Ants are ecologically important, environmentally sensitive, widespread, and abundant, yet ant assemblages of many habitats remain poorly understood. Ants in inland barrens of New York State (NY) barely have been studied, but the limited data suggest such habitats are likely to support uncommon ant species and high species density for the region. To increase knowledge of these assemblages, I systematically collected ants from three inland barrens systems in NY, to create species lists and measure species density. I also investigated how hiking trails — a common man-made disturbance — may be impacting ant assemblages in these early-successional, disturbance-dependent ecosystems. My data strongly indicate uncommonly high densities of ant species in NY pine barrens, including the most northern known occurrences of some species, and show that ant assemblage composition and species density are altered on hiking trails relative to managed barrens habitat bordering the trails. I conclude that monitoring ants on hiking trails could provide valuable information, particularly on disturbance-tolerant species, and an opportunity for visitor participation and citizen science programs that could detect additional rare species.

Bird migration is a poorly-known yet important phenomenon, as understanding movement patterns of birds can inform conservation strategies and public health policy for animal-borne diseases. Recent advances in wildlife tracking technology, in particular the Motus system, have allowed researchers to track even small flying birds and insects with radio transmitters that weigh fractions of a gram. This system relies on a community-based distributed sensor network that detects tagged animals as they move through the detection nodes on journeys that range from small local movements to intercontinental migrations. The quantity of data generated by the Motus system is unprecedented, is on its way to surpass the size of all other centralized databases of animal detection and requires novel statistical methods. Building from the bsam package in R, I propose two new biologically informed Bayesian state-space models for animal movement in JAGS that include informed assumptions about songbird behavior. I evaluate the models using a simulation study in realistic conditions of data missingness. One of these models is generalized to a hierarchical version that fits population-level movement through joint estimation of movement parameters over multiple animal tracks. To apply the models, I then employ a localization routine on a Motus data set from migrating songbirds (Red-eyed Vireos - Vireo olivaceus) from the Eastern coast of North America. This allows me to apply the new hierarchical model and its predecessor to estimate unobserved locations and behaviors. Migratory flights were observed to occur mostly in the evenings along the coast and directed migratory flights were detected over water over e.g. the Bay of Fundy, the Long Island Sound and the New York Bight. Area-restricted searches were confined to coastal areas, in particular the Gulf of Maine, Long Island and Cape May.

In recent decades, New England agriculture has become increasingly characterized by small, diversified farming operations with values deeply rooted in community and conservation. In sharp contrast to large-scale, high-intensity agriculture currently typified by the majority of North American farms, New England farmers commonly prioritize ecologically beneficial production practices such as reduced chemical inputs, integrated pest management (IPM), low tillage, cover cropping and crop rotation, and retention of natural habitats like woody hedgerows and herbaceous strips. Public support and demand for local, sustainable food, evidenced by the success of CSA (Community Supported Agriculture) in the region, has helped to bolster this movement and increase the viability of these farms. In addition to boosting regional food production and self-sufficiency, these farms also present an opportunity for wildlife conservation. Shrubland bird species in particular may benefit from habitat created on these farms because of their preference for heterogeneous shrub and herbaceous vegetation and lower area sensitivity compared to other species, such as grassland obligates. In order to evaluate conservation potential and habitat associations of shrubland birds and other priority species on small, diversified farms, we conducted point counts and vegetation surveys across 23 farms in the Pioneer Valley, MA during the summers of 2017 and 2018. We used Poisson-binomial mixture models and canonical correspondence analysis (CCA) to assess the effects of a suite of microhabitat-, field-, and landscape-scale variables on the abundance of bird species. Our results confirmed that shrubland birds, including song sparrow (Melospiza melodia), gray catbird (Dumatella carolinensis), common yellowthroat (Geothlypis trichas) and American goldfinch (Spinus tristis), were the predominant species present, accounting for over 52% of the total observations. Species-habitat relationships were diverse; however, smaller field sizes, and increased cover of tall, dense, woody or nonproductive vegetation types were associated with higher abundance of shrubland species as well as lower abundance of crop pests such as European starling (Sturnus vulgaris) and house sparrow (Passer domesticus). These findings support the hypothesis that small, diversified farms are providing beneficial habitat for shrubland birds, as well as providing species-specific guidelines for farmers interested in conserving birds on their land. In order to place small, diversified farms into a regional conservation context and evaluate their contribution to shrubland bird conservation efforts in New England, we compared bird abundance, community composition, and conservation value of small, diversified farms to five established shrubland habitat types in the region: wildlife openings, two types of silvicultural openings (larger clearcuts and small forest openings created by group selection harvest), beaver meadows, and powerline rights-of-way. We compiled avian survey data from previous studies of each of the aforementioned habitat types conducted from 2002-2006 (powerline rights-of-way, wildlife openings, clearcuts, and beaver meadows) and 2014 (small forest openings). We then compared the relative bird abundance, community composition, and conservation value across all five habitat types (including farms) using Generalized Linear Models (GLMs), non-metric multidimensional scaling (NMDS) ordination, permutational multivariate analysis of variance (PERMANOVA), and calculated an Avian Conservation Significance (ACS) score for each habitat. The avian community composition of farms most closely resembled that of wildlife openings and harbored more open-habitat and generalist species such as American robin (Turdus migratorius), red-winged blackbird (Agelaius phoeniceus), and eastern kingbird (Tyrannus tyrannus). Several shrubland species were found to have higher relative abundances on farms than any of the other four habitats including song sparrow (Melospiza melodia), northern cardinal (Cardinalis cardinalis), and willow flycatcher (Empidonax traillii). Farms possessed a higher ACS score than powerline rights-of-way, small forest openings, and beaver meadows, but were lower than clearcuts and wildlife openings. Our results suggest that small, diversified farms support a unique suite of shrubland species, and while they certainly cannot replace managed shrubland habitats such as wildlife openings and clearcuts, they may complement these existing habitats in terms of their community composition and conservation value.

Observation of floral visitation is an accepted method to describe plant-pollinator interactions despite potential biases. Collecting pollen from pollinators offers new insights on the structure and function of plant-pollinator communities. Furthermore, the strength and frequency of plant-pollinator interactions can vary across temporal scales. However, within-season and within-day (morning vs. evening) variation in plant-pollinator networks has been little studied. By evaluating variation in network structure across these biologically relevant time scales, we will gain a better understanding of the factors that shape plant-pollinator communities. The objectives of this study are to 1) Compare the structure of plant-pollinator networks built on floral visitation and pollen transport data, 2) Evaluate intra-annual variation in plant-pollinator network structure and 3) Evaluate variation in plant-pollinator structure within a single day (i.e. morning vs. evening). Congruence and Temporal Variation of Floral Visitation and Pollen Transport Networks in Southern Appalachia

The objective of this study was to assess the above-ground factors affecting the establishment and recovery of the dominant chaparral shrub Adenostoma fasciculatum (chamise) on the Rocky Canyon granite mine. Attempts to restore the California chaparral have been challenging and few successful efforts have been documented. However, the California chaparral can fully recover from fire in as little as 10-15 years. Factors affecting chamise seedling establishment were tested by planting chamise seed in forty eight 1square meter plots managed to test the effects of interspecific competition with native postfire vegetation, post-mining volunteer vegetation, and intraspecific only competition. Plots were managed and observed from December 2004 to May 2006. Half of the 48 plots were summer irrigated throughout the first growing season to test the addition of water on growth and survivability. Nearly all chamise seedlings exposed to competition from either native or non-native vegetation perished within the first summer. Seedlings in plots where interspecific competition was removed experienced significantly higher germination and higher survival. Seedlings in the irrigated plots with competition removed, grew on average 18cm taller and had 27% more coverage per plot than non-irrigated with competition removed. They were also 4 times more likely to survive. The results of this study clearly show that chamise seedlings are not strong competitors when exposed to either native herbaceous post-fire vegetation or non-native weedy vegetation. Restoration of chamise at Rocky Canyon will require removal of all competing species during the first year of establishment. Irrigation is not required for successful germination and establishment but will result in higher survival and faster growth.

Increasing evidence documenting the decline of insect populations, resulting from increasing human disturbances has resulted in efforts to establish pollinator gardens to provide additional resources for insect populations. However, our understanding of biotic and abiotic garden characteristics important for attracting and sustaining pollinator diversity is limited. Here, we evaluated 17 pollinator gardens to evaluate the effect of five biotic and three abiotic garden characteristics on pollinator species richness, abundance, and proportional representation of four pollinator functional groups. Plant species richness positively influenced pollinator richness and negatively influenced flower visitation. Bombus proportional abundance responded to several variables (distance to vegetation, plant species richness, floral symmetry, floral native status, habitat type), and decreases in their proportional representation were accompanied by increasing proportions of other insect groups. Our results suggest any size, diverse, native pollinator gardens can improve pollinator diversity, and small-scale pollinator gardens should favor functional groups adapted for the habitat type.

Mesopelagic (open ocean, 200-1000 m depth) fishes are important consumers of zooplankton and are prey of oceanic predators. Some mesopelagic fishes (e.g., myctophids and stomiids) undertake a diel vertical migration where they ascend to the near-surface waters during the night to feed and descend into the depths during the day to avoid predators. Other mesopelagic fishes (e.g., Sternoptyx spp.) do not vertically migrate and remain at deep depths throughout the day. While in the epipelagic zone (surface – 200 m depth), vertically migrating fishes become prey to upper-trophic level predators, such as: tunas and billfishes. Benthic fishes (e.g., macrourids) often vertically migrate as well, ascending into the pelagic zone to feed on pelagic organisms. Fishes of different depths and vertical migration habits likely have a different ecological role in food webs. The relationship between parasites and gut contents provides insights into ecological processes occurring within assemblages, as prey items are often vectors for parasites. This study examined the differences between the prey items present in the gastrointestinal cavity and parasites of 26 mesopelagic fish species in the Gulf of Mexico. Results showed that based on the proportionally dominant prey items per species, six different feeding guilds existed within this assemblage, five based on planktivory: copepodivores, predators of copepods and other zooplankton, predators of copepods and euphausiids, gelatinivores, generalists, crustacean decapodivores, and upper-trophic level predators. Larger fishes preyed on larger prey items and harbored more parasites. Sigmops elongatus exhibited an ontogenetic diet shift at 75 mm standard length, progressing from eating primarily copepods at small sizes to eating primarily euphausiids at large sizes. Compared to similar studies, this study revealed a higher parasitic infestation by trematodes, an endoparasite (parasite within the host) class often restricted to nearshore hosts, in Gulf of Mexico fishes. Helicometrina nimia, the dominant parasite of the gempylid Nealotus tripes, has not previously been recorded in hosts below 200 m depth, suggesting a foodweb pathway that transitions from nearshore to offshore. These data can be used to develop and refine models aimed at understanding ecosystem structure and connectivity.

Pieris oleracea, formerly Pieries napi, was once a widespread pierid butterfly in New England until the introduction of a biological control agent, Cotesia glomerata. It has been suggested that C. glomerata is responsible for the range reduction of P. oleracea. There are been several introductions of a second more specialized biological control agent, Cotesia rubecula, to the United States since the 1960’s. My first goal was to determine the current distribution and status of P. rapae parasitoids and the effectiveness of C. rubecula as a biological control agent since its release. The findings of a survey I conducted of the parasitoid community of P. rapae indicate that C. rubecula now occurs as far west as North Dakota and has become the dominant parasitoid of P. rapae in the northeastern and north central United States and adjacent parts of southeastern Canada, where it has displaced C. glomerata, the previously dominant parasitoid. Survival of artificially established cohorts of P. rapae larvae was assessed in a collard patch on an organic vegetable farm in western Massachusetts. There was a significant drop in larval survival between the 4th and 5th instar due to parasitism by C. rubecula. This was change from survival curves of P. rapae from a 1985-1986 study, in which there was a significant drop in survival between the 5th instar and pupal stage due to C. glomerata. The final goal of my thesis work is to try to understand why P. oleracea was able to survive at the focal study site in Lenox, MA despite parasitoid pressure and range reduction elsewhere in New England. In olfactometer tests, there was no difference in attractiveness of naïve C. glomerata females to volatiles of either Cardamine pratensis (cuckooflower) foliage, the host plant of P. oleracea or Brassica olercea (collard) foliage (P = 0.51). In order to determine if overtopping by other vegetation may provide an enemy free space for P. oleracea by affecting detection by C. glomerata, cage experiments were conducted. Overtopping vegetation had a significant effect on parasitism by C. glomerata (F = 12.8, df = 3, PP. oleracea has been able to thrive at the Lenox, MA site.

The hills of Yazoo Basin have a long history of land use modification and subsequent erosion and flood control issues. In response, federal actions were taken to address these issues beginning after the Mississippi River flood of 1927. Four major flood control reservoirs were built in 1932-1957, and instream low-drop grade control structures (GCS) were installed beginning in the 1980s. The objective of my dissertation was to ascertain the long-term effects of these efforts on stream fish assemblages and channel morphology. To assess whether the reservoirs affected upstream fish assemblages as barriers to recolonization by fluvial fishes or as source population for invasion by lentic generalist fishes (Chapter 1), I used stream data collected 43-61 years after the rivers were impounded to test for differences in fish assemblages between sites upstream and downstream of the reservoirs. Analysis of catch per effort and diversity metrics displayed little influence of the reservoirs, but trait-based analysis revealed marginal increases in planktivores, herbivores, detritivores, and generalists in upstream assemblages. After determining that potential effects of reservoirs would not confound further analysis, I assessed the effects of GCS on channel morphology (Chapter 2) and fish assemblages (Chapter 3) 30 years post-installation. To assess GCS effects on channel morphology, stream cross-sections were used to calculate Bank Height Ratio, Width/Depth Ratio, and Entrenchment Ratio, while point estimates made along the transects were used to calculate the average sediment size distribution. Analyses revealed that the GCS were successful in checking channel incision moving headward in the streams: sites upstream of the GCS were less incised and had greater accumulations of fine substrates compared to downstream sites and sites on streams lacking erosion control structures. The GCS could potentially affect fish assemblages through habitat modification or by selectively filtering the assemblages as a barrier to upstream migration. Analysis of beta diversity revealed that diversity was driven by species replacement rather than nestedness, which indicates GCS were not acting as filters on the assemblages. Analysis of catch per effort data confirmed differences in assemblage structure that echoed the instream habitat differences revealed in Chapter 2.

The unifying concept of endosymbiosis and the ‘holobiont’ is that the interaction of the microbial community and the host’s biology can affect myriad processes from speciation to physiology to behavior. This study explored the role of the microbiome as a potential facilitator of rapid evolution of social behavior in a socially polymorphic species of spider, Anelosimus studiosus. Adult females were collected from solitary and social colonies at two geographically distinct locations, and behaviorally assayed to assign individuals to ‘docile’ or ‘aggressive’ phenotypes. Microbiomes of each individual were analyzed by 16s rRNA sequencing. Correlations were found with external influences on the microbiome (colony type, local environmental microbiota, and among colony), and also between the microbiome and individual’s behavioral phenotype. While causation has not yet been established, these data suggest that demographics and ecology affect the microbiome, and that behavior may be affected by the microbiome.