Grantee Research Project Results

2001 Progress Report: Multi-scale Effects of Forest Fragmentation and Landscape Context on Population Health of Birds

EPA Grant Number: R827673
Title: Multi-scale Effects of Forest Fragmentation and Landscape Context on Population Health of Birds
Investigators: Martin, Thomas E. , Redmond, Roland L.
Institution: University of Montana
EPA Project Officer: Packard, Benjamin H
Project Period: October 1, 1999 through September 30, 2002 (Extended to September 30, 2003)
Project Period Covered by this Report: October 1, 2000 through September 30, 2001
Project Amount: $659,331
RFA: Regional Scale Analysis and Assessment (1999) RFA Text |  Recipients Lists
Research Category: Aquatic Ecosystems , Ecological Indicators/Assessment/Restoration

Objective:

Data from intensive studies of avian demography at replicated sites and multiple spatial scales within and among geographic regions in the United States will be used to assess the effects of spatial scale and landscape context on major factors (nest predation and Brown-headed Cowbird parasitism) that influence the health of bird populations.

Progress Summary:

Our work continued to advance on several fronts: modeling avian population growth, cleaning up and expanding the Breeding Biology Research and Monitoring Database (BBIRD), and developing GIS databases needed to evaluate fragmentation and landscape context. Progress for each of these three topics is summarized below.

Population Modeling

Using the population growth projection model that we developed for a general passerine life cycle, we modeled population growth rates for a variety of different bird species in breeding populations in North America. We simulated the finite rate of population increase () from detailed information on reproductive effort and adult survival for 23 species breeding in northern Arizona from 1994-1999 to assess model validity and the relative contributions of fecundity, nest success, juvenile survival and adult survival to = l. Results indicated that model projections compared well with independent assessments of population growth from Breeding Bird Survey census counts, and that nest success and total reproductive effort explained most of the variation in values. When we extended these findings to a larger group of species by using life history parameters reported in the literature, the model indicated that values were more sensitive to changes in nest success than to changes in adult survival.

We also began projecting population growth rates from reproductive life history data compiled in the BBIRD database for more than 130 different species across North America. The focus here is on quantifying geographic variation in reproductive effort of neotropical migrants and other passerines and the degree to which this variation affects population growth. Preliminary results indicate a weak latitudinal gradient in clutch size but much greater variation in nest success that is uncorrelated with geography. Both clutch size and nest success are strong predictors of = l. These results suggest that if landscape patterns limit reproductive effort in passerine populations, they may ultimately explain geographic trends in population growth as well.

Updating/Expanding the BBIRD Database

Because the BBIRD database had never been used for such extensive and detailed analyses, it had never been subjected to rigorous quality control and assurance, especially its spatial components like the plot coordinates and the consistency of vegetation measurements at each plot. Thus, we had to devote considerable effort this year to check and update portions of the database. Improvements to existing data were made by: (1) preparing maps for each BBIRD site showing plot locations as originally stored in the database; (2) mailing copies of these maps, along with a standard list of questions about the plots and any site-specific questions to the appropriate BBIRD participants; and (3) following up by telephone and e-mail until all questions were resolved. Simultaneously, we took this opportunity to expand the database by adding plot locations and nesting data for 14 new sites. Both aspects proved time-consuming because of the need to repeatedly contact approximately 30 busy BBIRD participants. By the end of this reporting period, we had completed the updates and were ready to proceed with simulations and analyses of plots from 41 different sites around the country.

GIS Data Assembly, Processing, and Analyses

Once the plot locations were updated, we began calculating minimum convex polygons connecting all plots at each of the 41 BBIRD sites. These polygons will be the geographic areas compared among sites. We also began attributing sites and plots with other summary information such as Level III ecoregion (revised from Omernik 1987) and climate variables (from DAYMET; http://www.daymet.org ). Finally, we continued to build GIS databases and to investigate methods of landscape characterization as follows:

• Development of automated methods for downloading 7.5-minute digital elevation models (DEMs) from the U.S. Geological Survey's (USGS) Web Site and assembling them for 100-km radii surrounding individual BBIRD sites. This summer, we learned that the DEM dataset we had been using was no longer available for download from USGS, which meant that we had to find a new data source and modify our automated methodology accordingly. We now are using another and better USGS product: 7.5-minute DEMs from the National Elevation Dataset (NED, http://edcnts12.cr.usgs.gov/ned/ ). We wrote scripts to request and download data within 100 km of all BBIRD sites, and all DEMs are ready for use. We also have written a script to assemble individual DEMs for each BBIRD site based on user-specified radii, meaning that we are ready to conduct topographic analyses.

• Exploration of methods for calculating fragmentation metrics. Instead of FRAGSTATS, we will use APACK software (mladenoff and DeZonia 2001; http://landscape.forest.wisc.edu/projects/APACK/apack.htm ) to calculate metrics for landscapes defined as: (1) 100-km circles surrounding each forested BBIRD site, (2) 50-km circles surrounding individual plots within BBIRD sites, and (3) 10-km circles surrounding individual plots. To date, only preliminary runs have been made.

• Assembly and evaluation of land cover data. We are working with the National Land Cover Data set (NLCD; http://edcwww.cr.usgs.gov/programs/lccp/nationallandcover.html ). These data are consistently mapped and currently available for the conterminous 48 states at fine scale (30-m resolution) for 21 land cover classes. As noted in last year's report, one drawback is that canopy class and size class data are not available. NLCD data for all states have been downloaded and are ready for use.

• Assembly of digital hydrography and roads data from USGS 1:100,000 scale Digital Line Graphs (DLGs). Hydrography and roads databases have been compiled for nearly all sites, extending out 100 km as for other GIS data.

Future Activities:

Our next steps modeling avian population growth will be directed towards determining the multiscale effects of forest fragmentation and landscape context on the nesting success of forest birds. In particular, we will examine how nest predation and cowbird brood-parasitism vary in relation to land cover type and local (patch size and edge effects) and landscape-level fragmentation variables. Furthermore, we will investigate how the consequences of fragmentation on nest predation and parasitism may differ among broad geographic areas with different land-use practices and predator guilds. As we quantify relationships between landscape characteristics and nest success, we will be able to develop predictive models of avian demographic responses to forest fragmentation across North America. These models will enable us to predict changes in the finite growth rates of populations from changes in landscape composition and spatial configuration, and to identify landscape conditions that are more likely to support self-sustaining populations.

Finally, we hope to determine whether different functional groups of birds differ in their vulnerability to fragmentation-induced changes in nest predation/parasitism to identify high-risk species or species groups. Some preliminary results will be presented at the upcoming Partners in Flight Meeting.

Journal Articles:

No journal articles submitted with this report: View all 8 publications for this project

Supplemental Keywords:

ecosystem, terrestrial, habitat, biology, avian ecology, modeling, surveys, remote sensing, geographic information systems, GIS., RFA, Scientific Discipline, Ecosystem Protection/Environmental Exposure & Risk, Forestry, Ecology, Biology, Regional/Scaling, Environmental Monitoring, habitat protection, forest fragmentation, remote sensing imagery, demographic consequences, population sensitivity, parasitism, land use, birds, agriculture, landscape context

Relevant Websites:

http://pica.wru.umt.edu/BBIRD/

Progress and Final Reports:

Original Abstract

2000 Progress Report

2002 Progress Report

Final Report