Recipient Organization
UNIVERSITY OF WASHINGTON
4333 BROOKLYN AVE NE
SEATTLE,WA 98195
Performing Department
Wildlife Science
Non Technical Summary
After an 80 year absence, the gray wolf (Canis lupus) is recolonizing the Pacific Northwest of the United States, dispersing from populations in the Rocky Mountains and the British Columbia coast. The future of wolf recovery in the Pacific Northwest relies on understanding how wolves interact with other predators and prey, and whether wolves provide positive ecosystem services in managed landscapes where logging, cattle ranching and hunting are permitted. In 2012, our team embarked on a landmark study to assess how wolves shape ecosystem processes in managed landscapes. The study focuses on how deer respond to the presence of wolves in Northeast Washington and whether wolves indirectly increase survival of deer fawns by suppressing coyotes. Here, we propose to broaden the scope of our study by assessing how cougars (Puma concolor) respond to the natural recolonization of wolves. The cougar/wolf interaction is important: cougars are second only to wolves in the apex predator hierarchy, targeted by hunters and are frequently persecuted because of their perceived impacts on livestock and people. Little is known, however, of how wolves and cougars interact in managed landscapes. The recent recolonization of wolves into Northeast Washington offers the perfect opportunity to undertake this research. Should wolves out-compete or displace cougars, changes in cougar abundance and resource use are likely. Such changes could have important, but unknown, effects on ecological communities, especially if cougars change the way in which they interact with prey species such as white-tailed deer (Odocoileus virginianus) and mule deer (Odocoileus hemionus).The goal of this project is to undertake groundbreaking research to understand the interactions between wolves and cougars. We will do this by analyzing the activity and behavior of cougars in areas with and without wolves. We hypothesize that cougars will alter their activity and foraging behavior in areas where wolves are present because of the consumptive (wolves killing cougars) and non-consumptive (wolves displacing cougars) effects of competition. We also hypothesize that cougar impacts on deer will be reduced in areas where wolves are present. To complete the project, our objective is to fit up to 35 cougars with GPS collars to allow movement, habitat and foraging data to be collected and then analyzed using sophisticated statistical and mapping software. This sample size will enable us to draw robust conclusions on the interactions of wolves and cougars in managed landscapes.
Animal Health Component
(N/A)
Research Effort Categories
Basic
100%
Applied
(N/A)
Developmental
(N/A)
Goals / Objectives
We aim to provide information that will help predict how the return of the wolf to Washington state will influence ecosystems by examining how wolves will affect cougar movements and foraging behavior. Thus, the objective of this project is to understand the interactions between wolves and cougars. We will do this by analyzing the activity and foraging behavior of cougars in areas with and without wolves.
Project Methods
Our study is a large-scale natural experiment whereby the potential impacts of wolves will be quantified by contrasting the activity and foraging behavior of cougars at two replicated wolf-impact (treatment) sites with similar measures at two replicated control sites that lack wolves. The field work will be conducted between 2015 and 2019. In order to test whether wolves affect cougars, we will fit up to 35 cougars with GPS collars. Roughly equal numbers of cougars will be fitted with GPS collars in wolf and non-wolf areas. We will also aim to collar equal numbers of males and females. By collaring 35 cougars our study will be consistent (in terms of sample sizes) with published studies on cougar movements elsewhere (e.g. Cooley 2008; Maletzke et al 2014). Thus, we will have sufficient power for statistical analysis. To fit the GPS collars, we will use trained trailing hounds to track and locate cougars within our study site. This method is preferred over leg hold trapping because it reduces the risk of trap-related injuries to cougars. It is also a widely used method to capture cougars for scientific studies (e.g. Bartnick et al. 2013). Once the hounds locate and tree the cougar in a suitable location, we willadminister a first intramuscular injection (4.0-9.0 mg/kg at 100 mg/mL Ketamine) using a DanInjec® dart gun with DanInject® 3 cc (1 cc = 1 mL) darts. Only a certified practitioner will use the dart gun and handle the drugs. When the Ketamine has begun taking effect, a member of the capture crew will advance up the tree and use ropes to lower the cougar to the ground. The crew member will use a harness, climbing spikes and ropes for their own safety. Once up the tree, the crew member will place ropes under the fore and hind legs of the cougar to ensure the animal is safely lowered to the ground. A padded layer (blanket or towel) will be placed between the rope and cougar so that no injuries occur from rope burns while lowering the cougar. Once the cougar is on the ground, we will administer a second injection (0.07 mg/kg at 1.0 mg/mL Medetomidine) by hand. When the cougar is fully immobilized, we will record its heart rate,breathing rate and temperature (°F; 33.8 °F = 1 °C) every 2-5 minutes following protocols outlined in Quigley (1997).The sedated animal will then be weighed and sexed and a hair or scat sample may be taken for DNA analyses. The GPS collar weighs approximately 500-700 grams, and cougars will only be collared if the weight of the collar is less than 5% of the weight of the cougar (an average cougar weighs 53 kg). No juveniles (< 1 year old) will be collared. The GPS collar will be fitted once the cougar is fully immobilized. After approximately 45-50 minutes has passed another intramuscular injection will be given to reverse the effects of the original sedative. The cougar will then be monitored until we are confident that it has the strength and mobility to avoid confrontations with other predators in the area. Intensive radio-tracking and inspection of GPS fixes via email will then be undertaken for a week after capture to ensure the cougar is continuing normal daily movements. GPS fixes will then be monitored on a daily basis and if the GPS collar stops moving for more than 48 hrs, a member of our team will go and inspect the location where the GPS collar has stopped. If the cougar is dead, then the cause of mortality will be investigated by checking the condition of the animal and any injuries. The GPS collars will befitted with a drop-off mechanism (fabric cotton brake away with 6 layers) so that the collar will drop off the animal after approximately 2 years.We proposed to use Vectronics Iridium 2-way survey collars that will record and send up to 12 GPS locations per day to a registered email account via a satellite link for up to 2 years. The GPS location data will be used to investigate if there is a difference in habitat use by cougars in areas with and without wolves. Specifically, we propose to undertake a resource selection functionanalysis, which is a modelling approach that can be used to contrast habitat selection by different populations of cougars. This approach has been used successfully by members of our research team for contrasting behavior of different dingo (Canis dingo) populations in Australia (Newsome et al. 2013a, 2013b) as well as for populations of red wolves (Canis rufus) in southeastern U.S.A (Dellinger 2013). The modelling process will allow for a comparison of how different environmental factors influence cougars in wolf and non-wolf areas. For example, we will assess factors such as elevation, slope, aspect, and vegetation cover. Under our main hypothesis, we would expect cougars in non-wolf areas to select for areas with higher elevation. We will also conduct a series of comparative analyses to investigate differences in daily movement patterns and home-ranges of cougars in wolf and non-wolf areas.Finally, the GPS data will be used to locate and inspect potential kill (or predation) sites as indicated by a cluster of GPS points in a small area (i.e. consecutive daily locations in the same area). The kill site inspections will be primarily undertaken in winter when field teams andtechnicians will be permanently based at our field site. To minimise disturbance to the collared cougar, the aim will be to investigate the kill sites after the cougar has vacated the area. Thus, kill sites will be avoided if the collared cougar is still within 1 km of the area. At each kill site we will follow standard procedures (e.g. following Ruth and Buotte 2007) to determine that the kill was made by a cougar. We will then record the GPS location of the kill site, the species of prey and its age. A standard necropsy may be performed to assess if the prey species has any underlying injuries or to assess the condition of the prey species based on an inspection of bone marrow. If there are cougar feces near the kill site we will collect these for future dietary analyses. We will collect the feces so that a swab of DNA can be taken to assist in estimating the cougar population size. If den sites can be located from clustering of GPS points, DNA from kittens and additional scats may also be collected.Importantly, wolves in the wolf-treatment areas are already collared for ongoing monitoring by the Confederated Tribes of the Colville Reservation and there is existing monitoring of other predators and prey via a comprehensive camera trapping web in both wolf and non-wolf areas. These data will be used for comparative purposes. In particular, the GPS movement data from the ongoing wolf monitoring will be contrasted with the cougar GPS data. This comparison will ensure that any observed differences in cougar habitat use can be assessed relative to known behaviors of wolves.All GPS data will be collated in Microsoft Access or Excel. The analyses will be undertaken using statistical analysis software and quality assurance will be guaranteed by standard data screening processes. A power analyses is difficult to conduct using our nominated target ofcollaring 35 cougars. For these types of studies the best approach is to collar as many individuals as logistically and feasibly possible. However, we note that if 35 cougars are collared it will generate a sufficient number of GPS locations (up to 8760 for each cougar) to perform a resource selection function analyses, to inspect kill sites, and to assess variations in daily movements and home ranges. Indeed, the number of cougars collared is consistent with other studies (e.g. Cooley 2008; Maletzke et al 2014), and the total number of GPS locations from the 35 cougars will exceed 300,000 if 12 fixes per day are generated for two years per cougar.