, consumption of seed by fire, shallow soil) indicate that impacts of an intense fire could be quite severe, sending parts of the forest to a permanently altered state.Some stands would revert to bare rock (restarting primary succession), some to a grass and shrub mixture, and some (probably most) to aspen stands with a few jack pine and black spruce (Figure 3).˙ Little is known about how jack pine gets back into an aspen-dominated forest in stands that were previously dominated by jack pine, although in theory this should be slow process that would take at least 2 generations of trees to accomplish.˙˙ ˙˙˙˙˙˙˙˙˙˙ Response after a large fire during a normal moisture year, and response after prescribed fire. This type of fire has the potential to open the serotinous cones of fallen jack pine and black spruce (which will stay viable for several years), and to allow good regeneration of these species and perpetuation of the forest type that was present prior to the blowdown. The duration and intensity of heat from a fire with flame length of 1 m, on seeds in the cone-laden crowns that are a few m above the ground could be very similar to that experienced by seeds in the cones of standing mature trees in a natural crown fire. This scenario and prescribed burning gives the best chance for perpetuation of one of the major forest types that has been historically important in the BWCAW (Figure 3). ˙˙˙˙˙˙˙˙˙˙˙˙ Also note that some older stands in this forest type no longer have a large jack pine population, having succeeded to black spruce and fir.In these cases, paper birch and aspen will be able to enter the stand at the time of the fire and form a mixture with black spruce.˙ As mentioned above, jack pine could take a long time to re-enter such stands.As with white pine, detailed studies would be necessary to better show how the long-term response would unfold. Lowland black spruce ˙˙˙˙˙˙˙˙˙˙˙˙˙ Response without occurrence of fire. Much of the lowland black spruce in the BWCAW is in small pockets that are sheltered by the surrounding hills. This factor, combined with the small stature of lowland black spruce, led to minimal impact from the blowdown. This forest type can easily repair the blowdown damage through regeneration by layering and seeding in from the remaining trees.˙ In as much as other tree species cannot grow on most lowland black spruce sites, it appears that the forest will stay black spruce. ˙˙˙˙˙˙˙˙˙˙˙˙ Response after large fire during a drought year. During a major drought, black spruce lowlands can dry out enough to burn (Heinselman 1963), and with all the slash fuel in adjacent upland stands, it is likely that fires will carry through some of them.˙ However, in those lowland stands with minimal blowdown, such a fire should not be any different than a natural stand-killing fire. The forest should regenerate to black spruce afterwards.I predict minimal impact of fire in these forests. ˙˙˙˙˙˙˙˙˙˙˙˙˙˙ Response after a large fire during a normal moisture year and response after prescribed burns. The response would be similar for these two cases, because large fires normally skip around lowland black spruce, or burn it very lightly, as would a prescribed burn.In either case, a few smaller black spruce would be killed, but the essential composition and character of the forest would not change.˙ Fire would not be able to burn into the peat very much and would be unlikely to get into the crowns. Intense fires would be confined to those stands where the trees were relatively large for black spruce, and had significant wind damage. In any case, no change in the natural successional pathway is likely. LITERATURE CITED Bentley, M.L., and T.L. Mote.˙ 1998.˙ A climatology of derecho-producing mesoscale convective systems in the central and eastern United States, 1986-95. Part I: temporal and spatial distribution.Bulletin of the American Meteorological Society 79: 2527-2540. Buchert, G.P., O.P Rojora, J.V. Hood, and B.P. Dancik.1997.˙ Effects of harvesting on genetic diversity in old-growth eastern white pine in Ontario, Canada.˙ Conservation Biology 11: 474-758. Finney, M.A.˙ 1999.A spatial analysis of fire behavior associated with forest blowdown in the Boundary Waters Canoe Area, Minnesota. Pages 67-88 In: T. Leuschen, Team leader, Fuels risk assessment of blowdown in Boundary Waters Canoe Wilderness and adjacent lands. Flannigan, M.D., T.J. Lynham, and P.C. Ward.1989.˙ An extensive blowdown occurrence in northwestern Ontario.˙ Presented at the 10th Conference on Fire and Forest Meteorology, April 17-21, 1989, Ottawa, Canada. Friedman, S.K. and L.E. Frelich. 1999.˙ Forest type and age-class distribution in the BWCAW.Report to Minnesota Forest Resources Council, July 2, 1999. Frelich, L.E. 1992.The relationship of natural disturbances to white pine stand development.˙ Pages 27-37 in: R.A. Stine and M.J. Baughman, editors.˙ White pine symposium proceedings: history, ecology, policy and management.˙ Minnesota Extension Service, University of Minnesota, St.Paul, MN, USA. Frelich, L.E. 1999.˙ Range of natural variability in forest structure for the Northern Superior Uplands.Report to the Minnesota Forest Resources Council. Frelich, L.E. 2000 (in press).˙ Disturbance regimes and forest dynamics. ˙Cambridge University Press, Cambridge, England. Frelich, L.E. and C.G. Lorimer. 1991.˙ Natural disturbance regimes in hemlock-hardwood forests of the Upper Great Lakes Region.Ecological Monographs 61: 145-164. Frelich, L.E., and P.B. Reich.˙ 1995a. Neighborhood effects, disturbance, and succession in forests of the western Great lakes region.˙ Ecoscience 2: 148-158. Frelich, L.E. and P.B. Reich. 1995b.˙ Spatial patterns and succession in a Minnesota southern-boreal forest. Ecological Monographs 65: 325-436. Frelich, L.E. and P.B. Reich. 1999.˙ Neighborhood effects, disturbance severity, and community stability in forests.˙Ecosystems 2: 151-166. Fujita, T.T. 1978.Manual of downburst identification for project NIMROD.˙Satellite and Mesometeorology Research Paper No. 156.˙ University of Chicago, Department of Geophysical Sciences. Heinselman, M.L. 1963. Forest sites, bog processes, and peatland types in the glacial Lake Agassiz region, Minnesota.Ecological Monographs 33: 327-374. Heinselman, M.L. 1973.Fire in the virgin forests of the Boundary Waters Canoe Area, Minnesota. Quaternary research 3: 329-407. Heinselman, M.L. 1981.Fire intensity and frequency as factors in the distribution and structure of northern ecosystems. USDA Forest Service, Fire Regimes and Ecosystem Properties, General Technical Report WO-26:7-57. Heinselman, M.L. 1996.The Boundary Waters Wilderness Ecosystem.University of Minnesota Press, Minneapolis, MN, USA. Johns, R.H. and W.D. Hirt. 1987.˙ Derechos: widespread convectively induced windstorms.˙ Weather and Forecasting 2: 32-49. Johnson, E.A. 1992.Fire and vegetation dynamics.Cambridge University Press, Cambridge, England. Leuschen, T, T. Wordell, M.A. Finney, D. Anderson, T. Aunan and P. Tin‚. 1999. Fuels risk assessment of blowdown in Boundary Waters Canoe Area Wilderness and adjacent lands. Draft document, Superior National Forest, Duluth, MN, USA. |
