Abstract |
- Understanding how wetland birds use habitat is pivotal to developing successful and
beneficial conservation strategies. Although it has been an ardent topic in forest
research for some time, how species interact with the spatial patterning of habitat
across a landscape (i.e., landscape structure) has been more or less neglected in
wetland research. There is every reason to believe that landscape structure might
influence wetland birds, especially since most species are highly mobile and so many
wetland landscapes are inherently heterogeneous. Moreover, as a result of the
agricultural development of riverine valleys, many former wetland landscapes have
undergone profound changes in landscape structure. To ultimately determine how
such landscapes should be restored, researchers need to integrate information
generated from three basic questions: what was the historical nature of the landscape?,
what is the present-day distribution of habitat and resources?, and most importantly,
how do wetland species currently interact with local (within wetlands) and landscape
(within a region) habitat distribution?
This dissertation examines these questions for the Willamette Valley
("Valley") of northwestern Oregon, USA. The Valley is an excellent model landscape for such a study as wetland loss from agriculture and urban development has been
severe, waterbirds use the many unprotected agricultural wetlands distributed across
the region, and many agencies and coalitions are interested in restoring the Valley
using an integrated landscape approach. To understand how conservation efforts
should proceed at both a local and landscape scale, I assessed the: 1) historical (circa
1840) nature and waterbird use of Valley wetlands, 2) typical distribution of wetland
habitat and invertebrate resources today, and 3) current influences of wetland
landscape structure and local conditions on winter wetland habitat use by a highly
mobile but relatively understudied group of waterbirds - shorebirds
(Charadriformes). I began by compiling a profile of the historic Valley from
historical land cover data and primary accounts (e.g., from naturalists, settlers, fur
trappers). The distribution of 'shorebird habitat' (wet, <50% vegetation cover) during
two winters of vastly differing rainfall (1999-2000: wet winter, 2000-2001: dry
winter) was then mapped using a combination of remotely sensed imagery and data
acquired in the field. I estimated invertebrate resources of Valley wetlands by
sampling the benthos of sites potentially attractive to shorebirds. Finally, I monitored
wetland use (frequency of occurrence, abundances) by two focal shorebird species (Dunlin Calidris alpina, Killdeer Charadrius vociferus) over the same two winters
(wet and dry) and examined how use of regions and individual wetland sites were
related to wetland landscape structure (amount of shorebird habitat within regions) and
landscape context (amount of shorebird habitat surrounding sites), respectively. I also
used locally measured features (availability of shorebird habitat and invertebrate
abundance within wetlands) to assess evidence for hierarchical habitat use (influence of shorebird habitat within and surrounding sites) and to determine if landscape
context mediated the influence of local food abundance on bird use. I analyzed data
using multiple linear regression and information theoretic (AIC) approaches to model
selection.
Historical research revealed that Valley wetlands were of three valuable types
to nonbreeding waterbirds (i.e., emergent wetlands, riverine wetlands, and wetland
prairie) and the numbers and diversity of waterbirds using these wetlands was
generally much greater than they are today. The most extensive wetland habitat type
was wetland prairie, which functioned as valuable fall/winter habitat for waterbirds,
but only while native Kalapuyans managed the region with fire. Current land cover
data from the winters of 1999-2001 indicate that while prime shorebird habitat (wet,
unvegetated) on agricultural land can be abundant and widely distributed in wet years,
it is typically scarce and more aggregated during dry winters. Compared to resource
levels in winter at other important wintering regions in North America, invertebrates
of Valley wetlands were of moderate abundance for most sites sampled. Finally, model selection results indicated that landscape structure was an important positive
determinant of Dunlin abundances within regions and within wetlands, but only during
the dry winter. Dunlin abundances at sites were also positively associated with local
conditions (amount of available wet habitat, percent of soil exposed, invertebrate food
abundance) during the dry winter. Kilideer use of sites was unrelated to landscape
structure and local conditions during both winters.
Overall, results from this dissertation indicate that wetland landscapes like the
Valley hold great potential for restoration and that landscape planners should consider the spatially-explicit nature of wetland sites (where they are located in relation to
others) when prioritizing sites to conserve. Within individual wetlands prioritized for
conservation, managers should then promote certain local conditions (shallowly
flooded and exposed substrates, productive invertebrate communities) that benefit
foraging shorebirds and other waterbirds. Contributing to the field of landscape
ecology, this study is among the first to examine the relationship between landscape
pattern and ecological processes in wetland systems, specifically for wetland birds in
winter. It illustrates the potential influence of landscape structure in any wetland
region, and thus the importance of considering the spatial distribution of habitat for the
successful conservation of wetlands and waterbirds.
|