Historical change in fish species distribution: shifting reference conditions and global warming effects

Didier Pont 0 1 2 3 M. Logez 0 1 2 3 G. Carrel 0 1 2 3 C. Rogers 0 1 2 3 G. Haidvogl 0 1 2 3 0 G. Carrel C. Rogers Irstea UR HYAX , 3275 Route de Ce zanne-CS 4006, 13182 Aix-en-Provence Cedex 5 , France 1 M. Logez Irstea UR HYAX, Pole Onema-Irstea Hydroe cologie Plans d'eau , 3275, Route de Ce zanne-CS 4006, 13182 Aix-en-Provence Cedex 5 , France 2 D. Pont (&) Irstea UR HBAN , 1 rue Pierre-Gilles de Gennes-CS 10030, 92761 Antony , France 3 G. Haidvogl Institute of Hydrobiology and Aquatic Ecosystem Management, University of Natural Resources and Life Sciences Vienna , Max-Emanuel-Strasse 17, 1180 Vienna , Austria Species distributions models (SDM) that rely on estimated relationships between present environmental conditions and species presence-absence are widely used to forecast changes of species distributions caused by global warming but far less to reconstruct historical assemblages. By compiling historical fish data from the turn to the middle of the twentieth century in a similar way for several European catchments (Rhone, Danube), and using already published SDMs based on current observations, we: (1) tested the predictive accuracy of such models for past climatic conditions, (2) compared observed and expected cumulated historical species occurrences at sub-catchment level, and (3) compared the annual variability in the predictions within one sub-catchment (Salzach) under a future climate scenario to the long-term variability of occurrences reconstructed during an extended historical period (1800-2000). We finally discuss the potential of these SDMs to define a ''reference condition'', the possibility of a shift in baseline condition in relation with anthropogenic pressures, and past and future climate variability. The results of this study clearly highlight the potential of SDM to reconstruct the past composition of European fish assemblages and to analyze the historical ecological status of European rivers. Assessing the uncertainty associated with species distribution projections is of primary importance before evaluating and comparing the past and future distribution of species within a given catchment. - For decades, freshwater biodiversity has been recognized as highly threatened due to the long history of anthropogenic modifications of continental aquatic ecosystems (Dudgeon et al. 2006). Among aquatic species, fish react to almost every kind of water quality and habitat alterations (Ormerod 2003). Fish sensitivity to human pressures is a basis for using fish-based biological monitoring tools to assess environmental change (Fausch et al. 1990). More recently, numerous papers have also pointed out the influence of climate modifications on freshwater ecosystems (Webb 1996) and in particular on fish species distributions (Xenopoulos et al. 2005; Graham and Harrod 2009). Long-term changes in climate during the last millennia are well known, particularly those since the end of the little ice age in the mid-19th century (Schurer et al. 2013). Evidence is now accumulating that one of the main species responses to global warming is a shift poleward or upward in elevation to colonize favorable thermal habitats (Parmesan and Yohe 2003; Crimmins et al. 2011; Comte and Grenouillet 2013). In both cases, human alteration and climate change, the impact of environmental modifications on species distributions is based on the comparison between an observed or a predicted situation and a benchmark. In bioassessment methods, a reference condition is determined from sites undisturbed by anthropogenic stressors, thus representing continuity with a former condition (Bailey et al. 1998). In Europe, most catchment landscapes and rivers have undergone modifications during the last two centuries. Undisturbed habitats are becoming increasingly rare and bioassessment methods generally rely on what are now judged to be minimally-disturbed sites (Stoddard et al. 2006). Moreover, even for such sites, the distributions of species could continuously evolve in response to largescale anthropogenic disturbance, past changing thermal and rainfall conditions, and to the expected future climate trend (Tingley and Beissinger 2009). The term shifting baseline was developed to refer to such long-term changes over generations which are difficult to recognize (Pauly 1995). Historical reconstruction of species distribution could allow testing such shifting baseline effects. Numerous studies have used printed historical information on fish communities from the late eighteenth and nineteenth centuries to reconstruct the long-term evolution of fish faunas (Rinne et al. 2005; Maceda-Veiga et al. 2010), to estimate reference conditions (Carrel 2002; Wolter et al. 2005; Winter et al. 2009) or to define specific conservation programs (Worthington et al. 2010). Different types of historical sources have been identified depending on the possibility to gain quantitative or only qualitative data at different scales (Haidvogl et al. 2013a). From the late eighteenth century on, early scientific fish ecological surveys were conducted compiling scientific inventories and systematic or selective enquiries of fishermen and fishmongers. From the second half of the nineteenth century, fish distribution maps were also produced in several catchments across Europe, localizing species occurrences in precise rivers or river sections (Haidvogl et al. 2015; Carrel 2002). As opposed to most other written sources, historical maps are among the most informative because the aim of the historical maps was to describe the complete fish fauna and not only the species of commercial interest (Haidvogl et al. 2013a). Several authors have attempted to model historical fish species distribution. Lassalle and Rochard (2009) compiled literature on past occurrences of anadromous species to model their distribution before the twentieth century and to simulate potential change due to climate change. In a recent work, Labay et al. (2011)using both historical and present dataapplied species distribution models (SDM) to establish baseline conditions and to assess the current status of a river fish community. Such models are also widely used to predict the change of fish distributions under climate change scenarios (Austin 2007; Buisson et al. 2008). A classical approach is to model present fish distributions based on recent sampling surveys and to compare the currently forecasted distribution with the expected distribution under different climate conditions (Buisson et al. 2008). The main aim of the present paper was to examine the ability of SDM to reconstruct historical freshwater fish assemblages and discuss the implication of using baseline conditions in the context of predicting species distribution under changing environmental conditions. The already published SDM used here are based on current observations from European rivers qualified as current reference sites (Logez et al. 2012). The analyses we presented are based on histor (...truncated)