John P. Hayes

Statistical power analysis in wildlife research

Statistical power analysis can be used to increase the efficiency of research efforts and to clarify research results. Power analysis is most valuable in the design or planning phases of research efforts. Such prospective (a priori) power analyses can be used to guide research design and to estimate the number of samples necessary to achieve a high probability of detecting biologically significant effects. Retrospective (a posteriori) power analysis has been advocated as a method to increase information about hypothesis tests that were not rejected. However, estimating power for tests of null hypotheses that were not rejected with the effect size observed in the study is incorrect; these power estimates will always be ≤0.50 when bias adjusted and have no relation to true power. Therefore, retrospective power estimates based on the observed effect size for hypothesis tests that were not rejected are misleading; retrospective power estimates are only meaningful when based on effect sizes other than the observed effect size, such as those effect sizes hypothesized to be biologically significant. Retrospective power analysis can be used effectively to estimate the number of samples or effect size that would have been necessary for a completed study to have rejected a specific null hypothesis. Simply presenting confidence intervals can provide additional information about null hypotheses that were not rejected, including information about the size of the true effect and whether or not there is adequate evidence to accept a null hypothesis as true. We suggest that (1) statistical power analyses be routinely incorporated into research planning efforts to increase their efficiency, (2) confidence intervals be used in lieu of retrospective power analyses for null hypotheses that were not rejected to assess the likely size of the true effect, (3) minimum biologically significant effect sizes be used for all power analyses, and (4) if retrospective power estimates are to be reported, then the α-level, effect sizes, and sample sizes used in calculations must also be reported.

Temporal variation in activity of bats and the design of echolocation-monitoring studies

I used Anabat II bat detectors to monitor echolocation calls of bats over two streams in the Oregon Coast Range for a total of 195 detector-nights. Activity of bats was positively correlated with biomass of insects and minimum nightly temperature, and was negatively correlated with length of night; activity levels at the two streams were positively correlated. Activity of bats was not significantly correlated with either hours of moonlight or with phase of moon. Level of activity within a night generally peaked shortly after sunset with a second, smaller peak in activity shortly before sunrise, but patterns varied substantially among nights. Total nightly activity at a site also varied substantially among nights, sometimes varying several-fold on consecutive nights. To assess the implications of temporal variation in activity of bats on sampling, I randomly sampled subsets of the data using from 2- to 12-night sample periods and calculated mean levels of activity for each subset. For subsets with seven or more nights, >60% of the subsets had means that were within 20% of the mean of the entire dataset. Less than 50% of the subsets had means within 10% of the mean of the entire dataset for any number of nights subsampled. When comparing activity between sites, use of blocked or paired designs improved sampling efficiency by 20%. Failure to account for temporal variation in activity of bats when designing research projects and monitoring programs could result in biased estimates of activity of bats.

Bat activity in thinned, unthinned, and old-growth forests in Western Oregon

Many aspects of the influences of forest management activities on bats (Chiroptera) in the Pacific Northwest are poorly known, We compared thinned and unthinned forest stands of the same age and old-growth forest stands to determine potential differences in structure and amount of use by bats. We hypothesized that activity levels of bats would differ in stands differing in structure as a result of management bistory and that activity of bats would be similar in stands of similar structure. We used automated ultrasonic detectors (Anabat II) to record calls of bats in 50-100-year-old thinned and unthinned stands, and in old-growth (200 yr old) stands in the Oregon Coast Range during the summers of 1994 and 1995. Our median index of bat activity was higher in old-growth than in unthinned stands and higher in thinned than in unthinned stands, We were not able to detect a significant difference between the index of median bat activity for old-growth and thinned stands. More than 90% of identifiable passes were identified as calls from Myotis species. The 3 stand types we examined differed in certain structural characteristics such as density and size of trees, and amount of overstory and understory cover. We concluded that the structural changes caused by thinning may benefit bats by creating habitat structure in young stands that bats are able to use more effectively.

Effects of thinning on small mammals in Oregon Coastal Forests

Because of fires and intensive logging practices, young forest stands dominate much of the landscape of the Pacific Northwest. Most young stands were reforested with Douglas-fir (Pseudotsuga menzeisii) trees at high densities. Researchers have proposed thinning of the densely stocked young stands as a means to improve habitats for vertebrates. However, effects of thinning intensity on forest-floor small mammals are not well understood. During 1994-1996, we conducted experimental and retrospective studies using pitfall trapping to assess effects of thinning intensity on abundance and reproduction of small mammals in Douglas-fir forests of the Oregon Coast Range, USA. In the experimental study, we assessed the short-term effects of thinning stands to moderate and to low tree densities on small mammals during the first 2 years following thinning. In the retrospective study, we assessed potential long-term effects of thinning by comparing relative abundance and reproductive performance of small mammals in previously thinned (7-24 years prior to the study) and unthinned stands. Among the 12 species of small mammals we examined in the experimental study, number of captures increased for 4 species and decreased for 1 within 2 years of thinning. However, responses were similar between moderately and heavily thinned stands. Among the 9 species we examined in the retrospective study, number of captures was greater for 5 species and lower for none in previously thinned than in unthinned stands. Furthermore, total number of small mammals captured was higher in previously thinned than in unthinned stands. Effects of thinning on 2 species, creeping voles (Microtus oregoni) and Pacific jumping mice (Zapus trinotatus), were consistent in the short and long term. The number of captures for both species increased in the first 2 years following thinning and was greater in stands thinned 7-24 years previously than unthinned stands. Number of western red-backed voles (Clethrionomys californicus) captured decreased within 2 years of thinning but was similar in stands thinned 7-24 years previously and in unthinned stands. Reproductive performances of deer mice (Peromyscus maniculatus) and creeping voles improved following thinning in the short term. In the retrospective study, reproductive performance of western red-backed voles was higher in thinned than in unthinned stands. Overall, thinning did not have substantial detrimental effects on any of the species we investigated and had positive effects on several. We suggest that thinning is a viable option to enhance habitat quality for several species of forest-floor small mammals in densely stocked, young Douglas-fir stands.

RESPONSE OF BIRDS TO THINNING YOUNG DOUGLAS‐FIR FORESTS

Silvicultural practices, such as thinning, are increasingly performed both for commodity production and to help achieve biodiversity goals and promote ecological restoration. However, relatively little research has examined effects of thinning conifer forests on vertebrates. We experimentally manipulated stands using a randomized block design to evaluate influences of two thinning intensities on populations of diurnal breeding birds in western Oregon. We conducted point counts of birds seven times each year in 1994 (prior to treatment) and from 1995 through 2000 (subsequent to treatment). We analyzed data using multiple linear regression and information-theoretic approaches to model selection. Of the 22 species for which we had sufficient data for analysis, detections of nine species decreased and eight species increased in thinned stands relative to controls, and there was no strong evidence that thinning influenced numbers of five species. Of the 17 species that responded to thinning, the magnitude of response of eight species varied with thinning intensity; for each of these species, response was greatest in the more heavily thinned stands. Although no species was extirpated from stands following thinning, detections of Huttons Vireos (Vireo huttoni), Golden-crowned Kinglets (Regulus satrapa), Brown Creepers (Certhia americana), Black-throated Gray Warblers (Dendroica nigrescens), and Varied Thrushes (Ixoreus naevius) decreased to less than half of the detections in controls in one or more treatment types, suggesting thinning significantly reduces their numbers. In contrast, American Robins (Turdus migratorius), Townsends Solitaires (Myadestes townsendi), and Hammonds Flycatchers (Empidonax hammondii) were rare or absent in controls but regularly present in thinned stands, and detections of Western Tanagers (Piranga ludoviciana), Evening Grosbeaks (Coccothraustes vespertinus), and Hairy Woodpeckers (Picoides villosus) increased by threefold or more in thinned stands relative to controls. Only Pacific-slope Flycatchers (Empidonax difficilis), Warbling Vireos (Vireo gilvus), and Western Tanagers showed strong evidence of temporal trends in response. For these species, differences between numbers in controls and treated stands became more extreme through time. Our findings suggest that thinning densely stocked conifer stands in landscapes dominated by younger stands enhances habitat suitability for several species of birds, but that some unthinned patches and stands should be retained to provide refugia for species that are impacted by thinning.

THE INFLUENCE OF FOREST MANAGEMENT ON HEADWATER STREAM AMPHIBIANS AT MULTIPLE SPATIAL SCALES

Understanding how habitat structure at multiple spatial scales influences vertebrates can facilitate development of effective conservation strategies, but until recently most studies have focused on habitat relationships only at fine or intermediate scales. In particular, patterns of amphibian occurrence across broad spatial scales are not well studied, despite recent concerns over regional and global declines. We examined habitat relationships of larval and neotenic Pacific giant salamanders (Dicamptodon tenebrosus), larval and adult Pacific tailed frogs (Ascaphus truei) (hereafter “tailed frogs”), and torrent salamanders (Rhyacotriton spp.) at three spatial scales (2-m sample unit, intermediate, and drainage). In 1998 and 1999, we captured 1568 amphibians in 702 sample units in 16 randomly chosen drainages in the Oregon Coast Range. We examined species–habitat associations at each spatial scale using an information-theoretic approach of analysis to rank sets of logistic regression models developed a pri...

Simulated stand characteristics and wood product yields from Douglas-fir plantations managed for ecosystem objectives

Abstract Hundreds of thousands of hectares of Douglas-fir ( Pseudotsuga menziesii [Mirb.] Franco) plantations in coastal forests in the U.S. Pacific Northwest were established over the past 40 years. Density management regimes designed to increase structural and compositional diversity in these plantations are being tested and implemented on an operational scale. These regimes are designed to promote various tree and stand characteristics, such as trees with large limbs, stands with multi-layered canopies, and dense unthinned patches. Changes in management policy associated with these types of regimes raise questions about the potential to manage for both ecosystem values and timber production. We used state-of-the-art models to simulate stand growth and wood product yields under several silvicultural prescriptions. The results indicated that timing and intensity of early thinnings are critical in determining both stand structure and wood quality. We concluded that it should be possible to manage Douglas-fir plantations to provide a high degree of structural diversity and wood products with quality similar to that grown in many industiral plantations.

Day-roosts of female long-eared myotis in western Oregon.

Roosts are a critical habitat component for bats and may influence their survival and fitness. We used radiotelemetry to investigate characteristics of day-roosts of female long-eared myotis (Myotis evotis) in watersheds characterized by different forest conditions and the spatial relationships between day-roosts and available water. We tracked 21 bats to 73 roosts (n = 102 occasions) from June to August 1996 and 1997 on the western slope of the Cascade Mountains, Oregon. Bats primarily used conifer stumps as day-roosts in watersheds dominated by younger forests and used conifer snags, and to a lesser extent conifer stumps, in watersheds with greater proportions of older forests. Individual long-eared myotis used different types of structures as day-roosts, and type of structure used did not differ with reproductive condition. Day-roosts were primarily located in upslope habitat and averaged 0.59 ± 0.03 km from available water and 0.66 ± 0.02 km from capture sites. Roosts were not located closer to available water than random points, but were closer than random points to captures sites. Conifer snags used as day-roosts averaged 34 ± 5 m in height and 93 ± 12 cm diameter at breast height (dbh); snags in intermediate stages of decay had highest use. Use of conifer snags was positively associated with the number of snags within 20 m and negatively associated with distance from stand edge. Conifer stumps used as day-roosts averaged 133 ± 9 cm in height and 59 ± 4 cm dbh. Western hemlock and Douglas-fir stumps were used more often than western redcedar stumps as day-roosts. Odds of a stump being used as a day-roost increased with increasing height of the stump (downhill side) and whether it was situated in a gap in vegetation. We contend that management of day-roosts for forest-dwelling bats should focus on maintaining large conifer snags across landscapes through space and time. In landscapes where there are relatively few large conifer snags, stumps appear to provide important, but ephemeral, roosts for long-eared myotis.

Influence of Vegetation on Bat Use of Riparian Areas at Multiple Spatial Scales

Abstract Research on habitat use by bats typically occurs at a single fine spatial scale, despite recent work demonstrating the importance of considering multiple spatial scales when investigating vertebrate habitat selection. We assessed bat use of 118 stream reaches located throughout the Oregon Coast Range, USA, and measured vegetation characteristics at 3 spatial scales surrounding each of these locations. We used an information-theoretic approach to determine vegetation characteristics most closely related to bat activity and a multilevel modeling approach to evaluate variation in bat activity at different spatial scales. Characteristics of vegetation at the finest spatial scale explained more variation in bat activity than did characteristics of vegetation at broader spatial scales, suggesting that fine-scale anthropogenic or natural disturbance events that alter cover of shrubs or trees in riparian areas are likely to influence bat habitat use. The influence of vegetation on activity varied by species of bat and appeared to operate more strongly through constraints imposed by vegetation architecture than through influences on abundance of insect prey. This diversity of responses to vegetation characteristics among bat species suggests that the best strategy for biodiversity conservation over broad spatial scales is maintenance or creation of a diversity of riparian vegetation conditions. We recommend that land managers planning to manipulate riparian vegetation strive to create diversity in shrub coverage, canopy coverage, and open space above the stream channel to promote foraging habitat for all species.

The foraging ecology of cavity-nesting birds in young forests of the northern coast range of Oregon

Cavity-nesting birds are important components of the avifauna in Pacific Northwest forests and are of special concern to land managers because of their reliance upon snags for nesting habitat. Because most research and management has focused on nesting habitat of cavity-nesting birds, the foraging ecology of cavity-nesting birds has generally been neglected. We studied the foraging ecology of the Chestnut-backed Chickadee (Poecile rufescens), Red-breasted Nuthatch (Sitta canadensis), Brown Creeper (Certhia americana), and Hairy Woodpecker (Picoides villosus) in 30- to 45-year-old, Douglas-fir (Pseudotsuga menziesii) forests of the Coast Range of Oregon. All four species were selective in their use of foraging substrates. Hardwoods were selected over conifers by Chestnut-backed Chickadees and Hairy Woodpeckers. In relation to randomly chosen live trees, Chestnut-backed Chickadees, Red-breasted Nuthatches, and Hairy Woodpeckers selected trees that were larger in diameter, and Brown Creepers selected trees with deeper furrows in the bark. Large diameter, heavily decayed snags and logs were selected for foraging by the Hairy Woodpecker. Foraging needs should be taken into account when managing habitat for cavity-nesting birds.