September 2025
In this issue:
- MRG and Fordham’s Calder Center partner on 10-week deer study
- MRG Maps Wildlife Corridors for Town of Bedford
- New urban wildlife research published with Queens College and Mercy University
- Support MRG’s Science & Research
MRG and Fordham’s Calder Center partner on 10-week deer study
This summer, Chris Nagy mentored an undergraduate student from Fordham University as part of the Louis Calder Center’s Summer Undergraduate Research (CSUR) program.
Claire Renault, a sophomore at Fordham, spent the summer testing a new method of measuring the abundance of free-roaming wildlife using motion-activated camera traps. For this study, she looked specifically at white-tailed deer, but the method could be used to measure the abundance of any medium- or large-sized species that is relatively common.
The Instantaneous Sample (IS) method was originally developed by Moeller et al. (2018) but Claire and Dr. Nagy made a few modification to the technique that allows for easy implementation in the field. Among other requirements, the area in front of each camera that is photographed needs to be measured. The full area a camera trap can photograph resembles a pie slice-shaped sector (Figure 1a). Also, in reality this viewshed is a complex surface where the probability of detecting an animal in a photograph declines with distance from the camera. The real “observable” viewshed in front of a camera can also be complicated by terrain and objects like trees (Figure 1b).
Instead of using the entire possible viewable area in front of each camera, Claire restricted the area to some portion of the viewshed where the chance of seeing a photographed deer was effectively 100% (Figure 1c). Each camera was placed facing a linear feature like a log that could be used as a boundary visible in the final photographs. The triangle formed by the field of view of our cameras to this boundary feature represented the plot within which the deer would be counted. Any animals photographed outside of that plot are not counted (Figure 2).
Measuring the size of a smaller triangular field over a clear area is easier than estimating a much larger and spatially complex sector, and can be done in the field with just a tape measure and a large protractor. Time lapse photos taken at set intervals, rather than motion-triggered photos, are preferable when using the IS method, and any deer that happen to be in the designated plot area when the photo is taken are counted. Using these repeated counts-over-known area, one can estimate the density of the species of interest – in our case, deer / km2.
Claire ran cameras at the Louis Calder Center and at Ward Pound Ridge and Muscoot Farm County Parks, examining more than 100k photos taken every 15 minutes over 3 weeks at each park. She also incorporated data from MRG’s own survey performed by a WTP student Atiksh Bordia last September.
We estimated density at all of these sites and found a wide range of deer density (Figure 3), with some correlation with lower deer density and the number of years the sites have been managing deer (r = 0.52), and total deer harvested at each site (r = 0.58). All of these sites are implementing deer management of some form, and thus estimates of how many deer each site have are vital to assessing their programs.
Our plan is to refine the method a bit more this fall and implement it on our Preserve annually. In the coming year, we will be working on guidelines for camera density and deployment duration that will yield acceptable precision (i.e., narrowing those error bars around our estimates in Figure 3). MRG will offer the method to other regional stakeholders interested in monitoring their wildlife populations as well.
We also want to thank our 2025 CISE interns for helping Claire with this project!
Moeller, A. K., P. M. Lukacs, and J. S. Horne. 2018. Three novel methods to estimate abundance of unmarked animals using remote cameras. Ecosphere 9(8):e02331. 10.1002/ecs2.2331
MRG Maps Wildlife Corridors for Town of Bedford
In 2024, MRG generated maps of potential wildlife corridors in and around the Town of Bedford, NY, as part of a grant from the New York Hudson River Estuary Program. The Town initiated this project to gain a better understanding of the areas that are important for wildlife movement and landscape connectivity. These predictions will inform planning and conservation efforts across the Town.
In fragmented and developed landscapes, the ability for animals to move between patches of habitat is critically important for maintaining the long-term viability of populations. Young individuals need to be able to leave their natal areas and find new territories of their own. At local scales, individual patches may provide certain resources, but for larger animals such as bobcats, bears, and even deer, different food types, water, and refuge may be located in different patches or shift over time, and these animals need to be able to move between them. Resources can change over time because of disturbance, seasonally, and because of longer-term shifts such as climate change.
The need for connectivity also applies even to plants or small, less–mobile animals, but instead of the movement of individuals, populations can shift and interact across generations via small movements over time or seed dispersal. This brings a genetic element into the importance of connectivity.
Simply put, if organisms can move, they can adapt.
Direct measurement of how animals move in and around Bedford would require a major effort, involving trapping of live animals, equipping them with GPS collars, and monitoring and analyzing their movements for, ideally, many years. The resources and time for this type of study were not available.
However, ecologists can use existing knowledge on species’ spatial ecology to make predictions on where pathways probably exist in a landscape. Based on several other similar efforts at the town and regional level (notably, the Town of Red Hook’s 2014 project) MRG was tasked with predicting the likeliest corridors given existing knowledge and using spatial modeling techniques. This would generate maps of our best predictions of where wildlife tend to move in and through Bedford.
MRG landed on using least-cost path (LCP) analysis to map wildlife corridors. We developed maps for bobcats, flying squirrels, and forest bats (e.g., hoary, silver-haired, and eastern red bats). These are species that are present in the Town but may have some limitations in moving through suburban/urban development and/or crossing roads of various sizes. We also developed a generic model that illustrated the general patterns of many medium and large mammal species in our area.
MRG used existing studies on bobcats, flying squirrels, and bats to choose resistance values for each model, and for the “generic” wildlife model we prioritized natural areas, forest-grassland transitions, and riparian corridors. We implemented LCP analysis on the area within a 12 mile wide buffer around the Town (Figure 1). The generic model was a good composite of the 3 species-specific models, and identified several pinchpoints associated with, not surprisingly, the major highways.
MRG was also able to perform some field validation by putting trail cameras on some of the predicted corridors (Figure 2).
There were many sources of potential but un-mapped obstacles to wildlife movement. First, as we explored some of the corridors on the ground we found many private fences throughout the Town. Depending on the size, type, and extent, fences can certainly affect wildlife movements but are not mapped comprehensively across the Town or County. Also, thanks to the hard work of a group of volunteers, we were able to assess dozens of roadside locations where predicted corridors crossed a road. Roads are likely the most important restriction on animal movements in any developed area and examining microfeatures around roads can assist in figuring out how animals are crossing (or are failing to cross) streets and highways.
Because we expanded the area we evaluated so far beyond the Town, MRG will be able to use these maps to direct our own conservation and land acquisition efforts. We also hope to refine these models and assist other municipalities in their conservation planning.
New urban wildlife research published with Queens College and Mercy University
MRG’s long-running Gotham Coyote Project began a collaboration with researchers from Queens College, Mercy University, and the American Museum of Natural History in 2020. As part of her thesis at Queens College, master’s student Angelinna Bradfield analyzed GCP’s large 2016-2019 dataset of camera trap images collected across NYC, under the supervision of Dr. David Lahti and Dr. Bobby Habig.
While this work was first presented in Angelinna’s thesis, which she successfully defended in early 2022, the team was also able to publish two papers from her work.
The first paper detailed Angelinna’s examination of mammalian biodiversity across NYC. She found that geography and intensity of development affected what species were present in City Parks. Specifically, parks that were situated in less urbanized areas, had larger forest area, and were situated on the mainland had higher species richness and evenness. Species diversity was higher overall in the Bronx than on any of the islanded boroughs, indicating that geographic barriers can play a role in what species are present in different areas of the City.
The second paper looked at the occurrence of free-ranging cats in NYC greenspaces. Cats were found in smaller parks that were surrounded by more development, but lower human population densities (think commercial or industrial areas). It seems cats certainly use City greenspaces but, as a domestic species, in a near-opposite way compared to wild, native species.
From Bradfield et al. (2025): Associations of free-ranging cat occupancy with (A) patch area, (B) log human population density (humans/km2), and (C) percent developed land cover. Shaded gray areas indicate 95% confidence intervals.
Also, cameras that photographed coyotes were less likely to also photograph cats, even if cats were present in the area. Thus, while not excluding cats from a site, coyotes may alter their behavior.
From Bradfield et al. (2025): Associations of free-ranging cat detection probability with (A) patch area, (B) log human population density (humans/km2), (C) percent developed land cover, and (D) presence of coyotes. Shaded gray areas (A–C) indicate 95% confidence intervals. Points and whiskers on the plot (D) represent the mean and SE.
Chris Nagy and Dr. Habig are continuing to collaborate on several NYC-related projects going forward. MRG’s research program has made impressive progress due to collaborations such as these. Each partner can bring their specific expertise, various resources, and people-hours to bear on projects that none of us could do on our own. Learn more about our ecological studies, our students, and our partners at the research section of our main website.
Bradfield, A., Nagy, C., Weckel, M. et al. 2025. Cats in the city: urban cat distribution is influenced by habitat characteristics, anthropogenic factors, and the presence of coyotes. Urban Ecosyst 28, 106 (2025). https://doi.org/10.1007/s11252-025-01709-3.
Bradfield, A.A.; Nagy, C.M.; Weckel, M.; Lahti, D.C.; Habig, B. 2022. Predictors of Mammalian Diversity in the New York Metropolitan Area. Front. Ecol. Evol. 10, 903211.
More on MRG’s Research Program…
Our research program depends on donations. Please consider supporting our scientific work.
In addition to land acquisition and management, MRG conducts ecological research on our lands and in the region. Since 2005, we have offered three student-based research programs in the field of ecology. These include:
- The Wildlife Technician Program (WTP) – a 2.5-year high school mentorship program
- The College Internship in Suburban Ecology – a 10-week summer internship for college students
- The Research Assistantship Program – a grant of $5,000 / year for up to 3 years to support graduate student research
Our staff also conduct long-term research on wildlife, invasive species management, and habitat restoration. We area also partners in several national or global research partnerships, where members pool their expertise and data to explore big questions about biology and conservation. Some examples are:
Gotham Coyote Project – Our long-term study of the range expansion and ecology of coyotes in New York City
Environmental Monitoring and Management Alliance – A network of conservation organizations, land trusts, field stations, and others that seeks to coordinate environmental monitoring protocols and address large-scale questions in the Hudson Valley and larger region.
Hudson-to-Housatonic Regional Conservation Partnership – H2H is a land protection collaboration that integrates research, geographic modeling
More info is available on our website! Please consider helping us out!