Skip to main content

Genetic diversity of calcareous grassland plant species depends on historical landscape configuration



Habitat fragmentation is considered to be a main reason for decreasing genetic diversity of plant species. However, the results of many fragmentation studies are inconsistent. This may be due to the influence of habitat conditions, having an indirect effect on genetic variation via reproduction. Consequently we took a comparative approach to analyse the impact of habitat fragmentation and habitat conditions on the genetic diversity of calcareous grassland species in this study. We selected five typical grassland species (Primula veris, Dianthus carthusianorum, Medicago falcata, Polygala comosa and Salvia pratensis) occurring in 18 fragments of calcareous grasslands in south eastern Germany. We sampled 1286 individuals in 87 populations and analysed genetic diversity using amplified fragment length polymorphisms. Additionally, we collected data concerning habitat fragmentation (historical and present landscape structure) and habitat conditions (vegetation structure, soil conditions) of the selected study sites. The whole data set was analysed using Bayesian multiple regressions.


Our investigation indicated a habitat loss of nearly 80% and increasing isolation between grasslands since 1830. Bayesian analysis revealed a significant impact of the historical landscape structure, whereas habitat conditions played no important role for the present-day genetic variation of the studied plant species.


Our study indicates that the historical landscape structure may be more important for genetic diversity than present habitat conditions. Populations persisting in abandoned grassland fragments may contribute significantly to the species’ variability even under deteriorating habitat conditions. Therefore, these populations should be included in approaches to preserve the genetic variation of calcareous grassland species.


Calcareous grasslands are important hotspots of plant species diversity in central Europe. They contain many rare and endangered plant species and are of strong conservation interest [13]. However, due to land use changes, calcareous grasslands declined significantly in Europe during the last 150 years [4]. Agricultural intensification, increased fertilization and afforestation caused a drastic loss of grasslands. In some regions, up to 90% of the grasslands disappeared [5]. Today calcareous grasslands are thus often highly fragmented, with the area of the grassland patches continuously decreasing while their spatial isolation increases [6]. This process of habitat fragmentation is a general threat to biodiversity, reducing species richness within small and isolated habitat patches [7].

However, fragmentation has not only an impact on biodiversity at the species level, but also on the genetic diversity, due to geographic isolation of smaller populations [8]. In particular formerly widespread species are more susceptible to the effects of fragmentation than naturally rare and isolated species [9]. Genetic variation is directly related to population size [10] and often decreases during the process of fragmentation. Moreover, the exchange of pollen and seeds between populations is impeded by fragmentation, which decreases gene flow [11] and increases genetic drift. This results in a loss of genetic diversity and may lead to reduced generative [12] and vegetative performance [13]. As a consequence, there is an increase in the susceptibility to pathogens and herbivores in the short term [14] and the probability of extinction in the long term [8]. Finally, the loss of genetic diversity may explain the observed loss of species diversity in numerous studies analysing fragmentation effects [6].

However, the results of many studies dealing with the impact of fragmentation on genetic diversity are inconsistent. Some of these studies support the assumptions derived from the theory of island biogeography [15], whereas others do not [16]. This dilemma has recently been illustrated by a review of 259 fragmentation studies, which concluded that the broad generalisations on the effects of fragmentation are problematic [17].

One of the most important challenges of fragmentation studies is that habitat conditions may differ between remnant grassland patches [17]. Many calcareous grasslands have been abandoned because sheep grazing is no longer economical [1, 2]. In the absence of grazing, habitat conditions continuously deteriorate and typical open short-grass conditions get lost [18]. This process is enhanced by the spill-over of fertiliser from adjacent agricultural areas [18]. Absence of grazing and increased nitrogen deposition change the vegetation structure of calcareous grasslands [19], resulting in a decrease in species richness [20] and shifts in species composition [21]. Due to the lack of biomass removal and increasing productivity litter accumulates while gaps of open soil, which are indispensable for the germination of many dry grassland species, become rare [22]. Furthermore, litter acts as a trap, further reducing the number of germinating seeds [23]. Particularly for species which require light to germinate, ground shadowing caused by increasing vegetation height, dominance of grasses and litter accumulation leads to the regression those of species [24]. Therefore, successful reproduction is impeded when habitat conditions deteriorate, which may subsequently affect genetic diversity. This process is intensified by the loss of suitable dispersal vectors, such as migrating sheep, which also reduces the exchange of seeds [2] and subsequently, the immigration of genetically deviant individuals.

Hence, genetic diversity may be affected by the interfering effects of both fragmentation and habitat conditions, which might explain the inconsistent results of many fragmentation studies. Moreover, due to different biological traits, patterns of genetic variation differ between plant species [25] and these may react completely different to fragmentation, again explaining unclear results of many genetic fragmentation surveys.

In this study, we analysed the impact of fragmentation on genetic diversity of several calcareous grassland species in a comparative approach, including both habitat fragmentation and habitat conditions. We used the same analytical approach for all species and applied Bayesian multiple regressions, which enables a detailed interpretation of the data, while being more flexibly adaptable to the data structure than traditional frequentist methods. More specifically we ask the following questions: (i) has fragmentation of calcareous grasslands a significant impact on genetic diversity of plant populations? (ii) is historical landscape configuration more important for genetic diversity than present landscape configuration? (ii) is genetic diversity of grassland plant populations affected by habitat conditions?


Study sites, habitat fragmentation and habitat conditions

For our study, we randomly selected 18 remnant calcareous grasslands in the valleys of the rivers Naab and Laber on the Franconian Alb in south eastern Germany near Regensburg (Fig. 1). Within the study region, calcareous grasslands have been subjected to fragmentation due to afforestation, intensification and abandonment since the nineteenth century. The study sites and all other calcareous grasslands occurring within a radius of 3 km around these sites were vectorised using a Geographic Information System (Arc Info 10.0, Esri) based upon corrected aerial photos (orthophotos) from August to October 2013 to study the effects of this fragmentation process. Vectorised data was used to calculate the current area (HA2013) and perimeter (P) of each grassland fragment as well as the distance (D2013) to the nearest calcareous grassland within the 3 km radius. The shape of the study sites was characterised by the ratio of habitat area to perimeter (HA/P2013), which was small for narrow and elongated grasslands and large for round and compact grasslands.

Fig. 1

Geographic location of the 18 selected study sites (labelled dots) in the valleys of Naab and Laber on the Franconian Alb in south eastern Germany near Regensburg and all other calcareous grasslands (grey areas) within a radius of 3 km around the study sites in 1830 and 2013

Using historical cadastral maps, which were available from local land surveying offices, we determined the area of the study sites (HA1830) and the distance (D1830) to the nearest calcareous grassland within the 3 km radius in 1830. The maps include a detailed legend, which allows the identification of calcareous grasslands. We compared the area covered by calcareous grasslands in 1830 and 2013 and calculated the habitat loss (HL) within the 3 km radius around each of our study sites as a percentage of the grassland area lost since 1830. Additionally, connectivity of each fragment to all other fragments within the 3 km radius in 2013 and 1830 was calculated according to Hanski [26] as \(Si = \sum\nolimits_{{j \ne i}} {\exp \left( { - \upalpha d_{{{\text{ij}}}} } \right)A_{{\text{j}}} }\) where Si is the connectivity of the patch i, d ij is the distance between patches i and j, and Aj is the area of the patch j [27]. Based upon the historical maps, grasslands were classified for further analyses as historically old, when they were already grassland in 1830, and historically young grasslands, when they originated after 1830.

The selected grasslands date back at least to the period of the Roman Empire [28] and have been grazed frequently until the 1960s, as have most other grasslands in central Europe [4]. Today they are abandoned or infrequently grazed. However, detailed information about the grazing history since the 1960s is not available. In 2014, we established ten study plots with the size of 2 × 2 m at each of the selected grasslands to analyse the impact of the habitat conditions on genetic diversity of our study species. In each plot we determined the vegetation height (VH) as well as the cover of grass (CG), litter (CL) and bare soil (BS). Furthermore, we took five soil samples at each study site with a core sampler, which were then pooled, in order to analyse the nutrient content of the soil. Pooled samples were dried in a heating cabinet at 50 °C for several days, cleaned by sieving with 2 mm mesh size and then stored at 4 °C until they were subjected to a soil chemical analysis following the procedures described by Bassler et al. [29]. We determined the phosphorous (P) and potassium (K) content, as well as the carbon to nitrogen ratio (C/N) as described previously [30].

Study species and genetic variation

For our study we selected five typical and widespread calcareous grassland species (Primula veris L., Dianthus carthusianorum L., Medicago falcata (L.) Arc., Polygala comosa Schkuhr and Salvia pratensis L.), frequently occurring in calcareous grasslands in south eastern Germany. In the field we assessed population size (NI) by counting the number of present individuals (Table 1) at each site. For the analysis of genetic variation with a few exceptions (at three sites P. comosa could not be sampled), leaf material of 15 individuals per population and species was collected (Table 5). In total 1286 individuals from 87 populations were analysed. Leaf material was placed in plastic bags in the field and stored in a lab freezer at −20 °C until molecular analysis.

Table 1 Size of the studied populations

Genomic DNA was isolated from dry leaf material using the CTAB-based method [31] as described before [32]. Concentrations of the DNA extracts were measured photometrically. DNA solutions were diluted with water to 7.8 ng/μL and used for the analysis of Amplified Fragment Length Polymorphisms (AFLPs), which were conducted concordant with the protocol from Beckmann Coulter as described previously [33, 34].

DNA adapters were prepared by adding equal volumes of both single strands of EcoRI (4 µM) and MseI (40 µM) adaptors (MWG Biotech), following a 5 min heating at 95 °C with a final 10 min step at 25 °C. DNA restriction and adapter ligation were performed in one step by adding a 3.6 μL mixture per reaction containing 2.5 U EcoRI (MBI Fermentas), 2.5 U MseI (MWG Biotech), 0.1 μM EcoRI and 1 μM MseI adapter pair, 0.5 U T4 Ligase with 0.1× of its corresponding buffer (MBI Fermentas), 0.05 M NaCl and 0.5 μg BSA (New England BioLabs) to 6.4 μL of genomic DNA in a concentration of 7.8 ng/μL. Following an incubation at 37 °C for 2 h with a final enzyme denaturation step at 70 °C for 15 min, the restriction-ligation products were diluted tenfold with 1× TE buffer for DNA (20 mM Tris–HCl, pH 8.0; 0.1 mM EDTA, pH 8.0).

For preselective DNA amplification, 1 μL diluted DNA restriction-ligation product, 0.25 µM preselective EcoRI and MseI primers (MWG Biotech) were added to an AFLP Core Mix (PeqLab, Germany) containing 1x Buffer S, 0.4 mM dNTPs and 1.25 U/µL Taq-Polymerase. In a 5 μL reaction volume PCR was performed on at 94 °C for 2 min then 30 cycles of 20 s denaturation at 94 °C, 30 s annealing at 56 °C and 2 min elongation at 72 °C, a final 2 min 72 °C and 30 min 60 °C step for complete extension ending with a final cool down to 4 °C. After PCR, products were diluted 20 fold with 1× TE buffer for DNA.

After an extensive screening of 30 primer combinations, three primer combinations per species were chosen for a subsequent selective PCR reaction. For detection, EcoRI primers labelled with different fluorescent dyes (Beckman Coulter) were used (Table 2). Selective PCR was carried out in a total reaction volume of 5 μL containing an AFLP Core Mix (1× Buffer S, 0.4 mM dNTP’s, 1.25 U/µL Taq-Polymerase, PeqLab, Germany), 0.05 μM selective EcoRI (Proligo, France), 0.25 μM MseI (MWG Biotech) primers and 0.75 μL diluted preselecive amplification product. For detection, EcoRI primers labelled with different fluorescent dyes (D2, D3, D4) were used. PCR parameters used were treated 2 min at 94 °C, 10 cycles 20 s denaturation at 94 °C, annealing 30 s at 66 °C and 2 min elongation at 72 °C, where annealing temperature was reduced every subsequent step by 1 °C, additional 25 cycles of 20 s denaturation at 94 °C, 30 s annealing at 56 °C and 2 min elongation at 72 °C completed by a following 30 min step at 60 °C and a cool down to 4 °C.

Table 2 Selective primer pairs used for AFLP analysis of the study species

Selective PCR products were diluted with 1× TE0.1 buffer for AFLP and pooled. After pooling 5 μL of each selective PCR product of a given sample and adding them to a mixture of 2 μL sodium acetate (3 M, pH 5.2), 2 μL Na2EDTA (100 mM, pH 8) and 1 μL glycogen (20 mg/mL; Roche), DNA was precipitated in a 1.5 mL tube by adding 60 μL of 96% ethanol (−20 °C) and an immediate shaking. DNA was pelleted by 20 min centrifugation at 14,000g at 4 °C, the supernatant was poured off and the pellet was washed once by adding 200 μL 70% ethanol (−20 °C) and centrifugation at the latter conditions and was subsequently vacuum dried in a concentrator [33, 34].

After redissolving the pelleted DNA in a mixture of 24.8 μL Sample Loading Solution (SLS, Beckman Coulter) and 0.2 μL CEQ Size Standard 400 (Beckman Coulter), selective PCR products were separated by capillary gel electrophoresis on an automated sequencer (GeXP, Beckmann Coulter). Results were examined using the GeXP software (Beckman Coulter) and analysed using the software Bionumerics 4.6 (Applied Maths, Kortrijk, Belgium). From the computed gels, only those fragments that showed intense and articulate bands were taken into account for further analyses. Samples yielding no clear banding pattern or obviously representing PCR artefacts were repeated. Finally, 19 individuals were excluded from the analysis due to the lack of a clear banding pattern. Reproducibility of molecular analyses was investigated with 10% of all analysed samples by means of estimating the genotyping error rate [35], which was 3.8%.

From the AFLP bands, a binary (0/1) matrix was created for each species. Based upon this matrix, we calculated the genetic diversity of each population as Nei’s Gene Diversity [36] using the program AFLP SURV [37].

Bayesian multiple regressions

The impact of habitat fragmentation and habitat conditions on diversity was analysed using a robust hierarchical Bayesian multiple regression approach with regression parameters for the species level (not shown) and an overarching set of hyperparameters for the species-independent estimation of regression parameters. Predictor variables were grassland fragmentation parameters (fragment area, area/perimeter ratio as well as distance to the nearest calcareous grassland and habitat connectivity) and habitat condition parameters (vegetation height, cover of grass, litter and bare soil, contents of P, K and C/N ratio). This approach is equivalent, though not identical, to a linear mixed model with random slopes and random intercepts using species as a random effect. The overarching hyperparameters ensure a transfer of information between the species’ parameters that shrinks outliers on species-level towards the main trend. The hyperparameters themselves are therefore suitable to depict the species-independent trends in the data.

There was no credible influence of habitat age detectable by the model and, hence, impeded accurate parameter estimation of the remaining parameters. Accordingly, habitat age as a parameter hampered model interpretation, and therefore was excluded from the final analysis. All predictor variables were checked for strong correlations (Pearson’s correlation coefficient >0.8; observed maximum 0.7) to check for multicollinearity. Additionally, multicollinearity can be recognised in Bayesian models by extremely broad posterior distributions of correlated parameters. However, this was not the case in the presented analysis. Model stability was verified by re-running the analysis with and without the most strongly correlated predictors. Accordingly, habitat age as a parameter hampered model interpretation and was, therefore, excluded from the final analysis.

A Bayesian approach was chosen for being flexibly adjustable to the situation at hand, e.g. it can be easily modified to reduce false positives in parameter estimation or improved to accommodate outliers in the data. In the situation at hand, the applied model could easily be tailored to reflect the hierarchical structure of our data (Additional file 1). Furthermore, results from Bayesian models have a higher informative value than classical NHST methods as they provide full probability distributions on the estimated parameters. Modelling and interpretation were carried out using the software packages R 3.2.1 [38] and JAGS 3.2.0 for Markov Chain Monte Carlo (MCMC) sampling [39] as well as utility functions provided by Kruschke [40]. Errors were modelled as being t-distributed in order to accommodate outliers and conduct a robust regression. Regression parameters were regularised using mildly informed, double-exponential prior distributions with location parameter set to 0 and a fixed precision parameter set to 0.1, thereby reducing chances for false positive regression parameters. These settings, known as the Bayesian Lasso [41] avoid overfitting in complex models and reduce the overestimation of effects that can happen in AIC-based model selection procedures.

Sampling was carried out with four MCMC chains with 300,000 steps in total with thinning set to every 10th step, a burn-in period of 2000 steps and 1000 steps for adaption. All parameters were checked for chain convergence. Autocorrelation in the MCMC chains was assessed as the effective sample size (ESS) aiming at a lower limit of 10 k for the relevant parameters. Highest density intervals (HDIs) were computed for the regression coefficients to check if coefficients were credibly non-zero. The predictors’ relative influences were assessed using standardised regression coefficients. A graph was produced by fixing all but the predictors of credible influence to their mean, resulting in a two dimensional scatter plot.


Habitat fragmentation and habitat conditions

Our results indicated a strong decline of calcareous grasslands within the study region. The mean size of the selected grasslands patches decreased from 115,045 m2 in 1830–14,881 m2 in 2013 (Table 3). In contrast, the mean distance to the nearest grassland increased from 110 m in 1830–210 m in 2013. Confirming this observation, mean connectivity among grasslands decreased from 74.81 in 1830–27.90 in 2013. Mean loss of calcareous grasslands within the 3 km radius around each of our study sites was 78.62%.

Table 3 Habitat fragmentation data

Habitat conditions strongly differed between study sites. Vegetation height ranged from 0.51 to 1.18 m with an average of 0.93 m (Table 4). Large differences could also be observed for the cover of grass, which varied between 48.0 and 90.0% with a mean of 76.5%. The cover of litter ranged from 7.7 to 38.0% with a mean of 20.4%, whereas the proportion of bare soil was minimum 0% and maximum 5.5% with an average value of 0.7%.

Table 4 Habitat condition data

The content of phosphorous also varied between sites and ranged from 8.04 to 53.76 mg/kg soil with a mean of 26.47 mg/kg soil (Table 4). Similarly, the content of potassium varied between 101.22 and 319.02 mg/kg soil. On average we observed a potassium content of 211.06 mg/kg soil. Finally, we determined the carbon to nitrogen ratio, which ranged from 10.9 to 42.0 with a mean of 19.6 (Table 4).

Genetic variation and hierarchical Bayesian multiple regression

AFLP analysis resulted in 120 fragments for Primula veris, 148 fragments for Dianthus carthusianorum, 285 fragments for Medicago falcata, 166 fragments for Polygala comosa and 192 fragments for Salvia pratensis. The proportion of polymorphic bands per species was 88.3% (P. veris), 97.4% (D. carthusianorum), 97.8% (M. falcata), 98.7% (P. comosa) and 94.3% (S. pratensis). Genetic diversity varied between species and populations. Mean genetic diversity of populations (Table 5) was highest in M. falcata (0.29), followed by S. pratensis (0.26), P. comosa (0.25) and D. carthusianorum (0.23). The lowest level of genetic diversity was observed in P. veris (0.18).

Table 5 Genetic diversity of the study species

Considering each species separately and all species together in the hierarchical Bayesian multiple regressions, we observed a credible impact of the distance to the nearest calcareous grassland in 1830 on the genetic variation within populations of the studied grassland species (Fig. 2, Tables 6, 7). The distance to the nearest calcareous grassland in 1830 was negatively correlated to the genetic variation within populations. This means that high levels of genetic variation have been detected at study sites which were closely located to other fragments in 1830. However, habitat area, today’s distance to the nearest calcareous grassland, habitat shape, habitat conditions and population size had no impact on the genetic diversity of the species.

Fig. 2

Relationship between genetic diversity (GD) and distance to the nearest calcareous grassland in 1830 (D1830) on 18 selected calcareous grasslands in south eastern Germany displayed as two dimensional scatter plot based upon the results of the hierarchical Bayesian multiple regression. Dashed lines represent twenty randomly chosen steps from the MCMC chains and are added to depict the variability in the posterior distribution of the regression parameters. Note that while intercepts vary considerably due to different levels of gene diversity in the analysed species, slopes are uniformly negative

Table 6 Bayesian multiple regressions for each of the analysed plant species
Table 7 Hierarchical Bayesian multiple regression


Impact of habitat fragmentation on genetic diversity

In our study, we observed an impact of the historical landscape configuration on the genetic diversity of the study species, since it depended on the distance to the nearest calcareous grassland in 1830 in the hierarchical Bayesian regressions. During the process of fragmentation the area of the habitat patches usually decreases [8]. Consequently, the size of plant populations occurring in remnant calcareous grasslands also declines, which results in a decline in genetic diversity [10]. In our study, we also observed a strong reduction of the habitat size. The contemporary area of the study sites was on average only 10% of the area in 1830. However, we observed no significant relationship between genetic diversity of the five study species and population size. Neither the current habitat area, nor the actual number of individuals per grassland fragment had an impact on genetic diversity. Similar findings have been reported in other fragmentation studies [12, 42]. Lag effects, a delayed reaction of genetic diversity on the reduction of population size [43], which is comparable to the extinction debt reported for species diversity [44, 45], could be a reason for the observed lack of a relationship between genetic diversity and population size [46]. In this case, genetic diversity should then be related to the historical area of the study sites. However, our study provided no evidence for such a relationship, which means that genetic diversity may generally be determined by factors other than habitat area or population size. Stochastic gene flow and long term survival under highly fragmented conditions are often considered as reasons for this observation [9, 47]. It has also been stated that the absence of this relationship may occur when the habitat area rapidly changes relative to the generation time of the study species [15]. Indeed, the calcareous grasslands in this study were formerly widely distributed and may have exhibited more or less erratic gene flow due to grazing. In combination with the long term persistence of the grassland species [48] these factors may be the most likely explanation for the lack of relationship between habitat area or population size and genetic diversity in this study.

Aside from decreasing habitat area, isolation of calcareous grasslands is also an important consequence of habitat fragmentation [7, 8]. The continuous process of increasing isolation affects genetic variation both between and within populations since gene flow by seeds and pollen declines with increasing isolation [8]. As gene flow decreases, the effects of genetic drift and inbreeding are intensified [49]. This results in an increased level of genetic variation between populations and a progressive loss of genetic variation within populations [50].

Gene flow by pollen is normally restricted to the nearest vicinity of plant populations to distances of less than 1 km [51, 52]. However, rare pollination events may also allow gene flow over larger distances [53]. Therefore, gene flow among fragmented calcareous grasslands at a larger scale is mainly caused by endo- and ectozoochorous seed dispersal, especially from migration of sheep flocks [2]. It has been shown, that the genetic structure of plant populations depends on present landscape connectivity [54] and that genetic variation between populations is affected by geographic distance between populations [8]. Moreover, it has also been demonstrated that genetic diversity may depend on habitat age both in natural [55] and semi-natural [56] habitats. In contrast to previous studies, which reported higher levels of genetic diversity in populations from historically older habitat fragments of forests [57] or calcareous grasslands [58], we observed no impact of habitat age on genetic diversity in our analysis.

Surprisingly, neither historical nor present habitat connectivity had an impact on genetic diversity in our study. This may be traced back to the fact that the calculated connectivity reflects the spatial structure, but not the real gene flow, which may be strongly affected by the migration of sheep flocks [46]. However, we observed a relationship between genetic diversity of the grassland species and the distance to the nearest grassland patch in 1830. Therefore, the historical landscape configuration is more important for the genetic diversity of calcareous grassland species than the present landscape structure. The effect of historical landscape configuration on species diversity has so far been demonstrated in several studies [59, 60], whereas the impact on genetic diversity has scarcely been shown [27]. However, it has recently been reported that the genetic diversity of the grassland species Succisa pratensis depends on the historical landscape structure of the habitat [15], which supports the results of our study. Moreover, it has been shown that the genetic diversity of Dianthus carthusianorum depends on patch connectivity by shepherding [61] and that population disconnection can create a genetic bottleneck, even in the absence of a demographic collapse [62], which underlines the importance of historical gene flow for the level of current genetic diversity.

Impact of habitat conditions on genetic variation

The results of our analyses indicated no impact of habitat conditions on genetic diversity of the studied calcareous grassland species. It has been demonstrated that, alongside habitat fragmentation, changes in habitat conditions have a strong impact on the species richness and composition of remnant calcareous grasslands fragments [18, 21]. Lack of grazing and the accumulation of soil nutrients lead to the loss of the typical open short-grass vegetation structure, allowing for the existence of many less competitive herbs. Under the conditions of abandonment and due to increased levels of nutrients, grasses such as Brachypodium pinnatum become increasingly dominant [19] and litter accumulates [63]. As a consequence, species requiring light for germination decline due to the effects of ground shadowing [24]. For the calcareous grasslands studied here, it has been demonstrated, that species diversity strongly depends on vegetation height and litter cover. Lack of grazing is therefore the most important reason for the declining species diversity of the grasslands, whereas fragmentation aspects play no significant role.

In contrast to the impact of land use on species diversity, the relationship between land use and genetic diversity is much less clear. However, it has already been demonstrated, that seedling recruitment and establishment in grasslands are positively affected by grazing and the removal of litter [64]. Schleuning et al. [65] even stated that the grassland species Trifolium montanum is more threatened by the effects of habitat degradation in the short term, than by the effects of fragmentation. Since genetic diversity depends explicitly on the degree of sexual reproduction [66], it appears possible that habitat conditions may also have an effect on genetic diversity. For the grassland species Dianthus seguieri, it has recently been shown that increased vegetation height and coverage as well as a high proportion of graminoids due to land use abandonment reduce genetic diversity and seed set [67]. As previously reported, land use by grazing generally has a positive impact on genetic diversity [68] since it reduces the cover of litter, and therefore may stimulate sexual reproduction.

However, in this study genetic diversity depended neither on vegetation structure, nor on soil nutrient levels. One reason for this observation may lie in the life span of the investigated plant species. It has already been reported that the frequency of plant species in remnant calcareous grasslands depends mainly on their persistence [48]. All species included in this study are long-lived perennials [69] and since the process of habitat deterioration due to abandonment goes back only about 50 years, many individuals we analysed may have been established before litter accumulation reached a critical level. This means that impaired habitat conditions may not yet have resulted in decreased levels of genetic diversity. Another reason for the lack of relationship between habitat conditions and genetic diversity could be the persistence of seeds in the soil seed bank, which may have contributed to the regeneration of the populations and to the maintenance of genetic diversity within the studied populations [70].


The results of our study provide evidence that the genetic diversity of calcareous grassland plant species depends on historical landscape configuration, rather than on the present population size or habitat conditions. In practice, efforts to preserve calcareous grasslands mainly concentrate on large fragments exhibiting the typical open-shortgrass habitat conditions and high species diversity. However, a comprehensive conservation approach should also consider the genetic diversity of calcareous grassland plant species. From our results it can be concluded that populations in smaller grassland fragments may, depending on historical landscape configuration, substantially contribute to the genetic variation of the plant species even under conditions of habitat deterioration. Preferably, these populations should therefore be included in strategies to preserve calcareous grasslands as the local biodiversity hotspots they are.


  1. 1.

    Wallis De Vries MF, Poschlod P, Willems JH. Challenges for the conservation of calcareous grasslands in northwestern Europe: integrating the requirements of flora and fauna. Biol Conserv. 2002;104:265–73.

    Article  Google Scholar 

  2. 2.

    Poschlod P, Wallis De Vries MF. The historical and socioeconomic perspective of calcareous grasslands—lessons from the distant and recent past. Biol Conserv. 2002;104(3):361–76.

    Article  Google Scholar 

  3. 3.

    Kajtoch Ł, Cieslak E, Varga Z, Paul W, Mazur MA, Sramkó G, Kubisz D. Phylogeographic patterns of steppe species in Eastern Central Europe: a review and the implications for conservation. Biodivers Conserv. 2016;25:2309–39.

    Article  Google Scholar 

  4. 4.

    Poschlod P. Geschichte der Kulturlandschaft. Stuttgart: Eugen Ulmer; 2015.

    Google Scholar 

  5. 5.

    Cousins SAO, Ohlson H, Eriksson O. Effects of historical and present fragmentation on plant species diversity in semi-natural grasslands in Swedish rural landscapes. Landsc Ecol. 2007;22:723–30.

    Article  Google Scholar 

  6. 6.

    Krauss J, Klein A-M, Steffan-Dewenter I, Tscharntke T. Effects of habitat area, isolation, and landscape diversity on plant species richness of calcareous grasslands. Biodivers Conserv. 2004;13:1427–39.

    Article  Google Scholar 

  7. 7.

    Fahrig L. Effects of habitat fragmentation on biodiversity. Annu Rev Ecol Syst. 2003;34:487–515.

    Article  Google Scholar 

  8. 8.

    Ouborg NJ, Vergeer P, Mix C. The rough edges of the conservation genetics paradigm. J Ecol. 2006;94:1233–48.

    Article  Google Scholar 

  9. 9.

    Honnay O, Jacquemyn H. Susceptibility of common and rare species to the genetic consequences of habitat fragmentation. Conserv Biol. 2007;21(3):823–31.

    Article  PubMed  Google Scholar 

  10. 10.

    Leimu R, Mutikainen P, Koricheva J, Fischer M. How general are positive relationships between plant population size, fitness and genetic variation. J Ecol. 2006;94:942–52.

    Article  Google Scholar 

  11. 11.

    Listl D, Reisch C. Genetic variation of Sherardia arvensis L.—How land use and fragmentation affect an arable weed. Biochem Syst Ecol. 2014;55:164–9.

    CAS  Article  Google Scholar 

  12. 12.

    Schmidt K, Jensen K. Genetic structure and AFLP variation of remnant populations in the rare plant Pedicularis palustris (Scrophulariaceae) and its relation to population size and reproductive components. Am J Bot. 2000;87:678–89.

    CAS  Article  PubMed  Google Scholar 

  13. 13.

    de Jong T, Klinkhamer PGL. Plant size and reproductive success through female and male function. J Ecol. 1994;82:399–402.

    Article  Google Scholar 

  14. 14.

    Ellstrand NC, Elam DR. Population genetic consequences of small population size: implications for plant conservation. Annu Rev Ecol Syst. 1993;24:217–42.

    Article  Google Scholar 

  15. 15.

    Münzbergová Z, Cousins SAO, Herben T, Plačkova I, Mildén M, Ehrlén J. Historical habitat connectivity affects current genetic structure in a grassland species. Plant Biol. 2013;15:195–202.

    Article  PubMed  Google Scholar 

  16. 16.

    Heelemann S, Krug CB, Esler KJ, Poschlod P, Reisch C. Low impact of fragmentation on genetic variation within and between remnant populations of the typical renosterveld species Nemesia barbata in South Africa. Biochem Syst Ecol. 2014;54:59–64.

    CAS  Article  Google Scholar 

  17. 17.

    Ibánez I, Katz DSW, Peltier D, Wolf SM, Connor Barrie BT. Assessing the integrated effects of landscape fragmentation on plants and plant communities: the challenge of multiprocess-multiresponse dynamics. J Ecol. 2014;102:882–95.

    Article  Google Scholar 

  18. 18.

    Zulka KP, Abensberg-Traun M, Milasowszky N, Bieringer G, Gereben-Krenn B-A, Holzinger W, Hölzler G, Rabitsch W, Reischütz A, Querner P, et al. Species richness in dry grassland patches of eastern Austria: a multi-taxon study on the role of local, landscape and habitat quality variables. Agric Ecosyst Environ. 2014;182:25–36.

    Article  Google Scholar 

  19. 19.

    Bobbink R, Willems JH. Increasing dominance of Brachypodium pinnatum (L.) Beauv. in chalk grasslands: a threat to a species-rich grassland. Biol Conserv. 1987;40:301–14.

    Article  Google Scholar 

  20. 20.

    Bobbink R, Hicks K, Galloway J, Spranger T, Alkemade R, Ashmore M, Bustamante M, Cinderby S, Davidson E, Dentener F, et al. Global assessment of nitrogen deposition effects on terrestrial plant diversity: a synthesis. Ecol Appl. 2010;20:30–59.

    CAS  Article  PubMed  Google Scholar 

  21. 21.

    Diekmann M, Jandt U, Alard D, Bleeker A, Corcket E, Gowing DJG, Stevens CJ, Dupré C. Long-term changes in calcareous grassland vegetation in North-western Germany—no decline in species richness, but a shift in species composition. Biol Conserv. 2014;172:170–9.

    Article  Google Scholar 

  22. 22.

    Ruprecht E, Enyedi M, Eckstein RL, Donath TW. Restorative removal of plant litter and vegetation 40 years after abandonment enhances re-emergence of steppe grassland vegetation. Biol Conserv. 2010;143:449–56.

    Article  Google Scholar 

  23. 23.

    Ruprecht E, Szabó A. Grass litter is a natural seed trap in long-term undisturbed grassland. J Veg Sci. 2012;23:495–504.

    Article  Google Scholar 

  24. 24.

    Jensen K, Gutekunst K. Effects of litter on establishment of grassland plant species: the role of seed size and successional status. Basic Appl Ecol. 2003;4:579–87.

    Article  Google Scholar 

  25. 25.

    Reisch C, Bernhardt-Römermann M. The impact of study design and life history traits on genetic variation of plants determined with AFLPs. Plant Ecol. 2014;215:1493–511.

    Article  Google Scholar 

  26. 26.

    Hanski I. Patch occupancy dynamics in fragmented landscapes. Trend Ecol Evol. 1994;9:131–5.

    CAS  Article  Google Scholar 

  27. 27.

    Vandepitte K, Jacquemyn H, Roldán-Ruiz I, Honnay O. Landscape genetics of the self-compatible forest herb Geum urbanum: effects of habitat age, fragmentation and local environment. Mol Ecol. 2007;16:4171–9.

    CAS  Article  PubMed  Google Scholar 

  28. 28.

    Poschlod P, Baumann A. The historical dynamics of calcareous grasslands in the Central and Southern Franconian jurassic mountains—a comparative pedoanthracological and pollen analytical study. Holocene. 2010;20:13–23.

    Article  Google Scholar 

  29. 29.

    Bassler R, Schmitt L, Siegel O: Methodenbuch/Verband Deutscher Landwirtschaftlicher Untersuchungsanstalten, vol. Teillieferung 3, 4. Auflage edn. Darmstadt: VDLUFA-Verlag; 2003.

  30. 30.

    Karlík P, Poschlod P. History of abiotic filter: which is more important in determining the species composition of calcareous grasslands. Preslia. 2009;81:321–40.

    Google Scholar 

  31. 31.

    Rogers SO, Bendich AJ. Extraction of total cellular DNA from plants, algae and fungi. In: Gelvin SB, Schilperoort RA, editors. Plant molecular biology manual. Dordrecht: Kluwer Academic Press; 1994. p. 1–8.

    Google Scholar 

  32. 32.

    Reisch C, Kellermeier J. Microscale variation in alpine grasslands: aFLPs reveal a high level of genotypic diversity in Primula minima (Primulaceae). Bot J Linn Soc. 2007;155:549–56.

    Article  Google Scholar 

  33. 33.

    Bylebyl K, Poschlod P, Reisch C. Genetic variation of Eryngium campestre L. (Apiaceae) in Central Europe. Mol Ecol. 2008;17:3379–88.

    Article  PubMed  Google Scholar 

  34. 34.

    Reisch C. Glacial history of Saxifraga paniculata (Saxifragaceae)—molecular biogeography of a disjunct arctic-alpine species in Europe and North America. Biol J Linn Soc. 2008;93:385–98.

    Article  Google Scholar 

  35. 35.

    Bonin A, Belleman E, Eidesen PB, Pompanon F, Brochmann C, Taberlet P. How to track and assess genotyping errors in population genetic studies. Mol Ecol. 2004;13:3261–73.

    CAS  Article  PubMed  Google Scholar 

  36. 36.

    Nei M. Analysis of gene diversity in subdivided populations. Proc Natl Acad Sci USA. 1973;70:3321–3.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  37. 37.

    Vekemans X. Aflp-Surv Version 1.0. In. Belgium: Distributed by the Author. Laboratoire De Génétique Et Ecologie Végétale, Université Libre De Bruxelles; 2002.

  38. 38.

    R-Core-Team. R: a language and Environment for Statistical Computing. In: Vienna, Austria, R foundation for statistical computing; 2013.

  39. 39.

    Plummer M. JAGS: a program for analysis of Bayesian graphical models using Gibbs sampling; 2003.

  40. 40.

    Kruschke JK. Doing Bayesian Data analysis—a tutorial with R, JAGS, and Stan. Waltham: Academic Press; 2015.

    Google Scholar 

  41. 41.

    Park T, Casella G. The Bayesian lasso. J Am Stat Assoc. 2008;103:681–6.

    CAS  Article  Google Scholar 

  42. 42.

    Lauterbach D, Ristow M, Gemeinholzer B. Genetic population structure, fitness variation and the importance of population history in remnant populations of the endangered plant Silene chlorantha (Willd.) Ehrh. (Caryophyllaceae). Plant Biol. 2011;13:667–77.

    CAS  Article  PubMed  Google Scholar 

  43. 43.

    Tilman D, May RM, Lehman CL, Novak MA. Habitat destruction and the extinction debt. Nature. 1994;371:65–6.

    Article  Google Scholar 

  44. 44.

    Helm A, Hanski I, Pärtel M. Slow response of plant species richness to habitat loss and fragmentation. Ecol Lett. 2006;9:72–7.

    PubMed  Google Scholar 

  45. 45.

    Krauss J, Bommarco R, Guardiola M, Heikkinen RK, Helm A, Kuusaari M, Lindborg R, Öckinger E, Pärtel M, Pino J, et al. Habitat fragmentation causes immediate and time-delayed biodiversity loss at different trophic levels. Ecol Lett. 2010;13:597–605.

    Article  PubMed  PubMed Central  Google Scholar 

  46. 46.

    Honnay O, Coart E, Butaye J, Adriaens D, Van Glabeke S, Roldán-Ruiz I. Low impact of present and historical landscape configuration on the genetics of fragmented Anthyllis vulneraria populations. Biol Conserv. 2006;127:411–9.

    Article  Google Scholar 

  47. 47.

    Kuss P, Pluess AR, Ǽgisdóttir HH, Stöcklin J. Spatial isolation and genetic differentiation in naturally fragmented plant populations of the Swiss Alps. J Plant Ecol. 2008;1:149–59.

    Article  Google Scholar 

  48. 48.

    Maurer K, Durka W, Stöcklin J. Frequency of plant species in remnants of calcareous grassland and their dispersal and persistence characteristics. Basic Appl Ecol. 2003;4:307–16.

    Article  Google Scholar 

  49. 49.

    Slatkin M. Gene flow in natural populations. Annu Rev Ecol Syst. 1985;16:393–430.

    Article  Google Scholar 

  50. 50.

    Widén B, Svensson L. Conservation of genetic variation in plants: The importance of population size and gene flow. In: Hansson L, editor. Ecological principles of nature conservation: applications in temperate and boreal environments. New York: Springer; 1992. p. 113–61.

    Google Scholar 

  51. 51.

    Aavik T, Holderegger R, Bolliger J. The structural and functional connectivity of the grassland plant Lychnis flos-cuculi. Heredity. 2014;112:471–8.

    CAS  Article  PubMed  Google Scholar 

  52. 52.

    Kwak MM, Velterop O, van Andel J. Pollen and gene flow in fragmented habitats. Appl Veg Sci. 1998;1:37–54.

    Article  Google Scholar 

  53. 53.

    Ellstrand NC. Current knowledge of gene flow in plants: implications for transgene flow. Philos Trans R Soc Lond B Biol Sci. 2003;358:1163–70.

    Article  PubMed  PubMed Central  Google Scholar 

  54. 54.

    Helm A, Oja T, Saar L, Takkis K, Talve T, Pärtel M. Human influence lowers plant genetic diversity in communities with extinction debt. J Ecol. 2009;97:1329–36.

    Article  Google Scholar 

  55. 55.

    Powolny M, Poschlod P, Reisch C. Genetic variation in Silene acaulis increases with population age. Botany. 2016. doi:10.1139/cjb-2015-0195.

    Google Scholar 

  56. 56.

    Vellend M. Parallel effects of land-use history on species diversity and genetic diversity of forest herbs. Ecology. 2004;85(11):3043–55.

    Article  Google Scholar 

  57. 57.

    Jacquemyn H, Honnay O, Galbusera P, Roldán-Ruiz I. Genetic structure of the forest herb Primula elatior in a changing landscape. Mol Ecol. 2004;13:211–9.

    CAS  Article  PubMed  Google Scholar 

  58. 58.

    Prentice HC, Lönn M, Rosquist G, Ihse M, Kindström M. Gene diversity in a fragmented population of Briza media: grassland continuity in a landscape context. J Ecol. 2006;94:87–97.

    CAS  Article  Google Scholar 

  59. 59.

    Lindborg R, Eriksson O. Historical landscape connectivity affects present plant species diversity. Ecology. 2004;85:1840–5.

    Article  Google Scholar 

  60. 60.

    Reitalu T, Johannson LJ, Sykes MT, Hall K, Prentice HC. History matters: village distances, grazing and grassland species diversity. J Appl Ecol. 2010;47:1216–24.

    Article  Google Scholar 

  61. 61.

    Rico Y, Boehmer HJ, Wagner HH. Effect of rotational shepherding on demographic and genetic connectivity of calcareous grasslands. Conserv Biol. 2013;28:467–77.

    Article  PubMed  Google Scholar 

  62. 62.

    Broquet T, Angelone S, Jaquiery J, Joly P, Lena J-P, Lengagne T, Plenet S, Luquet E, Perrin N. Genetic bottlenecks driven by population disconnection. Conserv Biol. 2010;24:1596–605.

    Article  PubMed  Google Scholar 

  63. 63.

    Kelemen A, Török P, Valkó O, Miglécz T, Tóthmérész B. Mechanisms shaping plant biomass and species richness: plant strategies and litter effect in alkali and loess grasslands. J Veg Sci. 2013;24:1195–203.

    Article  Google Scholar 

  64. 64.

    Brys R, Jacquemyn H, Endels P, De Blust G, Hermy M. Effect of habitat deterioration on population dynamics and extinction risks in a previously common perennial. Conserv Biol. 2005;19:1633–43.

    Article  Google Scholar 

  65. 65.

    Schleuning M, Niggemann M, Becker U, Matthies D. Negative effects of habitat degradation and fragmentation on the declining grassland plant Trifolium montanum. Basic Appl Ecol. 2009;10:61–9.

    Article  Google Scholar 

  66. 66.

    Jongejans E, Soons MB, De Kroon H. Bottlenecks and spatiotemporal variation in the sexual reproduction pathway of perennial meadow plants. Basic Appl Ecol. 2006;7:71–81.

    Article  Google Scholar 

  67. 67.

    Busch V, Reisch C. Population size and land use affect the genetic variation and performance of the endangered plant species Dianthus seguieri ssp. glaber. Conserv Genet. 2015. doi:10.1007/s10592-10015-10794-10591.

    Google Scholar 

  68. 68.

    Rudmann-Maurer K, Weyand A, Fischer M, Stöcklin J. Microsatellite diversity of the agriculturally important alpine grass Poa alpina in relation to land use and natural environment. Ann Bot. 2007;100:1249–58.

    Article  PubMed  PubMed Central  Google Scholar 

  69. 69.

    Schweingruber FH, Poschlod P. Growth rings in herbs and shrubs: life span, age determination and stem anatomy. For Snow Landsc Res. 2005;79(3):195–415.

    Google Scholar 

  70. 70.

    Honnay O, Bossuyt B, Jacquemyn H, Shimono A, Uchiyama K. Can a seed bank maintain the genetic variation in the above ground plant population? Oikos. 2008;117:1–5.

    Article  Google Scholar 

Download references

Authors’ contributions

CR conceived and designed the study. SS, KM, QS, CM, CH and CP collected data. CS conducted the statistical analyses, CS and CR wrote the manuscript. All authors read and approved the final manuscript.


Special thanks go to Christine Kammel, Annika Sezi and Mathias Rass for molecular analyses, to Petra Schitko for technical assistance in the lab, to Sabine Fischer for her help with GIS, to Günther Kolb for technical assistance during soil analyses, to Ellen Pagel, Jose Valdez and Sharon Wirth for improving the language and to Peter Poschlod for his generous support.

Competing interests

The authors declare that they have no competing interests.

Availability of data and materials

The datasets supporting the conclusions of this article are included within the article. AFLP raw data are available from the corresponding author upon reasonable request.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Author information



Corresponding author

Correspondence to Christoph Reisch.

Additional file

Additional file 1.

Text file containing the specification for the JAGS model used for the Bayesian hierarchical multiple regression. The basic model definition was adapted from Kruschke [40].

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Reisch, C., Schmidkonz, S., Meier, K. et al. Genetic diversity of calcareous grassland plant species depends on historical landscape configuration. BMC Ecol 17, 19 (2017).

Download citation


  • AFLP
  • Dry grasslands
  • Habitat fragmentation
  • Genetic diversity
  • Grazing
  • Land use
  • Litter
  • Soil analysis