Abstract
Context
Riparian areas are considered to undergo major alterations under changing climate, making floodplain habitats targets for conservation and landscape planning. Protected areas might provide sanctuaries especially for sessile riparian plant species, but these niches are not always persistent over time.
Objectives
We investigate if plant species of floodplain forests are provided with suitable habitat within currently protected areas and if these refugia persist. A coupled-modelling approach is used to gain spatially explicit information on new areas for sanctuaries.
Methods
We use species distribution models to predict the niche of 12 Salicion albae and 7 Fraxinion floodplain forest species along rivers in Switzerland, under current, moderate and extreme climate change scenarios up to 80 years to the future (2100). The spread of plant species from current habitat to suitable future habitat is simulated using dispersal vectors and life history traits.
Results
Salicion albae species are more flexible under both climate change scenarios than Fraxinion species. The main limitation for the spread of species is their dispersal ability, as only a minority of the suitable cells is colonized during the simulation process. The predicted future presence within currently protected areas decreases under both climate change scenarios in the model.
Conclusions
Current protected floodplains do not provide persistent refugia for the plants studied, but might still be of importance to other organisms. Planning of sanctuaries for riparian plant species and communities need to focus on connectivity along rivers to maintain viable source populations in dynamic riverine landscapes under changing climate.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
Over the last decade, the loss of wetlands including riverine floodplains was more rapid than any other ecosystem (United Nations 2018), although these ecosystems provide important services (Costanza et al. 1997), show higher species diversity than surrounding uplands (Brown and Peet 2003) and harbor distinct plant communities (Sabo et al. 2005). Natural floodplain areas as well as typical floodplain vegetation depending on river dynamics are decreasing due to human impact e.g. by fragmentation due to dams (Jansson et al. 2000; Alldredge and Moore 2014; Pracheil et al. 2015) and altered river flow regime (Rivaes et al. 2015). Vulnerable riparian areas are a main target for conservation management and planning (Pullin et al. 2009), but few guidelines for riparian vegetation in a changing world exist (but see Dufour et al. 2019; Lind et al. 2019).
Riparian habitats along rivers are suspected to undergo major changes under climate change scenarios (Seavy et al. 2009; Mosner et al. 2015), since both climatic and hydrological factors will change and e.g. extreme floods might disrupt suitable habitat more frequently (Death et al. 2015). The local persistence of dynamic habitats is therefore expected to be reduced (Vittoz et al. 2013), which is especially disastrous for habitats which need several decades to establish, like floodplain forests of the plant community Fraxinion (Ellenberg 2010; Werth et al. 2012; Delarze and Gonseth 2015). These species rich plant communities (Ellenberg 2010; Delarze and Gonseth 2015) are crucial for the survival and spread of other organisms (Schnitzler-Lenoble and Carbiener 2007; Machar et al. 2019). Changing future climatic conditions also directly affect habitats close to the waterline, like floodplain forests of the plant community Salicion albae, which might result in modified plant communities due to e.g. longer drought periods followed by extreme floods (Ikeda et al. 2014; Chen et al. 2018).
Changing climate and its effect on riparian ecosystems has been investigated previously, indicating either thread of species loss or retreat to higher altitudes (e.g. see Garssen et al. 2014; Mosner et al. 2015). Conservation planning for vulnerable species and habitats within the diverse ecosystems is recommended to be done at the landscape context (Capon et al. 2013). Still, many processes responsible for riparian vegetation establishment and persistence along catchments remain unclear, although some models allow predictions by linking information on habitat formation and species abundance (Fink et al. 2017; Gostner et al. 2017; Caponi et al. 2019). Similarly, genetic analyses of riparian plants allow indirectly inferring processes of the past and present leading to the current distribution of species and gene pools as well as connectivity patterns along rivers (Pollux et al. 2009; Cushman et al. 2014; Werth and Scheidegger 2014).
For the survival of riparian plant species, connectivity is crucial for the recolonization of suitable habitat (Ward et al. 1999). Therefore, sessile riparian plant species along rivers show adaptions to various types of dispersal, e.g. animal-mediated dispersal (Müller 1955; Ellenberg 2010), water-mediated dispersal (hydrochory, Nilsson et al. 2010), and wind dispersal (Ellenberg 2010), maintaining connectivity over long and short distances (e.g. Ulmus laevis, Venturas et al. 2014). Wind-mediated upstream dispersal was shown to be important for gene flow in many species (Wubs et al. 2016), while mainly vegetative diaspores and only few seeds are dispersed by water (Boedeltje et al. 2003).
As dispersal plays a key role for riparian plant species (Catford and Jansson 2014), including information on species’ spread from currently occupied habitats to future potential areas allows to improve predictions for future species’ occurrences (Engler et al. 2009). Landscape and conservation planning therefore uses modelling of current and future species’ distribution to predict changes in functional connectivity among populations along a river (Inoue and Berg 2017) as well as to detect refugia, which are areas providing habitat for species during disturbance events and under changing climate (Morelli et al. 2016). While protected areas like parks and nature reserves might provide sanctuaries, which are temporarily suitable habitats for species e.g. under current climatic conditions, these might not be persistent over time and new protected areas might be necessary (D'Amen et al. 2011; Hannah et al. 2014a; Barrows et al. 2020; Morelli et al. 2020). Therefore, more riparian species have to be investigated to adjust conservation goals (Pullin et al. 2009; Driscoll et al. 2013).
This study aims at determining if plant species of floodplain forests find sanctuaries in protected areas and if these persist as refugia under both moderate and extreme climate change scenarios. If protected areas provide sanctuaries we hypothesize that suitable habitat is prevalent within the designated protected perimeter, and in the case of refugia, models should predict most of the future suitable habitat remaining inside the protected zone. We use species distribution models to investigate the niche and the current and future distribution pattern of 12 Salicion albae and 7 Fraxinion floodplain forest species along rivers in Switzerland with citizen science and research data provided by the Swiss database for plants (Info Flora). To gain information for landscape and conservation planning of potential long term refugia and future sanctuaries as well as functional connectivity for floodplain forests, we simulate the spread of plant species using information on dispersal to potentially suitable future habitats along river catchments in Switzerland.
Methods
Study species
Typical plant species of the two plant communities of floodplain forest of the lowland (Salicion albae) and floodplain forest of a later successional stage (Fraxinion) were investigated in this study (see Table 1). Salicion albae habitats are listed as endangered in Switzerland (BAFU 2011; Delarze et al. 2013), while Fraxinion floodplain forests are described as a habitat type worth protecting by the Federal Act of Protection of Nature and Cultural Heritage (1966) as they have very long regeneration time of 50 to 200 years (Delarze et al. 2013; Delarze and Gonseth 2015). Species for the current study were selected based on the list of Swiss national priority species (BAFU 2011), representing indicator species for habitat types and therefore include redlisted and target species for conservation projects, and data availability at the Swiss National data base Info Flora (www.infoflora.ch, as of December 19, 2016). This data base collects occurrence data obtained in research and monitoring projects under the supervision of the Federal Office for the Environment of Switzerland, Swiss cantonal authorities, non-profit organizations as well as data collected by citizens. We obtained data for twelve Salicion albae species (two tree species, six shrub species, and four herbaceous species) and seven Fraxinion species (two tree species, four sedges, and one herbaceous species, see Table 1 and Table 2) along all waterways in Switzerland. Data selection for all analyses included the following precautions: only coordinates with a precision of ≤ 100 m and collection dates between 1960 and 2015 were used. All locations of species’ occurrence were transformed to a raster of 100 × 100 m using the raster package in R (Hijmans and van Etten 2012) covering whole Switzerland including all catchments. Duplicates were removed and only one presence point per grid cell was kept to avoid spatial autocorrelation in the niche modelling approach (Guisan and Zimmermann 2000). The data was used in a hierarchical framework by first modelling suitable habitat and subsequently simulating the spread of species from recorded sources to predicted suitable habitat, also taking into account changing climate.
Modelling habitat suitability
Data from nine uncorrelated geological, topographic and climate predictors (see Fink and Scheidegger 2018) were applied to define the current ecological niche of the species in Switzerland: Mean annual temperature, mean annual precipitation (sourced by the Federal Office of Meteorology and Climatology MeteoSwiss, prepared by the division of Landscape Dynamics of WSL, following the methodology used in Thornton et al. 1997), wetness index (following Tarboton 1997, see in Camathias et al. 2013), nutrient-holding capacity, northness, slope, soil permeability, water-holding capacity and hydrometric properties (Camathias et al. 2013). To account for proximity of habitats to rivers in Switzerland, all predictor layers were masked to only cover 1000 m around rivers within Switzerland in arcGIS (ESRI 2015). Habitat suitability for the reference period (1960–2013) was modelled using the software Maxent (Phillips et al. 2006). The reference period is shorter than the species record period to ensure that a majority of species have reach maturity for subsequent dispersal simulation (see below). We sampled 10′000 background points and performed a ten-fold-crossvalidation (Phillips et al. 2009). Models were evaluated for predictive power by the area under the curve (AUC) of the receiver operating characteristics (see Phillips et al. 2006) and by Boyce index using the ecospat package in R, applying a sliding window (Hirzel et al. 2006). The same R package was used to calculate and visualize the correlation between the ratio of predicted to expected species presences (Hirzel et al. 2006).
Predicted species distribution and Floodplains of National Importance
The average value of ten independent modelling runs resulted in a habitat suitability map with continuous probabilities for each species, and was transformed in species presence-absence maps. The binary maps were calculated using the equal training sensitivity and specificity logistic threshold (Liu et al. 2013) calculated by the software Maxent (Phillips et al. 2006). Raster cells with predicted presences for Salicion albae species and for Fraxinion species independently were overlapped and core areas forming ideal niches for the respective plant communities were identified in the program R (R Core Team 2016). Moreover, the predicted core areas as well as single species predictions were overlapped with areas of current protected floodplains (Floodplains of National Importance, legislation origination from 1992; Hausammann et al. 2005). Number of cells within Floodplains of National Importance were compared to cells outside these protected areas using Welch t-tests in R (Welch 1947; R Core Team 2016).
Niche modelling under climate change
Future habitat suitability was modelled by projecting the model with the current niche definition to eight future time periods covering 10 years each (first period 2014–2023 up to last period 2084–2093) using Maxent (Phillips et al. 2006). While geological and topographic predictors remained constant in all simulations, layers of temperature and precipitation were once left unchanged (baseline simulation), but also modified according to two climate scenarios based on information of the coordinating body of CORDEX (see http://www.euro-cordex.net, as agreed on CMIP5 (IPCC Fifth Assessment, 2009). The first of two climate scenarios applied was the regional climatic model CLMcom-CCLM4-8–17, calculated and downscaled from the global climatic model CNRM-CERFECS-CNRM-CM5 using emission predictions of RCP 4.5. The second scenario was based on the regional climatic model CLMcom-CCLM4-8–17 calculated and downscaled from the global climatic model ICHEC-EC-EARTH by ICHEC using emission predictions of RCP 8.5. Both data sets were further downscaled to a 100 × 100 m grid over Switzerland applying the Delta change method and using measured data on temperature and precipitation as a reference (data sourced by the Federal Office of Meteorology and Climatology MeteoSwiss, data prepared by the division of Landscape Dynamics at WSL). The first model represents a moderate climate change with moderately increasing temperature, and only moderately increasing precipitation in winter compared to the reference period (1961–1990). The second scenario represents a more extreme climate change, with very dry summers and extreme temperature increases in comparison to the reference period (see in Fink and Scheidegger 2018). The two scenarios allow assessing the climate change impact on habitat availability by contrasting moderate and extreme changes in terms of temperature and precipitation.
Simulated dispersal to future habitat
For the simulation of the spread of species from current present points to future modelled suitable habitat along all Swiss rivers, we performed a literature search on life history traits of species including information on dispersal vectors and distances (see Table 1). Information on initial maturity and age dependent seed production were implemented in the model to simulate the spread under realistic assumptions. Literature data on dispersal vectors and distances was fit to dispersal models and dispersal kernels respectively (Table 1). Dispersal was simulated using the package MigClim 1.5 in R (Engler et al. 2012): There the kernel represents the probability of the spread of source cells to adjacent cells. For species displaying long-range dispersal (e.g. by water, birds or wind), it was included in the simulation with a low probability (0.01) up to distances of 10 km (Fink and Scheidegger 2018).
We simulated the dispersal from current species’ occurrences for 50 and 80 years to the future, based on the probability to reach the surrounding cells (dispersal kernel) and the probability of suitable habitat (habitat suitability maps). Starting with the reference time period (1960–2013), habitat suitability maps were updated five and eight times respectively during the dispersal simulation for each climate change scenario. Each simulation was repeated 100 times for each species and scenario. The average of the results of these simulations were used to compare the area of occupied and colonized cells after 50 and 80 years of dispersal in general and to the current areas of Floodplains of National Importance (Hausammann et al. 2005). Colonized cells are raster cells that became inhabited by the species in the course of the simulation, while occupied cells remain populated from the initial state to the end of the simulation (see in Engler et al. 2012). Contrary to that, decolonized cells were once inhabited by the species but do not remain populated till the end of the simulation due to habitat becoming unsuitable. Comparisons of occupied and colonized as well as decolonized cells at the end of simulations between different climate change scenarios were calculated using Welch’s t-Tests (Welch 1947) and repeated measure ANOVA in R (R Core Team 2016). To analyze differences between catchments, the results were grouped by the four main catchments (Inn, Rhine, Rhone, Ticino), and the number of suitable but not colonized cells were determined for each catchment area using information on each tributary in arcMap (ESRI 2015).
Results
Modelling habitat suitability for the reference period
Maximum entropy modelling using the software Maxent (Phillips et al. 2006) resulted in models with AUC values indicating good prediction accuracy (AUC > 0.75, see Table 2 and compare to Manel et al. 2001). Model evaluations using the ecospat package showed high correlations between predicted to expected presence points and habitat suitability (Spearman correlation, 0.883 – 0.999, see Table 2 and Appendix for graphical representations, Figs A1). Average habitat suitability maps for whole Switzerland revealed that all species have independent niches (Appendix, Figs A2). Still, most Salicion albae species showed main suitable habitats along Rhine, Rhone and Ticino, except for Epipactis rhodanensis, for which suitable habitats are mainly found along Rhone and Rhine. For two species of this plant community (Salix myrsinifolia and Salix purpurea), models revealed suitable habitats along all four catchments equally (Inn, Rhine, Rhone, Ticino, see Appendix). Similarly, Fraxinion species showed many areas with high suitability along Rhine, Rhone and Ticino, except for Carex strigosa and Malaxis monophyllos, for which areas were limited to the Rhine catchments (see Appendix, Figs A2).
Predicted species distribution under current climate scenarios
Predicted distribution maps were obtained based on modelled habitat suitability maps (Fig A3). Major catchments with predicted species distribution are Rhine, Rhone and Ticino. One species of the Salicion albae community, S. myrsinifolia, has a widespread predicted distribution along all four catchments (see Fig A3). Similarly, for two Fraxinion species, Carex pendula and Carex remota, over 20% of cells were predicted to be suitable (Table 2), while all other floodplain forest species display occupation of less than one fifth of all cells along rivers. Especially C. strigosa and M. monophyllos, two species of the Fraxinion community, reveal limited distribution mainly along Rhine catchment, and consequently low percentage of predicted distribution (9.5% and 8.3%, see Table 2 and Fig A3).
Pooled information on all 19 forest species revealed 46.1% of all cells without any predicted species and no cell with predicted presence for all species (Fig. 1). When only considering species of the Fraxinion community separately, no cells were detected were all 7 species were predicted, and only 1% of the cells were occupied by 6 out of 7 species. Additionally, 55.6% of all cells in Switzerland were predicted to be uninhabited by Fraxinion species. For species of the Salicion albae community, 1.1% of all cells of Switzerland were predicted to be occupied by all 12 species together, while 62.6% cells were predicted to be uninhabited.
Niche modelling and simulated dispersal under climate change
Moderate and extreme climate change scenarios revealed generally similar prediction patterns of habitat suitability for both floodplain forest plant communities (see Supplementary material, Figure A2). While for a majority of the species of Salicion albae the area represented as suitable habitat increased under both scenarios after 80 years from now, for all Salix species (except Salix triandra) the habitat suitability decreased (see Supplementary material, Figure A2). Similarly, for all Fraxinion species (except for C. pendula), habitat suitability decreased in predictions under both climate change scenarios.
Simulations for the spread of floodplain forest species under dispersal assumptions revealed that although dispersal took place, significantly less cells were colonized (compared to uncolonized cells) for both plant communities under both climate scenario assumptions (Fraxinion species t = − 126.60, p < 0.01; Salicion albae species t = − 92.27, p < 0.01). Comparisons between the two climate change scenarios for each plant community showed significant differences in cells that remained occupied throughout the whole simulation both after 50 years (t = − 10.65, p < 0.01) and after 80 years (t = − 10.85, p < 0.01) for Fraxinion species. No difference in occupied cells was detected for simulations after 50 years (t = − 0.92, p = 0.98) for Salicion albae species, but differences between occupied cells for simulations of different climate change scenarios were significant after 80 years of dispersal (t = − 20.06, p < 0.01). For colonized cells, both plant communities showed significant differences after 50 (Fraxinion species: t = − 15.97, p < 0.01; Salicion albae species: t = − 23.46, p < 0.01) and 80 years (Fraxinion species: t = − 24.14, p < 0.01; Salicion albae species: t = − 31.03, p < 0.01).
Repeated measure ANOVA analysis for each species showed that the climate change scenario and the years of dispersal simulation both had a significant effect on the colonized cells during the simulation (repeated measure ANOVA, years: F = 24.54, p < 0.001; climate change scenario: F = 12.45, p = 0.002). Each catchment shows differences in predicted cells which could not be reached (suitable, but not colonized cells, see Fig. 2). Still, not only dispersal limitations were responsible for limiting the spread of species, as the additional 30 years of dispersal (80 years compared to 50 years) do not result in less uncolonized cells for all species, for each catchment and also within the protected floodplain area (Figs. 2 and 3).
Predicted distribution Floodplains of National Importance and shifting sanctuaries
We checked for recorded and predicted species presence within currently protected Floodplains of National Importance, which make up 0.48% of the size of Switzerland (Fig. 3). For most of the floodplain forest species, approximately half the cells within floodplains were suitable habitat for the species under current climatic conditions, except for S. myrsinifolia and S. purpurea, where > 70% of the cells were suitable (Fig. 3, A). Contrary to that, C. strigosa and M. monophyllos, two species of the Fraxinion community, had < 10% of suitable cells within protected floodplain areas (Fig. 3, A).
The predictions for the cells being colonized within Floodplains of National importance under two climate change scenarios are much lower both after 50 and 80 years of dispersal (Fig. 3). The only exception is Solanum dulcamara, a species of the Salicion albae community, which shows a high number of colonized cells within the area of the current protected floodplains (Fig. 3), and which was also able to colonize all catchments (see Fig. 4).
The spatially explicit outputs of the simulations show which species is expected to colonize which areas along each catchment under each climate change scenario (for the extreme climate change scenario compared to the current climate change scenario, see Fig. 4). The Salicion albae species S. dulcamara is the only species which is able to colonize a majority of cells along all catchments even under the extreme climate change scenario (Fig. 4). Some species mainly colonize the Central plateau area, along the Rhine catchment (C. pendula, C. remota, Prunus padus and Symphytum officinale, see Fig. 4), while the two Populus species, P. alba and P. nigra, as well as Salix alba and S. purpurea colonized both the Rhine and the Ticino catchment under extreme climate change scenarios (Fig. 4).
Few species show areas of predicted distribution which persist under current and future climate scenarios (Fig. 4, e.g. C. pendula, C. remota, P. padus, P. alba, S. alba, S.purpurea, S. dulcamara). Contrary to that, it is predicted for many species that the area along the Central Plateau is not persistent habitat (Fig. 4, e.g. C. strigosa, Ulmus laevis, Salix fragilis), but few species can cope with changing climatic conditions and new sanctuaries appear (Fig. 4, e.g. P. padus, P. alba, P. nigra).
Discussion
Change between current and future predicted habitat
Models predicting the current and future habitat of plants within the typical floodplain forest of the lowland (Salicion albae) and the later successional stage floodplain forest (Fraxinion) reveal that a majority of habitats is predicted along the main rivers (Rhine, Rhone, Ticino) in Switzerland (Figs. 1 and 4). Still, our study revealed that already under current climatic conditions, only few areas were predicted for all Salicion albae species and for a majority of the Fraxinion species (Fig. 1). The ecological differences for each species’ niches are reflected in the modelled predictions: e.g. the largest Carex species of Switzerland, C. pendula, revealed a broad distribution reflecting the species’ ability to occur also in secondary habitat along forests in wet patches (Lauber 2018). Contrary to that, C. strigosa shows limited habitat availability, emphasizing that species of the same genus have different ecological traits and different occurrence potential. Therefore, predictions for landscape and conservation planning should focus on single target species both at the national level as well as at the catchment level (Reichert et al. 2007; Guisan et al. 2013).
The habitat availability of species of both floodplain forest plant communities is predicted to change dramatically both under the moderate and the extreme climate change scenarios, similar to results of other riparian plant species in Switzerland (Fink and Scheidegger 2018) and worldwide (Ikeda et al. 2014) as well as other riparian organisms (D'Amen et al. 2011). Given that the plant community Salicion albae is endangered in Switzerland (Delarze et al. 2013; Delarze and Gonseth 2015), the change in habitat availability is important for landscape and conservation planning. Species of the genus Salix are predicted to be more flexible towards changing temperature and precipitation patterns, and even show more suitable habitats in the future. The model predictions are overlapping with ecological information, as Salix species can occupy secondary habitat and cope with longer dry periods than other species (Chmelar and Meusel 1979; Neumann 1981; Kuzovkina et al. 2004; Francis et al. 2005). Additionally, their potential to survive even extreme conditions has also been found in a previous study (Fink and Scheidegger 2018; but see in Ikeda et al. 2014).
While the predicted suitable habitat increased for most Salicion albae species, there was still a shift in catchments providing modelled suitable habitat, e.g. for S. purpurea (from Rhine to Ticino catchment) and S. myrsinifolia (from Rhine to Inn catchment, see Figure A2). This information is valuable as conservation and landscape planning strategies frequently focus on the catchment level. Additional change in future habitat availability might be due to extreme water table changes and flooding, which might have an effect on some Salix as well as other softwood floodplain forest species in close river proximity (Kuzovkina et al. 2004; Francis et al. 2005). While future models certainly need to consider such information, the climatic, topographic and geological predictors of this study already provide key information on which areas are unsuitable independent of river dynamics and river morphology.
For the species of the Fraxinion plant community, the predictions reveal very few habitats as suitable under both climate change scenarios (see Figure A2 for individual species). This is especially crucial as Fraxinion habitats are biodiversity hotspots but have long regeneration times (50 to 200 years, see in Werth et al. 2012; Delarze and Gonseth 2015). Additionally, these habitats are already fragmented now (Delarze and Gonseth 2015), and loss of additional suitable sites might further accelerate the loss of some of its vulnerable plant members. Increasing temperatures are likely enhancing the succession of floodplain forests to deciduous forest by lack of moisture (Lang et al. 2014; Delarze and Gonseth 2015). Given that Fraxinion floodplain forests are only occasionally flooded and that the overflow is mainly providing fine sediments (Ellenberg 2010), future extreme floods transporting also larger sediments might change the composition of the plant community considerably. Moreover, as most species have limited dispersal abilities (Table 1), conservation planning requires to link dispersal and habitat suitability information to provide target areas for sanctuaries for plant communities (Engler et al. 2009).
Dispersal simulation under climate change assumptions
The presented approach allows to explore various future scenarios to reveal the main driver of species loss by implementing information on dispersal as well as different changes in climate conditions (Engler and Guisan 2009; Engler et al. 2009). Dispersal is more important in structuring metacommunities than environmental processes in high mountain streams (Dong et al. 2016), which is also visible in the spatially explicit output of the simulations for the floodplain forest species investigated here (Fig. 4). While many habitats are suitable (compare Figure A2), the dispersal limitations as well as the information on life history traits implemented in the simulation process limit the spread of the species. Within each plant community group, dispersal vectors varied (see Table 1), but while species which are only dispersed by gravity indeed showed the smallest spread (Myosoton aquaticum and S. officinale, both within Salicion albae, see Fig. 4), high differences were found between wind dispersed species. The spread of various Carex species was simulated under similar assumptions, but C. strigosa was the least successful colonizer under extreme climate change scenarios of all Carex species (Fig. 4). This fact was not due to lower initial presence points used in the simulation, as both C. strigosa and C. brizoides showed comparable starting points for dispersal simulation (see Table 2), but due to differences in habitat suitability.
Another important point to consider in the interpretation of the results of the simulation are the changes in habitat suitability between various periods in the future, as seen in the analysis on suitable, but not colonized cells (Fig. 2). There, the pattern at each catchment level is more complex than simple dispersal limitations, as not all species showed more cells not being colonized when considering shorter time periods (50 instead of 80 years of dispersal, see Fig. 3). This is in agreement with changing habitat suitability predictions for each catchment between various time periods for several species (see Figure A3, e.g. C. strigosa, habitat suitability for Ticino catchment, and compare to Fig. 2, Ticino catchment). Conservation planning is usually done at the catchment level, making information on predicted changes important, especially as riparian species form metacommunities along catchments (Patrick et al. 2014; Werth and Scheidegger 2014).
While at the plant community level differences between climate change scenarios where significant in terms of colonized cells after 50 and 80 years of dispersal, the significant difference were no longer detected at the individual species level (for softwood floodplain forest species mainly, see results). The individual species were therefore either able to cope with changing climate per se (moderate or extreme), or not able to colonize cells in the future. To predict the outcome of target species for conservation planning, forecasts need be based on individual species’ life history and dispersal traits, with a lower focus on changes in habitat suitability due to various climate change scenarios This is particularly important for restoration projects with post-restoration success control studies, as the prediction using the method presented here allows to distinguish more resilient and vulnerable species.
In our study, some plant species clearly reveal the potential to colonize many areas along catchments in Switzerland, even under extreme climate change scenarios (e.g. P. alba, P. padus, S. dulcamara, see Fig. 4). These species are all considered to be generalists and have a IUCN criteria conservation status of least concern (Delarze and Gonseth 2015), while the species for both softwood and hardwood floodplain forest which are threatened are the least successful also under future climate change scenarios (E. rhodanensis, conservation status vulnerable; M. monophyllos, conservation status vulnerable; U. laevis, conservation status endangered). Especially for these vulnerable species, future sanctuaries are crucial for conservation success.
Floodplains of National Importance and future refugial areas
Functional floodplains should consist of a mosaic of habitats, allowing different plant communities to co-occur in close proximity (Naiman et al. 2005; Ellenberg 2010). Our study reveals that Floodplains of National Importance are not exclusive sanctuaries for floodplain forest species in Switzerland, as for a majority of the plant species less than half of the cells within the designated areas are predicted to harbor these characteristic species (Fig. 3).
The percentage of cells remaining occupied or being colonized in the future within these protected areas further decreases both under moderate and extreme climate change scenarios (Fig. 3). Still, the current protected floodplains might be – or might even become—of importance for other species. Similar shifts have been found for other riparian species in Switzerland (e. g marginalized habitat for endangered Myricaria germanica and Salix daphnoides, see in Fink and Scheidegger 2018) and for other plant species in other parts of the world, where only low percentages of currently protected areas are refugia under climate change (Grahama et al. 2019; Hoffmann et al. 2019). This method therefore reveals the potential to plan new target areas for riparian plant conservation in Switzerland, both under changing climate as well as dispersal and life history traits (see also Ikeda et al. 2014; Fink and Scheidegger 2018).
The spatial distribution of current and future available habitat (Fig. 4) stresses the importance of connectivity of sanctuaries along river catchments: Given the frequent dependence on wet areas for germination (Schutz 1997; Ludewig et al. 2014; Sameel et al. 2014), riparian plant species can only establish along a relatively small corridor along rivers, and the spread of species is moderate due to limited dispersal abilities (Andersson et al. 2000; Boedeltje et al. 2003). Additionally, especially plants of the Salicion albae community need frequent sediment turnover allowing them to persist despite high competition on gravel bars (Ellenberg 2010). Conservation planning therefore needs to focus on establishing connected sanctuaries areas allowing species survival both under changing climate as well as changing river dynamics.
Limitations of the modelling approach
Species distribution modelling has widely been applied for conservation biology (Guisan et al. 2013), even for rare species (Breiner et al. 2015). Nevertheless, modelling output is highly dependent on data accuracy as well as choice of environmental predictors and modelling method (Franklin 2013), and therefore precautions should be taken. In this study, we used a data set without real absence data, and therefore applied the method most suitable for this type of data (Phillips et al. 2009; Elith et al. 2011). Environmental predictor choice was based on the species’ ecology (see also in Camathias et al. 2013; Breiner et al. 2015; Fink and Scheidegger 2018). Switzerland covers various diverse biogeographic regions (Wohlgemuth 1996), allowing for predictions of various ecological niches along all the catchments included in the study as well as a wide range of values for the predictor layers.
The scale of the modelling approach is not only crucial for predictor variation, but should also be suitable for conservation and planning actions (Elith and Leathwick 2009). We used predictor layers with 100 × 100 m resolution, as we investigate mainly low distance dispersers, for which the landscape planning for habitat conservation also depends on local stepping stones (Darby and Sear 2008; Hannah et al. 2014b). By using different plant species to represent each community (e.g. trees, shrubs and herbaceous plants), and by focusing on species with high record abundance in Switzerland, the predicted species ranges allow to draw general conclusions for landscape planning undisturbed by individual species’ requirements, e.g. of rare species.
Models help understanding core processes in complex natural settings (Gorban and Roose 2011). Species distribution modelling offers the possibility to predict the fate of habitats given changing climate, and therefore allow for a better landscape planning by taking into account future conditions (Drew et al. 2011). While species distribution models alone have been shown to overestimate conservation potential especially for poor dispersers like plants (van-Loon et al. 2011), coupling with dispersal strategies and also including life history traits allow for more realistic predictions (Engler et al. 2009; Fink and Scheidegger 2018). As results of statistical downscaling of global scale climate models to regional scale can reveal uncertainties in climatic predictions (Laflamme et al. 2016), we used the comparison of two extreme climatic change assumptions (moderate and extreme) and averages of predictions over ten year periods to balance outliers for single years. Still, future analyses need to refine modelling approaches by defining the ecological niche of species more precisely, as e.g. for riparian species, hydrological parameters such as inundation frequency or sediment dynamics are crucial for defining the niche at the smaller scale within the riverscape (Gostner et al. 2017), and also the impact of water-mediated dispersal needs to be assessed (Nilsson et al. 2010).
Conclusions
Coupled models linking dispersal and changing habitat under different future climate conditions show non-persistent refugia for floodplain forest species. Although Salicion albae species in general are more likely to cope with predicted changes in the climatic conditions than Fraxinion species, there are still shifts in target catchments providing sanctuaries. Planning of protected areas for riparian plant species and communities need to focus on connectivity along rivers in Switzerland, and models can help to plan sites suitable for changes in management or for restoration, to maintain viable source populations in dynamic riverine habitats under changing climate.
Availability of data and materials
The datasets analysed during the current study are available from the corresponding author on reasonable request.
References:
Alldredge B, Moore G (2014) Assessment of riparian vegetation sensitivity to river hydrology downstream of major texas dam. River Res Appl 30:230–244
Andersson E, Nilsson C, Johansson ME (2000) Plant dispersal in boreal rivers and its relation to the diversity of riparian flora Journal of biogeography 27:1095–1106
BAFU (2011) Liste der National Prioritären Arten. Arten mit nationaler Priorität für die Erhaltung und Förderung, Stand 2011 Bundesamt für Umwelt, Bern Umwelt-Vollzug Nr 1103
Barrows CW et al (2020) Validating climate-change refugia: empirical bottom-up approaches to support management actions. Front Ecol Environ 18:298–306
Bill H-C (2000) Besiedlungsdynamik und Populationsbiologie charakteristischer Pionierpflanzenarten nordalpiner Wildflüsse. Philipps-Universität Marburg, PhD
Boedeltje G, Bakker J, Bekker R, van Groenendael J, Soesbergen M (2003) Plant dispersal in a lowland stream in relation to occurrence and three specific life-history traits of the species in the species pool. J Ecol 91:855–866
Breiner F, Guisan A, Bergamini A, Nobis M, Anderson B (2015) Overcoming limitations of modelling rare species by using ensembles of small models. Methods Ecol Evol 6:1210–1218
Breitbach N, Boehning Gaese K, Laube I, Schleuning M, Böhning Gaese K, Rees M (2012) Short seed-dispersal distances and low seedling recruitment in farmland populations of bird-dispersed cherry trees. J Ecol 100:1349–1358
Brown R, Peet R (2003) Diversity and invasibility of southern Appalachian plant communities. Ecology 84:32–39
Bundesrat DS (1992) Verordnung über den Schutz der Auengebiete von nationaler Bedeutung (Auenverordnung) vol 451.31.
Camathias L, Bergamini A, Küchler M, Stofer S, Baltensweiler A (2013) High-resolution remote sensing data improves models of species richness. Applied Veg Sci 16:539–551
Capon SJ et al (2013) Riparian ecosystems in the 21st century: hotspots for climate change adaptation? Ecosystems 16:359–381
Caponi F, Koch A, Bertoldi W, Vetsch DF, Siviglia A (2019) When does vegetation establish on gravel bars? observations and modeling in the alpine rhine river frontiers in environmental science
Cappers RTJ (1993) Seed dispersal by water: a contribution to the interpretation of seed assemblages. Vegetation History and Archaeobotany:173–186
Catford J, Jansson R (2014) Drowned, buried and carried away: effects of plant traits on the distribution of native and alien species in riparian ecosystems. New Phytol 204:19–36
Chen C, Ma M, Wu S, Jia J, Wang Y (2018) Complex effects of landscape, habitat and reservoir operation on riparian vegetation across multiple scales in a human-dominated landscape. Ecolog Indicators 94:482–490
Chmelar J, Meusel W (1979) Die Weiden Europas : die Gattung Salix. Die neue Brehm-Bücherei, 2., unveränderte Aufl edn. A. Ziemsen, Wittenberg Lutherstadt
Costanza R et al (1997) The value of the world’s ecosystem services and natural capital. Nature 387:253–260
Cushman S et al (2014) Landscape genetic connectivity in a riparian foundation tree is jointly driven by climatic gradients and river networks. Ecol Appl 24:1000–1014
D’Amen M, Bombi P, Pearman P, Schmatz D, Zimmermann N, Damen M, Bologna M (2011) Will climate change reduce the efficacy of protected areas for amphibian conservation in Italy? Biol Cons 144:989–997
Darby S, Sear D (2008) River restoration: managing the uncertainty in restoring physical habitat. Reprinted with, corrections. Wiley, Chichester, West Sussex, Chichester, West Sussex
Death R, Fuller I, Macklin M (2015) Resetting the river template: the potential for climate-related extreme floods to transform river geomorphology and ecology. Freshw Biol 60:2477–2496
Debeljak M, Ficko A, Brus R (2015) The use of habitat and dispersal models in protecting European black poplar (Populus nigra L.) from genetic introgression in Slovenia. Biol Cons 184:310–319
Delarze R, Gonseth Y (2015) Lebensräume der Schweiz : Ökologie, Gefährdung, Kennarten vol 3., vollständig überarbeitete Auflage. 2., Ott, Thun
Delarze R et al. (2013) Liste Rouge des habitats de Suisse et Liste des habitats prioritaires de Suisse. Office fédéral de l’environnement OFEV,
Dong X et al (2016) Flow directionality, mountain barriers and functional traits determine diatom metacommunity structuring of high mountain streams. Scientific Reports 6:24711
Drew CA, Wiersma YF, Huettmann F (eds) (2011) Predictive Species and Habitat Modeling in Landscape Ecology: Concepts and Applications. 1st ed. 2011. edn. Springer New York, New York, NY
Driscoll D, Banks S, Barton P, Lindenmayer D, Smith A (2013) Conceptual domain of the matrix in fragmented landscapes. Trends Ecol Evol 28:605–613
Dufour S, Rodríguez González P, Laslier M, Rodriguez Gonzalez P (2019) Tracing the scientific trajectory of riparian vegetation studies: Main topics, approaches and needs in a globally changing world. Science Total Environ 653:1168–1185
Elith J, Leathwick J (2009) Species distribution models: ecological explanation and prediction across space and time Annual Review of Ecology. Evolution, and Systematics 40:677–697
Elith J, J, Phillips S, Hastie T, Dudik M, Chee Y, Dudík M, Yates C, (2011) A statistical explanation of MaxEnt for ecologists. Diversity Distributions 17:43–57
Ellenberg H (2010) Vegetation Mitteleuropas mit den Alpen in ökologischer, dynamischer und historischer Sicht. UTB für Wissenschaft : Uni-Taschenbücher, 6., vollst. neu bearb. und stark erw. Aufl edn. Ulmer, Stuttgart
Engler R, Guisan A (2009) MigClim: Predicting plant distribution and dispersal in a changing climate. Diversity Distributions 15:590–601
Engler R et al (2009) Predicting future distributions of mountain plants under climate change: does dispersal capacity matter? Ecography 32:34–45
Engler R, Hordijk W, Guisan A (2012) The MIGCLIM R package - seamless integration of dispersal constraints into projections of species distribution models. Ecography 35:872–878
ESRI (2015) ArcGIS Desktop: Release 10.2.2. Environmental Systems Research Institute, CA
Federal Act on the Protection of Nature and Cultural Heritage (1966). Berne
Fink S, Scheidegger C (2018) Effects of barriers on functional connectivity of riparian plant habitats under climate change. Ecol Eng 115:75–90
Fink S, Lanz T, Stecher R, Scheidegger C (2017) Colonization potential of an endangered riparian shrub species. Biodivers Conserv 26:2099–2114
Francis R, Gurnell A, Petts G, Edwards P (2005) Survival and growth responses of Populus nigra, Salix elaeagnos and Alnus incana cuttings to varying levels of hydric stress. Forest Ecology Management 210:291–301
Franklin J (2013) Species distribution models in conservation biogeography: developments and challenges. Divers Distrib 19:1217–1223
Gage E, Cooper D (2005) Patterns of willow seed dispersal, seed entrapment, and seedling establishment in a heavily browsed montane riparian ecosystem. Canadian J Botany 83:678–687
Garssen AG, Verhoeven JTA, Soons MB (2014) Effects of climate-induced increases in summer drought on riparian plant species: a meta-analysis. Freshw Biol 59:1052–1063
Gizaw A et al (2016) Vicariance, dispersal, and hybridization in a naturally fragmented system: the afro-alpine endemics Carex monostachya and C. runssoroensis (Cyperaceae). Alpine Botany 126:59–71
Gorban AN, Roose D (eds) (2011) Coping with Complexity: Model Reduction and Data Analysis. Lecture Notes in Computational Science and Engineering, 1st ed. 2011. edn. Springer, Berlin Heidelberg
Gostner W, Paternolli M, Schleiss AJ, Scheidegger C, Werth S (2017) Gravel bar inundation frequency: an important parameter for understanding riparian corridor dynamics. Aquat Sci 79:825–839
Grahama V, Baumgartnera JB, Beaumonta LJ, Esperón-Rodríguez M, Grech A (2019) Prioritizing the protection of climate refugia: designing a climate-ready protected area network. J Environ Planning Manage 62:2588–2606
Guisan A, Zimmermann N (2000) Predictive habitat distribution models in ecology. Ecol Model 135:147–186
Guisan A et al (2013) Predicting species distributions for conservation decisions. Ecol Lett 16:1424–1435
Hannah L, Flint L, Syphard A, Moritz M, Buckley L, McCullough I (2014a) Fine-grain modeling of speciesâTM response to climate change: holdouts, stepping-stones, and microrefugia. Trends Ecol Evol 29:390–397
Hannah L, Flint L, Syphard AD, Moritz MA, Buckley LB, McCullough IM (2014b) Fine-grain modeling of species’ response to climate change: holdouts, stepping-stones, and microrefugia. Trends Ecol Evol 29:390–397
Hausammann A, Gsteiger P, Roulier C, Righetti A, Thielen R (2005) Faktenblatt Auen 11: Das Aueninventar. Bundesamt für Umwelt, Wald und Landschaft
Heller N, Zavaleta E (2009) Biodiversity management in the face of climate change: A review of 22 years of recommendations. Biol Conserv 142:14–32
Hijmans RJ, van Etten J (2012) raster: Geographic analysis and modeling with raster data. R package version 2.0–12. http://CRAN.R-project.org/package=raster.
Hirzel A, Randin C, Guisan A, Le Lay G, Helfer V (2006) Evaluating the ability of habitat suitability models to predict species presences. Ecol Model 199:142–152
Hoffmann S, Beierkuhnlein C, Irl SDH (2019) Predicted climate shifts within terrestrial protected areas worldwide Nature. Communications 10:4787–4787
Ikeda D, Grady K, Shuster S, Whitham T, Boldgiv B (2014) Incorporating climate change and exotic species into forecasts of riparian forest distribution. PLoS ONE 9:e107037
Inoue K, Berg D (2017) Predicting the effects of climate change on population connectivity and genetic diversity of an imperiled freshwater mussel, Cumberlandia monodonta (Bivalvia: Margaritiferidae), in riverine systems. Glob Change Biol 23:94–107
Jansson R, Nilsson C, Renöfält B (2000) Fragmentation of riparian floras in rivers with multiple dams. Ecology 81:899–903
Jermakowicz E, Wróblewska A, Brzosko E, Mirski P, Hirse T (2015) Phylogeographical structure of the boreal-montane orchid Malaxis monophyllos as a result of multi-directional gene flow. Bot J Linn Soc 178:138–154
Jermakowicz E, Brzosko E, Kotowicz J, Wróblewska A (2017) Genetic Diversity of Orchid Malaxis monophyllos Over European Range as an Effect of Population Properties and Postglacial Colonization. Polish Journal of Ecology 65:69–86
Kuzovkina Y, Knee M, Quigley M (2004) Effects of soil compaction and flooding on the growth of 12 Willow (Salix L.) species. J Environ Horticulture 22:155–160
Laflamme EM, Linder E, Pan Y (2016) Statistical downscaling of regional climate model output to achieve projections of precipitation extremes. Weather Climate Extremes 12:15–23
Lang P, Ewald J, Schaminée J (2014) Predictive modelling and monitoring of Ellenberg moisture value validates restoration success in floodplain forests. Applied Veg Sci 17:543–555
Lauber K (2018) Flora Helvetica : illustrierte Flora der Schweiz : mit Artbeschreibungen und Verbreitungskarten von 3200 wild wachsenden Farn- und Blütenpflanzen, einschliesslich wichtiger Kulturpflanzen. 6., vollständig überarbeitete Auflage, Haupt Verlag, Bern
Lind L, Hasselquist E, Laudon H (2019) Towards ecologically functional riparian zones: A meta-analysis to develop guidelines for protecting ecosystem functions and biodiversity in agricultural landscapes. J Environ Manage 249:109391–109391
Liu C, White M, Newell G, Pearson R (2013) Selecting thresholds for the prediction of species occurrence with presence-only data. J Biogeography 40:778–789
Ludewig K, Zelle B, Eckstein RL, Mosner E, Otte A, Donath T (2014) Differential effects of reduced water potential on the germination of floodplain grassland species indicative of wet and dry habitats. Seed Sci Res 24:49–61
Machar I, Schlossarek M, Pechanec V, Uradnicek L, Praus L, Sıvacıoğlu A (2019) Retention Forestry Supports Bird Diversity in Managed. Temperate Hardwood Floodplain Forests Forests 10:300
Manel S, Williams HC, Ormerod SJ (2001) Evaluating presence-absence models in ecology: the need to account for prevalence. J Appl Ecol 38:921–931
Marquez Corro J et al (2017) Long-distance dispersal explains the bipolar disjunction in Carex macloviana. Am J Botany 104:663–673
Morelli T et al (2016) Managing Climate Change Refugia for Climate Adaptation. PLoS ONE 11:e0159909–e0159909
Morelli TL et al (2020) Climate-change refugia: biodiversity in the slow lane. Front Ecol Environ 18:228–234
Mosner E et al (2015) Climate change and floodplain vegetation—future prospects for riparian habitat availability along the Rhine River. Ecol Eng 82:493–511
Müller P (1955) Verbreitungbiologie der Blütenpflanzen. Veröffentlichungen des Geobotanischen Institutes Rübel in Zürich, vol 30. Verlag Hans Huber, Bern
Müller N, Scharm S (2001) The importance of seed rain and seed bank for the recolonization of gravel bars in alpine rivers. In: Okuda ARPDS (ed) Studies on the vegetation of alluvial plants. Yokohama National University, Yokohama, pp 127–140
Müller-Schneider P (1986) Verbreitungsbiologie der Blütenpflanzen Graubündens : Diasporology of the Spermatophytes of the Grisons (Switzerland), vol 85
Naiman RJ, Décamps N, McClain ME (2005) Riparia: Ecology, conservation, and management of streamside communities. Elsevier, New York
Neumann A (1981) Die mitteleuropäischen Salix-Arten. Oesterreichischer Agrarverlag, Wien, Mitteilungen der forstlichen Bundes-Versuchsanstalt Wien
Nilsson C, Brown R, Jansson R, Merritt D (2010) The role of hydrochory in structuring riparian and wetland vegetation Biological reviews 85:837–858
Patrick C, Cooper M, Uzarski D (2014) Dispersal Mode and Ability Affect the Spatial Turnover of a Wetland Macroinvertebrate Metacommunity. Wetlands 34:1133–1143
Phillips S, Anderson R, Schapire R (2006) Maximum entropy modeling of species geographic distributions. Ecol Model 190:231–259
Phillips S et al (2009) Sample selection bias and presence-only distribution models: implications for background and pseudo-absence data. Ecol Appl 19:181–197
Pijl Lvd (1982) Principles of dispersal in higher plants. 3rd., rev. and expanded ed. edn. Springer, Berlin, Heidelberg, New York
Pollux BJA, Luteijn A, Van-Groenendael JM, Ouborg NJ, Ouborg NJ (2009) Gene flow and genetic structure of the aquatic macrophyte Sparganium emersum in a linear unidirectional river. Freshw Biol 54:64–76
Pracheil BM, Mestl GE, Pegg MA (2015) Movement through dams facilitates population connectivity in a large river. River Res Appl 31:517–525
Pullin A et al (2009) Conservation focus on europe: major conservation policy issues that need to be informed by conservation science. Conservation Biol 23:818–824
R Core Team (2016) R: A language and environment for statistical computing. R Foundation for Statistical Computing. http://www.R-project.org/.
Rakosy D (2014) Zur Bestäubungsbiologie dreier seltener Orchideenarten in Kärnten: Epipogium aphyllum. Malaxis monophyllos und Hammarbya paludosa Carinthia II 204:589–618
Reichert P et al (2007) Concepts of decision support for river rehabilitation. Environ Modell Softw 22:188–201
Rivaes R, Rodriguez Gonzalez P, Egger G, Rodríguez González P, Albuquerque A, Pinheiro A, Ferreira MT (2015) Reducing river regulation effects on riparian vegetation using flushing flow regimes. Ecol Eng 81:428–438
Sabo J et al (2005) Riparian zones increase regional species richness by harbouring different, not more, species. Ecology 86:56–62
Sameel J, Janssen R, Rip W, Rip WJ, Bakker E, Hooftman D (2014) Windows of opportunity for germination of riparian species after restoring water level fluctuations: a field experiment with controlled seed banks. J Appl Ecol 51:1006–1014
Schnitzler-Lenoble A, Carbiener R (2007) Forêts alluviales d'Europe. Tec & Doc.
Schutz W (1997) Are germination strategies important for the ability of cespitose wetland sedges (Carex) to grow in forests? Can J Bot 75:1692–1699
Seavy NE et al (2009) Why climate change makes riparian restoration more important than ever: recommendations for practice and research. Ecol Restor 27:330–338
Thornton P, Running S, White M (1997) Generating surfaces of daily meteorological variables over large regions of complex terrain Journal of hydrology 190:214–251
United Nations (2018) Sustainable Development Goal 6. Synthesis Report on Water and Sanitation, United Nations, New York
van-Loon AH, van Loon AH, Soomers H, Schot PP, Bierkens MFP, Griffioen J, Wassen MJ, (2011) Linking habitat suitability and seed dispersal models in order to analyse the effectiveness of hydrological fen restoration strategies. Biol Cons 144:1025–1035
Venturas M, Nanos N, Gil L (2014) The reproductive ecology of Ulmus laevis Pallas in a transformed habitat. For Ecol Manage 312:170–178
Vittoz P, Cherix D, Gonseth Y, Lubini V, Maggini R, Zbinden N, Zumbach S (2013) Climate change impacts on biodiversity in Switzerland: A review Journal for nature conservation 21:154–162
Ward JV, Tockner K, Schiemer F (1999) Biodiversity of floodplains river ecosystems: ecotones and connectivity Regulated Rivers: Research and Management 15 doi:3.0.co;2-e
Welch BL (1947) The generalization of “Student’s” problem when several different population variances are involved. Biomerika 34:28–35
Werth S, Scheidegger C (2014) Gene flow within and between catchments in the threatened riparian plant Myricaria germanica. PLoS ONE 9:e99400
Werth S, Alp M, Karpati T, Gostner W, Scheidegger C, Peter A (2012) Biodiversität in Fliessgewässern vol Merkblatt 2. BAFU, Bern
Wohlgemuth T (1996) A biogeographical division of Switzerland based on a floristic approach. Bot Helv 106:227–254
Wubs ERJ et al (2016) Going against the flow: a case for upstream dispersal and detection of uncommon dispersal events. Freshw Biol 61:580–595
Acknowledgements
We thank three anonymous reviewers and the editor for comments to improve a previous version of the manuscript. The Swiss data base for plants, www.infoflora.ch, provided information on species occurrence. We acknowledge the World Climate Research Program’s Working Group on Regional Climate, and the Working Group on Coupled Modelling, former coordinating body of CORDEX and responsible panel for CMIP5 and the climate modelling groups for producing and making available their model output. Dirk Schmatz provided access to future climatic data downscaled for Switzerland. Funding was provided by the Swiss Federal Offices for the Environment for the project “riverscape – sediment dynamics and connectivity”.
Funding
Open Access funding provided by Lib4RI – Library for the Research Institutes within the ETH Domain: Eawag, Empa, PSI & WSL. Funding was provided by the Swiss Federal Offices for the Environment for the project “Riverscape – sediment dynamics and connectivity”.
Author information
Authors and Affiliations
Contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Sabine Fink. The first draft of the manuscript was written by Sabine Fink and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Conflict of interest
The authors have no conflicts of interest to declare.
Consent for publication
All authors approved the manuscript for publication.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Fink, S., Scheidegger, C. Changing climate requires shift from refugia to sanctuaries for floodplain forests. Landscape Ecol 36, 1423–1439 (2021). https://doi.org/10.1007/s10980-021-01224-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10980-021-01224-8