Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

National Parks in Northern Sweden as Refuges for Illegal Killing of Large Carnivores

National Parks in Northern Sweden as Refuges for Illegal Killing of Large Carnivores Brown bear; Eurasian lynx; management; Large protected areas are often considered to be as important as population poaching; survival; land sparing; wolverine. size in reducing extinction risk for large carnivores. However, the effectiveness Correspondence of protected areas for large carnivore survival has rarely been tested where Geir Rune Rauset, Department of Ecology, surrounding areas also provide suitable habitats. Using individual-based long- Swedish University of Agricultural Science, term data, we here show that three species of large carnivores all suffered Grimso¨ Wildlife Research Station, SE-730 91 higher risk of illegal killing inside three large national parks than in surround- Riddarhyttan, Sweden. ing unprotected areas in northern Sweden. We suggest that this illegal killing Tel: +46722210788; Fax +46581697310. is the result of low enforcement and public presence in these remote parks, E-mail: geir.rauset@slu.se which results in a low probability for poachers to be discovered. Our results Received demonstrate that size of protected areas alone may be a poor predictor of their 5 October 2015 conservation value for large carnivores. We warn against passive national park Accepted management and advocate considering the ecological and socioeconomic con- 18 December 2015 text present inside as well as outside protected areas. Editor Reed Noss doi: 10.1111/conl.12226 are especially vulnerable to human persecution, due to Introduction their high trophic levels, low population densities, and Originally, national parks were created to preserve mon- slow life histories (Purvis et al. 2000; Cardillo et al. 2004; uments and wonders of nature. Today, together with Ripple et al. 2014). other forms of protected areas (PAs), they are one of Woodroffe & Ginsberg (1998) analyzed historical data the most important tools in biodiversity conservation and suggested that “critical reserve size” could predict lo- (Margules & Pressey 2000; Hoffmann et al. 2010). The cal large carnivore population extinctions. This perspec- conservation success of PAs has spurred scientific ef- tive has had a great influence and, consequently, the fort into designing representative and persistent reserves. “land sparing” policy of securing large PAs and reduc- Still, areas with grand scenery and wilderness are often ing negative edge effects has become a contemporary prioritized for protection, as they usually include remote paradigm for conservation of large carnivores. In line and rugged areas, with little human use, and thus are of with this perspective, many African parks are protected lower economic value (Pressey 1994; Soule & Sanjayan islands that increase large carnivore survival in other- 1998; Margules & Pressey 2000). Nonetheless, large and wise human-influenced surroundings, but also experi- remote PAs can have substantial conservation value as ence substantial edge effects in terms of anthropogenic they often function as refuges for large carnivores, which mortality (Loveridge et al. 2007; Kiffner et al. 2009; Balme 334 Conservation Letters, September/October 2016, 9(5), 334–341 Copyright and Photocopying: 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. G. R. Rauset et al. Protected areas and poaching et al. 2010). In North America, gray wolf (Canis lupus) Stora Sjofallet, ¨ and Padjelanta; Figure 1) by studying spa- mortality risk was lower in the Greater Yellowstone Area tial variation in risk of illegal mortality inside these parks and Central Idaho, relative to northwest Montana, likely as well as in surrounding unprotected areas. due to larger core areas that offered stronger wolf protec- tion (Smith et al. 2010). Johnson et al. (2004) also found Materials and methods higher mortality risk for grizzly bears (Ursus arctos)inar- eas with high road densities outside Yellowstone National Study area Park. However, a long-term concern in conservation is Sarek and Stora Sjofallet were among the first national that some PAs appear on the map, but have little con- parks created in Europe in 1909, chosen for their scenic servation value (i.e., “paper parks”; Di Minin & Toivonen and recreational value on public land of low commer- 2015). The major limitations to effective management of cial value, but also established to protect the then crit- PAs arise from a lack of financial resources or deficiencies ically endangered Scandinavian brown bear population of management (Di Minin & Toivonen 2015), and such (Swenson et al. 1994). The area also contained most of conditions are especially evident during times of govern- the remnant Swedish wolverines during a bottleneck in mental and social instability (e.g., civil wars; Plumptre the mid-1900s (Haglund 1965). Together with other PAs, et al. 2007). In the case of Wolong National Reserve for these national parks and Padjelanta National Park form Giant Pandas, China, an unexpected negative effect fol- the 9,400 km Laponia UNESCO World Heritage Site, lowed the founding of the PA; habitat loss as the result of one of the largest PA networks in Europe. The justifica- increased economic activity (Liu et al. 2001). tion for World Heritage designation included its natural An alternative perspective is “land sharing” (Fischer qualities (criteria vii, viii, and ix, UNESCO Ref:774) and et al. 2014), where large carnivores coexist with human the indigenous Sami ´ reindeer herding culture (criteria iii activities in the landscape (Chapron et al. 2014). It is and v). The study area provides important spring-to-fall based on the fact that many large carnivore populations grazing areas and spring calving grounds for semidomes- in North America and Europe have been generally sta- tic reindeer both inside and outside the national parks. ble or increasing throughout recent decades, despite high Snowmobile access during winter is not allowed for the human population densities. This perspective advocates general public inside the parks, but is allowed for rein- that large carnivore conservation is possible at high hu- deer herders and provincial rangers (Swedish Environ- man densities when management is favorable and sta- mental Protection Agency, NFS 2013:10). The surround- ble political institutions ensure proper law enforcement ing areas experience substantial recreational snowmobile (Linnell et al. 2001). Sweden is a case in point; eradica- activity. There is limited human infrastructure and agri- tion programs exterminated or reduced large carnivore culture both inside and outside the national parks (road populations to very low numbers during the 1800s and density 0.017 km road/km inside the parks vs. 0.15 out- 1900s (Haglund 1965; Swenson et al. 1994; Linnell et al. side, human settlement and agriculture, 0.0044% of the 2001). Today, after decades of favorable management parks’ area vs. 0.024% outside). Large game hunting is policies, including protection in large national parks, the not allowed inside the parks. populations of brown bear (Ursus arctos), Eurasian lynx The study area (48,536 km ) contained a vegetation (Lynx lynx), and wolverine (Gulo gulo) have recovered and gradient from Scots pine (Pinus sylvestris)and Norway are now widely distributed in multiuse landscapes, of- spruce (Picea abies) forests starting at about 200 m a.s.l. ten on privately owned land outside PAs (Chapron et al. via mountain birch (Betula pubescens) forests, heaths, and 2014). In northern Sweden, the main large prey of large grass to mountain peaks and plateaus of bare rock and carnivores is semidomestic reindeer (Rangifer tarandus) glaciers above 2,000 m. The climate was continental and (i.e., private property), which has created a conflict be- the ground was usually snow covered from November tween large carnivore conservation and the indigenous until May. Sami ´ reindeer husbandry (Swenson & Andren ´ 2005; Mattisson et al. 2011; Hobbs et al. 2012). To mitigate the economic impacts and ensure carnivore persistence, Swe- Data den implemented a conservation performance payment system for large carnivores, combined with intensive pop- We used individual-based demographic and telemetry lo- ulation monitoring (Zabel & Holm-Muller 2008; Persson cation data from radio-marked brown bears, Eurasian et al. 2015). lynx, and wolverines from 1984 to 2010. In total, the We have evaluated the protection from persecu- data included 455 large carnivore individuals, moni- tion provided to the three large mammalian carnivores tored over 132,119 radio-days (Table S1). For details on (brown bears, Eurasian lynx, and wolverines) present in capture and monitoring, see Supplementary Meth- the largest national parks in northern Sweden (Sarek, ods (online Supporting Information). Individuals were Conservation Letters, September/October 2016, 9(5), 334–341 Copyright and Photocopying: 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. 335 Protected areas and poaching G. R. Rauset et al. Figure 1 Location of the study area, including the national parks within it (gray shading), for brown bears, Eurasian lynx, and wolverines in northern Sweden, 1984–2010 (left); with animal locations (upper right; red = brown bears, black = lynx, blue = wolverines), vegetation and road distribution (lower right) in relation to the study area and national park borders. classified by age group (juveniles, subadults, and adults) Risk modeling and sex. Cause of death was established, based on a To estimate the risk of mortality among individuals of list of predefined methods and criteria (Supplementary the three species, we applied the Andersen–Gill model, Methods; Andren ´ et al. 2006; Bischof et al. 2009; Persson a counting process formulation of the Cox proportional et al. 2009), and the animals’ fates were grouped into the hazards model (Andersen & Gill 1982). The Andersen– classes “censored/alive,” “illegally killed,” or “other mor- Gill model allows for left (i.e., staggered entry) and right tality” (Table S2). censoring of data (i.e., animals lost for follow-up or out- Each telemetry location was assigned several measures living the study), time-varying continuous and categori- of habitat features (vegetation class, forest, terrain steep- cal variables, multiple events (i.e., fate of animals), and ness, and elevation) as well as human land use (national discontinuous time intervals of risk (Johnson et al. 2004; parks and reindeer calving grounds). Locations were also Murray 2006). We treated each time interval between se- split by season (snow cover vs. no snow) and by telemetry quential locations as a unique interval of risk, which was method (very high frequency [VHF] or global positioning attributed different strata (i.e., sex and age of the individ- satellite [GPS]) (Table S3). For details on environmental ual, Table S3 b.) and environmental variables (i.e., forest data handling, see Supplementary Methods (online Sup- and terrain steepness Table S3 a.) belonging to the end porting Information). 336 Conservation Letters, September/October 2016, 9(5), 334–341 Copyright and Photocopying: 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. G. R. Rauset et al. Protected areas and poaching Table 1 The effect of national parks in Andersen–Gill models of the risk of illegal mortality, the risk of mortality from other sources, and the total mortality among radio-marked brown bears, Eurasian lynx, and wolverines in northern Sweden, 1984–2010. Parameter estimates (β ± SE), effect size (exp(β)), and Akaike’s information criterion relative to the null models (AIC) a. Brown bear Mortality source β* ± SE exp(β) AIC Illegal 1.013 ± 0.456 2.75 −2.9 Other −0.458 ± 0.467 0.63 1.0 Total mortality 0.232 ± 0.307 1.26 1.4 b. Lynx Mortality source β ± SE exp(β) AIC Figure 2 Monthly distribution of illegally killed radio-marked brown bears, Illegal 0.925 ± 0.463 2.52 −2.3 Eurasian lynx, and wolverines in the study area in northern Sweden, 1984– Other −0.310 ± 0.716 0.73 1.8 Total mortality 0.549 ± 0.365 1.73 −0.3 c. Wolverine Mortality source β ± SE exp(β) AIC point of the time interval (Johnson et al. 2004). Events Illegal 0.841 ± 0.364 2.32 −3.5 were coded as 1 for mortality and 0 for right censoring. Other −0.084 ± 0.310 0.92 1.9 We excluded monitoring intervals > 30 days, consider- Total mortality 0.308 ± 0.233 1.36 0.3 ing these as discontinuous time intervals (Johnson et al. Positive coefficient indicates increased risk of mortality inside national 2004). We added the additional number of days equal parks. to 40% of the mean monitoring interval when right- censoring data (Johnson 1979). We built models in R clusion of season (snow vs. no snow) improved model fit library Survival (Therneau 2011) separately for each of substantially for all species (Table S4). Wolverines were the three species, and we built separate hazard models especially vulnerable, with 3.2 times higher risk of mor- for “illegal killing” (risk of being killed illegally), “other tality during the snow season. Brown bears varied some- mortality” (risk of mortality from other causes than ille- what from this general trend, as they also suffered il- gal killing), and “total mortality” (for details on mortal- legal killing in September during the regular bear and ity cause verifications, see Supplementary Methods). We moose (Alces alces) hunting season (Figure 2). Risk mod- stratified all models by sex, age class, season, and teleme- els stratified by sex also performed substantially better try method (VHF vs. GPS data, Table S3 b.), allowing for (Table S4), though the effect of sex differed substantially the calculation of separate baseline hazards for each stra- between species; wolverine females suffered 0.39 the risk tum. Our analysis at three spatial scales (i.e., point value, of illegal mortality relative to males, whereas this differ- 3,000 m buffer, and 6,000 m buffer, see Supplemen- ence was weaker in lynx (0.63) and brown bears (0.78). tary Methods) yielded comparable results (Table S5), thus The risk of illegal mortality was higher inside national we chose to keep the models with a buffer distance of parks for radio-collared individuals of all three species 3,000 m in our final analyses. For the best models, we (i.e., positive parameter estimates; Table 1). The mag- tested the model assumption of proportional hazards and nitude of this effect was consistent across all species, plotted Schoenfeld residuals along with the smoothed with 2.3–2.8 times higher risk inside the parks. However, hazard curves as well as Martingale residuals (Murray when correcting for varying habitat composition inside 2006). We assessed relative model fit by Akaike’s infor- and outside the parks (Table S6a), this effect was reduced mation criterion (AIC) and calculated model averaged pa- for lynx where the lower amount of forest inside the rameter estimates and variable relative importance (VRI) parks contributed more to the higher risk than the park among the models with AIC < 4 to the top models effect per se (Forest: VRI = 1.0, β =−0.09 ± 0.04 [SE]). (Burnham & Anderson 2002). Brown bear risk also interacted with forest availability where the nonforested areas inside the parks represented Results high-risk areas (NP Forest: VRI = 0.84, β =−0.21 ± 0.10 Illegal killing was an important source of mortality for all [SE]). For wolverines, national parks as well as terrain three large carnivore species (42–69% of total mortality, steepness contributed the most to risk (steep: VRI = 1.0, Table S2). We found an increased mortality risk for all β =−0.15 ± 0.07 [SE]). The presence of reindeer calv- species during the late snow season (Figure 2). This pat- ing grounds was included in several top ranking models. tern was also reflected in the risk models, where the in- However, model averaged parameters of reindeer calving Conservation Letters, September/October 2016, 9(5), 334–341 Copyright and Photocopying: 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. 337 Protected areas and poaching G. R. Rauset et al. grounds in all three species were associated with large er- conservation value represented by these parks therefore ror estimates, and no general conclusions could be drawn. has changed substantially since their establishment, not Common proxies of human disturbance, such as distance due to human development and habitat depletion (Liu to human infrastructure or permanent human activity, et al. 2001), but due primarily to improved poacher mo- had no predicative power on mortality risk at local or re- bility with modern snowmobiles. Thus, critical PA sizes gional scales in our study (Supplementary Methods). built on historical data may be a poor predictor of future The risk of mortality from other causes than illegal carnivore persistence. In fact, the network of national killing (“other mortality”) was not affected by national parks in our study is considerably larger than a suggested parks (negative risk parameters, but with relatively large PA size for grizzly bears in North America (Woodroffe & SEs; Table 1); thus, the risk of all sources of mortality (“to- Ginsberg 1998). tal mortality”) exhibited a weaker national parks effect Efficient law enforcement is necessary for all natural (positive risk parameters, but with relatively large SEs; resource management (Keane et al. 2008). However, law Table 1) than risk of illegal mortality alone. For wolver- enforcement measures often lead to the criticism that PAs ines, steep areas represented overall a lower risk of mor- are created by and for an elitist few, in conflict with local tality, and for brown bear and wolverines, there was a community interests (Wittemyer et al. 2008). Therefore, tendency for reindeer calving grounds to represent higher measures promoting acceptance of large carnivores by lo- overall risk (Table S6c). cal people often are of crucial importance for successful conservation programs (Treves & Karanth 2003). This is especially important in Laponia and similar areas hav- ing dual conservation goals, ensuring the conservation of Discussion native biodiversity and indigenous cultural heritage and Our results clearly show that the risk of illegal mortal- livelihood. Such measures include incentives to compen- ity was higher inside the national parks than it was in sate local costs of carnivores, such as direct compensation the surrounding and unprotected areas for the animals in of losses or conservation performance payment systems, our study. We also found an increased mortality risk for depredation prevention measures, controlled legal har- all species during the late snow season, potentially due vest, and by promoting activities giving carnivores a di- to increased human accessibility on snowmobiles dur- rect value (Treves & Karanth 2003; Dickman et al. 2011). ing this period, because of increased day length and in- The Swedish conservation performance payment system creased suitability of the snow for snowmobiling (Andren ´ has been instrumental in the recovery of wolverines in et al. 2006; Persson et al. 2009). Habitat factors that im- Sweden, by providing protection for adult female wolver- pair snowmobile use, steep terrain and forest, are gener- ines through a combination of direct monetary value ally selected by large carnivores in the area (Rauset et al. and indirect protection by monitoring denning activities 2013), and similarly were important factors reducing il- (Persson et al. 2015). The payment system for lynx and legal mortality. Permanent human activity, represented brown bears is not linked to denning females; lynx are by proximity to human infrastructure or agriculture, had paid per family group, based on tracks observed in the no explanatory power for illegal mortality, further sup- winter following reproductive events, whereas payment for bears is based on occurrence, regardless of density and porting our conclusion that large carnivores were killed reproduction (Zabel & Holm-Muller 2008). The greater mainly from snowmobiles capable of traversing through- out the parks. sex differences in risk of illegal mortality for wolverines We conclude that the national parks in our study had than lynx or brown bears thus suggest that there is a a negative effect on large carnivore conservation, con- link between payment system and risk of illegal mortal- trary to the general conservation paradigm and the his- ity, with a lesser inclination to kill individuals that the torical intention of the parks as a refuge for the brown payment system makes most “profitable.” Nonetheless, bear (Swenson et al. 1994). That the national parks seem- our results show that rates of illegal killing of large carni- ingly provided increased opportunities for illegally killing vores are still high, especially when poachers face a low carnivores was, however, not an effect of the parks per risk of detection. Hence, economic incentives may partly se, but we suggest that it was rather a result of snowmo- change poaching behavior, but are not a panacea (Treves bile restrictions for the general public. These restrictions & Bruskotter 2014). On the other hand, our results sug- were intended to limit human disturbance of wildlife and gest that presence by the general public has the poten- reindeer inside the parks. However, the combination of tial to reduce illegal killing of large carnivores. Poaching low public presence and low intensity of patrolling by does not happen in a social vacuum, and the general pub- provincial rangers in these large, remote areas also im- lic’s tolerance to carnivores, and to illegal killing of such, plied that poachers faced a low risk of detection. The strongly affect poaching behavior. 338 Conservation Letters, September/October 2016, 9(5), 334–341 Copyright and Photocopying: 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. G. R. Rauset et al. Protected areas and poaching Our results emphasize the importance of critically eval- Table S4. The effects of strata (refer to Table S3b.) on uating the conservation performance of PAs, even large risk of being killed illegally for brown bears, Eurasian and remote national parks in a highly developed country, lynx, and wolverines in northern Sweden, 1984– and considering the ecological and socioeconomic context 2010. Parameter estimates in Andersen–Gill models present inside as well as outside them. We caution against are (β ± SE), and exp(β) represents the effect size, passive management and high public expectations of pos- i.e., the difference in risk of group 1 relative to group itive effects of PAs on large carnivore conservation, based 0 of Sex (males=0/females=1), Season (snow=0/no on the assumption that sparing land is enough. snow=1), Age01 (juvenile=0/subadult = 1, Age02 (juvenile=0/adult =1), Age12 (subadult=0/adult =1), VHF(GPSdata=0/VHF=1). The AIC represents Acknowledgments the change in AIC when adding the stratum to the We thank Jon Arnemo for veterinary assistance in the nonstratified model. field, and numerous field workers for data collection. The Table S5. The effect of national parks on the risk of being killed illegally for radio-marked a) brown bears, b) study was funded by the Swedish Environmental Protec- tion Agency, the Norwegian Environment Agency, the Eurasian lynx, and c) wolverines in northern Sweden, World Wide Fund for Nature (Sweden), the Swedish Uni- 1984–2010, at three different scales (animal locations and versity of Agricultural Sciences, the Research Council of buffers of 3,000 m and 6,000 m around the locations). Norway, the European Association of Zoos and Aquaria Parameter estimates (β ± SE), effect size (exp(β)), and (EAZA), and the Swedish Research Council for Environ- model AIC relative to the best models (AIC) are for ment, Agricultural Sciences and Spatial Planning (FOR- Andersen–Gill models. MAS). All handling of animals were approved by the Table S6. Environmental variables influencing a) risk Animal Ethics Committee for northern Sweden, Umea. ˚ of illegal mortality, b) risk of mortality other than illegal Data and models reported in the article are available from killing, and c) total risk of mortality of large carnivores the corresponding author upon request. The authors de- in northern Sweden, 1984–2010. Parameter estimates clare no competing financial interests. Authorship: GRR, (mean β ± SE) and the Variable Relative Importance HA, JP, GS, and PS designed the project; all authors con- (VRI) are model-averaged parameters of Andersen–Gill tributed to the collection and preparation of data; GRR models with AIC < 4 relative to the best models. The analyzed the data and wrote the article with the help of set of candidate models included all combinations of the JES, JP, HA, AZ, and GS. environmental variables; “national park” (NP), “reindeer calving ground” (Calf), “steep terrain” (Steep), and “Forest “ and their two-way interactions. NP and Calf Supporting Information are binary variables, whereas Steep and Forest represent the area (range 0 to 28.3 km ) within a 3,000-m buffer Additional Supporting Information may be found in the around the location (refer to Table S3). The effects of the online version of this article at the publisher’s web site: strata (refer to Table S4) were corrected for and therefore Table S1. Summary of data used for survival analyses always included. of radio-marked brown bears, Eurasian lynx, and wolver- ines in northern Sweden, 1984–2010. Table S2. Cause-specific mortality of radio-marked References brown bears, Eurasian lynx, and wolverines in northern Andersen, P.K. & Gill, R.D. (1982). Cox’s regression model for Sweden, 1984–2010. counting processes: a large sample study. Ann. Stat. 10, Table S3. a) Environmental variables used and b) 1100-1120. strata included and corrected for in the survival analysis Andren, ´ H., Linnell, J.D.C., Liberg, O., et al. (2006). Survival of large carnivores in northern Sweden, 1984–2010. SMD rates and causes of mortality in Eurasian lynx (Lynx lynx) vegcode refers to the Swedish Land Cover vegetation map in multi-use landscapes. Biol. Conserv. 131, 23-32. from the National Land Survey of Sweden. “Definition” Balme, G.A., Slotow, R. & Hunter, L.T.B. (2010). Edge effects describes how the variables were estimated; “area” rep- and the impact of non-protected areas in carnivore resents the area within a buffer of either 3,000 m or conservation: leopards in the Phinda-Mkhuze Complex, 6,000 m around a location, “distance” represents the South Africa. Anim. Conserv. 13, 315-323. shortest distance between a location and a specific habi- Bischof, R., Swenson, J.E., Yoccoz, N.G., Mysterud, A. & tat; “binary” represents whether a location is within or Gimenez, O. (2009). The magnitude and selectivity of outside a specific habitat or whether a buffer contained natural and multiple anthropogenic mortality causes in the specific habitat or not. hunted brown bears. J. Anim. Ecol. 78, 656-665. Conservation Letters, September/October 2016, 9(5), 334–341 Copyright and Photocopying: 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. 339 Protected areas and poaching G. R. Rauset et al. Burnham, K.P. & Anderson, D.R. (2002). Model selection and Margules, C.R. & Pressey, R.L. (2000). Systematic multimodel inference, 2nd ed. Springer-Verlag, New York. conservation planning. Nature 405, 243-253. Cardillo, M., Purvis, A., Sechrest, W., Gittleman, J.L., Bielby, Mattisson, J., Odden, J., Nilsen, E.B., Linnell, J.D.C., Persson, J. & Mace, G.M. (2004). Human population density and J. & Andren, ´ H. (2011). Factors affecting Eurasian lynx kill extinction risk in the world’s carnivores. PLoS Biol. 2, rates on semi-domestic reindeer in northern Scandinavia: 909-914. can ecological research contribute to the development of a Chapron, G., Kaczensky, P., Linnell, J.D.C., et al. (2014). fair compensation system? Biol. Conserv. 144, 3009- Recovery of large carnivores in Europe’s modern 3017. human-dominated landscapes. Science 346, 1517-1519. Murray, D.L. (2006). On improving telemetry-based survival Di Minin, E. & Toivonen, T. (2015). Global protected area estimation. J. Wildlife Manage. 70, 1530-1543. expansion: creating more than paper parks. Bioscience 65, Persson, J., Ericsson, G. & Segerstrom, ¨ P. (2009). Human 637-638. caused mortality in the endangered Scandinavian Dickman, A.J., Macdonald, E.A. & Macdonald, D.W. (2011). wolverine population. Biol. Conserv. 142, 325-331. A review of financial instruments to pay for predator Persson, J., Rauset, G.R. & Chapron, G. (2015). Paying for an conservation and encourage human-carnivore coexistence. endangered predator leads to population recovery. Conserv. Proc. Natl. Acad. Sci. U.S.A. 108, 13937-13944. Lett., 8, 345-350. Fischer, J., Abson, D.J., Butsic, V., et al. (2014). Land sparing Plumptre, A.J., Kujirakwinja, D., Treves, A., Owiunji, I. & versus land sharing: moving forward. Conserv. Lett. 7, Rainer, H. (2007). Transboundary conservation in the 149-157. greater Virunga landscape: its importance for landscape Haglund, B. (1965). Jarv ¨ och varg.P.A.Norstedt&Soners ¨ species. Biol. Conserv. 134, 279-287. Forlag, ¨ Stockholm. (Wolverine and wolf; In Swedish) Pressey, R.L. (1994). Ad hoc reservations: forward or Hobbs, N.T., Andren, ´ H., Persson, J., Aronsson, M. & backward steps in developing representative reserve Chapron, G. (2012). Native predators reduce harvest of systems. Conserv. Biol. 8, 662-668. reindeer by Sami pastoralists. Ecol. Appl. 22, 1640-1654. Purvis, A., Gittleman, J.L., Cowlishaw, G. & Mace G.M. Hoffmann, M., Hilton-Taylor, C., Angulo, A., et al. (2010). (2000). Predicting extinction risk in declining species. Proc. The impact of conservation on the status of the world’s Biol. Sci. 267, 1947-1952. vertebrates. Science 330, 1503-1509. Rauset, G., Mattisson, J., Andren, ´ H., Chapron, G. & Persson, Johnson, C.J., Boyce, M.S., Schwartz, C.C. & Haroldson, M.A. J. (2013). When species’ ranges meet: assessing differences (2004). Modeling survival: application of the Andersen-Gill in habitat selection between sympatric large carnivores. model to yellowstone grizzly bears. J. Wildlife Manage. 68, Oecologia 172, 701-711. 966-978. Ripple, W.J., Estes, J.A., Beschta, R.L., et al. (2014). Status Johnson, D.H. (1979). Estimating nest success: the Mayfield and ecological effects of the world’s largest carnivores. method and an alternative. Auk 96, 651-661. Science 343 (6167), 1241484. Keane, A., Jones, J.P.G., Edwards-Jones, G. & Smith, D.W., Bangs, E.E., Oakleaf, J.K. et al. (2010). Survival Milner-Gulland, E.J. (2008). The sleeping policeman: of colonizing wolves in the northern Rocky Mountains of understanding issues of enforcement and compliance in the United States, 1982–2004. J. Wildlife Manage. 74, conservation. Anim. Conserv. 11, 75-82. 620-634. Kiffner, C., Meyer, B., Muehlenberg, M. & Waltert, M. Soule, M.E. & Sanjayan, M.A. (1998). Ecology - conservation (2009). Plenty of prey, few predators: what limits lions targets: do they help? Science 279, 2060-2061. Panthera leo in Katavi National Park, western Tanzania? Swenson, J.E. & Andren, ´ H. (2005). A tale of two countries: Oryx 43, 52-59. large carnivore depredation and compensation schemes in Linnell, J.D.C., Swenson, J.E. & Andersen, R. (2001). Sweden and Norway. Pages 323-339 in R. Woodroffe, S. Predators and people: conservation of large carnivores is Thirgood, A. Rabinowitz, editors. People and wildlife: conflict possible at high human densities if management policy is or coexistence? Cambridge University Press, New York. favourable. Anim. Conserv. 4, 345-349. Swenson, J.E., Sandegren, F., Bjarvall, ¨ A., Soderberg, ¨ A., Liu, J.G., Linderman, M., Ouyang, Z.Y., An, L., Yang, J. & Wabakken, P. & Franzen ´ R. (1994). Size, trend, Zhang, H.M. (2001). Ecological degradation in protected distribution and conservation of the brown bear areas: the case of Wolong Nature Reserve for giant pandas. Ursus-Arctos population in Sweden. Biol. Conserv. 70, 9-17. Science 292, 98-101. Therneau, T. (2011). Survival analysis, including penalised Loveridge, A.J., Searle, A.W., Murindagomo, F. & Macdonald, likelihood. R package’Survival. version 2.38, URL: “http:// D.W. (2007). The impact of sport-hunting on the CRAN.R-project.org/package=survival” population dynamics of an African lion population in a Treves, A. & Bruskotter, J. (2014). Tolerance for predatory protected area. Biol. Conserv. 134, 548-558. wildlife. Science 344, 476-477. 340 Conservation Letters, September/October 2016, 9(5), 334–341 Copyright and Photocopying: 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. G. R. Rauset et al. Protected areas and poaching Treves, A. & Karanth, K.U. (2003). Human-carnivore conflict Woodroffe, R. & Ginsberg, J.R. (1998). Edge effects and the and perspectives on carnivore management worldwide. extinction of populations inside protected areas. Science Conserv. Biol. 17, 1491-1499. 280, 2126-2128. Wittemyer, G., Elsen, P., Bean, W.T., Burton, A.C.O. & Zabel, A. & Holm-Muller, K. (2008). Conservation Brashares, J.S. (2008). Accelerated human population performance payments for carnivore conservation in growth at protected area edges. Science 321, 123-126. Sweden. Conserv. Biol. 22, 247-251. Conservation Letters, September/October 2016, 9(5), 334–341 Copyright and Photocopying: 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. 341 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Conservation Letters Wiley

National Parks in Northern Sweden as Refuges for Illegal Killing of Large Carnivores

Download PDF
8 pages

Loading next page...
 
/lp/wiley/national-parks-in-northern-sweden-as-refuges-for-illegal-killing-of-MNWEdXBExW

References (46)

Publisher
Wiley
Copyright
Copyright © 2016 Wiley Periodicals, Inc.
ISSN
1755-263X
eISSN
1755-263X
DOI
10.1111/conl.12226
Publisher site
See Article on Publisher Site

Abstract

Brown bear; Eurasian lynx; management; Large protected areas are often considered to be as important as population poaching; survival; land sparing; wolverine. size in reducing extinction risk for large carnivores. However, the effectiveness Correspondence of protected areas for large carnivore survival has rarely been tested where Geir Rune Rauset, Department of Ecology, surrounding areas also provide suitable habitats. Using individual-based long- Swedish University of Agricultural Science, term data, we here show that three species of large carnivores all suffered Grimso¨ Wildlife Research Station, SE-730 91 higher risk of illegal killing inside three large national parks than in surround- Riddarhyttan, Sweden. ing unprotected areas in northern Sweden. We suggest that this illegal killing Tel: +46722210788; Fax +46581697310. is the result of low enforcement and public presence in these remote parks, E-mail: geir.rauset@slu.se which results in a low probability for poachers to be discovered. Our results Received demonstrate that size of protected areas alone may be a poor predictor of their 5 October 2015 conservation value for large carnivores. We warn against passive national park Accepted management and advocate considering the ecological and socioeconomic con- 18 December 2015 text present inside as well as outside protected areas. Editor Reed Noss doi: 10.1111/conl.12226 are especially vulnerable to human persecution, due to Introduction their high trophic levels, low population densities, and Originally, national parks were created to preserve mon- slow life histories (Purvis et al. 2000; Cardillo et al. 2004; uments and wonders of nature. Today, together with Ripple et al. 2014). other forms of protected areas (PAs), they are one of Woodroffe & Ginsberg (1998) analyzed historical data the most important tools in biodiversity conservation and suggested that “critical reserve size” could predict lo- (Margules & Pressey 2000; Hoffmann et al. 2010). The cal large carnivore population extinctions. This perspec- conservation success of PAs has spurred scientific ef- tive has had a great influence and, consequently, the fort into designing representative and persistent reserves. “land sparing” policy of securing large PAs and reduc- Still, areas with grand scenery and wilderness are often ing negative edge effects has become a contemporary prioritized for protection, as they usually include remote paradigm for conservation of large carnivores. In line and rugged areas, with little human use, and thus are of with this perspective, many African parks are protected lower economic value (Pressey 1994; Soule & Sanjayan islands that increase large carnivore survival in other- 1998; Margules & Pressey 2000). Nonetheless, large and wise human-influenced surroundings, but also experi- remote PAs can have substantial conservation value as ence substantial edge effects in terms of anthropogenic they often function as refuges for large carnivores, which mortality (Loveridge et al. 2007; Kiffner et al. 2009; Balme 334 Conservation Letters, September/October 2016, 9(5), 334–341 Copyright and Photocopying: 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. G. R. Rauset et al. Protected areas and poaching et al. 2010). In North America, gray wolf (Canis lupus) Stora Sjofallet, ¨ and Padjelanta; Figure 1) by studying spa- mortality risk was lower in the Greater Yellowstone Area tial variation in risk of illegal mortality inside these parks and Central Idaho, relative to northwest Montana, likely as well as in surrounding unprotected areas. due to larger core areas that offered stronger wolf protec- tion (Smith et al. 2010). Johnson et al. (2004) also found Materials and methods higher mortality risk for grizzly bears (Ursus arctos)inar- eas with high road densities outside Yellowstone National Study area Park. However, a long-term concern in conservation is Sarek and Stora Sjofallet were among the first national that some PAs appear on the map, but have little con- parks created in Europe in 1909, chosen for their scenic servation value (i.e., “paper parks”; Di Minin & Toivonen and recreational value on public land of low commer- 2015). The major limitations to effective management of cial value, but also established to protect the then crit- PAs arise from a lack of financial resources or deficiencies ically endangered Scandinavian brown bear population of management (Di Minin & Toivonen 2015), and such (Swenson et al. 1994). The area also contained most of conditions are especially evident during times of govern- the remnant Swedish wolverines during a bottleneck in mental and social instability (e.g., civil wars; Plumptre the mid-1900s (Haglund 1965). Together with other PAs, et al. 2007). In the case of Wolong National Reserve for these national parks and Padjelanta National Park form Giant Pandas, China, an unexpected negative effect fol- the 9,400 km Laponia UNESCO World Heritage Site, lowed the founding of the PA; habitat loss as the result of one of the largest PA networks in Europe. The justifica- increased economic activity (Liu et al. 2001). tion for World Heritage designation included its natural An alternative perspective is “land sharing” (Fischer qualities (criteria vii, viii, and ix, UNESCO Ref:774) and et al. 2014), where large carnivores coexist with human the indigenous Sami ´ reindeer herding culture (criteria iii activities in the landscape (Chapron et al. 2014). It is and v). The study area provides important spring-to-fall based on the fact that many large carnivore populations grazing areas and spring calving grounds for semidomes- in North America and Europe have been generally sta- tic reindeer both inside and outside the national parks. ble or increasing throughout recent decades, despite high Snowmobile access during winter is not allowed for the human population densities. This perspective advocates general public inside the parks, but is allowed for rein- that large carnivore conservation is possible at high hu- deer herders and provincial rangers (Swedish Environ- man densities when management is favorable and sta- mental Protection Agency, NFS 2013:10). The surround- ble political institutions ensure proper law enforcement ing areas experience substantial recreational snowmobile (Linnell et al. 2001). Sweden is a case in point; eradica- activity. There is limited human infrastructure and agri- tion programs exterminated or reduced large carnivore culture both inside and outside the national parks (road populations to very low numbers during the 1800s and density 0.017 km road/km inside the parks vs. 0.15 out- 1900s (Haglund 1965; Swenson et al. 1994; Linnell et al. side, human settlement and agriculture, 0.0044% of the 2001). Today, after decades of favorable management parks’ area vs. 0.024% outside). Large game hunting is policies, including protection in large national parks, the not allowed inside the parks. populations of brown bear (Ursus arctos), Eurasian lynx The study area (48,536 km ) contained a vegetation (Lynx lynx), and wolverine (Gulo gulo) have recovered and gradient from Scots pine (Pinus sylvestris)and Norway are now widely distributed in multiuse landscapes, of- spruce (Picea abies) forests starting at about 200 m a.s.l. ten on privately owned land outside PAs (Chapron et al. via mountain birch (Betula pubescens) forests, heaths, and 2014). In northern Sweden, the main large prey of large grass to mountain peaks and plateaus of bare rock and carnivores is semidomestic reindeer (Rangifer tarandus) glaciers above 2,000 m. The climate was continental and (i.e., private property), which has created a conflict be- the ground was usually snow covered from November tween large carnivore conservation and the indigenous until May. Sami ´ reindeer husbandry (Swenson & Andren ´ 2005; Mattisson et al. 2011; Hobbs et al. 2012). To mitigate the economic impacts and ensure carnivore persistence, Swe- Data den implemented a conservation performance payment system for large carnivores, combined with intensive pop- We used individual-based demographic and telemetry lo- ulation monitoring (Zabel & Holm-Muller 2008; Persson cation data from radio-marked brown bears, Eurasian et al. 2015). lynx, and wolverines from 1984 to 2010. In total, the We have evaluated the protection from persecu- data included 455 large carnivore individuals, moni- tion provided to the three large mammalian carnivores tored over 132,119 radio-days (Table S1). For details on (brown bears, Eurasian lynx, and wolverines) present in capture and monitoring, see Supplementary Meth- the largest national parks in northern Sweden (Sarek, ods (online Supporting Information). Individuals were Conservation Letters, September/October 2016, 9(5), 334–341 Copyright and Photocopying: 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. 335 Protected areas and poaching G. R. Rauset et al. Figure 1 Location of the study area, including the national parks within it (gray shading), for brown bears, Eurasian lynx, and wolverines in northern Sweden, 1984–2010 (left); with animal locations (upper right; red = brown bears, black = lynx, blue = wolverines), vegetation and road distribution (lower right) in relation to the study area and national park borders. classified by age group (juveniles, subadults, and adults) Risk modeling and sex. Cause of death was established, based on a To estimate the risk of mortality among individuals of list of predefined methods and criteria (Supplementary the three species, we applied the Andersen–Gill model, Methods; Andren ´ et al. 2006; Bischof et al. 2009; Persson a counting process formulation of the Cox proportional et al. 2009), and the animals’ fates were grouped into the hazards model (Andersen & Gill 1982). The Andersen– classes “censored/alive,” “illegally killed,” or “other mor- Gill model allows for left (i.e., staggered entry) and right tality” (Table S2). censoring of data (i.e., animals lost for follow-up or out- Each telemetry location was assigned several measures living the study), time-varying continuous and categori- of habitat features (vegetation class, forest, terrain steep- cal variables, multiple events (i.e., fate of animals), and ness, and elevation) as well as human land use (national discontinuous time intervals of risk (Johnson et al. 2004; parks and reindeer calving grounds). Locations were also Murray 2006). We treated each time interval between se- split by season (snow cover vs. no snow) and by telemetry quential locations as a unique interval of risk, which was method (very high frequency [VHF] or global positioning attributed different strata (i.e., sex and age of the individ- satellite [GPS]) (Table S3). For details on environmental ual, Table S3 b.) and environmental variables (i.e., forest data handling, see Supplementary Methods (online Sup- and terrain steepness Table S3 a.) belonging to the end porting Information). 336 Conservation Letters, September/October 2016, 9(5), 334–341 Copyright and Photocopying: 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. G. R. Rauset et al. Protected areas and poaching Table 1 The effect of national parks in Andersen–Gill models of the risk of illegal mortality, the risk of mortality from other sources, and the total mortality among radio-marked brown bears, Eurasian lynx, and wolverines in northern Sweden, 1984–2010. Parameter estimates (β ± SE), effect size (exp(β)), and Akaike’s information criterion relative to the null models (AIC) a. Brown bear Mortality source β* ± SE exp(β) AIC Illegal 1.013 ± 0.456 2.75 −2.9 Other −0.458 ± 0.467 0.63 1.0 Total mortality 0.232 ± 0.307 1.26 1.4 b. Lynx Mortality source β ± SE exp(β) AIC Figure 2 Monthly distribution of illegally killed radio-marked brown bears, Illegal 0.925 ± 0.463 2.52 −2.3 Eurasian lynx, and wolverines in the study area in northern Sweden, 1984– Other −0.310 ± 0.716 0.73 1.8 Total mortality 0.549 ± 0.365 1.73 −0.3 c. Wolverine Mortality source β ± SE exp(β) AIC point of the time interval (Johnson et al. 2004). Events Illegal 0.841 ± 0.364 2.32 −3.5 were coded as 1 for mortality and 0 for right censoring. Other −0.084 ± 0.310 0.92 1.9 We excluded monitoring intervals > 30 days, consider- Total mortality 0.308 ± 0.233 1.36 0.3 ing these as discontinuous time intervals (Johnson et al. Positive coefficient indicates increased risk of mortality inside national 2004). We added the additional number of days equal parks. to 40% of the mean monitoring interval when right- censoring data (Johnson 1979). We built models in R clusion of season (snow vs. no snow) improved model fit library Survival (Therneau 2011) separately for each of substantially for all species (Table S4). Wolverines were the three species, and we built separate hazard models especially vulnerable, with 3.2 times higher risk of mor- for “illegal killing” (risk of being killed illegally), “other tality during the snow season. Brown bears varied some- mortality” (risk of mortality from other causes than ille- what from this general trend, as they also suffered il- gal killing), and “total mortality” (for details on mortal- legal killing in September during the regular bear and ity cause verifications, see Supplementary Methods). We moose (Alces alces) hunting season (Figure 2). Risk mod- stratified all models by sex, age class, season, and teleme- els stratified by sex also performed substantially better try method (VHF vs. GPS data, Table S3 b.), allowing for (Table S4), though the effect of sex differed substantially the calculation of separate baseline hazards for each stra- between species; wolverine females suffered 0.39 the risk tum. Our analysis at three spatial scales (i.e., point value, of illegal mortality relative to males, whereas this differ- 3,000 m buffer, and 6,000 m buffer, see Supplemen- ence was weaker in lynx (0.63) and brown bears (0.78). tary Methods) yielded comparable results (Table S5), thus The risk of illegal mortality was higher inside national we chose to keep the models with a buffer distance of parks for radio-collared individuals of all three species 3,000 m in our final analyses. For the best models, we (i.e., positive parameter estimates; Table 1). The mag- tested the model assumption of proportional hazards and nitude of this effect was consistent across all species, plotted Schoenfeld residuals along with the smoothed with 2.3–2.8 times higher risk inside the parks. However, hazard curves as well as Martingale residuals (Murray when correcting for varying habitat composition inside 2006). We assessed relative model fit by Akaike’s infor- and outside the parks (Table S6a), this effect was reduced mation criterion (AIC) and calculated model averaged pa- for lynx where the lower amount of forest inside the rameter estimates and variable relative importance (VRI) parks contributed more to the higher risk than the park among the models with AIC < 4 to the top models effect per se (Forest: VRI = 1.0, β =−0.09 ± 0.04 [SE]). (Burnham & Anderson 2002). Brown bear risk also interacted with forest availability where the nonforested areas inside the parks represented Results high-risk areas (NP Forest: VRI = 0.84, β =−0.21 ± 0.10 Illegal killing was an important source of mortality for all [SE]). For wolverines, national parks as well as terrain three large carnivore species (42–69% of total mortality, steepness contributed the most to risk (steep: VRI = 1.0, Table S2). We found an increased mortality risk for all β =−0.15 ± 0.07 [SE]). The presence of reindeer calv- species during the late snow season (Figure 2). This pat- ing grounds was included in several top ranking models. tern was also reflected in the risk models, where the in- However, model averaged parameters of reindeer calving Conservation Letters, September/October 2016, 9(5), 334–341 Copyright and Photocopying: 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. 337 Protected areas and poaching G. R. Rauset et al. grounds in all three species were associated with large er- conservation value represented by these parks therefore ror estimates, and no general conclusions could be drawn. has changed substantially since their establishment, not Common proxies of human disturbance, such as distance due to human development and habitat depletion (Liu to human infrastructure or permanent human activity, et al. 2001), but due primarily to improved poacher mo- had no predicative power on mortality risk at local or re- bility with modern snowmobiles. Thus, critical PA sizes gional scales in our study (Supplementary Methods). built on historical data may be a poor predictor of future The risk of mortality from other causes than illegal carnivore persistence. In fact, the network of national killing (“other mortality”) was not affected by national parks in our study is considerably larger than a suggested parks (negative risk parameters, but with relatively large PA size for grizzly bears in North America (Woodroffe & SEs; Table 1); thus, the risk of all sources of mortality (“to- Ginsberg 1998). tal mortality”) exhibited a weaker national parks effect Efficient law enforcement is necessary for all natural (positive risk parameters, but with relatively large SEs; resource management (Keane et al. 2008). However, law Table 1) than risk of illegal mortality alone. For wolver- enforcement measures often lead to the criticism that PAs ines, steep areas represented overall a lower risk of mor- are created by and for an elitist few, in conflict with local tality, and for brown bear and wolverines, there was a community interests (Wittemyer et al. 2008). Therefore, tendency for reindeer calving grounds to represent higher measures promoting acceptance of large carnivores by lo- overall risk (Table S6c). cal people often are of crucial importance for successful conservation programs (Treves & Karanth 2003). This is especially important in Laponia and similar areas hav- ing dual conservation goals, ensuring the conservation of Discussion native biodiversity and indigenous cultural heritage and Our results clearly show that the risk of illegal mortal- livelihood. Such measures include incentives to compen- ity was higher inside the national parks than it was in sate local costs of carnivores, such as direct compensation the surrounding and unprotected areas for the animals in of losses or conservation performance payment systems, our study. We also found an increased mortality risk for depredation prevention measures, controlled legal har- all species during the late snow season, potentially due vest, and by promoting activities giving carnivores a di- to increased human accessibility on snowmobiles dur- rect value (Treves & Karanth 2003; Dickman et al. 2011). ing this period, because of increased day length and in- The Swedish conservation performance payment system creased suitability of the snow for snowmobiling (Andren ´ has been instrumental in the recovery of wolverines in et al. 2006; Persson et al. 2009). Habitat factors that im- Sweden, by providing protection for adult female wolver- pair snowmobile use, steep terrain and forest, are gener- ines through a combination of direct monetary value ally selected by large carnivores in the area (Rauset et al. and indirect protection by monitoring denning activities 2013), and similarly were important factors reducing il- (Persson et al. 2015). The payment system for lynx and legal mortality. Permanent human activity, represented brown bears is not linked to denning females; lynx are by proximity to human infrastructure or agriculture, had paid per family group, based on tracks observed in the no explanatory power for illegal mortality, further sup- winter following reproductive events, whereas payment for bears is based on occurrence, regardless of density and porting our conclusion that large carnivores were killed reproduction (Zabel & Holm-Muller 2008). The greater mainly from snowmobiles capable of traversing through- out the parks. sex differences in risk of illegal mortality for wolverines We conclude that the national parks in our study had than lynx or brown bears thus suggest that there is a a negative effect on large carnivore conservation, con- link between payment system and risk of illegal mortal- trary to the general conservation paradigm and the his- ity, with a lesser inclination to kill individuals that the torical intention of the parks as a refuge for the brown payment system makes most “profitable.” Nonetheless, bear (Swenson et al. 1994). That the national parks seem- our results show that rates of illegal killing of large carni- ingly provided increased opportunities for illegally killing vores are still high, especially when poachers face a low carnivores was, however, not an effect of the parks per risk of detection. Hence, economic incentives may partly se, but we suggest that it was rather a result of snowmo- change poaching behavior, but are not a panacea (Treves bile restrictions for the general public. These restrictions & Bruskotter 2014). On the other hand, our results sug- were intended to limit human disturbance of wildlife and gest that presence by the general public has the poten- reindeer inside the parks. However, the combination of tial to reduce illegal killing of large carnivores. Poaching low public presence and low intensity of patrolling by does not happen in a social vacuum, and the general pub- provincial rangers in these large, remote areas also im- lic’s tolerance to carnivores, and to illegal killing of such, plied that poachers faced a low risk of detection. The strongly affect poaching behavior. 338 Conservation Letters, September/October 2016, 9(5), 334–341 Copyright and Photocopying: 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. G. R. Rauset et al. Protected areas and poaching Our results emphasize the importance of critically eval- Table S4. The effects of strata (refer to Table S3b.) on uating the conservation performance of PAs, even large risk of being killed illegally for brown bears, Eurasian and remote national parks in a highly developed country, lynx, and wolverines in northern Sweden, 1984– and considering the ecological and socioeconomic context 2010. Parameter estimates in Andersen–Gill models present inside as well as outside them. We caution against are (β ± SE), and exp(β) represents the effect size, passive management and high public expectations of pos- i.e., the difference in risk of group 1 relative to group itive effects of PAs on large carnivore conservation, based 0 of Sex (males=0/females=1), Season (snow=0/no on the assumption that sparing land is enough. snow=1), Age01 (juvenile=0/subadult = 1, Age02 (juvenile=0/adult =1), Age12 (subadult=0/adult =1), VHF(GPSdata=0/VHF=1). The AIC represents Acknowledgments the change in AIC when adding the stratum to the We thank Jon Arnemo for veterinary assistance in the nonstratified model. field, and numerous field workers for data collection. The Table S5. The effect of national parks on the risk of being killed illegally for radio-marked a) brown bears, b) study was funded by the Swedish Environmental Protec- tion Agency, the Norwegian Environment Agency, the Eurasian lynx, and c) wolverines in northern Sweden, World Wide Fund for Nature (Sweden), the Swedish Uni- 1984–2010, at three different scales (animal locations and versity of Agricultural Sciences, the Research Council of buffers of 3,000 m and 6,000 m around the locations). Norway, the European Association of Zoos and Aquaria Parameter estimates (β ± SE), effect size (exp(β)), and (EAZA), and the Swedish Research Council for Environ- model AIC relative to the best models (AIC) are for ment, Agricultural Sciences and Spatial Planning (FOR- Andersen–Gill models. MAS). All handling of animals were approved by the Table S6. Environmental variables influencing a) risk Animal Ethics Committee for northern Sweden, Umea. ˚ of illegal mortality, b) risk of mortality other than illegal Data and models reported in the article are available from killing, and c) total risk of mortality of large carnivores the corresponding author upon request. The authors de- in northern Sweden, 1984–2010. Parameter estimates clare no competing financial interests. Authorship: GRR, (mean β ± SE) and the Variable Relative Importance HA, JP, GS, and PS designed the project; all authors con- (VRI) are model-averaged parameters of Andersen–Gill tributed to the collection and preparation of data; GRR models with AIC < 4 relative to the best models. The analyzed the data and wrote the article with the help of set of candidate models included all combinations of the JES, JP, HA, AZ, and GS. environmental variables; “national park” (NP), “reindeer calving ground” (Calf), “steep terrain” (Steep), and “Forest “ and their two-way interactions. NP and Calf Supporting Information are binary variables, whereas Steep and Forest represent the area (range 0 to 28.3 km ) within a 3,000-m buffer Additional Supporting Information may be found in the around the location (refer to Table S3). The effects of the online version of this article at the publisher’s web site: strata (refer to Table S4) were corrected for and therefore Table S1. Summary of data used for survival analyses always included. of radio-marked brown bears, Eurasian lynx, and wolver- ines in northern Sweden, 1984–2010. Table S2. Cause-specific mortality of radio-marked References brown bears, Eurasian lynx, and wolverines in northern Andersen, P.K. & Gill, R.D. (1982). Cox’s regression model for Sweden, 1984–2010. counting processes: a large sample study. Ann. Stat. 10, Table S3. a) Environmental variables used and b) 1100-1120. strata included and corrected for in the survival analysis Andren, ´ H., Linnell, J.D.C., Liberg, O., et al. (2006). Survival of large carnivores in northern Sweden, 1984–2010. SMD rates and causes of mortality in Eurasian lynx (Lynx lynx) vegcode refers to the Swedish Land Cover vegetation map in multi-use landscapes. Biol. Conserv. 131, 23-32. from the National Land Survey of Sweden. “Definition” Balme, G.A., Slotow, R. & Hunter, L.T.B. (2010). Edge effects describes how the variables were estimated; “area” rep- and the impact of non-protected areas in carnivore resents the area within a buffer of either 3,000 m or conservation: leopards in the Phinda-Mkhuze Complex, 6,000 m around a location, “distance” represents the South Africa. Anim. Conserv. 13, 315-323. shortest distance between a location and a specific habi- Bischof, R., Swenson, J.E., Yoccoz, N.G., Mysterud, A. & tat; “binary” represents whether a location is within or Gimenez, O. (2009). The magnitude and selectivity of outside a specific habitat or whether a buffer contained natural and multiple anthropogenic mortality causes in the specific habitat or not. hunted brown bears. J. Anim. Ecol. 78, 656-665. Conservation Letters, September/October 2016, 9(5), 334–341 Copyright and Photocopying: 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. 339 Protected areas and poaching G. R. Rauset et al. Burnham, K.P. & Anderson, D.R. (2002). Model selection and Margules, C.R. & Pressey, R.L. (2000). Systematic multimodel inference, 2nd ed. Springer-Verlag, New York. conservation planning. Nature 405, 243-253. Cardillo, M., Purvis, A., Sechrest, W., Gittleman, J.L., Bielby, Mattisson, J., Odden, J., Nilsen, E.B., Linnell, J.D.C., Persson, J. & Mace, G.M. (2004). Human population density and J. & Andren, ´ H. (2011). Factors affecting Eurasian lynx kill extinction risk in the world’s carnivores. PLoS Biol. 2, rates on semi-domestic reindeer in northern Scandinavia: 909-914. can ecological research contribute to the development of a Chapron, G., Kaczensky, P., Linnell, J.D.C., et al. (2014). fair compensation system? Biol. Conserv. 144, 3009- Recovery of large carnivores in Europe’s modern 3017. human-dominated landscapes. Science 346, 1517-1519. Murray, D.L. (2006). On improving telemetry-based survival Di Minin, E. & Toivonen, T. (2015). Global protected area estimation. J. Wildlife Manage. 70, 1530-1543. expansion: creating more than paper parks. Bioscience 65, Persson, J., Ericsson, G. & Segerstrom, ¨ P. (2009). Human 637-638. caused mortality in the endangered Scandinavian Dickman, A.J., Macdonald, E.A. & Macdonald, D.W. (2011). wolverine population. Biol. Conserv. 142, 325-331. A review of financial instruments to pay for predator Persson, J., Rauset, G.R. & Chapron, G. (2015). Paying for an conservation and encourage human-carnivore coexistence. endangered predator leads to population recovery. Conserv. Proc. Natl. Acad. Sci. U.S.A. 108, 13937-13944. Lett., 8, 345-350. Fischer, J., Abson, D.J., Butsic, V., et al. (2014). Land sparing Plumptre, A.J., Kujirakwinja, D., Treves, A., Owiunji, I. & versus land sharing: moving forward. Conserv. Lett. 7, Rainer, H. (2007). Transboundary conservation in the 149-157. greater Virunga landscape: its importance for landscape Haglund, B. (1965). Jarv ¨ och varg.P.A.Norstedt&Soners ¨ species. Biol. Conserv. 134, 279-287. Forlag, ¨ Stockholm. (Wolverine and wolf; In Swedish) Pressey, R.L. (1994). Ad hoc reservations: forward or Hobbs, N.T., Andren, ´ H., Persson, J., Aronsson, M. & backward steps in developing representative reserve Chapron, G. (2012). Native predators reduce harvest of systems. Conserv. Biol. 8, 662-668. reindeer by Sami pastoralists. Ecol. Appl. 22, 1640-1654. Purvis, A., Gittleman, J.L., Cowlishaw, G. & Mace G.M. Hoffmann, M., Hilton-Taylor, C., Angulo, A., et al. (2010). (2000). Predicting extinction risk in declining species. Proc. The impact of conservation on the status of the world’s Biol. Sci. 267, 1947-1952. vertebrates. Science 330, 1503-1509. Rauset, G., Mattisson, J., Andren, ´ H., Chapron, G. & Persson, Johnson, C.J., Boyce, M.S., Schwartz, C.C. & Haroldson, M.A. J. (2013). When species’ ranges meet: assessing differences (2004). Modeling survival: application of the Andersen-Gill in habitat selection between sympatric large carnivores. model to yellowstone grizzly bears. J. Wildlife Manage. 68, Oecologia 172, 701-711. 966-978. Ripple, W.J., Estes, J.A., Beschta, R.L., et al. (2014). Status Johnson, D.H. (1979). Estimating nest success: the Mayfield and ecological effects of the world’s largest carnivores. method and an alternative. Auk 96, 651-661. Science 343 (6167), 1241484. Keane, A., Jones, J.P.G., Edwards-Jones, G. & Smith, D.W., Bangs, E.E., Oakleaf, J.K. et al. (2010). Survival Milner-Gulland, E.J. (2008). The sleeping policeman: of colonizing wolves in the northern Rocky Mountains of understanding issues of enforcement and compliance in the United States, 1982–2004. J. Wildlife Manage. 74, conservation. Anim. Conserv. 11, 75-82. 620-634. Kiffner, C., Meyer, B., Muehlenberg, M. & Waltert, M. Soule, M.E. & Sanjayan, M.A. (1998). Ecology - conservation (2009). Plenty of prey, few predators: what limits lions targets: do they help? Science 279, 2060-2061. Panthera leo in Katavi National Park, western Tanzania? Swenson, J.E. & Andren, ´ H. (2005). A tale of two countries: Oryx 43, 52-59. large carnivore depredation and compensation schemes in Linnell, J.D.C., Swenson, J.E. & Andersen, R. (2001). Sweden and Norway. Pages 323-339 in R. Woodroffe, S. Predators and people: conservation of large carnivores is Thirgood, A. Rabinowitz, editors. People and wildlife: conflict possible at high human densities if management policy is or coexistence? Cambridge University Press, New York. favourable. Anim. Conserv. 4, 345-349. Swenson, J.E., Sandegren, F., Bjarvall, ¨ A., Soderberg, ¨ A., Liu, J.G., Linderman, M., Ouyang, Z.Y., An, L., Yang, J. & Wabakken, P. & Franzen ´ R. (1994). Size, trend, Zhang, H.M. (2001). Ecological degradation in protected distribution and conservation of the brown bear areas: the case of Wolong Nature Reserve for giant pandas. Ursus-Arctos population in Sweden. Biol. Conserv. 70, 9-17. Science 292, 98-101. Therneau, T. (2011). Survival analysis, including penalised Loveridge, A.J., Searle, A.W., Murindagomo, F. & Macdonald, likelihood. R package’Survival. version 2.38, URL: “http:// D.W. (2007). The impact of sport-hunting on the CRAN.R-project.org/package=survival” population dynamics of an African lion population in a Treves, A. & Bruskotter, J. (2014). Tolerance for predatory protected area. Biol. Conserv. 134, 548-558. wildlife. Science 344, 476-477. 340 Conservation Letters, September/October 2016, 9(5), 334–341 Copyright and Photocopying: 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. G. R. Rauset et al. Protected areas and poaching Treves, A. & Karanth, K.U. (2003). Human-carnivore conflict Woodroffe, R. & Ginsberg, J.R. (1998). Edge effects and the and perspectives on carnivore management worldwide. extinction of populations inside protected areas. Science Conserv. Biol. 17, 1491-1499. 280, 2126-2128. Wittemyer, G., Elsen, P., Bean, W.T., Burton, A.C.O. & Zabel, A. & Holm-Muller, K. (2008). Conservation Brashares, J.S. (2008). Accelerated human population performance payments for carnivore conservation in growth at protected area edges. Science 321, 123-126. Sweden. Conserv. Biol. 22, 247-251. Conservation Letters, September/October 2016, 9(5), 334–341 Copyright and Photocopying: 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. 341

Journal

Conservation Letters – Wiley

Published: Sep 1, 2016

Keywords: ; ; ; ; ; ;

There are no references for this article.