European bison conservation cannot afford to ignore alternative hypotheses: a commentary on Perzanowski et al. (2019)

Abstract

Conservation science is about developing a robust understanding of how to reduce threats and extinction risks for biodiversity, as expressed in the International Union for the Conservation of Nature (IUCN) Conservation Action Plan approach (IUCN/SSC, 2013). As a branch of science, it must therefore follow the scientific best practice of considering alternative hypotheses in order to minimise the risk of erroneously increasing threats or extinction risks for species (Caughley, 1994). The consequences of failure to consider alternative hypotheses are not trivial. Conservation interventions that are developed from artificially constrained and limited perspectives of a species’ ecological requirements likely increase the risk of population and species losses, the latter being irreversible. We are concerned that Perzanowski et al. (2019), in modelling an approach for conserving the European bison Bison bonasus in Poland, ignored an alternative hypothesis that identifies the European bison as a refugee species due to its current confinement in forest habitat (Kerley, Kowalczyk & Cromsigt, 2012). Perzanowski et al. (2019) choose to frame their modelling approach within the paradigm that the European bison is a forest specialist, with occurrence data that reflect the prevailing confinement of the species to forest habitats, where it is managed through supplementary feeding and culling to mitigate movements and range expansion to open habitats (Kerley et al., 2012). The Refugee Species Hypothesis (RSH) for the European bison was developed in 2012, and reflects the evolutionary background, palaeoecology, life history, feeding morphology, diet choice, and habitat selection of the European bison (Kerley et al., 2012). All of these point to the European bison being adapted to more open, grass-rich habitats. Furthermore, Kerley et al. (2012) demonstrate that the current confinement of European bison to forests reflects the trapping of this species in forest habitat by human pressures and management, a process that has been ongoing for hundreds of years (Samojlik et al., 2019). The RSH predicts that confining the European bison to forest habitat has an array of fitness and management consequences. These predictions include reduced access to preferred/optimal resources (see habitat preferences below) and decreased fitness, as well as the costs of supplementary feeding that forest-confined bison require, increased vulnerability to disease and parasites, and human–wildlife conflicts when bison attempt to move out of forests (Hofman-Kamińska & Kowalczyk, 2012; Kerley et al., 2012; Kowalczyk et al., 2013; Kołodziej-Sobocińska et al., 2016). This problem is exacerbated in coniferous forests where bison density is lower than in mixed forests (Table 1), and levels of human–bison conflict are elevated (Hofman-Kamińska & Kowalczyk, 2012). Perzanowski et al. (2019) emphasize that ‘The majority of forested area is coniferous’, which suggests that applying their model will lead to low bison densities and heightened human–wildlife conflict. Additionally, the forest specialist paradigm will unduly restrict the global European bison population because it constrains it to low densities in areas of lower-quality habitat (Kerley et al., 2012). Subsequently, as informed by the RSH, additional independent evidence has emerged to support the proposition that the landscape ecology of the European bison should include sufficient open habitat. This includes pre-refugee ecology evidence indicating utilisation of open habitats in the early Holocene before human expansion (Bocherens et al., 2015; Hofman-Kamińska et al., 2018a,b; Hofman-Kamińska et al., 2019). Moreover, contemporary habitat selection indicates the species’ preference for open and wet/open habitats and abandoned fields over the proportionally (in relation to available area) less-used forest habitats (Kuemmerle et al., 2010, 2018, Zielke, Wrage-Mönnig, Müller & Neumann, 2019). Broeze (2018) found preference for grasslands and open habitats and avoidance of coniferous forests in three populations of European bison. When allowed to occupy grass-rich landscapes, European bison exhibit foraging similar to that of acknowledged grazer species such as free-ranging cattle and horses (Cromsigt et al., 2018), and the American bison Bison bison (Plumb, White & Aune, 2014). Revealingly, none of these studies have been able to support the alternative hypothesis that the European bison is a forest specialist. Based on this accumulated evidence, the RSH is now recognised by other groups working on the ecology and management of both the European bison (Balčiauskas & Kazlauskas, 2014; Németh et al., 2016; Balčiauskas, Kazlauskas & Balčiauskienė, 2017; Cromsigt et al., 2018) and the American bison (Plumb & McMullen, 2018). In adopting the ‘European bison is a forest specialist’ paradigm, Perzanowski et al. (2019) have given expression to the shifting baseline syndrome (Pauly, 1995), whereby succeeding generations accept changed ecological circumstances as their reference base. Caughley (1994, 229) pointed out that for threatened species that have suffered a range contraction, it is safer to hypothesize ‘that the species ends up, not in the habitat most favourable to it, but in the habitat least favourable to the agent of decline [threat]’. Cromsigt, Kerley & Kowalczyk (2012) argued that assuming that forest habitat is the preferred habitat for the European bison leads to a flawed assessment of Europe-wide habitat availability for this species. We are also concerned that in following Perzanowski et al. (2019), the pending revision of the IUCN conservation action plan for the European bison (Olech, Klich & Perzanowski, 2019) will adopt the same shifted baseline framework. This would perpetuate the last 300 years of management that confines the bison to forests (Samojlik et al., 2019), and requires ongoing food supplementation, as forest habitats are not able to support bison forage needs year-round (Kowalczyk et al., 2011; Kerley et al., 2012). Such management does not reflect wildlife conservation as defined by the IUCN (IUCN/SSC, 2013), but is more akin to farming. A constrained focus on forest habitats artificially limits consideration of potentially available open habitat that could contribute to European bison conservation. There is an estimated 1.5 Mha of agricultural land that has been abandoned in Poland in the last few decades (Alcantara et al., 2013). Ignoring the potential ecological restoration of bison to these lands, that comprise c. 75% of the area of forest included in Perzanowski et al.’s (2019) model, clearly misses a very important conservation opportunity. The European bison has already gone extinct in the wild once (Kerley et al., 2012), and it would be a tragedy if we were to place it at risk again through incomplete conservation science. We urge the IUCN Bison Specialist Group to consider the full range of alternate hypotheses of the species’ ecological requirements in developing an updated Conservation Action Plan for the full ecological recovery of the species. As we have historically placed the European bison at risk, so must we now avail our efforts to save the species with the full capacity of our best available and rigorous science. We thank the Editor and two anonymous reviewers for input that considerably improved this manuscript.