Graduation date: 1970

The diverse lichen flora of the Pacific Northwest is being impacted by population growth and by forest management practices. Accumulating information about our lichen flora will improve our conservation strategies. This dissertation first collects information to improve our understanding of how lichen communities vary among forests of differing structure, and across the western Oregon landscape. It then proposes a method to predict species occurrence in unsampled sites by utilizing the information on forest characters and environmental gradients at sampled sites. Macrolichen communities sampled in coniferous forests revealed that old-growth stands (>200 yrs old) harbored communities that differed from those in young forests (50-110 yrs old). Even more atypical communities occurred in macrolichen hotspots, which were primarily in riparian zones. Many macrolichen species were associated with these hotspots, including numerous nitrogen-fixing cyanolichens. Macrolichen species associated with old-growth forested plots included the nitrogen-fixing lichen Lobaria oregana and several forage-providing alectorioid lichens. The presence of remnant old trees apparently increased the occurrence of old-growth associates in young stands. The calicioids, a group of microlichens investigated only in the Cascades, had a strong association with old growth forest and remnant trees. Diversity of calicioids may also be increased by legacy structures such as old snags and wolf trees. These structures increase continuity between current and previous stands. Macrolichen communities varied between the Coast and Cascade Mountain Ranges, following climatic gradients, particularly annual precipitation. Successional patterns in macrolichen communities appeared to differ between the mountain ranges. The modeling method proposed for using habitat associations to predict occurrence has several advantages over common modeling methods, such as regression. The method is simple, avoids parametric assumptions, provides easy updating of models as additional sites are sampled, and automatically accounts for interactions among predictor variables. It can be linked with GIS data and software to map estimated probability of occurrence across landscapes. The data on calicioids from the Cascades, supplemented with additional stand inventories, were used to test and demonstrate the modeling method. / Graduation date: 2001

In an initial research proposal of December 1969, the scientists of the Coniferous Forest Biome (CFB), an ecosystem study centered in the Pacific Northwest and part of the larger International Biological Programme (IBP), expressed optimism that computer simulations and systems modeling could transform empirical knowledge of the carbon, water, and nutrient flows turned into viable forest management practice. The CFB's strategy aimed to use projections of the computer simulations and data from field study to constantly check and direct each other, resulting in a flexible, refined, and accurate understanding of forest ecosystems, as well as a reliable guide to forest management. To what extent did the CFB's research program, centered on a total system model, complete its cycle of field study, modeling, and validation? Despite the innovative strategies of the CFB modelers, ecosystem modeling lost its preeminent status among the goals of the CFB, due to different interpretations of the purpose and philosophy of ecosystem modeling and the practical limitations of administering a large research program. Instead, small field-based studies during the CFB yielded a number of ground-breaking discoveries. Although they diverged from the modeling objectives, these areas of fieldwork emerged from questions the forest's functions and cycling processes that the modeling efforts of the CFB required. Focusing on the work of CFB participants from Oregon State University and the USDA Forest Service in the H. J. Andrews Experimental Forest, this thesis addresses the relationship between the marginalization of the modeling objectives and the rising centrality of field-based forest studies in the CFB from 1969 to 1980. Given the ongoing legacy of CFB research at the Andrews Long-Term Ecological Research (LTER) site and the later implications of CFB findings in debates over forest policy and management, this thesis also seeks to evaluate the Coniferous Forest Biome as a whole and discuss the role of modeling and field work within large ecological research endeavors more generally. / Graduation date: 2006

The effect of contiguous forested habitat area on local avian diversity and species richness in the Tualatin Mountain area of northwestern Oregon was investigated. Observations of eight forested stands representing seven area values (1, 2, 7, 14, 18, 24 and 40 hectares) were made during the spring and summer of 1991 and 1992. The variables measured were chosen in an attempt to show possible relationships between vegetation factors, spatial patterns and bird communities. Kendall's rank correlation coefficients were used to analyze the data. Avian species richness and diversity were significantly correlated with forest stand (patch) size. The only significant correlation between avian species richness and diversity and vegetation measures was with percent shrub layer cover. It seems likely that avian diversity and richness are increased due to the presence of species that can utilize the interior and edges of forest stands along with species which depend upon true forested interior. When forested patch size drops below a critical area, the patch becomes all edge. Interior species are absent due to increased predation and the inability to compete with interior-edge species. Edge effect may be a contributing factor to variation in diversity of birds. The correlation of percent shrub layer cover with avian measures is accompanied by a correlation of percent shrub layer cover with distance from edge. This suggests further investigation is required to assess this relationship. Studies conducted in the northeastern and north central United States have shown a similar relationship between bird communities and forest patch size.

Intermittent, headwater streams recently have been recognized as important components of forest ecosystems and have been provided increased protection by the Northwest Forest Plan. However, few studies have examined their distribution, dynamics, and ecological roles, such as habitat for wildlife. My goal was to provide additional information on the ecology of intermittent streams in the Pacific Northwest. I examined and compared hydrologic, water quality, and physical characteristics of 16 intermittent streams in old-growth and young forest stands in the central Cascade Range in western Oregon. I documented amphibian communities and habitat associations in these streams during spring and summer. I used comparisons of current habitat conditions and amphibian communities between stand types to gain insight into potential impacts of timber harvesting on these stream systems. Of the streams surveyed in old-growth and young forest stands, relatively few (23%) were designated as intermittent based on my definition which included presence of a definable channel, evidence of annual scour and deposition, and lack of surface flow along at least 90% of the stream length. Intermittent streams in old-growth stands exhibited the following characteristics: (1) annual flow pattern in which streams started to dry in May and June and were mostly dry by July; (2) lengthy annual flow durations (range 6-11 months); (3) cool and stable daily stream temperatures; (4) primarily coarse substrates, such as cobbles and pebbles; (5) streamside vegetation comprised of predominantly coniferous overstories, and plant species associated with uplands or dry site conditions, such as Oregon-grape and salal, as well as riparian areas or wet site site conditions, such as Oregon-grape and salal, as well as riparian areas or wet site conditions, such as red alder, oxalis, red huckleberry, and vine maple (Steinblums et al. 1984, Bilby 1988); and (6) low to moderate densities of large wood, mostly moderately- and well-decayed. Study streams in young forest appeared to dry about one to two months later than the streams in old growth but had similar annual flow durations. They also were characterized by higher daily stream temperatures, similar diel fluctuations, finer substrates, more deciduous overstory and herbaceous understory cover, and lower densities of moderately-decayed large wood. Differences in habitat conditions between stand types may be attributed to timber harvesting as well as discrepancies in physiographic and geological factors, such as elevationgradient, and soil type. Amphibian communities in spring and summer were comprised primarily of the Cascade torrent salamander (Rhyacotriton cascadae), Dunn's salamander (Plethodon dunni), and Pacific giant salamander (Dicamptodon tenebrosus). Amphibian communities in streams in young forest stands exhibited different species composition and seasonal patterns in total density from those in old growth. Cascade torrent salamanders and Dunn's salamanders maintained similar densities and biomass between spring and summer by potentially adopting drought avoidance strategies. Species differed in their use of habitat types and associations with habitat features. In general, amphibian species were positively correlated with percent surface flow, water depth, intermediate-sized substrates and negatively associated with overstory canopy cover, elevation, and wood cover. Results of my study suggest that intermittent streams may warrant protection for their potential effects on downstream habitat and water quality and for their role as habitat for aquatic species, such as amphibians. Streamside vegetation should be maintained along intermittent channels to provide shade protection for water temperature regulation and sources of large woody debris and other allochthonous energy input, to help stabilize slopes, and to minimize erosion and sedimentation. At a minimum, intermittent stream channels should receive protection from physical disturbance during timber harvesting operations. However, since intermittent stream systems are highly variable, management should address individual site conditions and vary accordingly. / Graduation date: 1997

Graduation date: 1992

Monitoring wildlife habitats has become important to forest ecosystem management because it provides valuable information about the response of forests and their species to harvest practices, impacts from recreational use, conservation efforts, and natural and human-caused disturbances. Monitoring is a complex task that requires a variety of abiotic and biotic measurements and decisions about what should be measured, and when and where measurements should be taken. Riparian habitats contain unusually high diversity and are important to land managers. Wildlife assessments of riparian areas have focused on vertebrate species such as amphibians, birds, and mammals, but have largely ignored the arthropod components of the habitats. Arthropods constitute over 85% of all species and posses characteristics that make them valuable for tracking environmental changes. The purpose of this study was to gather site-specific data about epigaeic, riparian beetle community composition of the H.J. Andrews Experimental Forest (HJA). The patterns of beetle distribution, abundance, and diversity were analyzed and the results were used to characterize and compare the riparian and adjacent upslope beetle communities. Almost 8,000 beetle specimens representing about 250 species were collected from 141 pitfall traps placed along 10 transects in 3 different channel morphologies along Lookout Creek in the HJA. Traps were opened during six 30-day sampling periods over 2 years. Riparian and adjacent upslope beetle communities had high diversity measurements. The average difference of the calculated Simpson's Diversity Index between the two communities was 0.0116 and represented about 1% of the average riparian diversity. Analysis of species-curves indicated that the riparian habitats contained a higher total number of species. Multivariate Principal Coordinate Analysis indicated that the two habitats had distinctly different beetle communities. Multigroup Discriminant Analysis correctly classified 89.7% of the sampling units as the habitat group into which they were assigned a priori. Detailed recommendations for monitoring riparian habitats were discussed. / Graduation date: 2000

This study describes the composition of forest landscapes surrounding northern spotted owl (Strix occidentalis caurina) nests in the central Cascade Mountains of Oregon. I compared forest composition around 126 owl nests in 70 pair territories with forest composition around 119 points drawn randomly from all terrestrial cover-types, and around 104 points drawn randomly from the old-forest (closed canopy, > 80 yrs) cover type. All nest sites and random points were drawn from U.S. Forest Service lands and were not drawn from privately owned lands or Wilderness Areas. Forest cover was classified on a Landsat Thematic Mapper image. I quantified the percentage of old-forest within 200 concentric circular plots (0.04-5.0-km radii), centered on each analyzed point, using a geographic information system. I used logistic regression to make spatially-explicit inferences. Owl nests were surrounded by more old-forest when compared to points drawn randomly from all terrestrial cover types: there was significantly (P<0.05) more old-forest around the owl nests in plots as large as 1.79 km in radius. When compared to points drawn randomly from the old-forest cover type, owl nests were surrounded by significantly (P<0.05) more old-forest in plots with 0.17-0.80-km radii. Exploratory analyses suggest that the landscape scales most pertinent to northern spotted owl nest site positioning in this study area appear to be (in descending order): the surrounding 10-15 ha (~200-m radius), the surrounding 25-30 ha (~300-m radius), the surrounding 200 ha (800-m radius), and possibly the surrounding 700 ha (1,500-m radius). This study supports the assertion that northern spotted owls are strongly associated with older forests. The results also indicate that owl nests are most associated with higher proportions of old-forest near the nest implying that the arrangement of habitat is important for nest-site selection/positioning Since spotted owls in the central Cascade Mountains of Oregon are known to have home-ranges that average 1,769 ha, it is important to recognize that these results apply to nest-site selection/positioning on the landscape and not to the amount of habitat necessary for pair persistence or successful reproduction. / Graduation date: 1998

Ten species of bats occur in the Oregon Coast Range and are hypothesized to be associated with late-successional forests. The development of characteristics of late-successional forests in young forest stands can be accelerated through silvicultural practices such as thinning I examined the effects of thinning on the use of forests by bats in the Oregon Coast Range. I used automated ultrasonic detectors to record bat calls in 50- to 100-year-old thinned and =thinned stands as well as in old-growth (2200-year-old) stands in 11 sites in the Oregon Coast Range during the summers of 1994 and 1995. I compared bat activity levels among the 3 stand types. In addition, I classified bat calls into 1 of 5 species groups: Eptes/Las, Myev/vo, Myyu/ca, MythCory, and Mysp. I measured selected vegetation and environmental variables in conjunction with bat activity. I also compared bat activity on roads with activity in the stand interior at 1 site. Bat activity was higher in old growth than in young stands, and higher in thinned than in =thinned stands in 1995 and over both seasons combined. I did not detect a difference in bat activity among stand types in 1994, until I removed 1 site from the analysis. The Mysp and MythCory species groups exhibited differences among stand types. Bat activity along roads was higher than activity within stands. Tree density, tree diameter, tree height, shrub cover, and shrub height varied significantly between old-growth and young stands. Tree density, tree diameter, shrub cover, canopy cover, and crown height varied significantly between thinned and unthinned stands. Bat activity, overall or by species group, was significantly related to structural variables, including mean snag diameter, mean distance from the detector to snags, and percent shrub cover. My results suggest that bats are sensitive to stand structure and that silvicultural practices, such as thinning, which promote development of structural characteristics found in old-growth stands, would benefit bat populations. Further study is needed to clarify the habitat preferences of separate bat species and to specify habitat elements required by bat species. / Graduation date: 1997

This thesis examines factors limiting understory herb presence and flowering in young second-growth Douglas-fir (Pseudotsuga menziesii) forests on the west side of the Cascade Mountains, Oregon, USA. I studied the belowground effects of canopy trees on understory herbs and shrubs in old-growth forests using trenched plots from which tree roots were excluded. Effects of tree density and stand age were tested by comparing the understory community composition of old-growth stands and pre-commercially thinned and unthinned young second-growth stands. I also examined the effect of conifer basal area on understory herb presence and flowering within one young second-growth watershed. In young stands, I focused on three groups of understory herb species: disturbance-responsive (release), forest generalist and old-growth associated. The effects of root trenching on vegetation and soil moisture were tested in closed-canopy and gap locations in two old-growth Douglas-fir (Pseudotsuga menziesii) forests. Ten years after installation, trenched plots averaged 92% total understory cover while untrenched plots averaged 47% cover. Trenched plots under closed canopies were moister than control plots throughout the growing season; the trenching effect on soil moisture became apparent in the generally wetter gaps only at the end of the growing season. Vegetation responses to trenching were concomitantly larger under closed canopies than in gaps. Stands that had been pre-commercially thinned 20 years earlier exhibited understory composition more similar to old growth than did unthinned stands. Thinned stands exhibited higher frequencies, abundances and density of flowering of old-growth associated herbs than did unthinned stands, but lower than did old-growth stands. Forest generalist and release species showed mixed responses to thinning. I used both general linear models and classification and regression tree models to explore the association of herb species presence and flowering with conifer basal area and abiotic variables. Both modeling approaches yielded similar biological insights. Flowering was more sensitive than presence to current stand basal area. Flowering of old-growth associated and release species was negatively correlated with conifer basal area. Linear models allowed clearer hypothesis tests, while tree-based models had greater explanatory power and provided information about interactions between variables. / Graduation date: 2004