Source: UNIVERSITY OF MICHIGAN submitted to NRP
TERRITORY SIZE AND ESTABLISHMENT IN BREEDING FOREST SONGBIRDS: IMPLICATIONS FOR FOREST MANAGMENT AND CONSERVATION
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
0207649
Grant No.
(N/A)
Cumulative Award Amt.
(N/A)
Proposal No.
(N/A)
Multistate No.
(N/A)
Project Start Date
May 1, 2005
Project End Date
Oct 31, 2005
Grant Year
(N/A)
Program Code
[(N/A)]- (N/A)
Recipient Organization
UNIVERSITY OF MICHIGAN
(N/A)
ANN ARBOR,MI 48109
Performing Department
RESOURCE ECOLOGY & MANAGEMENT
Non Technical Summary
Understanding the relationship between birds and their habitat is necessary to assess the effects of forestry management on the size and health of breeding songbird populations. The goal of this project is to improve our ability to estimate population densities and predict habitat choice; we will accomplish this by developing better methods to measure breeding territory size and considering the relationship between territory placement and the structure of forest vegetation.
Animal Health Component
75%
Research Effort Categories
Basic
25%
Applied
75%
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
1230613310030%
1230680310040%
1237310209030%
Goals / Objectives
We are developing methods that rigorously measure the size, shape, and boundaries of breeding territories defended by forest songbirds and to compare territory sizes across a gradient of management history. Improving techniques for delineating and describing territories within forests is a key step toward clarifying the relationship between vegetation structure and patterns of songbird habitat occupancy. Our study will be of immediate use to forest land managers who are working to improve stand composition and structure for the maintenance of viable songbird populations; the methods we are developing will increase manager's ability to measure and predict bird densities in heterogeneous landscapes. This research has two primary aims: (1) to develop sampling methods that accurately estimate territory size, and (2) to better understand the micro-habitat features that affect space use and breeding densities of songbirds in northern hardwood forests. In the period covered by this proposal, we are laying the groundwork for developing a comprehensive evaluation of how the management history of hardwood forests affects the distributions of two indicator bird species. We are currently focusing on collecting temporally intensive location measurements on a small number of black-throated blue warblers Dendroica caerulescens, as well as recording locations for a large number of American redstarts Setophaga ruticilla on twelve intensively studied survey plots. We will consider how territory size depends on vegetation characteristics, behavior, habitat requirements, and time in the breeding season. Future work will use our new methods to assess patterns of songbird territory size and distributions across a wide range of forest conditions. This research will allow managers to better predict the number of breeding birds that different managed forest stands can support, information that is critical for managing forest landscapes to meet long-term wildlife conservation goals.
Project Methods
Field methods. We will collect data on plots in the eastern Hiawatha National Forest in Michigan's Upper Peninsula. All forest plots are in even-age, managed hardwood forests and provide a range of understory and groundcover characteristics, overstory densities, and species. The territory locations and reproductive success of black-throated blue warblers have been recorded on these sites since 1998 and of American redstarts since 2004. In summer 2005, we will make intensive observations of 6 black-throated blue warblers and will also collect data on 24 American redstarts. The current fieldwork has 2 components: 1. Habitat data. Vegetation maps will be updated and supplemented with data on vertical structure of the understory. We will also trap flying insects and conduct leaf surveys; 2. Territory data. We are mapping breeding territories of a representative sample of males by recording locations where banded males sing, counter-sing, or aggressively interact with neighbors. These data are being used to measure territory size, shape, and spatial distribution, and are being combined with observations of blue warblers from 2002-4 and Hall's previous observations to consider temporal changes in territory size for returning birds and to explore the effects of forest stand characteristics, bird density, and demographic parameters, such as bird age, on territory size. Analytical methods. We are using the ADE-Habitat functions in the R-language to compare methods of territory estimation (nonparametric kernel estimation/minimum convex polygon). We are focusing on 3 issues in the estimation of territory sizes: (1) the best field methods to collect good territory maps, (2) the optimal number of locations needed for each species - considering both bout length and number bouts per bird and (3) the appropriate analytical methods for determining territory boundaries from field maps. We are addressing the inherent autocorrelation in the data and explore rules for excluding/including locations. Our approach includes simulation modeling and bootstrapping to improve territory size estimates and considers the sensitivity of territory size estimates to the estimation method and sample size. Once appropriate territory boundaries and sizes are determined, we will analyze on the relationship between vegetation structure and territory location for both species. The analyses will identify vegetation characteristics, such as percent cover, overstory and understory species composition, density, and height, which are likely to be important to each species. Using our habitat dataset, we will compare vegetation variables within territories to random locations outside territories; we will also employ boundary analysis to identify the vegetation structures marking territorial boundaries, such as the edges of dense understory patches. These analyses will suggest habitat variables that managers may wish to monitor or promote. We will also compare territory size in areas of high and low deer browse. Our work will provide important information on how forest management and deer densities interact to affect the distribution and territory sizes of breeding songbirds.

Progress 05/01/05 to 10/31/05

Outputs
OUTPUTS: Our analyses comparing the 2003 and 2004 sampling methods, applied to black-throated blue warblers, Dendroica caerulescens, and American redstarts, Setophaga ruticilla, demonstrated that fixed interval sampling provides a more comprehensive method for data collection and consistent territory measurement than continuous sampling methods. In particular, use of continuous methods resulted in sample sizes that varied substantially between individual birds. Using bootstrap analysis, we determined that the accuracy and precision of territory sizes estimated with both kernel density procedures and minimum convex polygon methods are sensitive to sample size differences. An important goal of sampling design should be to attain similar sample sizes per bird. Based on our bootstrapping analysis, we determined that larger sample sizes (numbers of locations recorded for each bird) were needed to fully evaluate the effect of sample size on territory size estimation. To achieve the necessary sample sizes, we planned for more intensive sampling of fewer birds in 2005 and developed a simulation model based on the 2003 behavioral data in order to take arbitrary large "samples" for individual birds. Preliminary results indicated that the model realistically mimicked bird movements and the resulting locations produced reasonable territory sizes, when compared to actual 2004 territories. Result of initial model runs highlighted the critical importance of bandwidth estimation to any assessment of the effect of sample size on territory measurement with the kernel-based measure. Thus, we began an investigation of bandwidth estimation methods. For the small number of total birds sampled, we have not detected evidence that microhabitat characteristics related to management history and bird age affect territory size in black-throated blue warblers (unpublished data). In a related forest stand-level study focused on heavily-deer browsed sites that varied primarily in understory vegetation characteristis, we found that no blue warblers were detected in the low understory density sites, and abundance was significantly higher on plots with greater than 20% balsam fir (moderate and high fir sites; Kearns et al. 2006). Following the 2004 field season, additional analyses were also underway using data collected in 2003 (and from 1998-2002) to understand the relationship between territory placement and habitat characteristics, but we do not yet have enough samples collected with our improved methodology to conduct the suite of tests of effects on territory size described above. We received a grant from the US Fish and Wildlife Service to explore habitat selection by American redstarts; territory measurement and comparison of redstart territory sizes in areas of high and low conspecific density will be considered, as will the effect of vegetation differences between study plots. Baseline data from the current study will be used to inform the sampling methods of this closely related work. PARTICIPANTS: Nothing significant to report during this reporting period. TARGET AUDIENCES: Nothing significant to report during this reporting period. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
The data we are collecting on forest birds and vegetation structure can serve as baseline information for further studies in the region. We have just completed another associated project that provides recommendations for forest understory management. In addition, we monitor and record species of special interest to the Forest Service biologists during data collection and field activities. We will continue our collaborations with the Forest Service during our third field season and provide them with updates of our results, as well as access to our data.

Publications

  • Hall, K.R. Comparing geographic boundaries in songbird demography data with vegetation boundaries: A new approach to evaluating habitat quality. Ecological and Environmental Statistics. 2008, vol. 15, no4, pp. 491-521.
  • Kearns, L.J., E.D. Silverman, and K.R. Hall. Black-throated blue warbler and Veery abundance in relation to understory composition in northern Michigan forests. Wilson Journal of Ornithology. Vol. 118, No. 4 (Dec., 2006), pp. 461-470