Short-term Effects of Timber Harvest and Forest Edges on Ground-layer Bryophytes in the Pacific Northwestern United States

Cara R. Nelson and Charles B. Halpern College of Forest Resources Box 352100 University of Washington Seattle, WA 98195-2100 crnelson@u.washington.edu We provide an overview of this work below. For full text see: Nelson, C. R., and C. B. Halpern. 2005. Short-term effects of timber harvest and forest edges on ground-layer mosses and liverworts. Canadian Journal of Botany 83:610-620.

Introduction

There is widespread interest in the ecological effects of timber harvest, and the resulting fragmentation of forest habitat and creation of edges. However, there has been surprisingly little research devoted to the effects of these activities on forest bryophytes (mosses and liverworts). Yet bryophytes contribute greatly to the biological diversity of temperate and boreal forest ecosystems. For example, >450 species of mosses and 170 species of liverworts inhabit old-growth forests of the Pacific Northwest, with roughly 20% endemic to the region or to western North America. In addition, bryophytes contribute substantially to net primary productivity, enhance retention of nutrients and moisture, and provide food and habitat for invertebrates. They are also economically important as special forest products. Thus, understanding the consequences of timber harvest and forest fragmentation for bryophytes is of critical conservation and management concern.

Bryophytes are sensitive to the physical disturbances and changes in microclimate associated with timber harvest. However, their persistence and long-term recovery may be enhanced if harvest includes the retention of undisturbed forest patches (or “aggregates”). Whether these forest remnants can serve as refugia (and as potential sources of propagules for adjacent harvest areas) depends on the degree to which they are compromised by edge effects. In this study we examine the immediate effects of disturbance and the short-term persistence of bryophytes in forest patches created by aggregated retention harvest. We pose the following questions:

• Do bryophyte species richness and abundance decline after timber harvest?
• If so, are declines correlated with harvest-related ground disturbance or elevated light levels, or moderated by the presence of residual herb or shrub cover?
• Do 1-ha forest aggregates adjacent to harvested areas provide refugia for bryophytes?
• Within these aggregates, does proximity to edge explain variation in bryophyte richness or abundance, and if so, are edge-related patterns correlated with ground disturbance, light availability, or cover of herbs or shrubs?
• Do bryophyte responses within forest aggregates vary as a function of edge orientation?

Field Methods and Analyses

Studies were conducted in the 40% aggregated retention treatments at the Butte and Paradise Hills blocks (see Study Areas and Experimental Design). Pre-treatment sampling was conducted in 1996 and post-treatment sampling in 1998. At each site, two of the five 1-ha aggregates were randomly selected (Figure 1, a and b). In each aggregate we established four 81-m-long transects (Figure 1c), extending from the center and ending 25 m into the surrounding area to be harvested. Twelve bands of permanent plots were established along each transect. Each band (Figure 1d) consisted of five, 0.1-m2 microplots (Figure 1e), within which we estimated cover of each bryophyte species. Sampling was limited to within 1 m of the forest floor. We also measured total cover of herbs and tall shrubs (Figure 1f), cover of logging slash and disturbed soil (Figure 1g), and cover of open sky (digital hemispherical images) (Figure 1h).

Figure 1.  Sampling design.  Transects originated at the center of each aggregate (b, c) and extended 25 m into the adjacent harvest area.  Twelve, 1 x 5 m bands (d) were spaced at 5- to 10-m intervals.  Each band consisted of 5, 0.1-m2 microplots (e) for sampling bryophytes and five, 1-m2 subplots (f) for sampling herbs and tall shrubs.  Harvest-related disturbance (cover of logging slash and disturbed soil) was sampled along the interior edge of each band (g), and cover of open sky at the two endpoints (h).

We used changes (post- minus pre-treatment values) in species richness, total bryophyte cover, and individual species frequency as measures of response to harvest or creation of forest edge. Statistical tests of individual species’ responses were limited to a subset of relatively common taxa. Paired t-tests were used to compare responses in forest aggregates and harvest areas, and Spearman rank correlations to test for correlations with distance to forest edge or with other habitat variables (cover of slash, disturbed soil, open sky, herbs and shrubs).

Results

Prior to treatment, a total of 30 bryophyte taxa (23 mosses and seven liverworts) were identified in sample bands.

Responses within harvested areas and forest aggregates

• After one year, harvested areas showed significant declines in bryophyte richness and total cover, but changes were small within forest aggregates (Figure 2).
• In harvest areas, declines in richness and cover correlated significantly with open sky, logging slash, and/or disturbed soil. However, declines were reduced significantly at higher levels of tall shrub cover.

Figure 2.  Mean changes (± 1 SE) in (a) species richness (number of taxa per band) and (b) total bryophyte cover in harvest areas (filled circles) and forest aggregates (open circles) 1 yr after treatment.  P values are shown where pre- and post-treatment means differed significantly based on a paired t-test (n = 4).

• In harvest areas, ~90% of taxa found before treatment declined in frequency, and of the eight taxa tested, declines were significant for five; in forest aggregates, changes in frequency were comparatively small (Figure 3).

Figure 3.  Mean frequency of individual bryophyte taxa before and after treatment in (a) harvest areas (25 taxa) and (b) forest aggregates (28 taxa).  Common taxa (those present prior to treatment in ≥3 forest aggregate/harvest area pairs and with >10% frequency) are plotted with closed circles; less common taxa are plotted with open circles.  Common taxa are:  Bra_spp = Brachythecium spp., Dic_fus = Dicranum fuscescens, Dic_tau = Dicranum tauricum, Eur_ore = Eurynchium oreganum, Hyp_cir = Hypnum circinale, Pti_cal = Ptilidium californicum, Rhi_rob = Rhytidiopsis robusta, and Sca_spp = Scapania spp.  Probabilities of significance (paired t-tests, n = 4) are coded as:  ** =  P ≤0.01 or * = 0.01< P ≤ .05.

Edge-related gradients within forest aggregates

• Within forest aggregates, declines in species richness were correlated to proximity to edge, but changes in total cover were not (Figure 4).
• Of eight taxa considered, only Scapania spp. showed a significant decline in frequency with proximity to edge.
• No significant differences in richness, total cover, or species’ frequencies were found among transects representing contrasting edge orientations (one-way ANOVAs: 0.116< P <0.963).

Figure 4.  Changes in (a) species richness and (b) total cover with proximity to forest aggregate edge.  Values represent mean differences (±1SE) (n = 16) between pre- and post-treatment measurements at each distance.  Points in the shaded region represent bands within forest aggregates.

Discussion

Our results clearly show the sensitivity of ground-layer bryophytes to logging disturbance. One year after treatment, species richness, total cover, and frequency of common taxa declined significantly within harvested areas. Harvest-related declines were greatest in areas with heavy slash, soil disturbance, and low tree cover. In the short term, burial by slash and soil disturbance, processes by which bryophytes may be immediately destroyed, appear to play larger roles in declines than does overstory removal. However, changes in microclimate associated with overstory removal may result in longer-term depression of bryophyte abundance.

In contrast to harvested areas, bryophyte communities within forest aggregates were relatively stable. Although edge environments were characterized by increased light availability and ground disturbance to a distance of 15 m (see Edge-related responses of understory plants), bryophyte cover and the frequency of common taxa did not show strong edge-related responses one year after timber harvest. However, we did detected significant edge-related declines in species richness, reflecting local loss of relatively uncommon taxa.

Implications for conservation and management

Ground-layer bryophytes appear more sensitive to the direct effects of timber harvest than do vascular plants and likely require longer periods for recovery from disturbance. Thus, management strategies specific to bryophyte conservation are necessary to ensure adequate protection for these organisms in managed landscapes. Retention of live trees may moderate the effects of clearcut logging; however, some forest-floor bryophytes may require conditions that are only met in undisturbed forest and may disappear from areas where even moderate timber extraction has occurred.

Forest aggregates may provide a temporary refugium for disturbance-sensitive species and dispersal sources for re-colonization of harvested areas once microclimatic conditions become suitable for establishment. However, the efficacy of these forest patches will depend on their size and the degree to which habitat conditions are not compromised by edge phenomena. In the mature coniferous forests that we studied, 1-ha remnant patches appear sufficiently large to buffer common ground-layer bryophytes from the direct effects of harvest, but may not be large enough to prevent declines of less common species. We expect that remnants smaller than 0.5 ha will be strongly influenced by microclimatic and disturbance effects (elevated levels of light, disturbed soil, and deposition of logging slash).

These results suggest that current standards for structural retention on federal “matrix” lands in the Pacific Northwest — which allow for aggregates as small as 0.1 ha — may not be adequate to retain the diversity and abundance of bryophyte species found in older, undisturbed forests. Longer-term studies in these and other fragmented forests will be critical for improving standards for bryophyte protection in managed landscapes.