Search USGS⌕ Search

Geology topics

Therese M. Donovan

Publications and source records attributed to Therese M. Donovan.

At least 37 records · Page 2Linked to original sources
USGS2020

Wildlife resistance and protection in a changing New England landscape

Rapid changes in climate and land use threaten the persistence of wildlife species. Understanding where species are likely to occur now and in the future can help identify areas that are resistant to change over time and guide conservation planning. We estimated changes in species distribution patterns and spatial resistance in five future scenarios for the New England region of the northeastern United States. We present scenario-specific distribution change maps for nine harvested wildlife species, identifying regions of increasing, decreasing, or stable habitat suitability within each scenario. Next, we isolated areas of greatest resistance across all future scenarios. Resistance was also evaluated relative to current land protection to identify patterns in and out of Protected Areas (PAs). Generally, species distributions declined in area over the 50-year assessment period (2010-2060), with the greatest declines occurring for moose (62.4%) and wild turkey (24.2%). Species resistance varied considerably across the region with coyote demonstrating the highest regional resistance (91.81% of the region) and moose demonstrating the lowest (0.76 % of the region). At the state level, average focal species resistance was highest in Maine and lowest in Massachusetts. Many of the focal species showed high overlap in resistance and land protection. Coyote, white-tailed deer, and black bear had the highest representation of resistance within PAs, while moose and wild turkey had the largest proportions of regional resistance occurring within PAs. Overall, relatively small portions of New England – ranging between 0.25% and 21.12% – were both protected and resistant for the focal species. Our results provide estimates of resistance that can inform conservation planning for commonly harvested species that are important ecologically, economically, and culturally to the region. Expanding protected area coverage to include resistant areas may provide longer term benefits to these species.

Connecticut, Maine, Massachusetts, New Hampshire, ↗
USGS2020

Drivers and consequences of alternative landscape futures on wildlife distributions in New England, United States

In an era of rapid climate and land transformation, it is increasingly important to understand how future changes impact natural systems. Scenario studies can offer the structure and perspective needed to understand the impacts of change and help inform management and conservation decisions. We implemented a scenario-based approach to assess how two high impact drivers of landscape change influence the distributions of managed wildlife species (n = 10) in the New England region of the northeastern United States. We used expert derived species distribution models (SDMs) and scenarios developed by the New England Landscape Futures Project (NELFP) to estimate how species distributions change under various trajectories (n = 5) of landscape change. The NELFP scenarios were built around two primary drivers – Socio-Economic Connectedness (SEC) and Natural Resource Planning and Innovation (NRPI) – and provide plausible alternatives for how the New England region may change over fifty years (2010 to 2060). Our models generally resulted in species occurrence and richness declines by 2060. The majority of species (7 of 10) experienced declines in regional occurrence for all NELFP scenarios, and one species experienced a projected increase in mean regional occurrence for all scenarios. Our results indicate that the NRPI and SEC drivers strongly influenced projected distribution changes compared to baseline projections. NRPI had a greater impact on distribution change for five species (coyote, moose, striped skunk, white-tailed deer, and wild turkey), while SEC had a greater impact on four species (American black bear, bobcat, raccoon, and red fox); one species (gray fox) was equally influenced by both NRPI and SEC. These results emphasize the importance of integrating both natural resource planning and socio-economic factors when addressing issues of distribution change and offer insights that can inform proactive management and conservation planning.

Connecticut, Maine, Massachusetts, New Hampshire, ↗
USGS2020

AMMonitor: Remote monitoring of biodiversity in an adaptive framework with R

Ecological research and management programs are increasingly using autonomous monitoring units (AMUs) to collect large volumes of acoustic and/or photo data to address pressing management objectives or research goals. The data management requirements of an AMU-based monitoring effort are often overwhelming, with a considerable amount of processing to translate raw data into models and analyses that have research and management utility. We created the r package AMMonitor to simplify the process of moving from remotely collected data to analysis and results, using a comprehensive SQLite database for data management that tracks all components of a remote monitoring program. This framework enables the tracking of analyses and research/management objectives through time. We illustrate the AMMonitor approach with the example of evaluating an occurrence-based management objective for a target species. First, we provide an overview of the database and data management approach. Next, we illustrate a few available workflows: temporally adaptive sampling, automated detection of species sounds from acoustic recordings and aggregation of automated detections into an encounter history for use in an occupancy analysis, the outcome of which can be analysed with respect to the motivating management objective. Without a comprehensive framework for efficiently moving from raw remote monitoring data collection to results and analysis, monitoring programs are limited in their capacity to systematically characterize ecological processes and inform management decisions through time. AMMonitor provides an option for such a framework. Code, comprehensive documentation, and step-by-step examples are available online at https://code.usgs.gov/vtcfwru/AMMonitor

Methods in Ecology and Evolution↗
USGS2020

Predicting wildlife distribution patterns in New England USA with expert elicitation techniques

Understanding the impacts of landscape change on species distributions can help inform decision-making and conservation planning. Unfortunately, empirical data that span large spatial extents across multiple taxa are limited. In this study, we used expert elicitation techniques to develop species distribution models (SDMs) for harvested wildlife species (n = 10) in the New England region of the northeastern United States. We administered an online survey that elicited opinions from wildlife experts on the probability of species occurrence throughout the study region. We collected 3396 probability of occurrence estimates from 46 experts, and used linear mixed-effects methods and landcover variables at multiple spatial extents to develop SDMs. The models were in general agreement with the literature and provided effect sizes for variables that shape species occurrence. With the exception of gray fox, models performed well when validated against crowdsourced empirical data. We applied models to rasters (30 × 30 m cells) of the New England region to map each species’ distribution. Average regional occurrence probability was highest for coyote (0.92) and white-tailed deer (0.89) and lowest for gray fox (0.42) and moose (0.52). We then stacked distribution maps of each species to estimate and map focal species richness. Species richness ( s ) varied across New England, with highest average richness in the least developed states of Vermont ( s = 7.47) and Maine (s = 7.32), and lowest average richness in the most developed states of Rhode Island (s = 6.13) and Massachusetts (s = 6.61). Our expert-based approach provided relatively inexpensive, comprehensive information that would have otherwise been difficult to obtain given the spatial extent and range of species being assessed. The results provide valuable information about the current distribution of wildlife species and offer a means of exploring how climate and land-use change may impact wildlife in the future.

Connecticut, Maine, Massachusetts, New Hampshire, ↗
USGS2019

Temporally adaptive acoustic sampling to maximize detection across a suite of focal wildlife species

Acoustic recordings of the environment can produce species presence–absence data for characterizing populations of sound-producing wildlife over multiple spatial scales. If a species is present at a site but does not vocalize during a scheduled audio recording survey, researchers may incorrectly conclude that the species is absent (“false negative”). The risk of false negatives is compounded when audio devices have sampling constraints, do not record continuously, and must be manually scheduled to operate at pre-selected times of day, particularly when research programs target multiple species with acoustic availability that varies across temporal conditions. We developed a temporally adaptive acoustic sampling algorithm to maximize detection probabilities for a suite of focal species amid sampling constraints. The algorithm combines user-supplied species vocalization models with site-specific weather forecasts to set an optimized sampling schedule for the following day. To test our algorithm, we simulated hourly vocalization probabilities for a suite of focal species in a hypothetical monitoring area for the year 2016. We conducted a factorial experiment that sampled from the 2016 acoustic environment to compare the probability of acoustic detection by a fixed (stationary) schedule versus a temporally adaptive optimized schedule under several sampling efforts and monitoring durations. We found that over the course of a study season, the probability of acoustically capturing a focal species (given presence) at least once via automated acoustic monitoring was greater (and acoustic capture occurred earlier in the season) when using the temporally adaptive optimized schedule as compared to a fixed schedule. The advantages of a temporally adaptive optimized acoustic sampling schedule are magnified when a study duration is short, sampling effort is low, and/or species acoustic availability is minimal. This methodology presents the opportunity to maximize acoustic monitoring sampling efforts amid constraints.

Colorado↗
USGS2019

Statistical learning mitigation of false positives from template-detected data in automated acoustic wildlife monitoring

Audio sampling of the environment can provide long-term, landscape-scale presence-absence data to model populations of sound-producing wildlife. Automated detection systems allow researchers to avoid manually searching through large volumes of recordings, but often produce unacceptable false positive rates. We developed methods that allow researchers to improve template-based automated detection using a suite of statistical learning algorithms when false positive rates are problematic. To test our method, we acquired 668 hours of recordings in the Sonoran Desert, California USA between March 2016 and May 2017, and created spectrogram cross-correlation templates for three target avian species. We trained and tested five classification algorithms and four performance-weighted ensemble classifier methods on target signals and false alarms from March 2016, and then selected high-performing ensemble classifiers from the train/test phase to predict the class of new detections thereafter. For three target species, our ensemble classifiers were able to identify 98%, 81%, and 100% of false alarms compared with the baseline template detection system, and comparative positive predictive values improved from 6% to 69%, 87% to 95%, and 2% to 77%. We show that statistical learning approaches can be implemented to mitigate false detections acquired via template-based automated detection in automated acoustic wildlife monitoring.

California↗
USGS2019

Dynamic wildlife occupancy models using automated acoustic monitoring data

Automated acoustic monitoring of wildlife has been used to characterize populations of sound-producing species across large spatial scales. However, false negatives and false positives produced by automated detection systems can compromise the utility of these data for researchers and land managers, particularly for research programs endeavoring to describe colonization and extinction dynamics that inform land use decision-making. To investigate the suitability of automated acoustic monitoring for dynamic occurrence models, we simulated underlying occurrence dynamics, calling patterns, and the automated acoustic detection process for a hypothetical species under a range of scenarios. We investigated an automated species detection aggregation method that considered a suite of options for creating encounter histories. From these encounter histories, we generated parameter estimates and computed bias for occurrence, colonization, and extinction rates using a dynamic occupancy modeling framework that accounts for false positives via small amounts of manual confirmation. We were able to achieve relatively unbiased estimates for all three state parameters under all scenarios, even when the automated detection system was simulated to be poor, given particular encounter history aggregation choices. However, some encounter history aggregation choices resulted in unreliable estimates; we provide caveats for avoiding these scenarios. Given specific choices during the detection aggregation process, automated acoustic monitoring data may provide an effective means for tracking species occurrence, colonization, and extinction patterns through time, with the potential to inform adaptive management at multiple spatial scales.

Ecological Applications↗
USGS2019

Bayesian statistics for beginners: A step-by-step approach

Bayesian statistics is currently undergoing something of a renaissance. At its heart is a method of statistical inference in which Bayes' theorem is used to update the probability for a hypothesis as more evidence or information becomes available. It is an approach that is ideally suited to making initial assessments based on incomplete or imperfect information; as that information is gathered and disseminated, the Bayesian approach corrects or replaces the assumptions and alters its decision-making accordingly to generate a new set of probabilities. As new data/evidence becomes available the probability for a particular hypothesis can therefore be steadily refined and revised. It is very well-suited to the scientific method in general and is widely used across the social, biological, medical, and physical sciences. Key to this book's novel and informal perspective is its unique pedagogy, a question and answer approach that utilizes accessible language, humor, plentiful illustrations, and frequent reference to on-line resources. Bayesian Statistics for Beginners is an introductory textbook suitable for senior undergraduate and graduate students, professional researchers, and practitioners seeking to improve their understanding of the Bayesian statistical techniques they routinely use for data analysis in the life and medical sciences, psychology, public health, business, and other fields.

Book↗
USGS2018

Public acceptability of development in the Northern Forest of Vermont, USA – the influence of wildlife information, recreation involvement, and demographic characteristics

Increasing development such as roads and houses will alter future landscapes and result in biological, social, and economic trade-offs. Managing development requires information on the public’s acceptability of development and understanding which factors shape acceptability. In this study, we examined three questions: 1) What is the public’s acceptability of development? 2) Is acceptability of development influenced by wildlife information? and 3) Is the maximum amount of acceptable development influenced by views about wildlife, involvement in outdoor recreation, and demographic factors? We conducted a visual-preference survey of 9,000 households in Vermont, USA that asked about acceptable levels of development, acceptability of wildlife, involvement in recreation, and individual and town demographics. The survey response rate was 44%. Maximum acceptable condition (MAC) for development was 41 houses/km 2 and not meaningfully influenced by broader consequences of development on seven common wildlife species. MAC was influenced by views on individual species, including bear and coyote, but not by other species such as deer, fox, and bobcat. Respondents with a positive attitude toward bear favored less development, whereas the opposite relationship existed for coyote. Similarly, MAC was negatively influenced by involvement in birding and hunting, but not by other common recreational activities. Among demographic factors, respondents that were younger and not born in Vermont were more accepting of development. Population density also positively influenced development acceptability. Results provide measures of the public’s acceptability of development that can help guide decision-making about development, wildlife, and recreation management.

Vermont↗
USGS2018

Estimating distribution and connectivity of recolonizing American marten in the northeastern United States using expert elicitation techniques

The American marten Martes americana is a species of conservation concern in the northeastern United States due to widespread declines from over‐harvesting and habitat loss. Little information exists on current marten distribution and how landscape characteristics shape patterns of occupancy across the region, which could help develop effective recovery strategies. The rarity of marten and lack of historical distribution records are also problematic for region‐wide conservation planning. Expert opinion can provide a source of information for estimating species–landscape relationships and is especially useful when empirical data are sparse. We created a survey to elicit expert opinion and build a model that describes marten occupancy in the northeastern United States as a function of landscape conditions. We elicited opinions from 18 marten experts that included wildlife managers, trappers and researchers. Each expert estimated occupancy probability at 30 sites in their geographic region of expertise. We, then, fit the response data with a set of 58 models that incorporated the effects of covariates related to forest characteristics, climate, anthropogenic impacts and competition at two spatial scales (1.5 and 5 km radii), and used model selection techniques to determine the best model in the set. Three top models had strong empirical support, which we model averaged based on AIC weights. The final model included effects of five covariates at the 5‐km scale: percent canopy cover (positive), percent spruce‐fir land cover (positive), winter temperature (negative), elevation (positive) and road density (negative). A receiver operating characteristic curve indicated that the model performed well based on recent occurrence records. We mapped distribution across the region and used circuit theory to estimate movement corridors between isolated core populations. The results demonstrate the effectiveness of expert‐opinion data at modeling occupancy for rare species and provide tools for planning marten recovery in the northeastern United States.

Animal Conservation↗
USGS2018

Timber harvest as the predominant disturbance regime in northeastern U.S. forests: Effects of harvest intensification

Harvesting is the leading cause of adult tree mortality in forests of the northeastern United States. While current rates of timber harvest are generally sustainable, there is considerable pressure to increase the contribution of forest biomass to meet renewable energy goals. We estimated current harvest regimes for different forest types and regions across the U.S. states of New York, Vermont, New Hampshire, and Maine using data from the U.S. Forest Inventory and Analysis Program. We implemented the harvest regimes in SORTIE‐ND, an individual‐based model of forest dynamics, and simulated the effects of current harvest regimes and five additional harvest scenarios that varied by harvest frequency and intensity over 150 yr. The best statistical model for the harvest regime described the annual probability of harvest as a function of forest type/region, total plot basal area, and distance to the nearest improved road. Forests were predicted to increase in adult aboveground biomass in all harvest scenarios in all forest type and region combinations. The magnitude of the increase, however, varied dramatically—increasing from 3% to 120% above current landscape averages as harvest frequency and intensity decreased. The variation can be largely explained by the disproportionately high harvest rates estimated for Maine as compared with the rest of the region. Despite steady biomass accumulation across the landscape, stands that exhibited old‐growth characteristics (defined as ≥300 metric tons of biomass/hectare) were rare (8% or less of stands). Intensified harvest regimes had little effect on species composition due to widespread partial harvesting in all scenarios, resulting in dominance by late‐successional species over time. Our analyses indicate that forest biomass can represent a sustainable, if small, component of renewable energy portfolios in the region, although there are tradeoffs between carbon sequestration in forest biomass and sustainable feedstock supply. Integrating harvest regimes into a disturbance theory framework is critical to understanding the dynamics of forested landscapes, especially given the predominance of logging as a disturbance agent and the increasing pressure to meet renewable energy needs.

Ecosphere↗
USGS2018

Landscape connectivity for bobcat (Lynx rufus) and lynx (Lynx canadensis) in the Northeastern United States

Landscape connectivity is integral to the persistence of metapopulations of wide ranging carnivores and other terrestrial species. The objectives of this research were to investigate the landscape characteristics essential to use of areas by lynx and bobcats in northern New England, map a habitat availability model for each species, and explore connectivity across areas of the region likely to experience future development pressure. A Mahalanobis distance analysis was conducted on location data collected between 2005 and 2010 from 16 bobcats in western Vermont and 31 lynx in northern Maine to determine which variables were most consistent across all locations for each species using three scales based on average 1) local (15 minute) movement, 2) linear distance between daily locations, and 3) female home range size. The bobcat model providing the widest separation between used locations and random study area locations suggests that they cue into landscape features such as edge, availability of cover, and development density at different scales. The lynx model with the widest separation between random and used locations contained five variables including natural habitat, cover, and elevation—all at different scales. Shrub scrub habitat—where lynx’s preferred prey is most abundant—was represented at the daily distance moved scale. Cross validation indicated that outliers had little effect on models for either species. A habitat suitability value was calculated for each 30 m 2 pixel across Vermont, New Hampshire, and Maine for each species and used to map connectivity between conserved lands within selected areas across the region. Projections of future landscape change illustrated potential impacts of anthropogenic development on areas lynx and bobcat may use, and indicated where connectivity for bobcats and lynx may be lost. These projections provided a guide for conservation of landscape permeability for lynx, bobcat, and species relying on similar habitats in the region.

PLoS ONE↗
USGS2018

AMModels: An R package for storing models, data, and metadata to facilitate adaptive management

Agencies are increasingly called upon to implement their natural resource management programs within an adaptive management (AM) framework. This article provides the background and motivation for the R package, AMModels . AMModels was developed under R version 3.2.2. The overall goal of AMModels is simple: To codify knowledge in the form of models and to store it, along with models generated from numerous analyses and datasets that may come our way, so that it can be used or recalled in the future. AMModels facilitates this process by storing all models and datasets in a single object that can be saved to an .RData file and routinely augmented to track changes in knowledge through time. Through this process, AMModels allows the capture, development, sharing, and use of knowledge that may help organizations achieve their mission. While AMModels was designed to facilitate adaptive management, its utility is far more general. Many R packages exist for creating and summarizing models, but to our knowledge, AMModels is the only package dedicated not to the mechanics of analysis but to organizing analysis inputs, analysis outputs, and preserving descriptive metadata. We anticipate that this package will assist users hoping to preserve the key elements of an analysis so they may be more confidently revisited at a later date.

PLoS ONE↗
USGS2018

Monitoring Least Bitterns (Ixobrychis exilis) in Vermont: Detection probability and occupancy modeling

Ixobrychus exillis (Least Bittern) is listed as a species of high concern in the North American Waterbird Conservation Plan and is a US Fish and Wildlife Service migratory bird species of conservation concern in the Northeast. Little is known about the population of Least Bitterns in the Northeast because of their low population density, tendency to nest in dense wetland vegetation, and secretive behavior. Urban and agricultural development is expected to encroach on and degrade suitable wetland habitat; however, we cannot predict the effects on Least Bittern populations without more accurate information on their abundance and distribution. We conducted surveys of wetlands in Vermont to assess the efficacy of a monitoring protocol and to establish baseline Least Bittern abundance and distribution data at a sample of 29 wetland sites. Surveys yielded detections of 31 individuals at 15 of 29 sites across 3 biophysical regions and at 5 sites where occupancy had not been previously reported. Probability of occupancy was positively related to wetland size and number of patches, though the relationships were not strong enough to conclude if these were true determinants of occupancy. Call—response broadcast surveys yielded 30 detections, while passive surveys yielded 13. Call—response broadcasts ( P = 0.897) increased the rate of detection by 55% compared to passive surveys ( P = 0.577). Our results suggest that call—response broadcast surveys are an effective means of assessing Least Bittern occupancy and may reduce bias in long-term monitoring programs.

Vermont↗
USGS2017

Estimating Mudpuppy (Necturus maculosus) abundance in the Lamoille River, Vermont, USA

The Mudpuppy (Necturus maculosus) is classified as a Species of Greatest Conservation Need by the state of Vermont. There is concern regarding status of populations in the Lake Champlain basin because of habitat alteration and potential effects of 3-trifluromethyl-4-nitrophenol (TFM), a chemical used to control Sea Lamprey (Petromyzon marinus). The purpose of our research was to assess Mudpuppy capture methods and abundance in the Lamoille River, Vermont, USA. We sampled Mudpuppies under a mark-recapture framework, using modified, baited minnow traps set during two winter-spring periods. We marked each Mudpuppy with a passive integrated transponder (PIT) tag and released individuals after collecting morphological measurements. We collected 80 individuals during 2,581 trap days in 2008–2009 (year 1), and 81 individuals during 3,072 trap days in 2009–2010 (year 2). We estimated abundance from spring trapping periods in 2009 and 2010, during which capture rates were sufficient for analysis. Capture probability was low (< 0.04), but highest following precipitation events in spring, during periods of higher river flow, when water temperatures were approximately 3 to 6° C. During October 2009, management agencies treated the Lamoille River with TFM. Surveyors recovered more than 500 dead Mudpuppies during the post-treatment assessment. Overall, Mudpuppy captures did not change between sampling periods; however, we captured fewer females during year 2 compared to year 1, and the sex ratio changed from 0.79:1 (M:F) during year 1 to 3:1 (M:F) during year 2. Our data may help wildlife managers assess population status of Mudpuppies in conjunction with fisheries management techniques.

Vermont↗
USGS2016

Predicting the stability of endangered stonecats in the LaPlatte River, Vermont

Stonecats Noturus flavus in Vermont conform to a rare distribution pattern (as designated by Rabinowitz 1981) because their known distribution within the state is limited to the LaPlatte and Missisquoi rivers. We focused on Stonecats in the LaPlatte River to predict the stability of the population. During 2012&ndash;2014, we captured Stonecats via backpack electrofishing; fish were PIT-tagged (>90 mm TL) and marked with visible implant elastomer. Among the 1,671 Stonecats that were captured, 1,252 were PIT-tagged. Only 156 (12%) of the PIT-tagged fish were recaptured, and only 22 of those individuals were recaptured more than once. The Pradel model in Program MARK was used to estimate apparent survival (&Phi;) and seniority, which were used to derive the rate of population change (&lambda;) for the Stonecat encounter histories we studied. We examined a total of 64 models in our candidate set, with the following covariates: TL at first capture, maximum temperature, season, maximum discharge, and area sampled. Survival estimates were highest in the spring (range of daily &Phi; = 0.9993&ndash;0.9995) and increased with greater TL at first capture. We also estimated increases in capture probability with increasing area sampled. We derived an annual &lambda; of 0.9794, which indicates a slightly decreasing population. However, our &lambda; estimate contained uncertainty that was likely increased due to the low recapture rates. Additional years of data could increase the accuracy of the &lambda; estimate. In the meantime, we have provided insight into Stonecat population parameters that were otherwise unknown.

Vermont↗
USGS2014

Normative standards for land use in Vermont: Implications for biodiversity

The conversion of natural lands to developed uses poses a great threat to global terrestrial biodiversity. Natural resource managers, tasked with managing wildlife as a public trust, require techniques for predicting how much and where wildlife habitat is likely to be converted in the future. Here, we develop a methodology to estimate the “social carrying capacity for development” – SK d – for 251 towns across the state of Vermont, USA. SK d represents town residents’ minimum acceptable human population size and level of development within town boundaries. To estimate SK d across towns within the state of Vermont (USA), as well as the average state-wide SK d , we administered a visual preference survey ( n = 1505 responses) to Vermont residents, and asked respondents to rate alternative landuse scenarios in a fictional Vermont town on a scale of +4 (highly acceptable) to −4 (highly unacceptable). We additionally collected demographic data such as age and income, as well as ancillary information such as participation in town-planning meetings and location of residence. We used model selection and AIC to fit a cubic function to the response data, allowing us to estimate SK d at a town scale based on town demographic characteristics. On average, Vermonters had a SK d of 9.1% development on the landscape; this estimate is 68% higher than year 2000 levels for development (5.4%). Respondents indicated that management action to curb development was appropriate at 9.4% development (roughly the statewide SK d average). Management by local, regional, and state levels were considered acceptable for curbing development while federal level management of development was considered unacceptable. Given a scenario where development levels were at SK d , we predicted a 16,753 km 2 reduction in forested land (−11.16%) and a 1038 km 2 reduction in farmland (−60.45%). Such changes would dramatically alter biodiversity patterns state-wide. In a companion paper, we estimate how these changes would affect the distribution of wildlife species.

Vermont↗