Understanding the mechanisms driving biodiversity patterns amidst an era of global environmental change
is the core of modern ecological research. The magnitude of biodiversity losses associated with anthropogenic
activities has prompted resource managers and ecologists alike to identify strategies to address conservation
issues. Broadly, two types of approaches are employed to answer ecological research questions: 1) single-species
and 2) ecosystem-based approach. Single-species approaches are often useful to elucidate mechanisms driving
population trajectories of individual species. On the other hand, ecosystem-based approaches can help in
identifying general patterns that may be useful for multi-species management.
Here, I used both approaches in assessing broad-scale patterns and mechanisms driving count trends of
migrating raptors recorded at Hawk Mountain Sanctuary (HMS), Pennsylvania. In the first chapter, I used a
hierarchical breakpoint model to identify the assemblage-wide and species-specific timing of the shifts in
count trends. Then I evaluated if changes in trend directionality of counts were linked to species’ traits
(body size, population size, migratory behavior, tolerance to human presence, DDT (dichlorodiphenyltrichlorethane)
susceptibility, habitat or dietary specialization). I found that an assemblage-wide shift in counts occurred
around 1974, and this timing was common among 14 of the 16 species in the assemblage. Moreover, I found that
habitat specialization appeared to explain the synchronous positive and negative count trends of multiple
species. Other traits that I evaluated were not consistently associated with either types of trends. The temporal
shift in trends in 1974 emphasized the relative importance of DDT, an organochlorine known to have adversely
influenced several wildlife species and was banned in the US around the 1970s, in driving population dynamics of
raptor species. However, because the counts of species susceptible to DDT were highly variable after 1974, this
may suggest that a suite of additional factors, acting together, affected the recovery of species from DDT-associated
declines. Additionally, the potential role of habitat specialization in count trends may suggest important
linkages between habitat use and demography.
In the second chapter, I used a generalized linear mixed-effects model to assess the relationships between
changes in the count totals and total proportional cover of major land-use types in nine states located in the
northeastern US (Connecticut, Maine, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode
Island, Vermont). The hierarchical modelling approach that I used allowed me to identify average and
species-specific responses to the proportional cover of forested and urban area. These land-use variables
were not associated with overall raptor counts. However, species-specific responses were variable and
significant. I found that counts of Northern Goshawk, American Kestrel, Rough-legged Hawk, Sharp-shinned
Hawk, and Red-tailed Hawk were positively associated with forest cover. On the other hand, Turkey and Black
Vultures, Bald Eagle, and Peregrine Falcon were positively associated with urban cover. Moreover,
Red-shouldered Hawk, Broad-winged Hawk, and Northern Harrier were not significantly associated with forest
cover but were negatively associated with urban cover. Merlin and Cooper’s Hawk exhibited similar
non-significant associations to forest but positive associations with urban cover. Finally, Golden Eagle
and Osprey were not significantly associated with either land-use variables. These results provided insights
on the potential influence of land-use changes on the demography of migrating raptors. Thus, these findings
may be useful in improving our predictions of the population trajectories of these species in future landscape
scenarios.
These results illustrate the utility of evaluating species-level and assemblage-wide patterns in long-term
count data. In this case, it allowed me to identify general patterns in counts of migrating raptors and gain
detailed insights on the responses of individual species to land-use changes. In doing so, I was able to
better understand the potential drivers of their ecological dynamics. By integrating information from these
two approaches, we can expect to obtain a better understanding of natural systems and consequently, increase
the probability of successful conservation outcomes.
REFORESTATION PLANNING BASED ON PLANT HARDINESS ZONES IN VITILEVU ISLAND, FIJI
