Search

Search Results

• 1. [Article] Density, ages, and growth rates in old-growth and young-growth forests in coastal Oregon

  We studied the ages and diameter growth rates of trees in former Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) old-growth stands on 10 sites and compared them with young-growth stands (50-70 years ...

  Citation × Citation Citation Abstract Copy Title: Density, ages, and growth rates in old-growth and young-growth forests in coastal Oregon Author: Huffman, David W. (William), 1962-, Tappeiner, John C, Spies, Thomas A, Marshall, David, Bailey, John D We studied the ages and diameter growth rates of trees in former Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) old-growth stands on 10 sites and compared them with young-growth stands (50-70 years old, regenerated after timber harvest) in the Coast Range of western Oregon. The diameters and diameter growth rates for the first 100 years of trees in the old-growth stands were significantly greater than those in the young-growth stands. Growth rates in the old stands were comparable with those from long-term studies of young stands in which density is about 100-120 trees/ha; often young-growth stand density is well over 500 trees/ha. Ages of large trees in the old stands ranged from 100 to 420 years; ages in young stands varied by only about 5 to 10 years. Apparently, regeneration of old-growth stands on these sites occurred over a prolonged period, and trees grew at low density with little self-thinning; in contrast, after timber harvest, young stands may develop with high density of trees with similar ages and considerable self-thinning. The results suggest that thinning may be needed in dense young stands where the management objective is to speed development of old-growth characteristics. Title: Density, ages, and growth rates in old-growth and young-growth forests in coastal Oregon Author: Huffman, David W. (William), 1962-, Tappeiner, John C, Spies, Thomas A, Marshall, David, Bailey, John D We studied the ages and diameter growth rates of trees in former Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) old-growth stands on 10 sites and compared them with young-growth stands (50-70 years old, regenerated after timber harvest) in the Coast Range of western Oregon. The diameters and diameter growth rates for the first 100 years of trees in the old-growth stands were significantly greater than those in the young-growth stands. Growth rates in the old stands were comparable with those from long-term studies of young stands in which density is about 100-120 trees/ha; often young-growth stand density is well over 500 trees/ha. Ages of large trees in the old stands ranged from 100 to 420 years; ages in young stands varied by only about 5 to 10 years. Apparently, regeneration of old-growth stands on these sites occurred over a prolonged period, and trees grew at low density with little self-thinning; in contrast, after timber harvest, young stands may develop with high density of trees with similar ages and considerable self-thinning. The results suggest that thinning may be needed in dense young stands where the management objective is to speed development of old-growth characteristics. Close

• 2. [Article] The effects of natural fire and recreational disturbance on montane forest ecosystem composition, structure and nitrogen dynamics, Crater Lake National Park, Oregon

  Disturbance, whether natural or of human origin, modifies to varying degrees many ecosystem attributes. Fire is a natural process in the montane forests of southern Oregon but for much of the 20th century ...

  Citation × Citation Citation Abstract Copy Title: The effects of natural fire and recreational disturbance on montane forest ecosystem composition, structure and nitrogen dynamics, Crater Lake National Park, Oregon Author: Wilson, John K. (John Kevin) Disturbance, whether natural or of human origin, modifies to varying degrees many ecosystem attributes. Fire is a natural process in the montane forests of southern Oregon but for much of the 20th century fire was viewed as an apocalypse and thus fervently suppressed. Effective natural resource management requires an understanding of how ecosystems function, including the ecological response to the natural disturbance regime or anthropogenically altered conditions. To describe the effects of natural fire and ecosystem changes with time following fire, I examined plant composition, structure, biomass and soil nitrogen using a chronosequence of sites ranging in age from 1 to >300 years. The ecological importance of human activity is a function of the magnitude and permanence of its effects which were described at active and abandoned recreation sites. Stand-replacing fires killed virtually 100% of the overstory trees, consumed 87% of the forest floor and resulted in a 27% loss in total aboveground biomass. A pulse of mineralized nitrogen was still apparent 2 years following fire. Two years following fire, little plant colonization was established. The availability of local seed sources, particularly seedbanking species, appears to have been an important determinant of the early-successional community composition. The developing community was dominated by nitrogen-fixing species which may play an important role in the post-fire nitrogen dynamics in these ecosystems. Conifer seedling densities measured 667 individuals/ha 8 years following fire and 1714 individuals/ha 24 years after fire. Extended tree maturation and variation in composition characterized the forest structure and composition during secondary succession. Total aboveground biomass remained high and accumulation followed a general U-shaped curve. Recreational use resulted in a reduction in tree density and biomass at all active sites. The greatest reductions occurred at sites set in late-successional forests. Compositional change was also greatest at sites established in late-successional forest with increased dominance of lodgepole pine and graminoids. Within 40 years the abandoned, mid-successional sites had recovered from most of the recreational impacts; however 40 years was not sufficient for late-successional forest recovery. These results suggest that the stage of succession should be considered when assessing potential recreational impacts. Title: The effects of natural fire and recreational disturbance on montane forest ecosystem composition, structure and nitrogen dynamics, Crater Lake National Park, Oregon Author: Wilson, John K. (John Kevin) Disturbance, whether natural or of human origin, modifies to varying degrees many ecosystem attributes. Fire is a natural process in the montane forests of southern Oregon but for much of the 20th century fire was viewed as an apocalypse and thus fervently suppressed. Effective natural resource management requires an understanding of how ecosystems function, including the ecological response to the natural disturbance regime or anthropogenically altered conditions. To describe the effects of natural fire and ecosystem changes with time following fire, I examined plant composition, structure, biomass and soil nitrogen using a chronosequence of sites ranging in age from 1 to >300 years. The ecological importance of human activity is a function of the magnitude and permanence of its effects which were described at active and abandoned recreation sites. Stand-replacing fires killed virtually 100% of the overstory trees, consumed 87% of the forest floor and resulted in a 27% loss in total aboveground biomass. A pulse of mineralized nitrogen was still apparent 2 years following fire. Two years following fire, little plant colonization was established. The availability of local seed sources, particularly seedbanking species, appears to have been an important determinant of the early-successional community composition. The developing community was dominated by nitrogen-fixing species which may play an important role in the post-fire nitrogen dynamics in these ecosystems. Conifer seedling densities measured 667 individuals/ha 8 years following fire and 1714 individuals/ha 24 years after fire. Extended tree maturation and variation in composition characterized the forest structure and composition during secondary succession. Total aboveground biomass remained high and accumulation followed a general U-shaped curve. Recreational use resulted in a reduction in tree density and biomass at all active sites. The greatest reductions occurred at sites set in late-successional forests. Compositional change was also greatest at sites established in late-successional forest with increased dominance of lodgepole pine and graminoids. Within 40 years the abandoned, mid-successional sites had recovered from most of the recreational impacts; however 40 years was not sufficient for late-successional forest recovery. These results suggest that the stage of succession should be considered when assessing potential recreational impacts. Close

• 3. [Article] An examination of several methods of quantifying forest structure in headwater riparian forests of western Oregon

  Headwater streams are generally small, first-order streams that can comprise up to 80% (by stream length) of the drainage network in mountainous areas of the Pacific Northwest. These streams are intimately ...

  Citation × Citation Citation Abstract Copy Title: An examination of several methods of quantifying forest structure in headwater riparian forests of western Oregon Author: Haxton, Zane A. Headwater streams are generally small, first-order streams that can comprise up to 80% (by stream length) of the drainage network in mountainous areas of the Pacific Northwest. These streams are intimately connected with downstream reaches, serving as a source of sediment, woody debris, organic matter and nutrients. The surrounding forests can strongly influence the ecology of headwater streams by regulating microclimate, influencing nutrient and organic matter cycling process, and providing habitat for wildlife species. Central to the provision of ecosystem services by these headwater forests is their structure, the three-dimensional arrangement of trees, shrubs, snags and down wood in space. This thesis explored several methods of quantifying forest structure in headwater riparian forests of western Oregon. N-tree distance sampling, a sampling method where the nearest n trees to a sample point are measured, was compared with fixed plot and variable plot sampling for estimation of density and basal area in a simulation study using stem-mapped data taken from 8 headwater sites across western Oregon. In general, variable plot sampling gave the best statistical performance for estimation of basal area, while fixed plot sampling gave the best statistical performance for estimation of density. While n- tree distance sampling gave reasonable performance on some stem maps, the sampling method performed poorly for estimating density of populations with a clumped spatial pattern. Relative root mean square error of the best n-tree distance sampling estimator was up to 75% higher than for fixed plot sampling under these conditions. Neighborhood-based indices are a set of diversity measures that are based on the relationship between a reference tree and a certain number of nearest neighbors (i.e. trees to which it has the lowest horizontal distance). I compared the bias and relative root mean square error of several different methods of choosing reference trees for neighborhood-based index calculation, using stem-mapped data from the same 8 headwater sites. The selection of a random tree from a fixed plot gave the lowest absolute relative bias, but results with this method were quite variable, with relative root mean square error ranging from 33-67% for the species mingling index and 21-25% for the diameter differentiation index. The nearest-tree (selection of the nearest tree to the sample point) and azimuth (selection of the first tree from north in a fixed plot) methods were biased, but had lower variability, with relative root mean square error ranging from 31-64% for the species mingling index and 18-22% for the diameter differentiation index. The variable plot method gave highly variable results for all species combined, but performed well for some individual species. I close with an exploration of the use of measures of forest structure in characterizing the Trask Watershed, located in the headwaters of the northern Oregon Coast Range. The structure of the Trask Watershed is essentially even-aged, having regenerated in the aftermath of the Tillamook Burn. Douglas-fir and red alder are the most prevalent species, with western hemlock, western redcedar and noble fir locally abundant. While neighborhood-based indices were somewhat informative, they offered an impoverished picture of structural diversity in the Trask Watershed by ignoring the contribution made by tall shrubs such as vine maple and western hazelnut. Title: An examination of several methods of quantifying forest structure in headwater riparian forests of western Oregon Author: Haxton, Zane A. Headwater streams are generally small, first-order streams that can comprise up to 80% (by stream length) of the drainage network in mountainous areas of the Pacific Northwest. These streams are intimately connected with downstream reaches, serving as a source of sediment, woody debris, organic matter and nutrients. The surrounding forests can strongly influence the ecology of headwater streams by regulating microclimate, influencing nutrient and organic matter cycling process, and providing habitat for wildlife species. Central to the provision of ecosystem services by these headwater forests is their structure, the three-dimensional arrangement of trees, shrubs, snags and down wood in space. This thesis explored several methods of quantifying forest structure in headwater riparian forests of western Oregon. N-tree distance sampling, a sampling method where the nearest n trees to a sample point are measured, was compared with fixed plot and variable plot sampling for estimation of density and basal area in a simulation study using stem-mapped data taken from 8 headwater sites across western Oregon. In general, variable plot sampling gave the best statistical performance for estimation of basal area, while fixed plot sampling gave the best statistical performance for estimation of density. While n- tree distance sampling gave reasonable performance on some stem maps, the sampling method performed poorly for estimating density of populations with a clumped spatial pattern. Relative root mean square error of the best n-tree distance sampling estimator was up to 75% higher than for fixed plot sampling under these conditions. Neighborhood-based indices are a set of diversity measures that are based on the relationship between a reference tree and a certain number of nearest neighbors (i.e. trees to which it has the lowest horizontal distance). I compared the bias and relative root mean square error of several different methods of choosing reference trees for neighborhood-based index calculation, using stem-mapped data from the same 8 headwater sites. The selection of a random tree from a fixed plot gave the lowest absolute relative bias, but results with this method were quite variable, with relative root mean square error ranging from 33-67% for the species mingling index and 21-25% for the diameter differentiation index. The nearest-tree (selection of the nearest tree to the sample point) and azimuth (selection of the first tree from north in a fixed plot) methods were biased, but had lower variability, with relative root mean square error ranging from 31-64% for the species mingling index and 18-22% for the diameter differentiation index. The variable plot method gave highly variable results for all species combined, but performed well for some individual species. I close with an exploration of the use of measures of forest structure in characterizing the Trask Watershed, located in the headwaters of the northern Oregon Coast Range. The structure of the Trask Watershed is essentially even-aged, having regenerated in the aftermath of the Tillamook Burn. Douglas-fir and red alder are the most prevalent species, with western hemlock, western redcedar and noble fir locally abundant. While neighborhood-based indices were somewhat informative, they offered an impoverished picture of structural diversity in the Trask Watershed by ignoring the contribution made by tall shrubs such as vine maple and western hazelnut. Close