Feral Pig Eradication Campaign on Santiago Island, Galapagos1 Marc Patry, Charles Darwin Research Station Casilla 17-01-3891 Quito, ECUADOR Introduction Feral pigs (Sus scrofa) are considered one of the most ecologically destructive introduced vertebrates in Galapagos, where they are effective predators of giant tortoise and sea turtles nests, possibly prey upon young tortoises, and in certain circumstances, compete for food resources with young giant tortoises (MacFarland et al, 1974a, Green et al, 1982, Coblentz, 1987). On Santiago island (58,465 ha) pigs have been responsible for a near zero recruitment rate of giant tortoises, whose numbers had already been critically depleted after centuries of human exploitation (ibid). The Galapagos National Park Service (GNPS), and the Charles Darwin Foundation (CDF), both created in 1959, focused on the conservation of the giant tortoises as one of their first priorities – the CDF carrying out research to support management activities carried out by the GNPS. In 1964, the CDF established the Charles Darwin Research Station (CDRS) in Galapagos and within a year established an ex-situ tortoise breeding centre. The first Santiago island tortoise eggs were brought to the breeding centre in 1970 for hatching and rearing (Cruz et al., 1999). By 1973, active pig control campaigns were established on Santiago in an effort to increase natural recruitment rates. The past 30 years have seen a waxing and waning feral pig control efforts. These variations were related to funding availability, El Niño events, morale fluctuations, or lack of training. Despite periods of frustration, the GNPS is now very close to having reached its objective. Only one pig has been observed (dead) since May 2000 despite intensive on-going monitoring. The campaign will not be officially completed until an arbitrarily set 18 month monitoring period for pig sign reveals no further evidence of their presence. This paper provides a descriptive account of the pig eradication effort since 1997, when both the GNPS and the CDRS made the program a bi-institutional priority, allocating it necessary funding and staff. Fundamentals of an eradication campaign Many efforts to eradicate an introduced species have failed for a variety of reasons. Several factors have been identified as contributing to a successful eradication campaign (Myers, J. et al, 2000). For the Santiago pig eradication, the relevant factors are: i) Sufficient resource to fund the campaign to its conclusion: The Santiago pig control/eradication campaign, from 1973 to 1996, had been plagued by unstable funding. Until 1998, GNPS budgets were allocated centrally, in Quito, in competition with the demands of many other government institutions, resulting in an inability to ensure long term stable 1 Referenced as: Patry, M. 2001. Feral Pig Eradication Campaign on Santiago Island, Galapagos. In Informe Galapagos 2001. Fundación Natura, Quito (in press). funding to specific programs. Growing support from the CDF fundraising efforts, along with important budget allocation changes at the GNPS helped provide stable funding starting in the late 1990’s. ii) The species must be detectable at low densities: Throughout the first 24 years of the pig campaign, densities never reached such low values as to create problems in detection by the methods used (fig. 2 illustrates how hunting effort per pig removed remained relatively low until 1996). Groups of hunters and their dogs were sufficiently trained and covered enough territory to be able to detect pigs on the island by direct observation of pigs, or pig sign, at all times. However, as densities declined, hunting effort per pig removed began to climb rapidly by 1997. By 1999, densities reached levels so low as to require the use of widespread baiting in order to detect pigs. This method was successful in detecting a pig in November 2000, 6 months after the baiting campaign was established, and during which time no visible sign of pig had been observed by hunters or dogs. This detections suggests the ability to detect possible a single pig on the entire island. iii) All individuals must be at risk: Extensive areas of dense vegetation on Santiago effectively created refuges where pigs were safe from hunters and their dogs. The development and maintenance of a extensive trail system throughout these areas increased vulnerability of the pigs throughout their range, permitting hunters and their dogs to more easily flush pigs out from their cover. iv) The risk of re-invasion must be controllable: The only means through which pigs could re-establish themselves on Santiago island is through deliberate introduction by humans. This risk is being managed by the GNPS. Pre-1997 By the end of 1996, sporadic control efforts had been on going for 24 years. During that time, 18,903 pigs were removed from Santiago (Calvopiña, J. 1985, Calvopiña, L. 1989, Isabela Project, 2000) and a valuable study on pig distribution, reproduction and territoriality was conducted (Coblentz, B. and D.W. Baber, 1987). However, during the few years prior to 1997, especially from 1991 to 1994, funds were cut back and enthusiasm waned, resulting in dramatic reduction in hunting effort. Figure one illustrates that the average trimester hunter effort (measured as [#hunters + #dogs]x #hunting days) during the 1991-94 period was 419 hunter-dog-days (hdd’s) compared to average for all other years of 1,034 hdd’s. Figure 1 also illustrates how the low hunting effort during the 91-94 was followed by a large increase in pigs killed in 1995, when hunting effort was more than doubled. These numbers may suggest how rapidly the pig population responded to reduced hunting pressures. 1997- Renewed bi-institutional commitment In mid-1997, the Santiago pig eradication program was given renewed priority status by both the GNPS and the CDRS. The project’s management was transferred to the Isabela Project, a newly created bi-institutional project whose longer-term objective was the eradication of goats from northern Isabela Island. This management structure helped ensure the commitment of both institutions, including allocations of staff and funds. These changes resulted in a near doubling of hunting effort from 1996 to 1997 (from 711 to 1,368 hdd’s per trimester), though hunting strategies remained the same. Typically, a group of 10 to 12 GNPS hunters usually with one dog each were sent to Santiago on trips of 8-10 effective working days (not including travel time). Hunters alone, in pairs would carry out loosely coordinated searches of pig habitat on Santiago. Every evening, informal debriefing sessions would determine where hunters would focus efforts on the following day. 1998-99 Stable funding adoption of new technologies Stable Funding The passing of the Special Law for Galapagos in 1998 provided the GNPS with steady and guaranteed funding for its operations. These changes allowed the GNPS to make and keep longer term financial commitments to specific projects. Also, by 1998, a portion of CDF fund-raising efforts were dedicated specifically to the campaign. Combined, these efforts helped ensure stable financing of the pig eradication program, a basic requirement for a sucessful eradication campaign. New technologies By the second half of the 1990’s, the CDRS had begun to apply global positioning system (GPS) and geographic information system technologies (GIS) to field work. The Isabela Project, with the support from the CDRS department of Vertebrate Ecology and Monitoring, became one of the first to systematically apply GIS and GPS to a specific park management challenge. By 1999, hunters had become proficient in the use of GPS units and regularly returned from the field to download GPS data into ARC-INFO GIS software on GNPS desk-top computers. The information revealed exactly what terrain had been covered in their search for pigs (fig. 3). Field computers were acquired in late 1999; with further training, hunters were able to download and evaluate GPS data on site daily. Hunters who had previously boasted at having covered “every square metre of terrain” now saw graphic evidence of huge gaps where no hunter had been. This information helped the hunting team plan the following day’s outing, ensuring that indeed, all of the pig habitat was being monitored. GPS and GIS technology was also helpful to re-establish and maintain a 300 km network of trails through some of the densest vegetation on the island (fig. 3). The trail network, first used 10 years earlier but abandoned and completely overgrown, helped hunters access areas of dense vegetation previously used by pigs as refuges to hide from dogs and hunters. With trails, radios and GPS units, hunters could now range freely throughout pig habitat, coordinate their movements with extreme precision, and increase the overall pig hunting effectiveness. Increased Effort per Pig Removed Typical of an eradication campaign, as the density of the target species drops, the effort required per individual removed increases. Figure 2 illustrates the rapid increase in effort required to remove individuals in the final years of the Santiago campaign. The figures show that 450 times more effort was required to remove a pig in 2000 than in 1988. Obviously, this effort comes with a high relative price tag; though the annual cost of the pig eradication campaign increased by approximately 20% from 1997 to 2000, the actual cost per pig removed increased by approximately 2,000% during that same period. 1999 Monitoring for pig presence By 1999, the number of pigs killed per outing was never greater than 3 (see Fig. 1). In order to increase pressure on the remaining pigs and also, as the start up of a monitoring phase, the systematic poisoned baiting of pig habitat was initiated. Baiting would contribute to increasing the pressure on any remaining pigs and provide an indication the number of pigs left – as pig disturbance of baits is easily noted (no Santiago island animal besides pigs has either the ability or an interest in moving the baits about). 1080 was first applied, with a combination of an anti-emetic (metaclopramide) to ensure that the ingested poison was not later regurgitated. The poison, in solution, was injected uniformally via a series of lethal doses throughout goat carcasses, and these were then placed in areas where hunters thought pigs frequented. Results of the baiting were conflicting – hunters knew of the on-going presence of pigs on Santiago via the occasional footprint or characteristic soil disturbance from rooting behaviour, yet the baiting work often did not reflect a similar presence. Project staff concluded that the pigs had become bait shy. Early onset of symptoms, characteristic of 1080, are thought to discourage the pigs from eating a lethal dose andsurviving pigs learn to associate discomfort with the bait, thus develop bait shyness. In December 1999, after careful consideration of the poisoning campaing results, the Isabela Project switched from 1080 to Warfarin, a poison commonly used for rats, but to which pigs were known to show an unusually high sensitivity. In contrast to 1080, Warfarin is slow acting, with symptoms appearing well-after a lethal dose is ingested, thus reducing the chances of pigs associating the baits with negative effects on their health. Finally, added advantages to using Warfarin included the availability of a simple antidote (vitamin K), and a much reduced toxicity to dogs, humans and other non-target species. Warfarin proved to be more effective and practical than 1080 and remains the poison currently in use. 2000 On-going monitoring and hunting; equidistant baiting In 2000, the initial practice of placing baits haphazardly in areas where pigs signs had been observed was replaced by an equidistant baiting approach to ensure that all pig habitat was uniformly covered. This change marked the transition from erradication activities to monitoring activities. Using GIS, 500m equidistant georeferenced points were superimposed on a map of Santiago (fig. 3). Baits were placed and within two weeks, field staff return to check every bait for tampering. As Warfarin remains active at all stages of bait decomposition, field staff could afford to let a reasonable amount of time pass before returning to check for tampering, thus increasing the effectiveness of their monitoring efforts (Harwood et al, 2000). Between November 2000 and May 2001, all possible pig habitat was baited twice and each of the approximately 15,000 baits applied were monitored. The process will be repeated once more in the following 12 months. Should no further pig sign be detected, Santiago will be considered pig free, though annual monitoring trips will continue. This 18 month period has been arbitrarily chosen, but is thought to be fairly conservative given a pig’s gestation period and seasonal changes in food availability for pigs (e.g during droughts, they woud be more likely to take the bait). The time frame is flexible to take into consideration uncontrollable events such as El Niño, which would increase monitoring difficulties and likelihood of survival and reproduction of individual pigs. May 2001 Despite on-going intensive hunting pressure, no pigs have been shot since May 2000. Intensive monitoring through systematic baiting was begun in November 2000. Equidistant poisoned baits were twice applied to all pig habitats, (including areas considered borderline habitat – the vegetated arid zones of the island) of Santiago, thus covering a total of 38,344 hectares (66% of total land area) – only lava fields were excluded. Each bait was monitored approximately two weeks after having been placed. These efforts led to the detection of pig sign in late November 2000; one pig was subsequently found dead two weeks later, near the disturbed bait. No other pigs have been detected to date. Should the on-going monitoring efforts reveal no further pig sign, the GNPS would declare Santiago island pig free in May 2002 – the first time in at least 127 years, and the largest ever island from which pigs will have been successfully eradicated. Observations The Santiago island pig eradication, though not formally completed, illustrates a series of elements of value for those wishing to contemplate similar projects elsewhere. 1. During start up, much is accomplished with relatively little effort. Initial removal rates are very high and relatively little effort is required to remove pigs (see fig. 2). During this stage, staff is enthusiastic and institutional support strong. 2. As pig numbers drop, hunting effort per pig increases. Typical outings result in far fewer pigs removed. Over time, no definite trend towards fewer pigs removed is evident (1990 – 1996), resulting in a realization that on-going control or eradication will require an increased and sustained effort. Institutional support may falter as there is “no end in sight” of the pig control effort and this may trigger a drop in morale by the project leaders. There is a risk of projet abandonment at this stage. Fig. 3: SPOT image of Santiago island showing: a) a sample of georeferenced 500m equidistant points, for placement of poisoned baits; b) trail system (black) and c) 4th of October 1998 tracking of a hunter on a day’s outing. 3. Project proponents determine realistic levels of hunting effort required for eradication. Institutional support is secured for a protracted effort. 4. Hunter effort increased (1995-present): Recognizing that the effort required to remove the last individuals is far greater than at the outset (fig. 2), the institutions involved allocate the necessary resources for completion of the eradication campaign. Crucial institutional support is maintained despite extremely high effort required / pig killed (roughly 450 times increase over 1988). Acknowledgements I would like to thank Gonzalo Banda and Timo Koester, self-taught Isabela Project GIS techies, for organizing and processing field data. The success of the Santiago pig eradication campaign is largely due to the indefatigable efforts of Felipe Cruz and the professional hunters of the Isabela Project, led by Roberto Ballesteros and Wilson Cabrera. Additional thanks is due to Dr. Howard Snell of the Charles Darwin Research Station for supporting the Isabela Project in its efforts to adapt GIS and GPS technologies to field requirements. References: Cayot, L.J., H.L. Snell, W. Llerena, and H.M. H.M. Snell. 1994. Conservation biology of Galapagos reptiles: twenty-five years of successful research and management, p. 297- 305. In J.B. Murphy, IK.Adler and J.T. Collins (eds.), captive Management and Conservation of Amphibians and Reptiles. Society for the Study of Amphibians and Reptiles, Ithaca (New York). Contributions to Herpetology, volume 11. Coblentz, B.E. and D. W. Baber. 1987. Biology and control of feral pigs on isla Santiago, Galapagos, Ecuador. Journal of Applied Ecology. 24, 403-418. Enriquez, E.C. 1984. Aspectos importantes del ciclo de vida de la tortuga de galapagos Geochelone elephantopus darwini de la isla Santiago. Tesis doctoral, Universidad Central del Ecuador. Green, D., F. Ortiz-Crespo. 1982. Status of sea turtle populations in the central eastern Pacific. In K. Bjornadal (ed.), Biology and Conservation of Sea Turtles, pp. 221-233, Smithsonian Institution Press, Washington, D.C. Harwood, P., K. Campbell, M. Patry y F. Cruz. 2000. Residual toxicity of warfarin on pigs. Internal report. Proyecto Isabela, Charles Darwin Research Station, Galapagos. Isabela Project. 2000. Santiago island pig eradication field reports, 1989-97. Galapagos National Park Service, Puerto Ayora, Galapagos. MacFarland, C.G., Villa, J. and B. Toro. 1974a. The Galapagos giant tortoises (Geochelone elephantopus) Part I: Status of Surviving Populations . Biological Conservation 6(3): 198-212. MacFarland, C.G., Villa, J. and B. Toro. 1974b. The Galapagos giant tortoises (Geochelone elephantopus) Part II: Conservation Methods. Biological Conservation 6(3): 198-212. Marques, C., Cayot, L. and S. Rea. 1999. La crianza de tortugas gigantes en cautiverio: un manual operativo. Fundación Charles Darwin, Quito, Ecuador. Merlen, G. 1999. Restoring the tortoise dynasty. Charles Darwin Foundation, Quito. Myers, J.H., D. Simberloff, A.M. Kuris and J.R. Carey. 2000. Eradication revisited: dealing with exotic species. TREE 15(8). Snell, H.I., Snell, H.M. and C.R. Tracy. 1984. Variation among populations of Galapagos land iguanas (Conolophus): contrasts of phylogeny and ecology. In: Evolution in the Galapagos Islands. Edited by R.J. Berry. Academy Press. London, pp. 185-207.