Abstract
Trap-neuter-return (TNR) is promoted as a “humane” alternative to lethal methods for population control of feral domestic cats (Felis catus). This paper explores feedbacks between feral domestic cats, coyotes (Canis latrans), raccoons (Procyon lotor), and skunks (Mephitis mephitis) at a TNR feral cat colony in Rhode Island, USA. A total of 12,272 photographs from a motion-activated camera were analyzed. Cat population size and visitation frequency of wildlife were estimated during three different feeding regimes. Abundant food on the ground was associated with increased wildlife visits, while elevated or limited food was associated with decreased wildlife visits. During the two-year study period, the population of cats dropped from 17 to 12 individuals and the cats appeared to have short life spans, which could have been due to predation by coyotes. Our results suggest that wildlife confluence and predation risks can be influenced by feeding regime.
Domestic cats (Felis catus) are descendants of a subspecies of the wildcat (F. silvestris libyca) and were tamed by early grain farmers in the “Fertile Crescent” to help control rodents (Ottoni et al., 2017). Genetic evidence suggests that cats were domesticated again in early ancient Egypt, but Egyptian domestic cats were probably tamer and more sociable than the cats domesticated earlier (Ottoni et al., 2017). Domestic cats (hereafter “cats”) held these dual roles – companion animal and self-reliant hunter – for millennia. Today, however, their hunting skills are less in demand.
There are an estimated 60–90 million cats living with humans as companion animals in the US today, but there are an additional 10–50 million feral cats living in outdoor environments, i.e., cats that have minimal or no support from humans (Baker et al., 2010). As they are forced to integrate with local ecosystems, feral cats are often thought of as victims of circumstance, but also as a threat to native wildlife (Peterson et al., 2012). Calls for population control of feral cats – in particular, lethal methods – are understandably being met with resistance and appeals for context-specific measures and humane treatment (Lynn et al., 2019; Wandesforde-Smith et al., 2021).
Trap-neuter-return (TNR) is promoted and widely accepted as a “humane” way of reducing the population of feral cats (Wolf & Hamilton, 2020). In TNR projects, feral cats are captured, neutered, sometimes vaccinated, and reintroduced to the colony where they were caught (Barrows, 2004; Wolf & Hamilton, 2020). The success of TNR programs in reducing the number of feral cats has been mixed, however, and high neutering rates (Boone et al., 2019; Coe et al., 2021; Foley et al., 2005; Hostetler et al., 2020; Kreisler et al., 2019) and sometimes additional measures of population reduction (indoor adoption, or in some cases euthanasia; Hostetler et al., 2020) are needed. In a study in Tel Aviv, Israel, Finkler et al. (2011) found that communities with more resources to direct to TNR efforts had more success than less affluent communities.
Neutered free-ranging cats may not reproduce, but they continue to interact with their surrounding environment, causing a range of concerns, including predation and disease transfer (Barrows, 2004). Cats kill small mammals, birds, reptiles, amphibians, and invertebrates (Baker et al., 2005; Blancher, 2013; Loss et al., 2022; McDonald et al., 2015; van Heezik et al., 2010; Woolley et al., 2020), and it was found that well-fed cats continue to show high predation rates on small wildlife (Herrera et al., 2022). Additionally, their mere presence has been shown to reduce nesting success of birds (Bonnington et al., 2013). TNR colonies may also create undesirable interactions, including disease transfer, with other mesopredators (Hernandez et al., 2018; Jessup, 2004).
It is possible that feeding and neutering at TNR colonies may reduce the size of cats’ home ranges, thus reducing the spatial effects of TNR colonies on surrounding wildlife (Kays et al., 2020), but a meta-analysis found no such effects (Hall et al., 2016) and more research is needed. When it comes to effects on local wildlife, the location of the TNR colony may be crucial, too, because cats in dense human settlements have smaller home ranges and kill less wildlife than in more rural areas (Herrera et al., 2022; Pirie et al., 2022; Piontek et al., 2021). The impact on wildlife is not unidirectional, however, and cat colonies in North America may face predation by coyotes (Canis latrans), who are attracted to the cat food and prey opportunistically on cats (Grubbs & Krausman, 2009; Nation & Clair, 2019; Poessel et al., 2017; Quinn, 1997).
In this paper, we report and reflect upon findings from a camera trap survey carried out at a TNR cat colony in Rhode Island, USA. Over the course of two years, colony tenders used three distinctly different methods to feed the cats, which created an opportunistic experiment into how feeding regime affects (a) timing of visitation and visitation rates by cats; (b) visitation rates by coyotes (Canis latrans), raccoons (Procyon lotor), and skunks (Mephitis mephitis); and (c) interactions between cats and wildlife. We also report on cat population structure and age. Based on these findings, we make suggestions on feeding regime and monitoring at TNR cat colonies.
Materials and Methods
Research Partnership and Study Site
In 2012, with the goal of developing joint solutions for the large Rhode Island (RI) feral cat population, a group of stakeholders formed the RI Feral Cat Working Group (RIFCWG) headed by the State Veterinarian (SM) at the Department of Environmental Management/Division of Agriculture. Members included four veterinarians, six animal welfare organizations, four animal control officers, and three conservation biologists. PawsWatch, a RI TNR organization and member of RIFCWG, located a large colony of managed cats for this study, captured cats, and assisted with data collection.
Work was conducted at a feral cat colony behind the Bainbridge Apartment Complex in Johnston, RI. The feeding station was located in a mixed evergreen/deciduous woodland with a dense understory of briars (predominantly Rosa multiflora and Smilax glauca), close to the residential apartments. The colony was monitored by PawsWatch, whose members had been working at the site for several years prior to the onset of our study in April 2012.
Photo Analysis and Cat Identification
We used a single Bushnell 12MP Trophy Camera fixed at a height of 1.5 m. Prior to July 2012, the camera was placed on a trail leading to the feeding station, after which it was moved to a location four meters from the feeding station where cats were observed to congregate. The camera took motion-activated photos 24 hours per day (full color in daylight, grey scale from dusk to dawn) with a one-second delay.
We used PhotoMechanic 5 (Camera Bits, Inc.) to sort and save animal-triggered shots for further evaluation. We scored each animal present in a photo frame as one “visit” to the feeder and recorded total visits per day. Visits by wildlife were summed by species (e.g., six raccoon visits). Cats could generally be individually identified (see below); therefore cat visits were summed separately by individual (e.g., three visits by Cat 4, five visits by Cat 8). Unidentifiable animals (blurred or partial photos) were ignored. For this study, we examined 12,272 camera-trap images.
We identified individual cats by markings and characteristics, which we compiled in a spreadsheet identification key. Each cat’s record contained a thumbnail photo of the cat, known characteristics – color (generally useful in full color daylight photos only) or lightness, distinctive spot or stripe patterns, fur length, distinctive injuries, sex, and ear-clips by TNR personnel – and a link to a companion folder of full-color digital photos showing, if possible, lateral, anterior, and posterior views of the cat.
Cat Colony Feeding Regime
During the study, the colony tenders in Johnston modified the initial feeding regime twice, in response to recommendations from PawsWatch, based on a preliminary report of our findings to RIFCWG at the end of 2012 (Fletcher, 2013). This created an opportunity to study the impact of three different feeding regimes on wildlife confluence. We monitored the feeding regime from 2012–2013. Cats were fed Mondays, Wednesdays, and Fridays at 08:00–08:30. The food was placed in a wooded area adjacent to a quiet parking lot. Both the quantity of food and method of delivery varied as follows:
Feeding Regime 1, large quantities at ground level: Initially and during all three sampling sessions in 2012 – April 6–11 (568 photos), July 26–August 2 (1773 photos), and Aug 15–22 (3145 photos) – colony tenders delivered approximately 1.3–2.2 kg (3–5 lb) of dry food per visit. The food was heaped in a large salad bowl, or two “deli” trays sheltered by a plastic bin (Figure S1 in supplemental material).
Feeding Regime 2, large quantities elevated: During the April 11–16, 2013, sampling period (1796 photos), the feeding method changed to an “on-demand” gravity feeder secured on top of a one-meter-high table (Figure S2 in supplemental material). Animals had to climb or jump on the table and push open a flap at the bottom of the feeder’s storage bin to access dry food. Feeders could hold up to 3.2 kg (7 lb) of food; the feeder was never left empty.
Feeding Regime 3, small quantities at ground level: During the July and August 2013 sampling periods – June 27–July 18 (1195 photos) and July 30–Aug 29 (3795 photos) – the feeding system was altered again. At this point, about 0.6–0.7 kg (1.25–1.5 lb) of dry food was placed in one bowl on the ground.
Age Determination and Reproductive Status
PawsWatch volunteers determined age and reproductive status when cats were captured, and we used this information for the identified individuals. Cats were determined to be <1 year old if they had the following characteristics: clean, unbroken teeth; good coat condition; lack of scarring; small genitalia; presence of other juvenile characteristics (proportionally long limbs vs. body mass, proportionally large eyes and ears); and evidence of maturation in date-stamped photos. Cats 1 – <2 years old were identified by lack of tooth wear and breakage with some tartar, and/or a shallow chest and abdomen (stomachs tight), and relatively good coat condition. Cats identified as 2 – <3 years old were known 2-year-olds who had survived to the second study year. The one cat >3 years old at the Johnston colony was identified at capture by tooth breakage and tartar, a deep chest and abdomen, sinewy body, scarring and poor coat condition. In some cases, additional information about the ages of the cats was available from PawsWatch captures or from volunteers who knew the history of the cats. Reproductive status (intact, neutered, or pregnant) was determined via capture and examination, interviewing colony tenders familiar with the cats, camera trap photos revealing ear clips (a TNR symbol for neutering), observed genitalia, and observation of heavily pregnant females.
Population Estimates
We used photographic data from the feeding station to make visual counts of individuals, and performed “mark and recapture” population estimates in order to examine the utility of our camera counts for assessments of feral colony population size. The individually recognizable cats were considered “marked” when they were first seen in a photograph and given an identification number. A “recapture” would be scored on any subsequent day an identified cat was photographed. Only presence on a particular day, not the individual visits, were used for analyzing population size.
For an estimate of the total population, we used loglinear models, implemented in the function closedp in the R package Rcapture (Rivest & Baillargeon, 2014). We followed the approach described by Bengsen et al. (2011). After the best-fitting model was identified based on the Akaike Information Criteria (AIC), 95% confidence intervals were calculated with the function closedpCI of the R package Rcapture (Rivest & Baillargeon, 2014). To estimate how quickly confidence intervals for the modeled population size closed as we lengthened the trapping effort, we ran the calculation for three days, four days, etc., until the total number of trapping days for the respective periods was reached. We then plotted the resulting population estimate and the 95% confidence interval.
Visitation by Cats and Other Mammals
Subsamples of photographic data were used to quantify daily visits by cats, raccoons, skunks, and coyotes to feral cat feeding stations. Unfortunately, due to a hardware failure, the visitation data for April 2012 were not available for coyotes, skunks, or raccoons, but there was one sample period for each of the three feeding regimes.
Results
Camera Trap Population Estimates and Trapping Effort
Daytime photos (in color) were found to be particularly useful for developing cat identification keys. All photos contributed information about visitation frequency and activity patterns. Different age classes and ear-clips were recognizable. Resident cats (observed on multiple days) and non-resident cats (observed for a single day) were easily distinguished. For the first three periods, models without a time effect and a Poisson effect were used because they showed the lowest AIC, while for the last three periods, models without a time effect and the Darroch distribution showed the lowest AIC and were used instead.
In all cases, the modeled population size in our six sample periods converged on an estimate and stabilized within several days (Figure 1A–C). During the first four periods, when abundant food was provided to cats, estimates stabilized within 3–5 days (Figure 1A–C, large amounts of food on the ground and Figure 1D, elevated gravity feeder). In the two periods where food was limited, it took 8–11 days for population estimates to stabilize (Figure 1E–F, small amounts of food on ground). Confidence intervals closed most quickly for the three periods with abundant food provided on the ground (Figure 1A–C). The total population of cats dropped from 17 individuals in early 2012 to 12 in late 2013.
Modeled population size estimates (with 95% confidence interval) versus trapping effort at a Johnston, Rhode Island, TNR Colony Note: The time required for an accurate estimate was dependent on how the cats were fed. The confidence intervals closed on an estimated value after 3–5 days with large amounts of food provided on the ground (A–C) or in an elevated feeder (D). With limited food provided on the ground the estimate stabilized in 8–11 days (E–F). Number of cats declined over the study by about 40%.
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
Modeled population size estimates (with 95% confidence interval) versus trapping effort at a Johnston, Rhode Island, TNR Colony Note: The time required for an accurate estimate was dependent on how the cats were fed. The confidence intervals closed on an estimated value after 3–5 days with large amounts of food provided on the ground (A–C) or in an elevated feeder (D). With limited food provided on the ground the estimate stabilized in 8–11 days (E–F). Number of cats declined over the study by about 40%.
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
Modeled population size estimates (with 95% confidence interval) versus trapping effort at a Johnston, Rhode Island, TNR Colony Note: The time required for an accurate estimate was dependent on how the cats were fed. The confidence intervals closed on an estimated value after 3–5 days with large amounts of food provided on the ground (A–C) or in an elevated feeder (D). With limited food provided on the ground the estimate stabilized in 8–11 days (E–F). Number of cats declined over the study by about 40%.
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
Population Structure
According to colony tenders, the Johnston colony contained 20–30 cats in 2011. Table 1 shows that by 2012, the start of the study, the population consisted of 17 known individuals and a maximum of 20 individuals (95% confidence interval). We recorded that the population continued to decline over the study period even though 50–70% of the cats were not neutered (Table 1, Figure 2).
Age structure, reproductive status, and longevity (horizontal bars) of cats, by sample period and feeding regime, at a Johnston, Rhode Island, TNR Colony Note: Feeding methods: 1) large quantities on ground, 2) large quantities elevated, 3) small quantities on ground. This figure shows that 30% of cats present disappeared during feeding regime 1; 40% had disappeared by the end of feeding regime 2, no cats disappeared during feeding regime 3. Only one cat was more than 3 years old.
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
Age structure, reproductive status, and longevity (horizontal bars) of cats, by sample period and feeding regime, at a Johnston, Rhode Island, TNR Colony Note: Feeding methods: 1) large quantities on ground, 2) large quantities elevated, 3) small quantities on ground. This figure shows that 30% of cats present disappeared during feeding regime 1; 40% had disappeared by the end of feeding regime 2, no cats disappeared during feeding regime 3. Only one cat was more than 3 years old.
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
Age structure, reproductive status, and longevity (horizontal bars) of cats, by sample period and feeding regime, at a Johnston, Rhode Island, TNR Colony Note: Feeding methods: 1) large quantities on ground, 2) large quantities elevated, 3) small quantities on ground. This figure shows that 30% of cats present disappeared during feeding regime 1; 40% had disappeared by the end of feeding regime 2, no cats disappeared during feeding regime 3. Only one cat was more than 3 years old.
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
Cat population changes and demographics at a Johnston, Rhode Island, TNR Colony
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
Cat population changes and demographics at a Johnston, Rhode Island, TNR Colony
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
Cat population changes and demographics at a Johnston, Rhode Island, TNR Colony
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
Cat population changes and demographics at a Johnston, Rhode Island, TNR Colony
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
Cat population changes and demographics at a Johnston, Rhode Island, TNR Colony
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
By the end of the two-year study, 40% of cats (8/20) observed at the Johnston site were no longer present. Three were gone after the first sample period during feeding regime 1, and a total of five were gone by the start of the feeding regime 2 sample period (Table 1, Figure 2). A total of eight were gone before the start of feeding regime 3 sample period, after which no further cats disappeared during the course of the study (Table 1, Figure 2). Throughout the study, young cats (less than three years old) made up at least 70–80% of the population at the Johnston colony (Table 1, Figure 2).
Cat and Wildlife Visitation
Visitation Frequency
We observed many wildlife species at feeding stations over the course of the study, including blue jays (Cyanocitta cristata), turkeys (Meleagris gallopavo), white-footed mice (Peromyscus leucopus), grey squirrels (Sciurus carolinensis), grey fox (Urocyon cinereoargenteus), red fox (Vulpes vulpes), opossums (Didelphis marsupialis), skunks, raccoons, and coyotes. The last three species were most frequently seen; other wildlife visits were occasional and excluded from analyses (Figure 3).
Boxplots of cat, coyote, raccoon, and skunk visits per day during three different feeding regimes at a Johnston, Rhode Island, TNR Colony Note: Feeding methods: 1) large quantities on ground, 2) large quantities elevated, 3) small quantities on ground. All animals decreased visitation as accessibility to food dropped. Coyotes were largely present during feeding regime 1; cats and raccoons were the only animals that used the elevated feeder in regime 2; coyotes and other mesopredators essentially stopped visits during feeding regime 3.
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
Boxplots of cat, coyote, raccoon, and skunk visits per day during three different feeding regimes at a Johnston, Rhode Island, TNR Colony Note: Feeding methods: 1) large quantities on ground, 2) large quantities elevated, 3) small quantities on ground. All animals decreased visitation as accessibility to food dropped. Coyotes were largely present during feeding regime 1; cats and raccoons were the only animals that used the elevated feeder in regime 2; coyotes and other mesopredators essentially stopped visits during feeding regime 3.
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
Boxplots of cat, coyote, raccoon, and skunk visits per day during three different feeding regimes at a Johnston, Rhode Island, TNR Colony Note: Feeding methods: 1) large quantities on ground, 2) large quantities elevated, 3) small quantities on ground. All animals decreased visitation as accessibility to food dropped. Coyotes were largely present during feeding regime 1; cats and raccoons were the only animals that used the elevated feeder in regime 2; coyotes and other mesopredators essentially stopped visits during feeding regime 3.
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
Over the two-year study period in Johnston, cats were responsible for 64% of the total visits by the four focal species to the feeding station, averaging 133 visits per day. Raccoons were the second-most common, accounting for 22% of visits, followed by both coyotes with 9% of visits, and skunks with 5% of visits (Figure 3). In general, the number of cat visits per day was lower in 2012 than in 2013. The number of raccoon visits per day was high in 2012 through April 2013 but declined in the last two sampling periods of 2013.
Coyote, raccoon, and skunk visits were highest when large quantities of food were presented at ground level (feeding regime 1, Figure 3). In contrast, cat visits were highest when large quantities of food were presented in an elevated on-demand feeder; raccoons could also exploit the elevated feeder while coyotes and skunks could not (feeding regime 2, Figure 3). Limited food on the ground reduced coyote and raccoon visitation and skunks did not visit at all (feeding regime 3, Figure 3).
Activity Patterns
In general, cats visited less when coyotes, raccoons, and skunks were visited (Figure 4A–C). During feeding regime 1, cats tended to visit the feeding station during the day (Figure 4A), and coyotes, raccoons, and skunks were more commonly observed at night. During feeding regime 2 (Figure 4B), cats visited the feeding station at night when raccoons were present but coyotes and skunks were absent. During feeding regime 3 (Figure 4C), cats visits were more evenly spread over the 24 hours but peaked in the morning (just before and after food was replenished by colony tenders). During this period coyote and raccoon visits were minimal and tended to be nocturnal; skunks did not visit (Figure 4C).
Visitation times by cats, coyotes, raccoons, and skunks at a Johnston, Rhode Island, TNR Colony during each of three feeding regimes Note: (A) Large quantities on ground (B) Large quantities elevated (C) Small quantities on ground. Cats were largely diurnal when coyotes and other mesopredators were present at night during feeding regime 1 (A). Cats increased nocturnal activity as coyote, and skunk (regime 2) and raccoon visits (regime 3) declined (B–C).
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
Visitation times by cats, coyotes, raccoons, and skunks at a Johnston, Rhode Island, TNR Colony during each of three feeding regimes Note: (A) Large quantities on ground (B) Large quantities elevated (C) Small quantities on ground. Cats were largely diurnal when coyotes and other mesopredators were present at night during feeding regime 1 (A). Cats increased nocturnal activity as coyote, and skunk (regime 2) and raccoon visits (regime 3) declined (B–C).
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
Visitation times by cats, coyotes, raccoons, and skunks at a Johnston, Rhode Island, TNR Colony during each of three feeding regimes Note: (A) Large quantities on ground (B) Large quantities elevated (C) Small quantities on ground. Cats were largely diurnal when coyotes and other mesopredators were present at night during feeding regime 1 (A). Cats increased nocturnal activity as coyote, and skunk (regime 2) and raccoon visits (regime 3) declined (B–C).
Citation: Society & Animals 30, 7 (2022) ; 10.1163/15685306-bja10112
Discussion
Impacts on Cats
Cats, who are primarily nocturnal (Baker et al., 2010), generally avoided feeding stations at night when coyotes were present but made nocturnal visits when they were not. About 30% of cats disappeared in 2012 when the feeding regime attracted coyotes with abundant food on ground. Hence, leaving abundant food in bowls on the ground may set the cats up for coyote predation. While we found no direct evidence in this study, cats are a regular part of coyote diets (Grubbs & Krausman, 2009; Nation & Clair, 2019; Poessel et al., 2017; Quinn, 1997). Additionally, the Narragansett Bay Coyote Study (NBCS) in Rhode Island, USA, initiated in 2005, reports that coyotes rapidly discover reliable food resources like farm carcass piles, garbage or composting facilities, and feral cat feeding stations (Mitchell, 2010; Mitchell et al., 2015).
Impacts to Wildlife and Humans
All feeding methods enable the interspecific transfer of disease from saliva on feeding bowls and feeders, perhaps the most significant being parasites, feline leukemia virus (Jessup, 2004), and the raccoon rabies virus variant (RVV) prevalent in Rhode Island (Krebs et al., 1996; Ma et al., 2020; Wilson et al., 1997). While raccoons and skunks are highly susceptible to the prevalent raccoon RVV (Ma et al., 2020), spillover to coyotes is rare (Wang et al., 2010). Rabies is usually transmitted through saliva that comes into contact with wounds or mucous membranes (Ma et al., 2020), so direct transfer through saliva on food bowls at feral cat colonies could transfer raccoon RVV to coyotes and cats – and from there, to humans. For example, in September 2012, a rabid kitten from a feral cat colony in Jamestown, RI, USA, was adopted by a person who was then exposed to rabies (Riel, 2012). While the risks for humans are very low, humane treatment of wildlife should consider the spread of diseases within wildlife populations facilitated by feeding stations. Our study provides evidence that feeding regime has an influence on frequency and timing of interactions. Coyote, raccoon, and skunk visits were highest when cats were fed abundant food at ground level; only raccoon visits were common when food was just as abundant but elevated to table height (other mesopredators, including coyotes, could not or would not jump or climb up); coyote, raccoon, and skunk visits were low when food was limited on the ground and generally gone before they arrived at night.
A secondary impact of outdoor feeding that attracts wildlife is habituation (loss of fear) and food conditioning. It is a common observation that delivery of food causes wildlife to become conditioned and even bold and aggressive around people (Hadidian et al., 2010; Murray & Clair, 2017; White & Gehrt, 2009). Habituated nocturnal species sometimes forage openly during the day (White & Gehrt, 2009), exhibit aggressive or otherwise objectionable behaviors, and then are subject to lethal control or relocation (if possible) by nuisance wildlife specialists (Curtis & Hadidian, 2010; Hadidian et al., 2010; White & Gehrt, 2009). Habituation of coyotes and other mesopredators and the resulting human-wildlife interactions may also increase the transfer of zoonotic diseases to humans (Wieczorek Hudenko et al., 2010). Finally, feeding in a way that attracts coyotes to residential areas also puts owned companion animals at risk (White & Gehrt, 2009). While we have not studied habituation and food conditioning here, our results show that feeding does attract wildlife to a location close to residential housing. Our results also indicate that wildlife visits can be reduced by changing the feeding regime to elevated or limited resources on the ground.
Implications for TNR Colony Tenders
Our findings support calls for feeding stations that exclude other species (Hernandez et al., 2018). In our study, elevated platforms and well-timed feeding reduced potential cat-wildlife-interactions, including encounters with coyotes. Since TNR cat colonies are usually driven by an appreciation and love for cats (Centonze & Levy, 2002; Peterson et al., 2012), decreasing the risk of predation should be a common-sense measure.
We interpret the age structure of the cat population at the TNR station as a result of coyote predation: Over 90% of the cats at the colony were agile 1–3-year-olds. Only one of the 20 study cats was older than 3 years, a strong contrast to the 10-year average lifespan of owned cats (Baker et al., 2010). Such short lifespans were also reported at other TNR cat colonies (Hostetler et al., 2020). Our camera data demonstrated that in 2012, when abundant food was provided on the ground, coyotes accounted for 14% of all visits, making it probable that approaching cats would be encountered and killed at the feeding station. There could be other reasons for the short lifespan of the cats, as our study was not set up to identify the source of mortality. However, if cat colonies are managed in a way that attracts coyotes, the likelihood of encounters increases. Without changes in feeding protocols, euthanasia might be a humane alternative to releasing cats into, as we perceive it, this fraught environment.
Measuring the Efficacy of TNR Programs
A lack of scientifically applied monitoring (i.e., repeatable, transparent, objective) at TNR cat colonies has been criticized (Hostetler et al., 2020). Our study shows that setting up a motion-activated camera at a feeding location can provide a good estimate of population size in less than two weeks by counting the number of distinct individual cats captured on photographs. While this may still be a lot of work for volunteers, automatic image recognition may be a means forward (Yousif et al., 2019). Alternatively, RFID (Radio-frequency identification) technology has become very affordable and could be used to monitor captured and released animals (Ciaglo, 2015). RFID systems consist of a tiny radio transponder that operates without a battery and can be activated by a reader that sends out an electromagnetic pulse. RFID systems are already used for companion animal microchips, which have inspired technology such as microchip-activated cat doors (Mohr, 2015). Setting up such a reader at a feeding site would help keep track of released animals. Ideally, it should be set up in a way that allows for the identification at more than one colony, so possible movement could be tracked. RFID technology could also be used for designing cat-specific feeding stations, although this would exclude unmarked cats. Compared to RFID, camera trapping is a much easier and less invasive way of keeping track of populations at TNR colonies. Quarterly, bi-annual, or at least annual surveys at each colony would allow colony tenders to document population size and changes that have occurred. Documenting the feeding regime should be part of such efforts. In our opinion, valid population estimates should be a requirement to justify any funding for ongoing TNR.
Conclusion
Our work indicates that feeding outdoor cats ultimately may have negative impacts on wildlife due to cat-wildlife-encounters, wildlife confluence, and habituation of wildlife to anthropogenic food. Our work also suggests that feeding outdoor cats can have negative impacts on the cats themselves. Placing food for cats outdoors where coyotes can access it greatly increases the presence of coyotes, both during the day and at night. In turn, this increases the chance of cats encountering coyotes, who are well-known to prey on cats (Grubbs & Krausman, 2009; Nation & Clair, 2019; Poessel et al., 2017; Quinn, 1997). Others have reported high predation pressure put on small wildlife in residential areas with many outdoor cats (Herrera et al., 2022; Kays et al., 2020) and around TNR feeding stations (Herrera et al., 2022; Kays et al., 2020). Ignoring these impacts ignores our responsibility for both cats and wildlife. Based on our observations, feeding stations that limit access for wildlife, by elevating or reducing food, are strongly encouraged. This may better protect cats and reduce wildlife interactions (and the deleterious effects that follow), which is generally aligned with animal protection, animal welfare, and wildlife conservation goals. We therefore encourage state and local governments to adopt laws such as the “No Feeding” ordinances implemented in Rhode Island (Jamestown, Newport, Middletown, and Portsmouth), which prohibit placing food attractants where wildlife can access the food (City of Newport, 2013; Town of Jamestown, 2016; Town of Middletown, 2011; Town of Portsmouth, 2013). As a synthesis of animal welfare and wildlife conservation perspectives, the best feral cat management includes no outdoor feeding accessible to wildlife.
Acknowledgements
Thanks are due to the members of the Rhode Island Feral Cat Working Group, and the late Dr. Ernest Finocchio, RI Society for the Prevention of Cruelty to Animals, for advice and support. Thanks also to others who assisted in this study: Peter Auger and students from Wheaton and Leslie Colleges, Anne Fleming and Eliza Gentilli Lloyd (Tufts Veterinary School), Sarah Appleton and Samantha Bucklin (University of Rhode Island, Preveterinary Program), Wendy Finn Peter Paton (Natural Resources Science, University of Rhode Island), and Charles Brown, former Furbearer Biologist, RI Department of Environmental Management. We are grateful for the partnership of PawsWatch personnel, led by Gil Fletcher, who provided access to feral cat colonies and shared life history information about colony cats. Finally yet importantly, we would like to thank our colleague Kyle Hess, two anonymous reviewers and the associate editor Kristin Stewart for their much appreciated comments and recommendations.
References
Baker, P. J., Bentley, A. J., Ansell, R. J., & Harris, S. (2005). Impact of predation by domestic cats Felis catus in an urban area. Mammal Review, 35(3–4), 302–312. https://doi.org/10.1111/j.1365-2907.2005.00071.x.
Baker, P. J., Soulsbury, C. D., Iossa, G., & Harris, S. (2010). Domestic Cat (Felis catus) and Domestic Dog (Canis familiaris). In S. D. Gehrt, S. P. D. Riley, & B. L. Cypher (Eds.), Urban carnivores: Ecology, conflict, and conservation (pp. 157–171). Johns Hopkins University Press.
Barrows, P. L. (2004). Professional, ethical, and legal dilemmas of trap-neuter-release. Journal of the American Veterinary Medical Association, 225(9), 1365–1369. https://doi.org/10.2460/javma.2004.225.1365.
Bengsen, A., Butler, J., & Masters, P. (2011). Estimating and indexing feral cat population abundances using camera traps. Wildlife Research, 38(8), 732–739. https://doi.org/10.1071/wr11134.
Blancher, P. (2013). Estimated number of birds killed by house cats (Felis catus) in Canada. Avian Conservation and Ecology, 8(2), Article 3. https://doi.org/10.5751/ACE-00557-080203.
Bonnington, C., Gaston, K. J., & Evans, K. L. (2013). Fearing the feline: Domestic cats reduce avian fecundity through trait‐mediated indirect effects that increase nest predation by other species. Journal of Applied Ecology, 50(1), 15–24. https://doi.org/10.1111/1365-2664.12025.
Boone, J. D., Miller, P. S., Briggs, J. R., Benka, V. A. W., Lawler, D. F., Slater, M., Levy, J., & Zawistowski, S. (2019). A long-term lens: Cumulative impacts of free-roaming cat management strategy and intensity on preventable cat mortalities. Frontiers in Veterinary Science, 6, 238. https://doi.org/10.3389/fvets.2019.00238.
Centonze, L. A., & Levy, J. K. (2002). Characteristics of free-roaming cats and their caretakers. Journal of the American Veterinary Medical Association, 220(11), 1627–1633. https://doi.org/10.2460/javma.2002.220.1627.
Ciaglo, M. (2015). Build your own RFID. Mongabay. Retrieved August 8, 2022, from https://wildtech.mongabay.com/2015/07/build-your-own-rfid/.
City of Newport. (2013). Code of ordinances, title 6, animals, chapter 6.2, feeding of non-domesticated animals. Municode. Retrieved August 8, 2022, from https://library.municode.com/ri/newport/codes/code_of_ordinances?nodeId=COOR_TIT6AN_CH6.12FENOMAN.
Coe, S. T., Elmore, J. A., Elizondo, E. C., & Loss, S. R. (2021). Free-ranging domestic cat abundance and sterilization percentage following five years of a trap-neuter-return program. Wildlife Biology, 2021(1), wlb.00799. https://doi.org/10.2981/wlb.00799.
Curtis, P. D., & Hadidian, J. (2010). Responding to human-carnivore conflicts in urban areas. In S. D. Gehrt, S. P. D. Riley, & B. L. Cypher (Eds.), Urban carnivores: Ecology, conflict, and conservation (pp. 201–211). Johns Hopkins University Press.
Finkler, H., Hatna, E., & Terkel, J. (2011). The impact of anthropogenic factors on the behavior, reproduction, management and welfare of urban, free-roaming cat populations. Anthrozoös, 24(1), 31–49. https://doi.org/10.2752/175303711x12923300467320.
Fletcher, G. (2013). Best practices for feeding feral cats. Pawswatch Newsletter, Spring-Summer 2013, 1.
Foley, P., Foley, J. E., Levy, J. K., & Paik, T. (2005). Analysis of the impact of trap-neuter-return programs on populations of feral cats. Journal of the American Veterinary Medical Association, 227(11), 1775–1781. https://doi.org/10.2460/javma.2005.227.1775.
Grubbs, S. E., & Krausman, P. R. (2009). Observations of coyote–cat interactions. Journal of Wildlife Management, 73(5), 683–685. https://doi.org/10.2193/2008-033.
Hadidian, J., Prange, S., Rosatte, R., Riley, S. P. D., & Gehrt, S. D. (2010). Raccoons (Procyon lotor). In S. D. Gehrt, S. P. D. Riley, & B. L. Cypher (Eds.), Urban carnivores: ecology, conflict, and conservation (pp. 35–47). Johns Hopkins University Press.
Hall, C. M., Bryant, K. A., Haskard, K., Major, T., Bruce, S., & Calver, M. C. (2016). Factors determining the home ranges of pet cats: A meta-analysis. Biological Conservation, 203, 313–320. https://doi.org/10.1016/j.biocon.2016.09.029.
Hernandez, S. M., Loyd, K. A. T., Newton, A. N., Gallagher, M. C., Carswell, B. L., & Abernathy, K. J. (2018). Activity patterns and interspecific interactions of free-roaming, domestic cats in managed trap-neuter-return colonies. Applied Animal Behaviour Science, 202, 63–68. https://doi.org/10.1016/j.applanim.2018.01.014.
Herrera, D. J., Cove, M. V., McShea, W. J., Flockhart, D. T., Decker, S., Moore, S. M., & Gallo, T. (2022). Prey selection and predation behavior of free-roaming domestic cats (Felis catus) in an urban ecosystem: Implications for urban cat management. Biological Conservation, 268, 109503. https://doi.org/10.1016/j.biocon.2022.109503.
Hostetler, M., Wisely, S. M., Johnson, S., Pienaar, E., & Main, M. (2020). How effective and humane is trap-neuter-release (TNR) for feral cats? EDIS, 2020(2), 8. https://doi.org/10.32473/edis-uw468-2020.
Jessup, D. A. (2004). The welfare of feral cats and wildlife. Journal of the American Veterinary Medical Association, 225(9), 1377–1383.
Kays, R., Dunn, R. R., Parsons, A. W., McDonald, B., Perkins, T., Powers, S. A., Shell, L., Mcdonald, J. L., Cole, H., Kikillus, K. H., Woods, L., Tindle, H., & Roetman, P. (2020). The small home ranges and large local ecological impacts of pet cats. Animal Conservation. https://doi.org/10.1111/acv.12563.
Krebs, J., Strine, T., Smith, J., Noah, D., Rupprecht, C., & Childs, J. (1996). Rabies surveillance in the United States during 1995. Journal of the American Veterinary Medical Association, 209(12), 2031–2044.
Kreisler, R. E., Cornell, H. N., & Levy, J. K. (2019). Decrease in population and increase in welfare of community cats in a twenty-three year trap-neuter-return program in Key Largo, FL: The ORCAT Program. Frontiers in Veterinary Science, 6, 7. https://doi.org/10.3389/fvets.2019.00007.
Loss, S. R., Boughton, B., Cady, S. M., Londe, D. W., McKinney, C., O’Connell, T. J., Riggs, G. J., & Robertson, E. P. (2022). Review and synthesis of the global literature on domestic cat impacts on wildlife. Journal of Animal Ecology, 91(7), 1361–1372. https://doi.org/10.1111/1365-2656.13745.
Lynn, W. S., Santiago‐Ávila, F., Lindenmayer, J., Hadidian, J., Wallach, A., & King, B. J. (2019). A moral panic over cats. Conservation Biology, 33(4), 769–776. https://doi.org/10.1111/cobi.13346.
Ma, X., Monroe, B. P., Cleaton, J. M., Orciari, L. A., Gigante, C. M., Kirby, J. D., Chipman, R. B., Fehlner-Gardiner, C., Gutiérrez Cedillo, V., Petersen, B. W., Olson, V., & Wallace, R. M. (2020). Public veterinary medicine: Public health: Rabies surveillance in the United States during 2018. Journal of the American Veterinary Medical Association, 256(2), 195–208. https://doi.org/10.2460/javma.256.2.195.
McDonald, J. L., Maclean, M., Evans, M. R., & Hodgson, D. J. (2015). Reconciling actual and perceived rates of predation by domestic cats. Ecology and Evolution, 5(14), 2745–2753. https://doi.org/10.1002/ece3.1553.
Mitchell, N. (2010). Best management practices for coexistence with coyotes on Aquidneck and Conanicut Islands. The Conservation Agency. Retrieved August 8, 2022, from http://theconservationagency.org/wp-content/uploads/Best-Management-Practices-for-Coexistence-with-and-Management-of-Coyotes-current-vers.pdf.
Mitchell, N., Strohbach, M. W., Pratt, R., Finn, W. C., & Strauss, E. G. (2015). Space use by resident and transient coyotes in an urban–rural landscape mosaic. Wildlife Research, 42(6), 461–469. https://doi.org/10.1071/WR15020.
Mohr, M. (2015). Open doors with a Raspberry Pi and RFID module. Raspberry Pi Geek. Retrieved August 8, 2022, from https://www.raspberry-pi-geek.com/Archive/2015/14/Open-doors-with-a-Raspberry-Pi-and-RFID-module.
Murray, M. H., & Clair, C. C. S. (2017). Predictable features attract urban coyotes to residential yards. The Journal of Wildlife Management, 81(4), 593–600. https://doi.org/10.1002/jwmg.21223.
Nation, P. N., & Clair, C. C. S. (2019). A forensic pathology investigation of dismembered domestic cats: coyotes or cults? Veterinary Pathology, 56(3), 444–451. https://doi.org/10.1177/0300985819827968.
Ottoni, C., Neer, W. V., Cupere, B. D., Daligault, J., Guimaraes, S., Peters, J., Nikolay, S., Prendergast, M. E., Boivin, N., Morales-Nuñéz, A., Balasescu, A., Becker, C., Benecke, N., Boroneant, A., Buitenhuis, H., Chahoud, J., Crowther, A., Llorente-Rodriguez, Manaseryan, N., … Geigl, E.-M. (2017). The palaeogenetics of cat dispersal in the ancient world. Nature Ecology & Evolution, 1(7), 0139. https://doi.org/10.1038/s41559-017-0139.
Peterson, M. N., Hartis, B., Rodriguez, S., Green, M., & Lepczyk, C. A. (2012). Opinions from the front lines of cat colony management conflict. PLoS ONE, 7(9), e44616. https://doi.org/10.1371/journal.pone.0044616.
Piontek, A. M., Wojtylak-Jurkiewicz, E., Schmidt, K., Gajda, A., Lesiak, M., & Wierzbowska, I. A. (2021). Analysis of cat diet across an urbanisation gradient. Urban Ecosystems, 24(1), 59–69. https://doi.org/10.1007/s11252-020-01017-y.
Pirie, T. J., Thomas, R. L., & Fellowes, M. D. E. (2022). Pet cats (Felis catus) from urban boundaries use different habitats, have larger home ranges and kill more prey than cats from the suburbs. Landscape and Urban Planning, 220, 104338. https://doi.org/10.1016/j.landurbplan.2021.104338.
Poessel, S. A., Mock, E. C., & Breck, S. W. (2017). Coyote (Canis latrans) diet in an urban environment: variation relative to pet conflicts, housing density, and season. Canadian Journal of Zoology, 95(4), 287–297. https://doi.org/10.1139/cjz-2016-0029.
Quinn, T. (1997). Coyote (Canis latrans) food habits in three urban habitat types of western Washington. Northwest Science, 71(1), 1.
Riel, T. (2012). Rabies threat prompts town to trap feral cats. The Jamestown Press. Retrieved August 8, 2022, from https://www.jamestownpress.com/articles/rabies-threat-prompts-town-to-trap-feral-cats/.
Rivest, L. P., & Baillargeon, S. (2014). Rcapture: Loglinear models for capture-recapture experiments. In R package version 1.4–2. https://CRAN.R-project.org/package=Rcapture.
Town of Jamestown. (2016). The Jamestown code of ordinances, chapter 10, animals, article VII: Non-domestic animals, (2016). Municode. Retrieved August 8, 2022, from https://library.municode.com/ri/jamestown/ordinances/code_of_ordinances?nodeId=794840.
Town of Middletown. (2011). Code of ordinances, Title IX, general regulations, chapter 90A: Feeding non-domesticated animals. American Legal Publishing. Retrieved August 8, 2022, from https://codelibrary.amlegal.com/codes/middletown/latest/middletown_ri/0-0-0-10413.
Town of Portsmouth. (2013). Article IV, feeding of non-domesticated animals. Retrieved August 8, 2022, from https://www.ecode360.com/28381878.
van Heezik, Y., Smyth, A., Adams, A., & Gordon, J. (2010). Do domestic cats impose an unsustainable harvest on urban bird populations? Biological Conservation, 143(1), 121–130. https://doi.org/https://doi.org/10.1016/j.biocon.2009.09.013.
Wandesforde-Smith, G., Levy, J. K., Lynn, W., Rand, J., Riley, S., Schaffner, J. E., & Wolf, P. J. (2021). Coping with human-cat interactions beyond the limits of domesticity: Moral pluralism in the management of cats and wildlife. Frontiers in Veterinary Science, 8, 682582. https://doi.org/10.3389/fvets.2021.682582.
Wang, X., Brown, C. M., Smole, S., Werner, B. G., Han, L., Farris, M., & DeMaria, A. (2010). Aggression and rabid coyotes, Massachusetts, USA. Emerging Infectious Diseases, 16(2), 357–359. https://doi.org/10.3201/eid1602.090731.
White, L. A., & Gehrt, S. D. (2009). Coyote attacks on humans in the United States and Canada. Human Dimensions of Wildlife, 14(6), 419–432. https://doi.org/10.1080/10871200903055326.
Wieczorek Hudenko, H., Siemer, W. F., & Decker, D. S. (2010). Urban carnivore conservation and management: the human dimension. In S. D. Gehrt, S. P. D. Riley, & B. L. Cypher (Eds.), Urban carnivores: Ecology, conflict, and conservation (pp. 35–47). Johns Hopkins University Press.
Wilson, M. L., Bretsky, P. M., Cooper Jr, G. H., Egbertson, S. H., Van Kruiningen, H. J., & Cartter, M. L. (1997). Emergence of raccoon rabies in Connecticut, 1991–1994: Spatial and temporal characteristics of animal infection and human contact. The American Journal of Tropical Medicine and Hygiene, 57(4), 457–463.
Wolf, P. J., & Hamilton, F. (2022). Managing free-roaming cats in U.S. cities: An object lesson in public policy and citizen action. Journal of Urban Affairs, 44(2), 1–22. https://doi.org/10.1080/07352166.2020.1742577.
Woolley, L.-A., Murphy, B. P., Geyle, H. M., Legge, S. M., Palmer, R. A., Dickman, C. R., Doherty, T. S., Edwards, G. P., Riley, J., Turpin, J. M., & Woinarski, J. C. Z. (2020). Introduced cats eating a continental fauna: invertebrate consumption by feral cats (Felis catus) in Australia. Wildlife Research, 47(7–8), 610–623. https://doi.org/10.1071/wr19197.
Yousif, H., Yuan, J., Kays, R., & He, Z. (2019). Animal scanner: Software for classifying humans, animals, and empty frames in camera trap images. Ecology and Evolution, 9(4), 1578–1589. https://doi.org/10.1002/ece3.4747.