The influence of forest fragmentation and landscape pattern on American martens.
Dates
Publication Date
1999
Citation
Hargis, Christina D., Bissonette, John A., and Turner, David L., 1999, The influence of forest fragmentation and landscape pattern on American martens.: Journal of Applied Ecology 36, no.1 (1999): 157-172.
Summary
Conclusions: Marten capture rates were negatively correlated with increasing proximity of open areas and increasing extent of high-contrast edges. Forested landscapes were unsuitable for martens when the average nearest-neighbor distance between open, non-forested patches was <100m. Thresholds/Learnings: Synopsis: This study aimed to determine whether American marten abundance changed with incremental increases in habitat fragmentation caused by the combined effects of natural openings and timber clearcuts. Researchers evaluated differences in marten capture rates in 18 study sites with different levels of fragmentation. Martens appeared to respond negatively to low levels of habitat fragmentation, even with remaining patches of connectivity. [...]
Summary
Conclusions:
Marten capture rates were negatively correlated with increasing proximity of open areas and increasing extent of high-contrast edges. Forested landscapes were unsuitable for martens when the average nearest-neighbor distance between open, non-forested patches was <100m.
Thresholds/Learnings:
Synopsis: This study aimed to determine whether American marten abundance changed with
incremental increases in habitat fragmentation caused by the combined effects of natural openings and
timber clearcuts. Researchers evaluated differences in marten capture rates in 18 study sites with
different levels of fragmentation. Martens appeared to respond negatively to low levels of habitat
fragmentation, even with remaining patches of connectivity. Marten capture rates were negatively
correlated with increasing proximity of open areas and increasing extent of high-contrast edges.
Forested landscapes were unsuitable for martens when the average nearest-neighbor distance between
open, non-forested patches was <100m. Therefore, it is recommended that the combination of timber
harvests and natural openings constitute no more than 25% of a landscape greater than 9km2. Since the
spatial pattern of open areas determines the amount of forest interior habitat in a landscape, it is also
recommended that timber cuts are made progressively from a single cleared patch in order to retain the
largest amount of forest interior habitat as possible.
Caveats: 1) R2 values associated with all significant relationships were low. It may be unrealistic to
expect high R2 values in a study design where landscapes are the units of replication. 2) The size chosen
for study areas was small relative to marten home range scale. 3) Marten capture rates represent
relative, not absolute, marten numbers. Sites with no marten captures may have contained martens but
at densities too low to detect during the 6-night trapping period. 4) Researchers were unable to
determine whether martens in our samples were residents or transients. This is an important
consideration because high numbers of transients could indicate a population sink. 5) Martens in the
study sites may have exhibited a stronger response to low levels of fragmentation than would be
expected in geographical areas with less harsh conditions. The clearcuts in our sites provided no habitat
for martens because cut areas generally were stripped of both vegetation and logging slash after timber
harvest.
Added to ScienceBase on Thu Feb 20 15:25:46 UTC 2014 by processing file <b>LandscapePattern_Database_5_9_AMENDED_MattsBibEdits_v2.xlsx</b> Augmented by Bray J. Beltrán Feb 16, 2015
