Ecological Comparisons Between Wildfire and Logging -
The East Kootenay Songbird Project:
Interim Report

by
Kari Stuart-Smith, Forest Wildlife Ecologist,
Crestbrook Forest Industries / Department of Forest Science, Oregon State University

Introduction

Historically, frequent fires have played a major role in shaping the forests of the Invermere Forest District. Emulating natural disturbances such as fire is advocated in the Biodiversity Guidelines (BC Environment 1995) as a method to maintain biodiversity in managed forests. How many biological communities respond to fire, however, and which elements of fire regimes are important to emulate to maintain these communities is poorly understood.

Fires are partially characterized by the structural elements they leave behind, including live and dead trees, down logs, and understory vegetation (Swanson et al. 1993). These structural legacies add complexity to burned stands and provide habitat for many species (e.g., Ruggerio et al. 1993, Stelfox 1995). Stands intensively managed for wood production, however, are often structurally simple and support lower levels of vertebrate diversity (Hansen et al. 1991).

Consequently, guidelines have been established in British Columbia for the amounts of structure (in the form of wildlife trees and wildlife tree patches) to be left after harvest (Biodiversity Guidebook, BC Environment 1995). Yet, there is little information on residual structure in naturally-disturbed forests to provide an ecological baseline for the amounts, types, and patterns of structural features that should be retained after harvest (but see Stuart-Smith and Hendry 1998). Nor is wildlife response to different amounts and distributions of residual structure well understood. Whether the stand structure requirements specified in the guidelines are adequate to maintain biodiversity is unknown.

With funding support provided jointly by the EFMPP, Crestbrook Forest Industries Ltd. and Slocan Forest Products, the "East Kootenay Songbird Project" was initiated to compare the effects of logging and wildfire on songbirds, in relation to the type, amount, and distribution of structure left by the disturbance in the EFMPP study area.

Because they contain a wide variety of species with diverse habitat requirements, songbird communities provide a good measure of the structural diversity of a forest stand. In addition, many species of neo-tropical migrant birds (those that winter in central and South America) are thought to be declining due to loss or alteration of habitat, and are blue-listed in British Columbia. Understanding their habitat requirements and how they respond to disturbance is necessary to develop strategies to maintain their populations.

Recent studies have found that retaining live canopy trees in harvested areas can increase the number of bird species and the abundance of these species over that found in clearcuts (Hansen and Hounihan 1995, Norton and Hannon 1997, Schieck et al, submitted). However, these studies reported only the number of species and their abundance, and did not examine their reproductive success.

The long term viability of bird populations in managed forests requires successful reproduction. Currently, the reproductive success of songbirds in harvested areas relative to their success in naturally-disturbed areas is unknown.

Objectives

The main objectives of this project are to:

• Determine if logged and burned (by wildfire) stands of similar age and severity support similar songbird communities within the EFMPP study area. i.e., does a clearcut emulate a severe fire, or a partial cut a partial burn, in terms of the bird community? And how does this relationship change through time? i.e., do the bird communities in logged stands become more or less similar to those in burned stands as the stands age?
• Evaluate the influence of increasing numbers of residual trees in harvested areas on songbirds. Will leaving greater numbers of residual trees increase the numbers of birds using and breeding in the blocks? Will any old-growth associate species use cutblocks with high numbers of residual trees?
• Compare the reproductive success of open cup nesting birds in burned and logged areas, in relation to the number of residual trees.
• Evaluate the current stand structure guidelines in the biodiversity guidebook and provide management recommendations for the amount, type, and spatial distribution of stand structure that should be left in harvested stands to emulate fire patterns and maintain habitat quality for forest songbirds.
• Provide inventory data on forest birds in two of the main biogeoclimatic variants in the EFMPP study area (MSdk, ESSFdk), and determine the relationship between site series and avian diversity and abundance. To date, there have been no studies of songbird communities in the burns and harvested stands in these variants in the EFMPP study area.

Methods

Songbird Communities

In May 1997, 174 point count sites were established in logged and burned stands in the MSdk and ESSFdk variants of the EFMPP study area, more specifically in the White and Lussier River watersheds in the Rocky Mountains (see Braumandl and Curran 1992 for a description of biogeoclimatic variants in the Nelson Forest Region).

These sites were randomly selected from a pool of over 700 potential sites identified from aerial photographs and forest cover maps, which had been stratified into two disturbance classes (burn/log), 3 age classes (5-15, 16-30, 31-50 yr.), and 4 severity classes (0-10, 11-30, 31-60, 61-100 % live trees remaining after disturbance).

In May, 1998 an additional 34 sites were randomly selected from 100 new potential sites in the study area and in nearby Kootenay National Park.

In both 1997 and 1998, three point counts were conducted at each site between late May and early July. Counts lasted 12 minutes and were conducted from sunrise to 9:30 a.m. All birds seen or heard within a 150 m radius of the centre point were recorded and the distance to each bird estimated to enable density calculations for each species. Counts were not made during periods of wind or rain, because of the increased difficulty detecting birds under these conditions.

Stand Structure

Because each site covers nearly 7 ha and is not entirely homogeneous, four 20 m radius plots were established at each site. One plot was centred at the site centre (where the bird point counts were conducted), the other three centred 60 m away at 0, 120, and 240 degrees around the site centre. The maximum distance included in the plots was thus 80 m from the site centre, which covered the area in which 75 % of all bird detections were made.

Within each plot, the number, dbh class and species of all residual live trees was recorded. The species, dbh, height and decay class of all snags was noted, along with any observations of nest cavities or woodpecker feeding sign. Smaller sub-plots were established within the 20 m radius plots to record the dbh class and species of all seedlings and saplings, as well as, the percent cover and species of shrubs and herbs.

The total number of vertical vegetation layers and the percent cover of each was also recorded. Down woody debris was counted along transects laid out from the plot centre.

Finally, the elevation, aspect, slope, and slope position were recorded for each site. The site series was determined for a sub-set of the sites.

Songbird Reproductive Success

The reproductive success of open cup ground and shrub nesting songbirds is being examined through both experimental and observational approaches.

In 1998, artificial grass nests baited with plasticine and button quail eggs were placed out in over 30 logged and burned sites - half of these with low numbers of residual trees, the other half with high numbers of residual trees. Fifteen ground and 15 shrub/small tree nests were randomly placed out per site. Nests were checked at 6 and 12 days after placement for signs of predation. Nest predation rates will be compared between logged and burned sites and between sites with low and high residual trees.

To identify the type of predators from the marks left in the plasticine eggs and the eggshell remains, cameras equipped with triggers were set out in various locations. The eggs were placed on an electronic treadle such that when they were disturbed, a picture was taken. Teeth or beak marks were then matched with the photos and predators identified.

Because artificial nests may not always reliably represent patterns and rates of predation on actual nests (e.g., Storass 1988), nests of open cup ground and shrub nesting birds were located in logged and burned areas with low and high numbers of residual trees.

Over 60 nests have been found, mainly of dark-eyed junco, dusky flycatcher, warbling vireo, white-crowned sparrow and yellow-rumped warbler. Nests were monitored every 3-4 days until either the nest failed or the young were fledged. Reproductive success will be measured as daily survival rate of nests, number of young fledged per nest, and number of young fledged per successful nest, and then compared among all sites.

Future Work

Because avian populations often exhibit high annual variation, point counts will be repeated for a third year in 1999. New sites may also be established in old forests (> 140 yr.) in both the MSdk and ESSFdk variants to determine if stands with high amounts of residual structure support avian communities similar to those in old forests.

Work on reproductive success will also be repeated in 1999 to increase the sample size of real nests.

In addition, each point count site will be located using a GPS. Then performing GIS analysis through the EFMPP, landscape metrics such as the percentage of forest harvested, distance to nearest continuous forest edge, proportion of the forest in each seral stage, and amount of edge will be calculated for circles of varying radii around each site. The relative contribution of landscape vs. stand structure variables to the songbird diversity and abundance at each site will then be determined.

Finally, the site series will be determined for all sites and the relationship between site series and songbird communities elucidated.

Management Implications

Currently, the ecological consequences and management implications of the differences between wildfire and logging are poorly understood. Results from the East Kootenay Songbird Project will reveal how logged and burned stands in the MSdk and ESSFdk in the EFMPP study area differ in terms of their structure and songbird communities.

Recommendations detailing what type, how many, and in what distribution structural features should be retained in harvested areas for them to emulate naturally burned stands more closely will be offered.

In addition, results from this project can also be used to evaluate the stand structure guidelines outlined in the biodiversity guidebook, and provide a solid ecological basis to refine them if needed.

Since management objectives vary for different areas due to factors such as past management history and specific species concerns, approaches to forest management other than emulating natural disturbance may be selected for some areas. Data from this project will indicate how the various species of forest songbirds respond to different levels of residual tree retention and different stand-tending practices, and thus help forest managers determine practices coincident with their objectives.

This study is also testing whether the site series classification is a useful predictor of forest bird communities. This information will assist forest planners and biologists in using site series for forest planning decisions.

Project Partners

Invermere Pilot Project, Crestbrook Forest Industries Ltd., Slocan Forest Products - Radium Division, Oregon State University College of Forestry, Wings Over the Rockies Bird Festival, Kootenay National Park, and the National Science and Engineering Research Council of Canada.

References

BC Environment. 1995. Biodiversity Guidelines. Forest Practices Code of British Columbia.

Braumandl,T.F. and M.P.Curran. 1992. A field guide for site identification and interpretation of the Nelson Forest Region. BC Ministry of Forests, Research Branch, Victoria, British Columbia.

Hansen, A.J. and P. Hounihan. 1995. Canopy retention and avian diversity in the Oregon Cascades. in Szaro,R. Johnson,D. (eds.) Biodiversity in managed landscapes: Theory and practice. Oxford University Press, London.

Hansen, A.J. et al. 1991. Lessons from natural forests: Implications for conserving biodiversity in natural forests. Bioscience 41:382-392.

Norton,M. and S.J.Hannon. 1997. Songbird response to partial-cut logging in the boreal mixedwood forest of Alberta. Canadian Journal of Forest Research 27:44-53.

Ruggerio, L.F. et al. 1993. Wildlife and vegetation of unmanaged Douglas Fir forests. Gen. Tech. Rep. PNW-GTR-245. Pacific Northwest Research Station, USDA Forest Service.

Scheick, J., K. Stuart-Smith, and M. Norton. Residual trees and snags affect bird communities in harvested areas. Submitted for publication August 1998.

Storaas, T. 1988. A comparison of losses in artificial and naturally occurring Capercallie nests. Journal of Wildlife Management 52:123-126

Stelfox, J.B. (ed.) 1995. Relationships between stand age, stand structure, and biodiversity in aspen mixedwood forests in Alberta. Alberta Environment Centre, Vegreville, AB. 308 pp.

Stuart-Smith, A.K. and R. Hendry. 1998. Residual trees left by fire : final report. Enhanced Forest Management Pilot Project Report No. 7, Invermere Forest District, Box 189, Invermere, British Columbia.

Swanson, F.J. et al. 1993. Natural variability - Implications for ecosystem management. pp. 80-94 in Ecosystem Management: principles and applications. Volume II. USDA FS Gen. Tech. Rep. PNW-GTR-318, Portland, OR.

Vega, R.M. 1993. Bird communities in managed conifer stands in the Oregon Cascades: habitat associations and nest predation. M.S. thesis. Oregon State University. Corvallis, OR.