| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 United States Department of Agriculture, Animal and Plant Health Inspection Service, Wildlife Services, National Wildlife Research Center, Ohio Field Station, 6100 Columbus Avenue, Sandusky, Ohio 44870, USA
2 Corresponding author (email: bradley.f.blackwell{at}aphis.usda.gov)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key words: Bait density, forward-looking infrared camera, oral vaccination, population density, rabies, raccoon.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Using the US National Aeronautics and Space Administration (NASA) Plum Brook Station (PBS) in Erie County, Ohio, USA (41°27'N, 82°42'W), we evaluated the period of exposure of placebo vaccine baits placed at a density of 75 baits/km2 (a target ORV bait density currently used in Ohio; Ohio Department of Health, 2002,2003). There are no data relative to seasonal raccoon population density on the PBS. Our objectives were to 1) estimate raccoon population density within the fragmented forest, old-field, and industrial landscape at PBS; and 2) quantify the time that placebo Merial (Athens, Georgia, USA) RA-BORAL V-RG® vaccine baits were available to raccoons.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The 2,200-ha PBS is enclosed by a 2.4-m high chain-link fence with barbed-wire outriggers. Habitat within PBS differs from the surrounding mix of agricultural and suburban area, comprising canopy-dogwood (Cornus spp., 39%); old field and grasslands (31%); open woodlands (15%); and mixed hardwood forests (11%) interspersed with abandoned and actively-used structures relating to NASA and prior operations (Rose and Harder, 1985). Also, PBS comprises a network of paved roads, and raccoons on the facility are exposed to vehicle traffic at all hours. There are no consistent human-related food resources on PBS that would concentrate raccoons in a particular area.
Transect surveys
We established a 19.3-km survey route along five east-west paved-road transects across PBS, covering all habitat types. Traditional off-road transect surveys for moderate-size mammals like raccoons likely suffer from observer disturbance of target species causing flight or even attraction to the observer; either effect will potentially bias accurate measurement from the transect to the point of initial observation (Buckland et al., 1993). We conducted our surveys from a passenger van via infrared camera.
Infrared technology allows target animals to be discerned against background vegetation, an improvement over traditional sighting methods (e.g., spotlights and night-vision equipment; Belant and Seamans, 2000). We used a Raytheon (Dallas, Texas, USA) forward-looking infrared Nightsight Palm IR 250 Digital Camera (FLIR) mounted on the passenger side window to scan habitats on the right side of the vehicle. The camera was connected to a Sony (Park Ridge, New Jersey, USA) Video Walkman Digital-8 monitor.
When a raccoon or raccoon group was detected, we scanned to ensure a total count, and then recorded the number of individuals and perpendicular distance (meters) from the road to the point of the initial observation. For distances 
18 m, we obtained the perpendicular distance to the point of initial detection for a single animal (or initial center of a group of raccoons; Buckland et al., 1993) by illuminating the point with a spotlight and then quantifying the distance using a Bushnell (Overland Park, Kansas, USA) Yardage Pro 1000 Laser Ranging System. For animals observed within 18 m, the observer paced the distance to the initial observation point. We did not count animals moving from the driver side of the vehicle or those that responded to the vehicle approach (e.g., escape, avoidance, or attraction behavior; Buckland et al., 1993). We also recorded the habitat category at the point of initial detection (i.e., road, grassland, shrub, or wooded; Belant and Seamans, 2000).
We began our FLIR surveys on 5 August 2002 and conducted two per week (n=8) through 31 August. Raccoons generally mate from February through June (most frequent in March) and have a gestation period of 63–65 days (Wilson and Ruff, 1999); thus, the raccoon population on PBS was likely near its peak in August. We conducted one survey per week from 1 September through 6 November. Our protocol for each survey night comprised a random selection of the starting transect (i.e., either the extreme north or south transect) and the direction of travel; observations were made from one side of the vehicle and the same side of each transect. We began each survey between 1 and 2 hr after sunset and maintained a speed between 8 and 16 km/hr. A driver and at least one observer were present for each survey. Depending upon raccoon activity, each survey required from 2 to 4 hr to complete.
Data analysis
We quantified raccoon population density from the transect data by month (combining data for 1 October through 6 November) using Distance 3.5 software (Buckland et al., 1993). Distance software calculates a detection function, g(y)=probability (detection | distance y), which is the probability of detecting an object given that it is at distance y from a random line. The detection function is based on the fitting of a series of a priori models to the observed data (Buckland et al., 1993; Burnham and Anderson, 1998). Final model selection is based on best fit to the observed distributions (as per the Akaike Information Criterion; Buckland et al., 1993; Burnham and Anderson, 1998, 2002).
The underlying statistical theory of distance sampling (reviewed by Buckland et al., 1993) relies on three basic assumptions: 1) g(0)=1, or all animals on the transect were detected with certainty; 2) animals were detected at their initial location, prior to any movement in response to the vehicle; and 3) distances were measured accurately. Given these basic assumptions, the method theoretically provides accurate estimates of density, despite the fact that animals go undetected during transect surveys (Buckland et al., 1993).
Simulated ORV baiting
We delineated five 1-km2 bait grids on PBS using a Garmin (Olathe, Kansas, USA) GPSMap 76S with Garmin MapSource software and located the grids to maximize coverage of all habitat types, including or in proximity to bodies of water (Fig. 1
). Random location of grid corners was, however, not possible because of restrictions around certain NASA operations. The mean (SE) distance between centers of adjacent grids was 2.1 km (0.5 km). Each grid comprised nine 1-km transects with nine bait points at 125-m intervals (i.e., 81 points). Prior to distributing baits and because of inherent variability in repeatedly locating GPS waypoints, we marked each bait point with plastic flagging. We randomly selected six points per grid for omission of baits to yield 75 baits/km2. Also, where grid points fell on roads or in bodies of water, we placed baits on the periphery of the road or water at intervals less than 125 m (fewer than five of the 375 points).
|
) on north and south sides of PBS (2.8 km between grid centers) and, beginning at 8:00 AM on 30 September and 1 October 2002, hand-placed 75 baits on each. The Ohio Department of Health’s ORV program has conducted baiting during October (Ohio Department of Health, 2002, 2003). We placed wire flags approximately 1 m from the bait point to aid in locating baits on subsequent checks.
On 10 and 11 October, we baited KSITE grid on the north side of PBS and SPF grid on the south (4.9 km between grid centers; Fig. 1) at 75 baits/km2. We baited our last grid, MAG, on 24 October. We returned to MAG and RANGE after 3 days of bait exposure, then again after 1 wk of exposure. We checked all other grids after 5 and 7 days of exposure. Each grid check began by 9:00 AM. Upon returning to each grid point, we noted presence or absence of the bait; condition of the bait (if present); presence or absence of the sachet; condition of the sachet (if present); and any animal sign (e.g., scat or tracks) near the site. When a bait was missing and a sachet was not immediately noticed, each observer searched the immediate area around the point of bait placement (i.e., within a minimum 3-m radius) for evidence of the bait and sachet, as well as examining obvious trails to the bait point.
During field trials on PBS to evaluate bait flavor preferences for a new coated-vaccine sachet, Linhart et al. (2002) estimated that 22.9% of all baits exposed were taken by nontarget species, with opossums (Didelphis virginiana) the most common nontarget species. To provide an index of potential nontarget take in our ORV baiting simulation, we placed TrailMaster (Lenexa, Kansas, USA) TM35-1 camera kits with TM1550 and TM700v active infrared trail monitors at randomly selected bait points each week. During our first week of baiting, only six camera kits were available; we used 11 camera kits in subsequent weeks.
We considered each grid as an experimental unit (i.e., a sampling unit from PBS as a whole) and evaluated the period of bait exposure via descriptive statistics only. A comparison of bait exposure periods among grids would, by necessity, require replicates of habitat types (e.g., six grids per type) to gain insight into specific habitat effects. However, the addition of grids to meet sample size requirements would have placed our study area within NASA operational areas.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
One survey (22 August) was stopped after approximately 7.2 km due to heavy rain; however, we considered the habitat covered as representative of PBS and included these data in our analyses. The majority (68.8%) of raccoon sightings (n=269) were in grass or on paved road (transects and others), with 16.0% in wooded areas and 2.2% in shrubs. The majority of observations were of individuals, although group size ranged from one to five animals (mean=1.4 animals [SD=0.8]). For our analysis, we assumed each raccoon represented an individual data point. Also, based on the frequency distributions of the monthly transect data (i.e., number of observations versus perpendicular distance), we truncated our data at 110 m (removing a mean of 7.3% of observations per period [SD=0.06]). Generally, at least 5% of data are truncated to remove outliers (Buckland et al., 1993). To adjust for possible bias due to movement of raccoons ahead of the vehicle, which can lead to ‘‘heaping’’ of observations on or near the transect, we grouped the data within 10-m intervals. This type of grouping has little effect on efficiency as indicated by the sampling and can improve the robustness of the estimator (Buckland et al., 1993).
Our detection function, g(y)=key function (y)[1+series expansion (y)], for each survey period comprised a half-normal as the key function and a cosine series expansion as:
 | (1) |
cosine adjustments were of orders of two to five (i.e., the order of polynomial necessary for model fit), where y is the detection distance, w is the truncation point, and a is the area of interest (Fig. 2A,B,C). Estimated population density (raccoons per square kilometer) on PBS (Table 1) decreased from 1 September through 6 November, yielding a monthly mean (SD) of 24.5 (10.1) raccoons/km2.
|
|
After 3 days of exposure, 67% (n=50) and 73% (n=55) of baits had been removed from MAG and RANGE, respectively. Bait removals ranged from 48% to 81% per grid (EB, KSITE, and SPF) by day 5. A minimum of 68% of all baits (i.e., across five grids) had been removed after 5 days of exposure. By day 7, 64–91% of baits per grid had been removed (mean=83% [SD=11%]; Table 2). Further, of the baits removed within 7 days, on average 42% (SD=12.7%) were taken without leaving a discarded sachet; 52% (11.8%) of removals yielded a perforated sachet (Table 2). Evidence from sign at bait sites and our photographs (from 11 of 28 camera placements) indicated that, in addition to raccoons (16 individuals photographed, one to two individuals per camera placement, sign at one bait point); opossum (five individuals photographed, one to two individuals per camera placement); groundhog (Marmota monax; one individual photographed); fox squirrel (Sciurus niger; one individual photographed); and white-tailed deer (Odocoileus virginianus; sign at bait point) investigated baits. Equipment failure and varying TrailMaster sensitivity (e.g., adjusting for leaf litter vs. animal movement) contributed to fewer camera placements recording animal visits. In addition, during FLIR surveys, we commonly observed white-tailed deer and occasionally opossums, striped skunks (Mephitis mephitis), and coyotes (Canis latrans). Red (Vulpes vulpes) and gray (Urocyon cinereoargenteus) fox have also been observed on PBS (T.W.S. and J.D.C., unpubl. data).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Raccoon population densities vary widely (e.g., 1–100 animals/km2; Rosatte, 2000) and are primarily a function of available resources (e.g., increased food and shelter opportunities in suburban/urban areas; Gehrt and Fritzell, 1998; Rosatte, 2000). Urban (1970) estimated an average of 17.5 raccoons/km2 on a managed waterfowl marsh in Ohio on western Lake Erie. Results from spring live trapping in a semiurban area of northeastern Illinois indicated mean annual raccoon densities ranging between 13 and 49 animals/km2 (Gehrt et al., 2002). We estimated that the raccoon population on the PBS exceeded 700 animals (>33 raccoons/km2) during August 2002 and decreased to just under 300 animals (>13 raccoons/km2) through the first week of November. This pattern of decline in population density is consistent with a peak in the population annual cycle, reflecting adults and young-of-the-year present during late summer (Whitaker and Hamilton, 1998) followed by natal dispersal (Stuewer, 1943; Gehrt and Fritzell, 1998); emigration; and mortality (e.g., Gehrt and Fritzell, 1999).
As potential food resources changed with season (e.g., ripening mast [Quercus spp.], corn [Zea mays]), the frequency in which we observed animals on or close to the survey route decreased (Fig. 2A,B,C). However, our ability to detect animals with the FLIR increased as cover thinned with leaf drop (October to 6 November). Further, our data reflected consistent variation in density between months (mean percentage coefficient of variation=22.9% [SD=4.2%]). We contend, therefore, that our data are representative of the late summer and fall components of the annual population cycle for raccoons on PBS. Although a separate and independent density estimate might have provided additional information, mark-recapture methods would have introduced the confounding influence of baited traps and temporary removal of animals during the placebo-bait phase of the study.
Relative to potential bias due to visual attractants, we considered that scent cues from the bait and human presence within the grids during delineation and when checking bait points would serve as the predominant attractants, not flagging. The potential bias of attractants (i.e., decreased exposure time per bait) was uniform across grids. However, scent cues along baited operational flight lines are likely not uniform, and the discovery of ORV baits by raccoons and nontarget species are far from random. Specifically, scent, bait appearance, and bait density bias any assumption that individual baits are statistically independent units. We consider potential scent and visual biases due to revisiting bait points as unavoidable in the effort for a timely assessment of the period of bait exposure. For example, trapping within grids to obtain teeth for biomarker analysis introduces the potential confounding influence of baited traps and temporary removal of animals, as well potential age-related effects in the timing of marker binding (Linhart and Kennelly, 1967).
Thus, given an estimate of raccoon population density on PBS, an evaluation of baiting density relative to baits available per animal is possible. Making the unrealistic assumption that no nontarget species consumed baits, we found that on average 62 (SD=8.1) baits per grid were consumed by raccoons within 7 days. However, if we adjust for the percentage of intact sachets (i.e., across the total number of removed baits; 6.0% [SD=2.2%]) and potential nontarget take (i.e., assume 22.9% nontarget take as per Linhart et al., 2002), on average 45 baits/grid were consumed by raccoons and the sachet perforated. Therefore, given our estimate for raccoon population density on PBS during October 2002 and assuming 22.9% non-target take, 75 baits/km2 yielded 3.3 placebo vaccines consumed per raccoon and the sachet perforated.
We recognize, however, that in ORV programs bait consumption by nontarget species, effective delivery of the bait (Johnston and Tinline, 2002) and vaccine, and effective serologic conversion are all uncertainties. Still, an estimate of seasonal raccoon population density within the program area will allow for adjustments to bait distribution plans (e.g., bait density, frequency of baiting, and program duration) to ensure a reduction in the density of rabid raccoons. A cost-efficient ORV program is one that achieves a 70% population bait consumption level and with, on average, one bait per individual (Johnston and Tinline, 2002). To further evaluate the effectiveness of ORV bait density on PBS, sero-conversion rate in the raccoon population should be quantified relative to ORV bait density estimated from aerial distribution and relative to raccoon population density; this work began in spring of 2003.
|
ACKNOWLEDGMENTS |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
BELANT, J. L., AND T. W. SEAMANS. 2000. Comparison of 3 devices to observe white-tailed deer at night. Wildlife Society Bulletin 28: 154–158.
BUCKLAND, S. T., D. R. ANDERSON, K. P. BURNHAM, AND J. L. LAAKE. 1993. Distance sampling: Estimating abundance of biological populations. Chapman and Hall, London, UK.
BURNHAM, K. P., AND D. R. ANDERSON. 1998. Model selection and inference—A practical information-theoretic approach. Springer, New York, New York, 446 pp.
BURNHAM, K. P., AND D. R. ANDERSON. 2002. Model and multimodel inference: A practical information-theoretic approach, 2nd Edition. Springer, New York, New York, 488 pp.
CAREY, A. B., AND R. G. MCLEAN. 1983. The ecology of rabies: Evidence of co-adaptation. Journal of Applied Ecology 20: 777–800.
GEHRT, S. D., AND E. K. FRITZELL. 1998. Resource distribution, female home range dispersion and male spatial interactions: Group structure in a solitary carnivore. Animal Behaviour 55: 1211–1227.[Medline]
———, ———. 1999. Survivorship of a nonhar-vested raccoon population in south Texas. Journal of Wildlife Management 63: 889–894.
———, G. F. HUBERT JR., AND J. A. ELLIS. 2002. Long-term population trends of raccoons in Illinois. Wildlife Society Bulletin 30: 457–463.
HADIDIAN, J. S., R. JENKINS, D. H. JOHNSTON, P. J. SAVARIE, V. F. NETTLES, D. MANSKI, AND G. M. BAER. 1989. Acceptance of simulated oral rabies vaccine baits by urban raccoons. Journal of Wildlife Diseases 25: 1–9.[Abstract]
HANLON, C. A., J. E. CHILDS, AND V. F. NETTLES. 1999. Recommendations of a national working group on prevention and control of rabies in the United States. Journal of the American Veterinary Medical Association 215: 1612–1619.[Medline]
———, M. NIEZGODA, A. N. HAMIR, C. SCHUMACHER, H. KOPROWSKI, AND C. RUPPRECHT. 1998. First North American field release of a vaccinia-rabies glycoprotein recombinant virus. Journal of Wildlife Diseases 34: 228–239.[Abstract]
JOHNSTON, D. H., AND R. R. TINLINE. 2002. Rabies control in wildlife. In Rabies, A. C. Jackson and W. H. Wunner (eds.). Academic Press, New York, New York, pp. 445–471.
LINHART, S. B., AND J. J. KENNELLY. 1967. Fluorescent bone labeling of coyotes with demethylch-lortetracycline. Journal of Wildlife Management 31: 317–321.
———, J. C. WLODKOWSKI, D. M. KAVAHAUGH, L. MOTES-KREIMEYER, A. J. MONTONEY, R. B. CHIPMAN, D. SLATE, L. L. BIGLER, AND M. G. FEARNEYHOUGH. 2002. A new flavor-coated sachet bait for delivering oral rabies vaccine to raccoons and coyotes. Journal of Wildlife Diseases 38: 363–377.[Abstract]
MELTZER, M. I. 1996. Assessing the costs and benefits of an oral vaccine for raccoon rabies: A possible model. Emerging Infectious Diseases 2: 343–349.[Medline]
OHIO DEPARTMENT OF HEALTH. 2002, 2003, http//www.ody.state.oh.us/odhprograms/zoodis/orv/orv1.htm. Accessed April 2004.
OTIS, D. L., K. P. BURNHAM, G. C. WHITE, AND D. R. ANDERSON. 1978. Statistical inference from capture data on closed animal populations. Wildlife Monographs 62: 1–135.
PERRY, B. D., N. GARNER, S. R. JENKINS, K. MC-CLOSKEY, AND D. H. JOHNSTON. 1989. A study of techniques for the distribution of oral rabies vaccine to wild raccoon populations. Journal of Wildlife Diseases 25: 206–217.[Abstract]
POLLOCK, K. H., J. D. NICHOLS, C. BROWNIE, AND J. E. HINES. 1990. Statistical inference for capture-recapture experiments. Wildlife Monographs 54: 1–97.
ROBBINS, A. H., M. D. BORDEN, B. S. WINDMILLER, M. NIEZGODA, L. C. MARCUS, S. M. O’BRIEN, S. M. KREINDEL, M. W. MCGUILL, A. DEMARIA JR., C. E. RUPPRECHT, AND S. T. ROWELL. 1998. Prevention of the spread of rabies to wildlife by oral vaccination of raccoons in Massachusetts. Journal of the American Veterinary Medical Association 213: 1407–1412.[Medline]
ROSATTE, R. C. 2000. Management of raccoons (Procyon lotor) in Ontario, Canada: Do human intervention and disease have significant impact on raccoon populations? Mammalia 64: 369–390.
———, D. DONOVAN, M. ALLAN, L. A. HOWES, A. SILVER, K. BENNETT, C. MACINNES, C. DAVIES, A. WANDELER, AND B. RADFORD. 2001. Emergency response to raccoon rabies introduction into Ontario. Journal of Wildlife Diseases 37: 265–279.[Abstract]
ROSE, J., AND J. D. HARDER. 1985. Seasonal feeding habits of an enclosed high density white-tailed deer herd in northern Ohio. Ohio Journal of Science 85: 184–190.
RUPPRECHT, C. E., J. S. SMITH, M. FEKADU, AND J. E. CHILDS. 1995. The ascension of wildlife rabies: A cause for public health concern or intervention? Emerging Infectious Diseases 1: 107–114.[Medline]
SLATE, D., R. B. CHIPMAN, C. E. RUPPRECHT, AND T. DELIBERTO. 2002. Oral rabies vaccination: A national perspective on program development and implementation. Vertebrate Pest Conference 20: 232–240.
STUEWER, F. W. 1943. Raccoons: Their habits and management in Michigan. Ecological Monographs 13: 203–258.
URBAN, D. 1970. Raccoon populations, movement patterns, and predation on a managed waterfowl marsh. Journal of Wildlife Management 34: 372–382.
WHITAKER, J. O. SR., AND W. J. HAMILTON JR. 1998. Mammals of the Eastern United States. 3rd Edition. Comstock Publishing Associates, Cornell University Press, Ithaca, New York, 583 pp.
WIKTOR, T. J., R. I. MACFARLAN, K. J. REAGAN, B. DIETZCHOLD, P. J. CURTIS, W. H. WUNNER, M. KIENY, R. LATHE, J. LECOCQ, M. MACKETT, B. MOSS, AND H. KOPROWSKI. 1984. Protection from rabies by a vaccinia virus recombinant containing the rabies virus glycoprotein gene. Proceedings of the National Academy of Science USA 81: 7194–7198.
WILSON, D. E., AND S. RUFF. 1999. The Smithsonian book of North American mammals. Smithsonian Institution Press, Washington, DC.
Received for publication 28 April 2003.
This article has been cited by other articles:
|
|
 |
|
  R. Rosatte, M. Allan, P. Bachmann, K. Sobey, D. Donovan, J. C. Davies, A. Silver, K. Bennett, L. Brown, B. Stevenson, et al. PREVALENCE OF TETRACYCLINE AND RABIES VIRUS ANTIBODY IN RACCOONS, SKUNKS, AND FOXES FOLLOWING AERIAL DISTRIBUTION OF V-RG BAITS TO CONTROL RACCOON RABIES IN ONTARIO, CANADA J. Wildl. Dis., October 1, 2008; 44(4): 946 - 964. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. C. Ramey, B. F. Blackwell, R. J. Gates, and R. D. Slemons ORAL RABIES VACCINATION OF A NORTHERN OHIO RACCOON POPULATION: RELEVANCE OF POPULATION DENSITY AND PREBAIT SEROLOGY J. Wildl. Dis., July 1, 2008; 44(3): 553 - 568. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
