Annie Woods, M.S.

Quantifying the relationship between anthropogenic disturbance and biotic integrity in the Adirondack Park

My study investigated the response of biotic communities to anthropogenic disturbance in the Adirondack Park, and examined the relationship between land use and biotic integrity at the landscape scale. I developed an index of biotic integrity (IBI) for the Adirondack Park using data on bird guilds from the 2000-2005 New York State Breeding Bird Atlas (BBA). IBI was a better measure of biotic community condition than species richness, which was affected by sampling effort and responded non-linearly to disturbance. IBI was negatively related to development and open land covers and positively related to forest/wetland cover and elevation. IBI was predicted better by variables measured at the BBA block scale than larger scales. I found that, in the Adirondack Park, the biotic integrity of private lands used for natural resource management may be at risk of degradation from expanding development.

Zachary Danks, M.S.

Spatial, temporal, and landscape characteristics of moose-vehicle collisions in Maine

Motor vehicle collisions involving moose (Alces alces) pose safety risks and generate significant socio-economic costs. I analyzed records of moose-vehicle collisions (MVC) in Maine from 1992-2005 using spatial statistics and Geographic Information Systems (GIS). My objectives were to (1) describe temporal and spatial distributions of MVC, (2) determine relationships between landscape characteristics and MVC, and (3) identify geographic extents at which habitat management might effectively reduce MVC. Collisions clustered spatially at local and regional scales. Predicted probability of MVC was related to speed limit, traffic volume, cutover and coniferous forest, interspersion of land cover, and distance to wetlands and developed areas. Habitat covariates were most explanatory at spatial extents that approximate the home range size of moose (20 km²). My findings indicate that traffic and landscape-scale habitat, rather than topographic factors or moose abundance, are the primary determinants of MVC. Given that forest management is not practical at landscape scales, lowering speed limits during times of highest risk may be most effective for reducing MVC.

Sarah Nystrom, M.S.

Using artificial neural networks to minimize human and wildlife conflicts

My project entails using computer modeling, specifically Aritifical Neural Networks, to reduce the number of deer- vehicle accidents in Onondaga County, NY. I will be looking at how landscape characteristics, such as distance to vegetation, cover type, bank incline and level of development influence the number of deer vehicle accidents in an area. By measuring these variables at sites of previous accidents I will be able to build a model that will predict the likelihood of accidents in the future, providing a valuable tool for planning new roads and modifying current road networks.

Jeff Organ, M.S.

Linking white-tailed deer harvests to population and environmental processes through ecological modeling

Effective management of white-tailed deer requires scientific information on current population abundance and factors which will produce changes in abundance in the future. Harvest data efficiently provide some information for broad spatial areas. This research is an attempt to provide information on both harvest and non-harvest impacts through the use of ecological models. I integrate deer winter bio-energetic and weight models with deer range quality that predicts substantial mortality in areas with poor range quality, high winter snow depths, and low winter temperatures. I perform an analysis of deer-vehicle collisions that predicts substantial automobile mortality for suburban areas and areas containing major highways. I produce ecological models that predict changes in the deer population during the past and into the future for New York State. It appears that winter severity was responsible for most of the variation in annual buck harvest change from 1988 to 2005 throughout NYS. These models suggest that deer managers need greater certainty about hunter effort, population abundance, and non-harvest mortality to achieve precision levels of less than ±10% around management objectives.

Susan Walker, M.S.

Evaluation of microclimatic advantages within white-tailed deer wintering yards located in close proximity to residential communities in the Central Adirondacks of Northern New York State

White-tailed deer inhabiting the northern portion of their range annually migrate from summer to winter yards in response to changing winter climate. Within the Adirondack Park, deer historically wintered in lowland softwood areas. However, increasingly deer are now utilizing residential communities within the park. Documentation of Adirondack deer winter yards from over 30 years ago by the New York State Department of Environmental Conservation provided information on historical locations of social groups. Recent research comparing historical winter yards to current yards has shown that several social groups have shifted their locations closer to residential areas (Hurst 2004). While winter feeding (now banned throughout NYS) and increased conifer harvest on private land have likely influenced deer behavior and movement throughout the park, I was interested in examining microclimate (wind speed, ambient temperature, solar radiation, snow depth, and snow surface condition) variation within deer yards located in residential habitat and lowland softwood habitat. I examined microclimate differences that could potentially account for changes in deer movements and shifts in yard locations toward residences throughout the Adirondacks.

Carolyn Spilman, M.S.

The effects of lakeshore development on common loon productivity in the Adirondack State Park, New York

The common loon (Gavia immer) is a management Species of Special Concern in New York State. Studies suggest that the southern border of the loon's breeding range is slowly moving northward, indicating that loons are finding less suitable habitat than once existed in New York State (McIntyre 1988). A possible explanation for this trend is the loss of habitat resulting from lakeshore development. Located in northeastern New York, the Adirondack State Park represents a portion of the southern boundary of the Common loon's current breeding range. The periphery of a species' geographic range is usually characterized by less favorable habitat along with lower and more variable densities (Channell and Lomolino 2000). My primary research objective was to quantify shoreline development levels on a sample of 55 lakes in the Adirondack State Park and relate these measures to Adirondack Cooperative Loon Program (ACLP) records of common loon productivity on these lakes over the last 7 seasons. Field data were collected during the 2004 and 2005 loon breeding seasons to quantify current lakeshore development levels on a sample of 55 lakes. GIS analysis of Adirondack Park Agency land classification within buffered areas of study lakes was also used to predict maximum allowable levels of development surrounding study lakes.

Dawn Gorham, Ph.D.

Landscape-level analysis of deer overabundance

Many suburban communities are faced with problems associated with deer overabundance, including losses due to automobile-deer collisions and deer foraging of ornamental plants. Compounding the situation are the drastically differing opinions of what type of managment, if any, should be implemented in order to help alleviate the problem. I used a systems ecology approach to create a suburban deer habitat assessment model. I also examined patterns of suburban community design with regard to a suburban deer population. This information has the potential to be used in community planning to determine how landscapes might be better designed to minimize deer-human conflict.

Linda J. Gormezano, M.S.

Exploring the variance components of distance sampling to simplify abundance estimation of white-tailed deer

Jeremy Hurst, M.S.

Long-term change in spatial pattern of winter deer yards in the Adirondack Mountains

Alison Lott, M.S.

Quanifying the relationship between white-tailed deer and forest regeneration in the Adirondack Mountains

High white-tailed deer densities are a common problem throughout the eastern United States. Deer have been known to detrimentally affect the growth and survival of forest vegetation. I created a model that will identify the relative importance of deer density to northern hardwood forest regeneration. The model will serve to guide forest resource managers to obtaining the maximum desirable hardwood regeneration possible by identifying the deer density threshold for a specific area of the Adirondacks.

Karl Didier, Ph.D.

The role of white-tailed deer herbivory in changing Northern Hardwood ecosystems

For many decades, people have preceived major changes in the hardwood forest ecosystems of the Northeast. The most recent example is the perceived decrease in sugar maple abundance. The role of deer herbivory in these large scale and long term changes is unclear. Several questions are of interest to me: Does deer herbivory exhibit strong influence on forest composition at large scales (e.g., across the entire northern Adirondack Park)? If sugar maple is indeed decreasing in abundance, at what scale is that change occurring and what role are deer playin in that change? How do deer interact with other forces of ecosystem change, such as disturbance, competition among plants, and changes in soil fertility? Do deer affect basic ecosystem processes, such as nutrient cycling, and how strongly do these indirect effects influence change? My work is primarily centered around 2 projects, the first examining the role of deer in regeneration failure in the northern Adirondack Par, and the second examining how deer are affecting regeneration across a spectrum of disturbance caused by a large ice storm in the Adirondacks.

Kevin Douglass, M.S.

Effects of supplemental feeding on white-tailed deer behavior and reproductive biology

Supplemental feeding is becoming a common practice in many areas throughout New York State. Feeding programs are intended to provide additional food supplements during the winter when typical nutritional resources are limited. Even though some groups have been feeding deer for a number of years, there is no complete understanding of how it may affect the survival and well-being of deer throughout winter. I am focusing my research on privately owned land in the north central Adirondacks, New York. My objectives for this study are: 1) quantify the degree of access to food supplements by sex and age-group, and the duration of time spent in aggressive (non-foraging) behaviors and, 2) document the effect of supplemental feeding on the reproductive status of females, and to quantify the amount time spent in breeding behaviors. I wish to determine which type of individual is gainin access to the corn each day, and whether or not access to corn for fawns is limited compared to their access to hay. In addition, is the age or status of the dam important in determining forage activity of her fawns. Finally, I want to document the amount of breeding activity that is occuring around the feed site. Specifically, I want to examine how supplemental feeding programs may be linked to 'late' breeding behavior and if male behavior closely relates to changes in the reproductive cycle of female deer.

Jamie Fischer, M.S.

Investigating the role of the white-footed mouse (Peromyscus leucopus) in the distribution of the deer tick (Ixodes scapularis) and Borrelia burgdorferi

Lyme Disease has grown to epidemic proportions in the Northeas within the last 20 years. Our study site is located on Fire Island National Seashore, which is an Atlantic barrier island located along the southern coast of Long Island, New York. We have been monitoring the population dynamics of the small and large mammal communities for the last five years. My concentration was to understand how white-footed mice effect the distribution of the deer tick and the spirochete that causes Lyme disease.

Kathy Fleming, Ph.D.

Scale explicit spatial determinants of population structure in wild turkeys (Meleagris gallopavo silvestris)

In this study I investigated spatial patterns in a wildlife population arising from habitat and other environmental factors at multiple scales. I examined a wild turkey population to explore multi-scale spatial population structure and dynamics using a combination of field study, geospatial statistical analysis, and population modeling. Wild turkey populations undergo substantial annual fluctuations, primarily due to annual variation in nest success and survival of young. To determine the effect of hanitat variables at multiple spatial scales on nest predation, I placed 495 artificial turkey nests in 3 ecological zones across New York State. Nest predation was influenced by nest-site habitat variables: average cover at the nest site and shrub density; patch-level variables: distance to edge and forest age; and landscape variables: edge density and mean shape index. Landscape edge density showed the strongest relationship (positive) with nest predation, indicating that nest predation was heavily influenced by the degree of fragmentation in the surrounding landscape.

I used spatial autocorrelation analysis to determine if a spatial pattern existed in wild turkey abundance and population dynamics among townships in New York, and if this pattern was due to either weather or hanitat effects. I found that synchrony in annual dynamics existed between townships that were <150 km apart, but was probably not due to weather, which showed strong, large-scale synchrony. I found little evidence of spatially autocorrelated abundance patterns. Population synchrony was attributed to temporal variation in habitat such as mast availability.

I investigated the effect of landscape features and fragmentation on dispersal patterns with a source-sink population model for wild turkeys in a real landscape. I used cost-distance analysis to compare the cost of dispersing in fragmented vs. contiguous landscapes. Dispersal cost was lower in fragmented landscapes due to the high amount of edge habitat preferred by dispersing turkeys. From the results of the source-sink model, I predicted that the spatial arrangement of source and sink landscapes would be important in determining dispersal patterns; and that some landscapes may not be reachable by dispersers due to the distance of source populations from these landscapes, and the limited dispersal abilities of wild turkeys.

Kathy Schwager, M.P.S.

Estimating and modeling survival rates of a white-tailed deer population after 7 years of fertility control: combining Lefkovitch's and Siler's methods

Michale Glennon, Ph.D.

Assessing biodiversity in the Adirondack Park

The Adirondack Park provides habitat for numerous unique species and ecological communities. With its highly interspersed mix of public and privately owned lands, the Park is often cited as an example of successful application of sustainable development principles. However, the actual effects of land use management on wildlife populations in the Adirondacks are unknown. Little research has been done to assess the effects of development of forest industry practices in the Park. Often, no scientific basis exists for understanding possible effects of issues such as major state land purchases and other exchanges. I am using both field and historical data to investigate how distributions of vertebrates in the Adirondack park are related to land management. I am using breeding bird data from the New York State Breeding Bird Atlas and small mammal data from field collections to assess how community structures change along a gradient of human disturbance. At the stand level, I am comparing how small mammal communities change between old-growth forests, selectively-harvested forests, and forested areas within developed settings. using the historical data, I am also investigating breeding bird guilds and how their distributions are related to land management on the landscape scale. I hope that this research will lead toward ways of understanding the impacts of major land use changes in the Adirondack Park, and how the landscape of the Park as a whole plays a role in shaping the biodiversity of the region.

Ilana Cantrell, M.S.

Trends in land use and open space in New York State

The issue of urban sprawl has gained much attention in New York State in recent years due to the perception that open space and productive agricultural lands are being lost to urban and residential development. Thus far, no statewide studies have been conducted to determine the extent to which New York is losing open space and productive agricultural lands. Furthermore, if open space is disappearing, there have been no studies to assess the contribution of urban and residential development to this loss at a state level. I am studying whether New York State is losing open space and productive agricultural lands, and if so, how much of this loss can be attributed to urban and residential development. using biological, social, and economic drivers that may contribute to land-use change, I am projecting likely, alternative scenarios for New York'slanscape over the next 30 years. I am also assessing the impacts of current trends in land-use change and projected future land-use scenarios on wildlife populations in New York.

Sharon Goetz, M.S.

Statewide habitat assessment for wild turkeys in Arkansas

Traditionally, habitat assessments were conducted using ground-based data collection methods. As satellite imagery becomes increasingly available it provides an inexpensive and easily acquuired data source at the landscape level. In order to use satellite imagery for habitat assessment, ground-based variables need to be adapted to variables that can be quantified with satellite imagery. My research involves determining landscape characteristics that can become surrogate variables or an index of the actual biological nees of the wild turkey. These surrogate variables will then be used to perform a statewide assessment of habitat conditions for wild turkey in Arkansas using satellite imagery and geographical information systems (GIS).

Brian Miranda, M.S.

Landscape-scale habitat assessment for white-tailed deer in Arkansas

Increasing availability of satellite imagery, in combination with geographic information systems (GIS) and current computer processing power, provides opportunities for wildlife biologists to evaluate habitat quality at the landscape level. I am integrating these technologies to allow managers to assess habitat quality for white-tailed deer throughout the state of Arkansas. I am developing a habitat suitability index (HSI) model and pattern recognition (PATREC) model to be applied with currently available satellite imagery. The HSI model considers food and cover values of different vegetation types and their spatial distributions in evaluating habitat quality. The PATREC model evaluates which landscape characteristics are associated with high quality habitat, and allows the use to identify hight quality habitat by examining the landscape's characteristics. These tools, if found to be effective, will enable managers to quickly and inexpensively evaluate habitat quality over broad landscapes.

Anne Oyer, M.S.

Social behavior of white-tailed deer in the Adirondack Park, New York

I am studying white-tailed deer in Huntington Wildlife Forest (HWF) of the Adirondack Park. Previous studies have shown that white-tailed deer organize in social groups of females and that these groups are highly site-specific. In 1994, one of these social groups was removed from the center of the HWF and the response of the surrounding social groups was monitored. Deer were not found to be using the removal area until 1999 when 8 females were trapped there. My study will determine the home ranges of these deer and whether they are related to the surrounding social groups or may have dispersed from another area.

Blair Page, M.S.

Movements and survival of winter-fed white-tailed deer fawns

The main focus of our study is to quantify how winter feeding programs impact the survival and well-being of white-tailed deer with a specific emphasis on fawns. We are also intersted in understanding the seasonal movement patterns of deer using winter feeding sites. We are currently working on three privately-owned sites located in the north-central Adirondacks, New York. This area is contained within the southern portions of St. Lawrence and Franklin counties. We will be assaying a number of physiological parameters of captured deer to assess their relative state of well-being. From each animal we will collect blood, urine and fecal samples to be analyzed for protein, energy, and hormone levels. This will allow us to compare the physiologic, metabolic and reproductive status of deer within and between study sites. We will attempt to correlate a number of these indices with urine and fecal constituents so that comparisons can also be made to natural wintering areas where deer are not being fed.

Jen Woodard, M.S.

Movement behavior and population dynamics of white-tailed deer in a suburban environment

In my study, I use radiotelemetry to track a marked sample of deer in Irondequoit, NY. Forty-two females (as of September 1999) have been fitted with radiocollars and ear tags in 4 designated regions of town. I am testing the hypothesis that females in Irondequoit remain on a single range year-round. This is in contrast to females in rural areas, where seasonal migration is observed. If fidelity to home ranges is shown in Irondequoit, future management of overabundant deer in suburban environments could be implemented on individual neighborhoods. Managers could concentrate on actively controlling one family group without impacting deer in an adjacent neighborhood, thereby decreasing costs and safety concerns associated with management.

Genevieve Nesslage, M.S.

A geostatistical analysis of deer harvest in the Adirondack Park from 1954-1997

Success of the proposed wolf restoration to the Adirondack Park of northern New York will rely heavily on the status of local deer populations. However, no Park-wide assessment of Adirondack deer populations has been conducted since 1972. We used kriging, as geostatistical method of interpolation, to identify spatial patterns in Adirondack deer harvest and characterize how these patterns have changed between the 1950s and 1990s. Records of total annual buck harvest/ha were averaged by town over each decade. Patterns of spatial autocorrelation in these data were modeled and used to create kriged contour maps displaying spatial patterns in buck harvest/ha across the Park. A Park-wide decline in buck harvest/ha was onserved following a series of 3 severe winters from 1968-1971. By the 1980s and 1990s, buck harvest in most regions had recovered. An exception was the western/central Adirondacks. Private hunting club records that were coreected for hunter effort displayed similar spatial patterns to that of state buck harvest/ha data, indicating that our results are not an artifact of varying hunter effort across the Park. Winter severity differed among 5 geographic regions of the Adirondacks (P<0.0001). The failure of the western/central region to recover may reflect the combined effects of frequent severe winters and heavy antlerless harvests during the 1950s and 1960s. Restoration of a large predator such as the wolf to the Adirondacks should be carefully considered if current predation by coyotes is supplemented ny that of wolves. Additive mortality could cause negative long-term effects on an already low population of deer.

Brent Rudolph, M.S.

Evaluation of immunocontraception for managing free-ranging white-tailed deer

The Town of Irondequoit in Monroe County, New York, is an urban area of significant commercial and residential development which has experienced conflicts between human inhabitants and overabundant white-tailed deer. Culling and bowhunting have been used to manage deer in the Town, but disagreements abound regarding what constitutes an acceptable managment program. Interest exists in exploring the use of immunocontraception, specifically Porcine Zona Pellucida (PZP), as a means of deer population control. Previously, a computer model was generated to determine numbers of deer and associated effort required to manage the population at several sizes. Various combinations of removal and contraception were projected. Computer modeling required that assumptions be made regarding some unknown population dynamics parameters, but a limited, experimental application of immunocontraception will allow namagers to obtain the information necessary to eliminate these areas of uncertainty and more precisely determine the effort involved with various white-tailed deer management scenarios in the Town of Irondequoit. A customized managment plan integrating knowledge of the local deer population with data on efficiency and cost associated with various techniques may maximize public acceptance and support as well as maintain cost-effectivelness to operate within financial constraints.

Karl Didier, M.S.

The feasibility of restoring elk to New York State

Recently, strong national and local interest has been expressed in restoring elk to areas where they were once found in large numbers, including New York State. The objective of this project, funded by the Rocky Mountain Elk Foundation, was to determine the biological and sociological feasibility of the elk reintroduction in New York. SUNY-ESF undertook the biological component, while Cornell University conducted the sociological study. A habitat suitability index model was designed to identify and rank the habitat in New York most suitable for reintroduction. A set of recommendation concerning the feasibility of a reintroduction and management of a reintroduced population (if a reintroduction is recommended) were developed. The project was unique in that the area in question (i.e., New York State) is vastly larger than areas typically considered in such studies. Therefore, we made extensive use of existing digital data, satellite imagery, and geographic information systems (GIS) to analyze the habitat statewide.

Bobbi Jo McClain, M.S.

Evaluating the use of satellite imagery for deer in the Adirondack Park, NY

Landscape ecology has been limited in the past by the large effort necessary to qualify habitat at a useful resolution on a large scale. Improvements in quality and accessibility of satellite imagery have brought this technology into the view of many ecologists. My study tested the usefulness of satellite imagery for landscape-scale habitat evaluation. White-tailed deer were chosen as the subject of this study because of the abundance of information available regarding their habitat needs. Using classified imagery and habitat suitability models created from past studies, a model was created which to assess habitat throughout the Adirondack Park, NY. This model was then tested using a population index derived from hunter harvest data. Studies such as this can help to define the advantages and limitations of satellite imagery in ecological terms.

Jeff Villepique, M.S.

Urban habitat use by gray fox in Syracuse, NY

I investigated the use of urban greenspaces by gray fox in Syracuse, NY, using landscape-level and community-level metrics as predictors. Syracuse and its surrounding environs contain remnant and regenerating woodlands, industrial lands, and parks, which provide habitat for the gray fox. We identified 37 greenspaces bounded by residential or commercial development. These were surveyed for fox presence by snow-track searches during the winters of 1996-97 and 1997-98. PC Arc/Info and ArcView software was used to analyze 11 landscape-level variables, including patch size, cover type, within-patch heterogeneity, adjacent land use, and corridor connectivity. Sixteen vegetative and 10 physical factors were measured at the patch scale, including woody species density and basal area, canopy closure, and optical cover, in addition to indices of human use, prey species occurence, and den-site suitability. Discriminant analysis was used to determine the utility of predicting gray fox habitat from landscape-level and community-level metrics. Results suggest that proximity to adjacent patches, presence of corridors, dense vegetation, and lack of human intrusion are positive predictors of gray fox presence.

Michale Glennon, M.S.

Satellite imagery and wild turkey habitat in Southern New York

I investigated whether abundances of wild turkeys could be predicted using satellite imagery. A supervised classification was done of images from 1979, 1986, and 1993. Relationships between turkey abundance and land-cover types across 26 townships, as well as relationships over time were tested. I found that, within years, turkey abundance was related to the amount of linear edge between forest and open area, and to the proportion of open area in each township. In this largely forested landscape, heterogeneity and interspersion increase the potential for high densities of wild turkeys. Trends in land-cover changes in southwestern New York show that, in general, the amount of forest land is increasing, while the amount of cultivated agricultural land is decreasing. Over the long-term, this trend may lead toward a decline in numbers of wild turkeys in this part of New York.

Dr. Richard K. Lawrence, post-doctoral work

Dr. Lawrence's program involved investigating spatial patterns of plant/animal communities at ecosystem scales; for example, how size, distribution, and composition of habitat or topological features structure the dynamics and relationships of ecological communities. Specifically while Dr. Lawrence was here at ESF, he investigated the effects of spatial deer distributions and habitat use on forest regeneration.

Stacy McNulty, M.S.

Effect of habitat on movement behavior and social structure of white-tailed deer in the Adirondacks

I studied white-tailed deer in the central Adirondack region. Previous research has shown that deer exhibit strong site fidelity and that deer social organization is comprised of social groups of related females. Some social groups have many more members and higher recruitment than others. I hypothesized that this variation is a response to differences in habitat quality on summer range. Small-scale disturbances such as logging may influence deer at the social group. I used Arc/Info to examine social group dynamics over a 25-year period, comparing home ranges, numbers of individuals, and recruitment os social groups and related the information to forest composition and level of disturbance. The localized response exhibited by social groups has implications toward the impact of habitat alteration on wildlife.

Clayton Nielsen, M.S.

Feasibility of white-tailed deer management using immunocontraception in Irondequoit, NY

The recent proliferation of white-tailed deer in urban areas of the eastern U.S. has created a variety of management dilemmas. Irondequoit, New York, is one of many urban centers concerned with the management of deer populations. Immunocontraception (IC) for managing deer populations is a technique that has been tested in captive populations with encouraging results however costs, logistics, and efficacy are uncertain in an urban environment with a free-ranging population. We designed an experiment to test IC as a deer population management technique in Irondequoit. A stochastic, density-dependent population model that predicts removal regimes to initially reduce the population and subsequently maintain it with IC was created. We also estimate the person-hours required to complete IC treatment on the number of females prescribed by the population model. Finally, we present a cost:benefit analysis based on findings from demographic and logistical analysis.

Paul Quinlan, M.S.

An assessment of wildlife habitat characteristics in Adirondack selection system stands

My project involved an assessment of the impacts of uneven-aged silviculture on wildlife habitat. I am concerned with general habitat characteristics which can be manipulated using common silvicultural techniques. The study areas were five northern hardwood stands in the Huntington Wildlife Forest which are managed according to a 25-year cutting cycle selection system. Data on canopy closure, ground cover, and stand structure were collected and analyzed. The development of the stands was documented as well as projected into the future. The stands were compared to others managed with noth even-aged and uneven-aged silviculture. This study provided a first look at modeling wildlife habitat dynamics in uneven-aged, managed forest stands.

Sheila Sleggs, M.S.

Wildlife and vegetation response to wetland restoration in the Montezuma Marsh Complex of Central New York