Source: UNIVERSITY OF MAINE submitted to
INFLUENCE OF PARTIAL HARVESTING ON AMERICAN MARTEN AND PRIMARY PREY SPECIES
 
PROJECT DIRECTOR: Harrison, D. J.
 
PERFORMING ORGANIZATION
WILDLIFE ECOLOGY
UNIVERSITY OF MAINE
ORONO,ME 04469
 
NON TECHNICAL SUMMARY:
 
OBJECTIVES: To compare marten use of a forest landscape before and after partial harvesting and to evaluate landscape-scale selection for partially harvested areas. To document stand-scale patterns of habitat selection by marten whose home ranges have been influenced by partial harvesting. To evaluate patterns of sub-stand habitat selection by marten whose home ranges comprised 20% partially harvested forest. To compare densities of red-back voles, deer mice, shrews, red squirrels, and showshoe hares between partially harvested stands and residual forests. To develop a spatially explicit habitat supply model for marten on industrial forest lands in northern Maine.
 
APPROACH: Capture, radio-telemetry and census of marten (objectives 1 and 2). Snowtracking and vegetation measurements (objective 3). Live-trap grids, pellet transects, and incidental catch rates (objective 4). Geographic information systems (objectives 1, 2, and 5) and spatially explicit habitat modeling (objective 5).
 
CRIS NUMBER: 0135485 SUBFILE: CRIS
PROJECT NUMBER: ME09660 SPONSOR AGENCY: NIFA
PROJECT TYPE: MCINTIRE-STENNIS PROJECT STATUS: TERMINATED MULTI-STATE PROJECT NUMBER: (N/A)
START DATE: Oct 1, 1998 TERMINATION DATE: Sep 30, 2004

GRANT PROGRAM: (N/A)
GRANT PROGRAM AREA: (N/A)

CLASSIFICATION
Knowledge Area (KA)Subject (S)Science (F)Objective (G)Percent
135084010706.450%
135085010706.450%

CLASSIFICATION HEADINGS
KA135 - Aquatic and Terrestrial Wildlife
S0850 - Wildlife habitats
S0840 - Fur-bearing animals
F1070 - Ecology
G6.4 - Protect and Enhance Wildlife Habitat


RESEARCH EFFORT CATEGORIES
BASIC 100%
APPLIED (N/A)%
DEVELOPMENTAL (N/A)%

KEYWORDS: forest wildlife relations; wildlife; martens; resource inventory; wildlife management; forests; forest habitats; prey; harvesting systems; home range; shrews; timber; rodents; mice; wildlife population; logging; snowshoe hares; voles; forest ecology; squirrels; residual stands

PROGRESS: Oct 1, 2002 TO Sep 30, 2003
In the fifth year of our 6-year project we applied the results from our studies of the responses of marten to forest harvesting to develop and test statewide predictive models of marten occurrences. These investigations have indicated that marten are particularly sensitive to forest fragmentation and may serve as an indicator of forest-landscape integrity. Use of marten and other forest carnivores as umbrella species assumes that accommodation of their life history and habitat requirements at target levels will ensure the long-term viability of a diverse array of forest associated vertebrates, invertebrates, and plants; however, this critical assumption has not been rigorously tested and was our primary objective during the sixth and final year of our study. We developed predictive models of marten occurrences using logistic regression LANDSAT thematic mapper imagery, and long-term field studies of marten. Resulting models were accurate in predicting the occurrence and absence of marten home ranges on the landscape with > 75 % accuracy. Models incorporated spatial and social requirements of marten and the most reliable models, based on an information-theoretic approach, contained predictor variables describing extent of forest > 6 m tall with > 40% canopy closure, and variables that described habitat configuration and fragmentation. We used resulting model coefficients to map all the pixels throughout a 56,940 km2 area representing the geographic range of marten in Maine that had a >60%, >70%, and >80% probability of supporting marten. Based on each of these 3 probabilities of occupancy (i.e., representing a range of conservation objectives for marten), we calculated the proportion of northern Maine predicted to be occupied by marten and compared to the proportion of the predicted range (using statewide GAPS models) for 68 forest-generalist vertebrates and 43 forest-specialist vertebrates. Our results indicated that marten were an effective (quantified based on percentage of vertebrates with a disproportionate percentage of their predicted habitat conserved relative to the percentage of statewide marten habitat conserved) umbrella species for 81-84% of forest-dependent generalist species and for 70-74% of forest-dependent specialists. Based on the habitat requirements of forest species that were not disproportionately benefited by habitat conservation for marten, it appears that minor inclusion of some specific requirements that exceed the habitat needs of marten (e.g., requiring some mature conifer instead of either deciduous, mixed, or conifer patches) into landscape-scale conservation strategies could further enhance biodiversity conservation. We conclude that marten are an appropriate umbrella species on which to base landscape-scale biodiversity planning. Ensuring habitat to support viable marten populations would disproportionately benefit the vast majority of forest-dependent vertebrates whose geographic ranges overlap with marten in Maine.

IMPACT: 2002-10-01 TO 2003-09-30 This project has provided the tools neeeded to develop landscape-scale conservation strategies for marten and other forest-associated vertebrate wildlife throughout the commercial forestlands of northern Maine. Another application that is ongoing is the long-term monitoring of habitat for marten and other forest wildlife at a statewide scale. The project has further increased knowledge of the process of fragmentation in forest dominated landscapes the the relative importance of habitat loss versus fragmentation in determining occurrences of marten.

PUBLICATION INFORMATION: 2002-10-01 TO 2003-09-30
Payer, D.C. and Daniel J. Harrison. 2003. Influence of forest structure on habitat use by American marten in an industrial forest. Forest Ecology and Management 179:145-156

PROJECT CONTACT INFORMATION
NAME: (N/A)
PHONE: (N/A)
FAX: (N/A)