The African wild dog is Africa’s most endangered large carnivore and is listed as endangered in the IUCN Red List. Less than 6’000 free-ranging individuals survive in the wild and the species has been given very high conservation priority.
The African wild dog, like few other territorial land species, is characterized by the need for vast semi-pristine and undisturbed areas. This peculiarity makes it particularly vulnerable to habitat loss, deterioration, and fragmentation. This highlights the importance of landscape connectivity between subpopulations in and around protected areas.
A deeper understanding of where and how resident packs and dispersing individuals move, and the demographic consequences of dispersal, is therefore important for the management and conservation of the species over larger landscapes such as the Kavango-Zambesi Transfrontier Conservation Area (KAZA/TFCA).
On Christmas day 2018, we were able to observe the first high-resolution trans boundary dispersal from Botswana’s Okavango Delta to Zimbabwe’s Hwange National Park. This journey highlights the central role of Botswana in the KAZA/TFCA trans boundary conservation effort, as well as the role of the African wild dog as a flagship species.
On 16th December 2018, four sisters dispersed from their pack resident just outside Moremi Gamer Reserve in Botswana. The pack is part of a long term study and is regularly followed by our collaborators on site, the Botswana Predator Conservation Trust. This dispersing coalition is somehow unusual, as it is composed of females born to the same mother but in three different litters (in three consecutive years). The coalition dispersed in an almost straight line moving east and crossed the Zimbabwean border after covering an astonishing 280 km. They continued and covered additional 100 km within Hwange National Park, totalling 360 km in only nine days! Thanks to a collaboration with the Zimbabwean-based Painted Dog Conservation Program, we were able to confirm that all four females are still alive and in good conditions. As they seem to have slowed down a little, the daily covered distance much decreased in the past few days, we expect them to soon associate with unrelated males. The formation of a new pack and the settlement in a new territory is the ultimate goal of dispersal, as successful dispersal allow gene flow between subpopulations and for the recolonisation of empty areas.
We will be able to continue monitoring this dispersal event, thanks to the Satellite technology in the radio collar and to the support of the folks at Painted Dog Conservation in Zimbabwe. Stay tune for more updates on these girls and on future dispersers.
Trans boundary dispersal event from Botswana’s Okavango Delta to Zimbabwe’s Hwange National Park. The blue line represents the dispersal trajectory, the yellow line the international boundaries. The red square represents the last recorded location.
Well done team wild dog for two additional awards!
The Vectronics Aerospace photo contest 2018 and the British Ecological Society photography competition 2018, in the category “Ecology in Action”. What next…?
Stay tuned for some exciting findings from our African wild dog dispersal study.
The winter months of June, July, and August mark the coldest period of the year in Northern Botswana. It is this cool period that African wild dogs choose for whelping, as pup rearing is energetically costly for a pack. Typically, the dominant female looks for a suitable den site, such as an abandoned aardvark hole, where she will give birth. As the mother is obliged to stay at the den to nurse and protect the pups, so the survival of both the mother and pups is fully dependent on the other pack members during this critical period. The rest of the pack all leaves the den site once or twice per day to go hunt, and upon returning, they regurgitate part of their freshly caught meal to feed both mother and pups.
During the denning period, the normal ranging radius of a pack is drastically reduced. Consequently, denning season is the optimal time of the year to locate the study area packs and for pack monitoring. This July we tried our luck to locate the den site of the missing ‘Mula’ pack. This pack was formed in February 2016 by the joining of a male dispersing group with a female group, each of which we had monitored with satellite collars as part of our ongoing research project on African wild dog dispersal. However, these satellite collars both had dropped off by now. As we are also interested in the settlement and reproductive success of recently formed packs, it was critical to catch up with this pack again.
Having no working radio telemetry collars on the pack we were left with no other option than to track down the pack on the ground the old-fashioned way. We already knew that the new Mula pack’s home range extended from Xakanaxa all the way east along the Khwai river to North Gate of Moremi Game Reserve. This meant they could theoretically have been denning anywhere within about 400 km2 and along a stretch of about 50 km. We decided to start our search mission in Xakanaxa where frequent sightings of 12 wild dogs had been reported on the sightings board at Moremi South Gate. The last time we saw ‘Mula’ pack, at the end of 2016, they were ten adults plus two pups. Therefore, these reports sounded promising.
With enough supplies to spend a couple of days away from our research camp, Ed, a Research Technician at Botswana Predator Conservation Trust (BPCT), and I set out for the 50 km drive to Xakanaxa. Upon arrival, we bumped into some safari guides who told us that they had regularly seen a pack of 10 wild dogs and one male was wearing one of our radio collars. Luckily, one of the guides had just seen the pack earlier that morning. Using this information, we started looking for fresh tracks on the main road and soon found some. The prints were following the road for roughly 3km but eventually they left the road and we lost their tracks. We decided to park our LandRover close to a crossroad to see if the pack would use the same road on their way out of the presumed den site for the evening hunt.
Our car parked next to the road in anticipation of the ‘Mula’ pack coming along on their way out of the presumed den site.
Being impatient, we left our lookout before sunset and looped around to the North the nearby Paradise Pools to search there. With no success, we drove back to the crossroad to find out that the pack had just come by, as indicated by the many fresh tracks on the road. We rushed following the tracks and caught up with the pack shortly after. They were spaced apart and due to the progressing darkness, we were not able to count more than eight individuals. As we were taking photos of them for identification, one of the dogs who was lagging behind suddenly turned back and sprinted down the road toward where they just came from. We followed him as we guessed he might have changed his mind and decided to return to the den. He ran down the road at full speed, coming by our stake out place at the crossroad and then continued heading further south. Then suddenly, he turned east into the thick mopane bushes. It was too dark for us to try to follow him through the thick vegetation, we took a GPS fix of where he turned off and went off to find a suitable spot to spend the night. Later that evening we confirmed by examining our photos, that these were indeed the remaining members of ‘Mula’ pack.
Our first encounter of ‘Mula’ pack as they were trotting along to road for an evening hunt.
Early the next morning, we positioned the vehicle again at the crossroad in hopes that the pack would come by again. Unfortunately, no dogs appeared and we had to conclude they must have chosen a different route. Just as we were about to leave, another game drive vehicle pulled up and told us they had just seen the dogs back on the airstrip again heading east – which meant towards where we saw them last night. In anticipation that the dogs would continue running in a straight line, we drove south on the main road and parked close to where the extension of the airstrip would intersect the road. Luckily, we were at the right spot as a group of six of them came out of the thicket and crossed the road. We immediately followed them into the bush. Despite many obstacles, we managed to stick with the dogs for roughly 300m. But then, suddenly, we lost them in a relatively open area. We got out of the car and started looking for fresh tracks. There were quite a few prints on a small game trail. By following them, we ended up in a sandy rift with spaced out apple-leaf trees. Unfortunately, the sand was very deep which made it hard to surely identify any of the many tracks as being from wild dogs.
A subgroup of ‘Mula’ pack crossing the road on their way back to the den site
For the next three days, we continued our search but never managed to close in further to the potential den site. We usually lost track of the dogs in deep sand. As we were running low on supplies, we eventually decided to head back to our research camp. But this didn’t mean that we were giving up. Too close were we in finding the den site to call our mission a failure.
One of the identification photos we took of the ‘Mula’ pack dogs after we bumped into them early one morning.
A few days later we drove back to Xakanaxa early in the morning accompanied by ‘Tico’ McNutt, founder and director of BPCT, who has decades of experience in tracking wild dogs. He seemed quite confident in finding their den based on our preliminary efforts in narrowing down the search area. As we approached Xakanaxa, we drove to the spot where Ed and I lost sight of ‘Mula’ pack previously. Tico got out of the car with Ed and me following. He first circled around the spot in a wide loop in search of any fresh tracks. Once we closed the loop he headed for the direction where he must have seen the most promising tracks. Shortly after, he showed us fresh dog tracks and pointed towards the direction they were leading. After about 1.5 km we approached an open sand ridge. Tico told us to stick close together as he was confident of the den being nearby. We followed the fresh tracks in deep sand, and as we were approaching a few apple-leaf trees we suddenly heard the alarm bark of a wild dog. This was undoubtedly the den site we had been looking for. We crouched down but weren’t able to get a visual of the dog as the trees were quite dense. We decided to take a GPS fix and return with the vehicle.
Impressed by the way Tico tracked down the den in no time, we started heading back. After a few steps, he stopped and showed us animal tracks in deep sand. What seemed like it could have been any track to Ed and me, was definitely the spoor of a wild dog according to Tico. He pointed out that he could detect wild dogs in deep sand based on their gait. Two feet are always placed close together with a bigger gap between the next set of pads whereas the step lengths of hyena prints are generally more regularly spaced. This was an important lesson for us and definitely made the difference as Ed and I were constantly losing the dogs in soft substrate during our previous search efforts.
Wild dog tracks (left) are not always as distinguishable from hyena tracks (right). As the substrate gets softer, details of the paw vanish quickly. We learned from ‘Tico’ that it comes down to look for gait patterns such as spacing between the footprints in deep sand.
On our drive to the newly-discovered den site we were not able to see more than half a dozen of the adult dogs nor to get a visual of any pups. But when Ed and I drove back to the den a few days later, we managed to identify all ten adults, the two yearlings, and watch twelve new-born pups play youthfully outside the den. It was well worth the effort: No dogs had died since our last detailed encounter with the pack seven months ago and a promising number of pups was about to grow up.
Watch the ‘Mula’ pups as they emerge from the den:
I am an ecologist interested in exploring how predictions of animal populations can be improved by accounting for dispersal processes, species interactions, and environmental change. My motivation lies in developing reliable population forecasts, enabling the assessment of extinction risks for endangered species under various scenarios. By identifying key factors that drive these dynamics and exploring different intervention strategies, I aim to support effective wildlife management and conservation practices.
In my research career, I have acquired in-depth knowledge in dispersal and movement ecology, specifically examining how these factors influence the demography and spatial connectivity of individual populations. My primary focus has been on the African wild dog, through which I have gained valuable skills in demographic analyses, individual-based modeling, and mark-recapture analyses. More recently, my work has expanded to other species including meerkats in South Africa, giraffes in Tanzania, mouse lemurs in Madagascar, and both roe deer and wolves in Switzerland. With an interdisciplinary background in ecology and engineering, I strive to combine my understanding of ecological systems with quantitative modeling to support the long-term persistence of wildlife populations.
Research interests
Population ecology, movement ecology, conservation biology, population viability analysis, mark-recapture analysis, Bayesian statistics, decision science
2021–present, Postdoctoral scientist | Population Ecology Research Group, University of Zurich
2014–present, Tour guide | Zoo Zurich
2016–2021, PhD in ecology | Population Ecology Research Group, University of Zurich in collaboration with Botswana Predator Conservation (Maun, Botswana)
2014–2015, MSc in Environmental Sciences | University of Zurich
2014, Nature field guide qualification | Field Guide Association of Southern Africa (Limpopo, South Africa)
2008–2009, Various internships as an operations engineer in the private sector (USA & Switzerland)
2006–2009, MSc in Management, Technology and Economics | Swiss Federal Institute of Technology (ETH) Zurich
2005–2006, Research assistant & teaching assistant | Institute of Environmental Engineering, ETH Zurich
2003–2006, BSc in Environmental Engineering | ETH Zurich
Publications
Cozzi G, Reilly M, Abegg D, Behr DM, … McNutt JW (2023) An AI-based platform to investigate African large carnivore dispersal and demography across broad landscapes: A case study and future directions using African wild dogs. African Journal of Ecology➤
Behr DM, Hodel FH, Cozzi G, McNutt JW, Ozgul A (2023) Higher mortality is not a universal cost of dispersal: a case study in African wild dogs. American Naturalist ➤
Tucker MA, Schipper AM, … Behr DM, … Cozzi G, … Mueller T (2023) Behavioral responses of terrestrial mammals to COVID-19 lockdowns. Science➤
Hofmann DD, Cozzi G, McNutt JW, Ozgul A, Behr DM (2023) A three-step approach for assessing landscape connectivity via simulated dispersal: African wild dog case study. Landscape Ecology ➤
Hodel F*, Behr DM*, Cozzi G, Ozgul A (2023) A hierarchical approach for estimating state-specific mortality and state transition in dispersing animals with incomplete death records. Methods in Ecology and Evolution ➤
Jordan NR, Golabek KA, Behr DM, … , McNutt JW (2022) Priority of access to food and its influence on social dynamics of an endangered carnivore. Behavioral Ecology and Sociobiology➤
Hofmann DD*, Behr DM*, McNutt JW, Ozgul A, Cozzi G (2021) Bound within boundaries: Do protected areas cover movement corridors of their most mobile, protected species? Journal of Applied Ecology ➤
Behr DM, McNutt JW, Ozgul A, Cozzi G (2020) When to stay and when to leave? Proximate causes of dispersal in an endangered social carnivore. Journal of Animal Ecology ➤
Cozzi G, Behr DM, Webster H, Claase M, Bryce CM, Modise B, McNutt JW, Ozgul A (2020) African wild dog dispersal and implications for management. Journal of Wildlife Management ➤
Behr DM, Ozgul A, Cozzi G (2017) Combining human acceptance and habitat suitability in a unified socio-ecological suitability model: a case study for the wolf in Switzerland. Journal of Applied Ecology ➤
“Mit Katzen (nicht) kommunizieren: ein Hundeleben”
“In (no) communication with cats: A dog’s life”
At the end of his talk, Gabriele received the prestigious Price for Nature and Environmental Protection (2013) in recognition of his contribution to wildlife conservation in Africa. Congratulations to Gabriele for this well-deserved award!
