A Final Report
To: The Summerlee Foundation in Fulfillment of Grant Funding
Submitted by: Ivonne Cassaigne, DVM
CANINE DISTEMPER VIRUS TRANSMISSION -EFFECTS ON A PREDATOR-PREY RELATIONSHIPIN A HABITAT WITH A FERAL DOG PRESENCEAND SPECIFIC ACTIONS TO DECREASE ITSNEGATIVE EFFECTS
Figure 1. First-ever photo of a jaguar and her kitten in Sonora, Mexico
Funds in the amount of $15,000 were generously granted by The Summerlee Foundation and authorized for the following specific objectives:
1. Determine the seroprevalence of antibodies against canine distemper virus in pumas (Puma concolor), peccaries (Pecari tajacu), coyotes (Canis lantrans) and/or feral dogs (Canis familiaris)
2. Test for correlations between levels of antibodies from feral dogs and levels of antibodies in peccary, coyote and/or pumas
3. Deployment of 7 satellite collars on pumas in each study area to estimate the predation rate on cattle
4. Vaccination of the feral (ranch) dog population
5. Translocation of peccaries as alternative prey to decrease puma and jaguar (Panthera onca) predation on cattle
The funds were to be used in collaboration with funding from Primero Conservation (http://www.PrimeroConservation.org) (PC), a not-for-profit charity incorporated in the USA, created in 2011 to work with its partners in Sonora, Mexico to mitigate the local killing of carnivores, especially jaguars and pumas, on active cattle ranches within two established Management and Conservation Wildlife Units (UMAs) registered as; 1) the “Refugio Privado para Jaguars Silvestres”, and 2) “Programa de Conservacion del Jaguar en la Sierra Alta de Sonora”. The UMA’s objectives are to better manage wildlife in an attempt to decrease predation rates by pumas and jaguars on livestock. The following objective was added to this study since the interim report was submitted:
6. Determine the seroprevalence of antibodies against canine distemper virus in jaguars (Panthera onca) and deploy GPS collars opportunistically on any jaguars captured during the study
Field research invariably is about the past, whereas the reality of a species’ ability to survive or coexist as a viable and genetically significant population in the presence of humans is about the future. Annual and long-term survivorship of large carnivores that compete with food producers often depends on the real or perceived economic cost to the landowner of the habitat upon which the species needs to survive. Livestock producers and wildlife conservationists do not favor the same mix of generality, realism, and precision of the value of apex carnivores, and this problem is exacerbated by the real life economics of protein production by landowners driven by demands from the consumer. Communities, and the ecosystems upon which they depend, include traits that involve the collective behavior of all their parts, behavior that is not obvious from observations of specific ecosystem parts. The result of eliminating one species from an ecosystem often appears to have no immediate consequences. Thus, of the 5 species extirpated from the wild in Mexico, it is not surprising that 2 are apex carnivores (wolves and grizzly bears). Wildlife conservationists must be informed as to the collective forces within ecological communities that concern and motivate food producers and landowners who can be ultimately responsible for the conservation of multiple species. Therefore, it is believed that a manipulative study focusing on the provision of alternative prey for apex carnivores to benefit a single prey species (livestock) will have greater conservation implications, at least locally, for other species.
Jaguars and pumas are currently facing threats where they are not protected (Texas) and where law enforcement of regulations is minimal, (Mexico south throughout their range). They are considered across their range as species of conservation concern, apex carnivores, an umbrella, and keystone species. Thus, conservation programs focused on jaguars and pumas are likely to trigger greater and deeper conservation impacts. Recent studies suggest that jaguars are more abundant in the southern portions of their range than the northern portions (Cavalcanti and Gese 2010) and this species is considered endangered in both Mexico (SEMARNAT 2002) and the United States. Loss and fragmentation of habitat, and direct human-caused mortality due to livestock depredations and illegal human-caused mortality (trapping, poisoning, shooting) are the primary causes of declines in many populations (Lopez and Brown 2002, Rosas-Rosas et al. 2008, Rosas-Rosas and Valdez 2010).
As apex predators, both jaguars and pumas are intimately intertwined with many ecosystem processes and regulate proper functioning ecosystems through compensatory predation. Yet, the population dynamics and basic ecology of jaguars and their interactions with pumas and their prey species are poorly understood, particularly in the species’ northern sympatric distribution. Among the four large cats of the genus Panthera, jaguars are the least known. In the New World, the only cat of comparable size is the puma. Jaguars and pumas are sympatric across the entire jaguar range. Studies examining the ecology and conservation needs of coexisting jaguars and pumas are few and far between, and they have been typically carried out in tropical areas, biologically considered as the core habitats of jaguars. The jaguar range turns increasingly xeric in a south north transect, especially in its northern end, and decreasing population sizes have been assumed as prey richness decreases in dryer habitats.
The northern Mexico jaguar population may be an important source for animals occurring in the Southwest United States, and would be the logical source for reintroductions should that be deemed feasible and practical. Little is known about jaguar dispersal rates and use of corridors for connectivity, and what is known applies to tropical habitats, primarily tropical rainforest (Rabinowitz and Zeller 2010). Because our study area is thoroughly different from other jaguar habitats, and because the population in this area is likely to be the largest and closest to the U.S., this project could influence when and whether the United States can be repopulated through natural dispersal or whether translocations are necessary. Significantly more information is available for pumas (Logan and Sweanor 2001, Hornocker and Negri 2010). These types of data are of crucial importance now that the United States has erected many kilometers of impassable wall fences along the Mexican-United States border to halt illegal human immigration. These barriers impact the movements of jaguars and pumas in unknown ways and may significantly affect natural expected genetic exchange, dispersal, and movement of both species northward into the United States. International collaborations between both countries are essential for effective conservation of not only jaguars and pumas but also a wide variety of other species (Medellin 1998).
Mexico is embarked on a nation-wide effort to be the first in the world to have a countrywide jaguar survey that may support this assumption (R. Medellin, personal observation). Surprisingly, in northeastern Sonora, camera trap data is revealing that there exists a population of jaguars that is comparable to, and possibly larger than, many of the populations that have been studied
further south in tropical areas.
The Sonoran study area is located near the confluence of the Aros and Bavispe rivers in Sonora, Mexico (Figure 1.). Rosas-Rosas and Valdez (2010) provided a description of the study site and we paraphrase their description here. The area is located in the Sierra Madre Occidental and is composed of 11 ranches. It is located about 270 km south of the United States-Mexico
boundary. The area is about 30 km southwest of Nacori Chico, Mexico. Their study site was 400 km2 but movements of animals will determine the size of our study area. Movements out of the core study area will require cooperation from other land managers and conservation areas. The area has a wide diversity of vegetation such as Sinaloan Thornscrub, Sinaloan Tropical Deciduous Forest, Madrean Evergreen Woodland, and Montane Conifer Forest in additional to a number of riparian and other wetland communities (Brown 1994). The Aros and Bavispe rivers bisect the area. Free-ranging cattle ranching is the primary
land use and cattle are probably the most abundant large prey species. Other important species include white-tailed deer (Odocoilius virginianus), collared peccary, white-nose coati (Nasua narica), and various species of lagomorphs and skunks.
Figure 1. Peccary translocation area in Sonora, Mexico.
Comparative Study Area – Tans-Pecos southwest Texas
A second comparative study area was originally proposed to be located in southwestern Arizona. This site has been replaced with a similar size study area in southwest Texas. During the initial phase of this project an adaptive management strategy was adopted for pumas in and around the Kofa National Wildlife Refuge to restore desert bighorn sheep (Ovis canadensis) populations back to recent historic numbers (USFWS Kofa EA). The action was initially delayed due to an appeal to the ninth circuit court but eventually was fully implemented. The implementation has resulted in the consistent removal of pumas killing 2 or more desert bighorn sheep in a six-month period. These actions negated our ability to fully test the hypotheses stated in the original study design and to look at the impacts of CDV in a proper functioning predator/prey system. We selected a study area in southwest Texas that is currently fully funded by The Summerlee Foundation and titled as “Predator/Prey Relationships, Diet, and demography of Mountain Lions in the Davis Mountains”. We are working in collaboration with the Borderlands Research Institute at Sul Ross State University during this project. We hope this needed change is acceptable by The Summerlee Foundation Board in that The Summerlee Foundation’s full support of the Texas project greatly leverages our ability to complete a science based comparison between the two sites due to financial needs in Sonora.
The Trans-Pecos study area is located in the Davis Mountains and is centrally located within Jeff Davis County and represents the largest and highest land mass in Texas (approximately 400 km2). It is bounded on the east and south by Texas highway 17, on the west by Texas highways 166 and 118 and on the noth by Interstate 10. Annual average precipitation ranges from
45-58 cm with most of the precipitation occurring during the months of July-August. Other precipitation occurs, in the form of snow during December and January. Soils include shallow to deep, well drained, hilly to steep, noncalcareous, and loamy soils. Oak savannah vegetation dominates lower elevations. The dominant grasses in the area include bull grass (Mullenbergia spp.) cane bluestem (Bothriochloa barbinodis), side-oats grama (Bouteloua curtipendula), and blue grama (Bouteloua gracilis). Primary forbs in the area include Mexican sagewort (Artemisia spp.), bush sunflower (Asteraceae spp.), and crotons (Croton spp.). Typical woody species that occur include ponderosa pine (Pinus ponderosa), white pine (Pinus strobiformis), Mexican pinyon (Pinus cembroides), and alligator juniper (Juniperus deppeana) (Warnock 1977).
Initially a predator and prey survey was conducted in Sonora using camera traps at 38 locations continuously for 2 ½ years on the two UMAs between April 2009 and September 2011. Paired cameras by Cuddeback (http://www.Cuddeback.com) and Reconyx (http://ww.reconyx.com) were deployed at 38 locations in tropical thornscrub at elevations ranging from 440m to 1230m above sea level. Cameras were checked opportunistically during ranching operations. Memory cards were downloaded to a computer, pictures were labeled with date and time the image was recorded, and stored in location/species/number-of-individuals folders (Harris et al., 2010). For each camera location, independent pictures of a single species were defined to be
those pictures taken more than one hour apart. For each of 24 one-hour time segments, a species was considered active during an hour if any camera recorded its picture at least once during the hour. Pictures recorded in different cameras were considered independent. Sequential pictures of the same species at the same location during a one-hour segment were also considered redundant and not used in the analyses. Jaguar and ocelot were uniquely identified (hence the paired cameras) by coloration patterns to determine densities during the project. A similar camera trap survey has just been initiated in the Davis Mountains of southwest Texas by Borderlands Research Institute.
Pumas, jaguars, coyotes and peccaries were captured using box traps and leg-hold snares built to our specifications by Morgan Supply, Canada (morgansupply.com), set in areas of frequent lion visitation as determined by sign such as scrapes, tracks and scat. Snares were used in combination with a variety of battery-powered communication calls equipped with solar switches that played continuously during the night at preselected timed intervals constructed in collaboration with the company Wasatch Wild (wasatchwild.com). Pumas and jaguars were fitted with ATS G2110E Iridium/GPS location collars or North Star satellite collars scheduled to collect 6 locational fixes during a 24-hour period (daytime at 1200 hours and night time at 1800, 2100, 0100, 0400 and 0600 hours).
During captures blood collections of 2-3 milliliters of venous blood were collected from pumas, peccaries, coyotes, feral dogs and jaguars and sent to the University of Colorado Veterinary Lab and UMAM in Mexico City.
Kill site investigations
Kill and/or feeding sites were defined as locations consisting of two or more consecutive GPS coordinates less then 200 meters apart downloaded between the night-time hours of 1800 – 0600 hours and then uplinked to an Iridium satellite/website center. GPS location clusters were investigated using the center of cluster coordinates after it was determined that an animal moved
off from a suspected kill/feeding site, so as not to influence (increase) kill/feeding rates through our investigative presence. In the field we are differentiating cluster sites as either a kill site, determined by the presence of evidence of a lion-caused mortality (e.g. evidence of an attack, venous blood associated with an attack site), or a feeding site, defined as the site of a carcass
whose mortality was not directly attributed to a puma attack (e.g. feral hog carcass shot during control activities). Information recorded at kill/feeding sites includes: species, age class, breeding condition and gender of prey. Various habitat characteristics are also recorded such as slope, aspect, vegetation, elevation, percent canopy cover and distance from water. This same information is currently being collected in additional ongoing lion prey studies in at least three additional ecoregions in North America for later comparison. We feel it is important that wildlife researchers communicate with project personnel conducting similar studies in other locales so that the power of data sets can be increased when combined in future analysis. Results of kill site data investigations will be presented in a later report.
In preparation for a first-ever translocation of peccaries into Sonora, in 2010 a vaccination trial campaign was conducted using a government approved canary pox vectored canine distemper vaccine (Purevax® Ferret Distemper, Merial Limited) and determined to be safe and induced a serum antibody response in peccary in order to develop future strategies for the control of
population declines in captive and relocated peccary due to canine distemper outbreaks. Human conflict peccaries, often euthanized by state protocol, were permitted by the state to be penned at Southwest Wildlife Conservation Center (SWCC) and vaccinated against CDV and translocated to Sonora as a part of this project.
Camera trap survey of predators and prey
Trail cameras sampled an area of approximately 400 km2 for 8408 camera trap days. Results for species of interest for this project are shown in Figure 2 and listed as follows:
The number of jaguar was estimated to be 11 (σ = 0.439) resulting in a density of approximately 2.7 jaguar/100 sq. km. Puma were not uniquely identified and therefore a density cannot be calculated. Borderlands Research Institute will present additional camera trap data for Texas in a separate final report.
Figure 2. Number of camera sites with number of each felid species captured by camera (in parenthesis) (from data analysis by Jim Sanderson).
Vaccination/ CDV titers tests
No adverse reactions to the vaccine were observed at any time during the initial vaccinations. Serologic evidence of a response to the distemper antigen was assessed via blood drawn from all animals to be translocated at day 7, 14, 28, and on or after 56 days for antibody testing. Approximately 2-3 mls of blood from each animal was sent on cold packs to the Colorado State
University Veterinary Diagnostic Laboratory at Fort Collins, Colorado or UNAM Mexico City for titers determination. Additional analysis will be conducted after a larger comparative sample is obtained from wild populations of peccary, coyote, feral dog, puma and jaguar. Titer results from vaccinated peccary that will be translocated during this study are given in Table 1.
Results in Tables 1 – 4 are generally interpreted as:
• An animal has no protective field immunity from a virus if the titers level is detected at <1:2.
• Animals with titers levels between 1:2 and 1:16 are considered as having limited field immunity and above 1:16 the animal has had an exposure to CDV.
Table 1. Canine distemper virus titers in a captive peccary population (n = 30) after 1 and 2 vaccinations as determined from serological examinations.
Captured puma (n = 13), jaguar (n = 2), peccary (n = 15), feral dog (n = 5) and coyote (n = 5) blood samples have been tested and are being tested by the University of Colorado Veterinary Diagnostic Lab at Fort Collins and UNAM in Mexico City for CDV titers. Test results to date are presented as follows:
Table 2. Canine distemper virus titers of 17 wild peccaries captured in the Trans Pecos study site.
Table 3. Canine distemper virus titers of pumas from Sonora (SON) and Texas (TX) study sites.
Table 4. Canine distemper virus titers of feral dogs from Sonora (SOND) and coyotes from Texas (TXCOY) study sites.
A single male jaguar blood sample tested positive to CDV exposure at a high titers level of 1:256.
Kill site investigations
Depredations have been verified by our real-time GPS kill-site data. In Sonora we have investigated over 75 kill sites of recently collared jaguars (n = 2) and pumas (n = 7). The results of these and future ongoing investigations will be presented in a separate report acknowledging the Summerlee Foundation. However, we have verified that both pumas and jaguars kill and
consume CDV hosts including peccaries, coyotes, foxes and skunks. Livestock appear to be a prominent prey element in their diet.
On June 8, 2012, 29 peccaries previously vaccinated against CDV, eartagged, and micro-chipped were transported from Scottsdale, Arizona to the international border and crossed at Douglas/Agua Prieta after governmental inspection and permitting. The transport containers (modified horse trailers) were inspected and sealed by SAGARPA prior to transport from the
international border to the holding pens at the release site on Rancho Pueblo Viejo in Sonora, near the confluence of the Aros and Bavispe rivers. Peccaries were released into the pre-constructed pens that provide water and shade (Figure 3.) on June 9, 2012 for quarantine. An initial release of 40 peccaries with radio collars is planned for June 2013 during the beginning of the driest period of the biological year and highest period of livestock losses due to predators.
Figure 3. Peccaries are being held temporarily in holding facilities built on Ranch Pueblo Viejo. Releases are scheduled for June, 2013.
Depredations upon cattle by puma and jaguar are real and just now being documented by GPS technology as occurring on ranches comprising the UMA and additionally for surrounding area ranches. The natural forces that affect predator survivorship therefore affect area ranchers and how they manage their livestock. We hope our data will result in the illumination of statistically valid livestock predation rates and survival rates of pumas and jaguars. Only then can mitigation or reduction of livestock losses through a study-driven model to improve livestock husbandry and better management of important native buffer prey species be implemented. Using science-based data and modern geographic information displays (GIS) our goal is to clearly show livestock
operators how apex carnivores utilize native prey and associated private ranch habitats in the presence of livestock throughout the year. Periods of high livestock depredations by jaguars and pumas make it difficult to expand any type of protection for jaguars and pumas to even landowners adjacent to the UMA without the development of a mitigation model that offsets livestock
losses and offers the potential for additional sources of income in exchange for not poisoning or trapping offending animals
In Sonora, Rosas-Rosas et al. (2008) documented that beef calves constituted 38% and 9% of prey biomass for jaguars and pumas, respectively. We feel this data needs to be reevaluated using real time GPS data and not solely through scat analyses unsupported by DNA analysis. Collectively, the two predators accounted for <14% of known cattle mortalities but mortality
from the two species were suggested to be as high as 36% if missing calves were included, an assumption all local ranchers already suggest without additional evidence of confirmed causes of mortalities. Based upon an analysis by Rosas-Rosas et al. (2008) of just 27 scats, suspected to be from jaguars, but unverified by molecular methods such as those described by Naidu et al. (2010), the data indicated that cattle accounted for 58% of the biomass consumed by jaguars followed by white-tailed deer. Three individual jaguars were believed to be responsible for most of the depredation events, again unverified by DNA. In contrast, in the same study, biomass in puma scats consisted mostly of white-tailed deer (57.3%) followed by collared peccary
(Pecari tajacu: 11.3%) and then cattle (8.7%).
Rosas-Rosas et al. (2010) quantified gross habitat characteristics associated with cattle depredations by predators in northern Mexico. Most depredations during the dry season occurred in riparian areas in association with perennial water sources. Cattle management practices such as stocking rates, age class of livestock, year-round calving, movement patterns, and availability of water sources are all factors that are believed to influence jaguar and puma depredation.
The seroprevalence and zoonosis of canine distemper virus (CDV) is poorly understood in wildlands but is caused by a Morbillivirus virus in the family Paramyxoviridae (Budd, 1981). The disease is acute and highly contagious in dogs with 100% mortality and is transmitted via aerosol (Appel, 1987; Timoney et al., 1988). CDV infection is enzootic in many wild and domestic species throughout the world (Appel, 1987). All members of Canidae and Mustelidae are affected by CDV (Timoney et al., 1988). It is also known to be hosted by members of the Felidae, Procyonidae, Hyaenidae, Ailuridae, Ailuropodidae and Viverridae families, some of which very little is known as to the severity of the impacts during an outbreak of CDV (Appel 1987; Murphy 1999). In this study we have documented that jaguars and pumas have been exposed to CDV as have feral unvaccinated dogs and coyotes. We also know peccaries, coyotes and foxes are in the food chain as prey for jaguars and pumas. We have the first-ever video footage of both a puma and a jaguar scavenging on the same peccary carcass, but during different distinctive time periods.
Peccaries are considered a major prey item throughout the range of the jaguar (Medellin et al. 2002, Nunez et al. 2000) and where the range of the puma is sympatric with peccary (Rosas-Rosas 2008). In the Rio Yaqui watershed near the junction of the Aros and Bavispe rivers, peccaries are thought to be less plentiful now than prior to 2002 (personal communication by
Memo Galaz-Galaz). A sudden decline in peccaries is believed to have occurred due to unknown reasons in 2002, a drought year for the region as described by local inhabitants (personal communication by Memo Galaz- Galaz). It was approximately the same time period that local livestock depredations by jaguar and puma were also noted to be high and subsequently resulted in the removal of 11 jaguars (Rosas-Rosas 2008). Based on a review of camera trap photos from the project area during the period 2008-10 peccary populations are still considered to be very low when their detection rates by camera traps are compared to other prey items (see camera trap data in results).
Reasons for sudden past declines of peccary have been hypothesized to be the result of CDV outbreaks (Noon et al. 2003). In 1989 a disease epidemic of apparent high mortality occurred in collared peccary in southern Arizona. Serologic testing of clinically normal peccary at that time and in subsequent years found a relatively high and persistent prevalence of serum neutralizing antibodies to CDV (Noon et al. 2003). Our trail cameras documented that all four species of felids (jaguar, puma, bobcat and ocelot) share the same space in our study area. Results suggest that jaguar and puma are able to co-exist by avoidance, but not by segregation (dividing space) with pumas being more diurnal than jaguar. Our results differ from those found in the dry forest of southern Bolivia (Romero-Muñoz et al., 2010) and are significant relative the seroprevalence processes of CDV.
The study area appears to have a high density of jaguars and pumas as determined by camera traps since 2009. The ranches in the study are have a real or perceived high predation rate by puma and jaguars on livestock (personal communication by ranch owner Jesus Moreno). The relationship between density of apex carnivores and livestock predation is as yet undetermined.
Vaqueros who actively manage livestock accompany dogs in the majority of photographs. These dogs are considered to be feral when they are left alone to fend for themselves during extended vaquero absence. We do not know what impact unvaccinated feral dogs have on wildlife. Northeastern Sonora suffered a sharp decline in peccary numbers in the recent past (Memo Galaz - personal observation). We speculate that an outbreak of CDV, similar to the documented epizootic in Arizona (Noon, 2003), could have been responsible for the low number of peccary we recorded. Since peccary are known to be the prey of jaguar and puma, and are known to be relatively abundant in other ecoregions where they occur reintroduction of vaccinated
peccary might serve to mitigate the loss of calves to jaguar and puma that we have recorded.
Broad-scale conservation of species must take place on humandominated, heavily impacted landscapes such as ranchlands. In our study, cattle accounted for nearly 40% of all pictures and peccary just 0.12%. Thus, protected areas with few peccary and no cattle might well serve only as temporary sanctuaries. Our study demonstrates to date that ranchers and their ranchlands in northeastern Sonora, Mexico can and do act to protect the full richness of biodiversity, but not without negative impacts such as we hypothesize for peccary. Solutions may be as simple as vaccinating ranch dogs and as complex as the development of management plans for livestock on a landscape basis. We encourage cooperation among many different conservation organizations working in Sonora, Mexico, by publishing the left and right photographs of each individual jaguar and ocelot on our web site
Camera traps will be the primary method of monitoring peccary populations within the study area supplemented with translocated marked peccaries. Camera data will be compiled and analyzed using the program Renamer as described by Harris et al. in 2010 and analyzed by mark-recapture programs such as PRESENCE (MacKenzie et al. 2005). Kill cluster sites of
area pumas with GPS radio collars will be continue to be investigated to determine rate of predation on marked (DNA and tag presence) and unmarked peccaries. Scat of puma and jaguar will also be collected opportunistically from within the project area and analyzed for marked and unmarked peccary following procedures as described by Naidu et al. in 2011. A program using a
Peterson-Lincoln index analyses will be used to analyze capture/recapture data.
For the future we hypothesize that:
• An existing commercial vaccine against CDV will be successful in building significant antibodies in peccary against the canine distemper virus and that vaccinated peccaries will subsequently have a higher survivorship long term and comprise larger herd sizes than unvaccinated wild extant peccaries.
• Translocated vaccinated peccary will become alternative prey for project area puma and jaguar and marked animals will be detected in each carnivore’s diet.
• Puma and jaguar predation rates on livestock will be lower within the study site with translocated peccary than an adjacent site without translocated peccary.
• A dog vaccination program in the small communities within the sierras of Sonora would have indirect benefits to area wildlife.
Note: During past and future vaccinations all animals have been and will be observed for approximately 1-hour post vaccination and once during a 24-hour period and assessed for general vaccine safety. All observations were documented in a study log. Capture protocols were followed that have been reviewed and approved for animal safety standards established by UNAM and/or Sul Ross State University.
We wish to sincerely thank the Summerlee Foundation for its generous grant award that provided the initial funding for this project.