Regional Supervisor - Northeast Units - Cooperative Fish and Wildlife Research Units
Research Interests
Dr. Loftin's interests in systems, landscape, and wetlands ecology drive her research program involving wetlands restoration issues. Her research interest is aquatic ecology with an emphasis on conservation and restoration of perturbed, freshwater systems. The research program encompasses two concentrations: hydrological influences on aquatic vegetation dynamics and identifying effects of selected human-introduced stressors on Maine’s aquatic systems and their inhabitants. Current studies include identifying effects of fish introductions on invertebrate and amphibian communities in Maine's naturally fishless lakes; examining amphibian use of aquatic systems and the surrounding terrestrial landscape; identifying characteristics of wetlands supporting shrubby cinquefoil, host plant for Clayton's copper butterfly; developing algorithms with remote sensing data to estimate water clarity in Maine's lakes; examining uptake of marine derived nutrients by food web components in Atlantic salmon nursery streams; developing methods to remotely monitor disturbance of nesting seabirds; and, studying vegetation response to fire in Okefenokee Swamp, GA.
Teaching Interests
Dr. Loftin teaches a graduate level course (INT527 Applications of Remote Sensing and GIS in Natural Resource Management; 3 credits lecture and lab) each fall semester at the University of Maine. The course is team taught with Dr. Steven Sader, UMaine School of Forest Resources, and provides instruction on using remote sensing and GIS tools to conduct spatial analysis for projects with a conservation focus.
Professional Experience
Supervisor, Cooperative Research Units Program Headquarters, 2021-
Education and Certifications
Ph D University of Florida 1998
MS Auburn University 1987
BA University of Virginia 1984
Science and Products
Distribution Models Predicting Groundwater Influenced Ecosystems in the Northeastern United States
Globally, groundwater dependent ecosystems (GDEs) are increasingly vulnerable to groundwater extraction and land use practices. Groundwater supports these ecosystems by providing inflow, which can maintain water levels, water temperature, and chemistry necessary to sustain the biodiversity that they support. Many aquatic systems receive groundwater as a portion of base flow, and in some systems (e
Avian point count survey-specific data in Northern New England 2013 to 2015
Avian point count, vegetation, and management data from surveys in Northern New England at four National Wildlife Refuges (NWRs), Baxter State Park, and prviate lands in Northern Maine surveyed during the breeding and postbreeding season bewteen 1 June and 4 August 2013 to 2015. NWRs include Nulhegan (Silvio O. Conte), Umbagog, Moosehorn, and Aroostook. This data release includes five zip folders
Filter Total Items: 45
A manual for remote sensing of Maine lake clarity
The purpose of this manual is to support use of satellite-based remote sensing for statewide lake water-quality monitoring in Maine. The authors describe step-by-step methods that combine Landsat and MODIS satellite data with field-collected Secchi disk data for statewide assessment of lake water clarity. Landsat can be simultaneously used to assess more than Maine 1,000 lakes ≥ 8 ha, whereas MOD
Habitat associations of breeding conifer-associated birds in managed and regenerating forested stands
Forests are often affected by management that could influence demographics of breeding and post-breeding birds that reside within. Numerous studies have focused on immediate effects from management on wildlife soon after forestry treatment (e.g., 0–5 years), however, fewer studies have examined changes in focal species abundance over longer durations as a forest regenerates after disturbance. We e
Context-dependent deep learning
Explicitly representing an agent’s context has been shown to have many benefits, which should also apply to machine learning. In this paper, we describe an approach to do this called context-dependent deep learning (CDDL), which is based on earlier work in context-mediated behavior (CMB) that uses contextual schemas (c-schemas) to represent clas-ses of situations along with knowledge useful in the
Using landscape metrics to characterize towns along an urban-rural gradient
ContextUrban-rural gradients are useful tools when examining the influence of human disturbances on ecological, social and coupled systems, yet the most commonly used gradient definitions are based on single broad measures such as housing density or percent forest cover that fail to capture landscape patterns important for conservation.ObjectivesWe present an approach to defining urban–rural gradi
How well do proxy species models inform conservation of surrogate species?
ContextProxy species, which represent suites of organisms with similar habitat requirements, are common in conservation. Landscape Capability (LC) models aim to quantify the spatially-explicit capability of landscapes to support proxy species that represent suites of forest birds.ObjectivesWe evaluated the North Atlantic Landscape Conservation Cooperative (NALCC) proxy models of LC and represented
Testing prediction accuracy in short-term ecological studies
Applied ecology is based on an assumption that a management action will result in a predicted outcome. Testing the prediction accuracy of ecological models is the most powerful way of evaluating the knowledge implicit in this cause-effect relationship, however, the prevalence of predictive modeling and prediction testing are spreading slowly in ecology. The challenge of prediction testing is parti
Non-crop habitat use by wild bees (Hymenoptera: Apoidea) in a mixed-use agricultural landscape
Homogeneous, agriculturally intense landscapes have abundant records of pollinator community research, though similar studies in the forest-dominated, heterogeneous mixed-use landscape that dominates the northeastern United States are sparse. Trends of landscape effects on wild bees are consistent across homogeneous agricultural landscapes, whereas reported studies in the northeastern United State
Combined influence of intrinsic and environmental factors in shaping productivity in a small pelagic gull, the black-legged kittiwake Rissa tridactyla
While we have a good understanding in many systems of the effects of single variable changes on organisms, we understand far less about how variables act in concert to affect living systems, where interactions among variables can lead to unanticipated results. We used mixed-effect models to evaluate the effects of multiple variables that we expected to play a role in the early reproductive stages
Landscape genetics reveals unique and shared effects of urbanization for two sympatric pool-breeding amphibians
Metapopulation-structured species can be negatively affected when landscape fragmentation impairs connectivity. We investigated the effects of urbanization on genetic diversity and gene flow for two sympatric amphibian species, spotted salamanders (Ambystoma maculatum) and wood frogs (Lithobates sylvaticus), across a large (>35,000 km2) landscape in Maine, USA, containing numerous natural and anth
Survival and recruitment dynamics of Black-legged Kittiwakes Rissa tridactyla at an Alaskan colony
The majority of seabirds breed colonially and exhibit considerable site fidelity over the course of their long lifespans. Initial colony selection can therefore have substantial fitness consequences; however, factors contributing to recruitment into colonies and subsequent fidelity remain unclear. We used multi-state capture-recapture models to test several hypotheses related to apparent fledgling
Replicated landscape genomics identifies evidence of local adaptation to urbanization in wood frogs
Native species that persist in urban environments may benefit from local adaptation to novel selection factors. We used double-digest restriction-side associated DNA (RAD) sequencing to evaluate shifts in genome-wide genetic diversity and investigate the presence of parallel evolution associated with urban-specific selection factors in wood frogs (Lithobates sylvaticus). Our replicated paired stud
Satellite-detected forest disturbance forecasts American marten population decline: The case for supportive space-based monitoring
Limited monitoring resources often constrain rigorous monitoring practices to species or populations of conservation concern. Insufficient monitoring can induce a tautology as lack of monitoring resources makes it difficult to determine whether a species or population deserves additional monitoring resources. When in-situ monitoring resources are limited, remote habitat monitoring could be a usefu
Science and Products
- Data
Distribution Models Predicting Groundwater Influenced Ecosystems in the Northeastern United States
Globally, groundwater dependent ecosystems (GDEs) are increasingly vulnerable to groundwater extraction and land use practices. Groundwater supports these ecosystems by providing inflow, which can maintain water levels, water temperature, and chemistry necessary to sustain the biodiversity that they support. Many aquatic systems receive groundwater as a portion of base flow, and in some systems (eAvian point count survey-specific data in Northern New England 2013 to 2015
Avian point count, vegetation, and management data from surveys in Northern New England at four National Wildlife Refuges (NWRs), Baxter State Park, and prviate lands in Northern Maine surveyed during the breeding and postbreeding season bewteen 1 June and 4 August 2013 to 2015. NWRs include Nulhegan (Silvio O. Conte), Umbagog, Moosehorn, and Aroostook. This data release includes five zip folders - Publications
Filter Total Items: 45
A manual for remote sensing of Maine lake clarity
The purpose of this manual is to support use of satellite-based remote sensing for statewide lake water-quality monitoring in Maine. The authors describe step-by-step methods that combine Landsat and MODIS satellite data with field-collected Secchi disk data for statewide assessment of lake water clarity. Landsat can be simultaneously used to assess more than Maine 1,000 lakes ≥ 8 ha, whereas MODHabitat associations of breeding conifer-associated birds in managed and regenerating forested stands
Forests are often affected by management that could influence demographics of breeding and post-breeding birds that reside within. Numerous studies have focused on immediate effects from management on wildlife soon after forestry treatment (e.g., 0–5 years), however, fewer studies have examined changes in focal species abundance over longer durations as a forest regenerates after disturbance. We eContext-dependent deep learning
Explicitly representing an agent’s context has been shown to have many benefits, which should also apply to machine learning. In this paper, we describe an approach to do this called context-dependent deep learning (CDDL), which is based on earlier work in context-mediated behavior (CMB) that uses contextual schemas (c-schemas) to represent clas-ses of situations along with knowledge useful in theUsing landscape metrics to characterize towns along an urban-rural gradient
ContextUrban-rural gradients are useful tools when examining the influence of human disturbances on ecological, social and coupled systems, yet the most commonly used gradient definitions are based on single broad measures such as housing density or percent forest cover that fail to capture landscape patterns important for conservation.ObjectivesWe present an approach to defining urban–rural gradiHow well do proxy species models inform conservation of surrogate species?
ContextProxy species, which represent suites of organisms with similar habitat requirements, are common in conservation. Landscape Capability (LC) models aim to quantify the spatially-explicit capability of landscapes to support proxy species that represent suites of forest birds.ObjectivesWe evaluated the North Atlantic Landscape Conservation Cooperative (NALCC) proxy models of LC and representedTesting prediction accuracy in short-term ecological studies
Applied ecology is based on an assumption that a management action will result in a predicted outcome. Testing the prediction accuracy of ecological models is the most powerful way of evaluating the knowledge implicit in this cause-effect relationship, however, the prevalence of predictive modeling and prediction testing are spreading slowly in ecology. The challenge of prediction testing is partiNon-crop habitat use by wild bees (Hymenoptera: Apoidea) in a mixed-use agricultural landscape
Homogeneous, agriculturally intense landscapes have abundant records of pollinator community research, though similar studies in the forest-dominated, heterogeneous mixed-use landscape that dominates the northeastern United States are sparse. Trends of landscape effects on wild bees are consistent across homogeneous agricultural landscapes, whereas reported studies in the northeastern United StateCombined influence of intrinsic and environmental factors in shaping productivity in a small pelagic gull, the black-legged kittiwake Rissa tridactyla
While we have a good understanding in many systems of the effects of single variable changes on organisms, we understand far less about how variables act in concert to affect living systems, where interactions among variables can lead to unanticipated results. We used mixed-effect models to evaluate the effects of multiple variables that we expected to play a role in the early reproductive stagesLandscape genetics reveals unique and shared effects of urbanization for two sympatric pool-breeding amphibians
Metapopulation-structured species can be negatively affected when landscape fragmentation impairs connectivity. We investigated the effects of urbanization on genetic diversity and gene flow for two sympatric amphibian species, spotted salamanders (Ambystoma maculatum) and wood frogs (Lithobates sylvaticus), across a large (>35,000 km2) landscape in Maine, USA, containing numerous natural and anthSurvival and recruitment dynamics of Black-legged Kittiwakes Rissa tridactyla at an Alaskan colony
The majority of seabirds breed colonially and exhibit considerable site fidelity over the course of their long lifespans. Initial colony selection can therefore have substantial fitness consequences; however, factors contributing to recruitment into colonies and subsequent fidelity remain unclear. We used multi-state capture-recapture models to test several hypotheses related to apparent fledglingReplicated landscape genomics identifies evidence of local adaptation to urbanization in wood frogs
Native species that persist in urban environments may benefit from local adaptation to novel selection factors. We used double-digest restriction-side associated DNA (RAD) sequencing to evaluate shifts in genome-wide genetic diversity and investigate the presence of parallel evolution associated with urban-specific selection factors in wood frogs (Lithobates sylvaticus). Our replicated paired studSatellite-detected forest disturbance forecasts American marten population decline: The case for supportive space-based monitoring
Limited monitoring resources often constrain rigorous monitoring practices to species or populations of conservation concern. Insufficient monitoring can induce a tautology as lack of monitoring resources makes it difficult to determine whether a species or population deserves additional monitoring resources. When in-situ monitoring resources are limited, remote habitat monitoring could be a usefu