Vegetation changes in temperate ombrotrophic peatlands over a 35 year period

Vegetation changes in temperate ombrotrophic peatlands over a 35 year period a1111111111 a1111111111 Global changes in climate and land use are occurring at an unprecedented rate, often trig- a1111111111 gering drastic shifts in plant communities. This study aims to reconstruct the changes that a1111111111 occurred over 35 years in the plant communities of temperate bogs subjected to indirect human-induced disturbances. In 2015–17, we resurveyed the vascular flora of 76 plots located in 16 bogs of southern Que ´ bec (Canada) first sampled in 1982. We evaluated changes in species richness, frequency of occurrence and abundance, while considering OPENACCESS species shade-tolerance and preferential habitat. We calculated beta diversity as between- Citation: Pinceloup N, Poulin M, Brice M-H, site similarities in composition, and evaluated differences between the two surveys using Pellerin S (2020) Vegetation changes in temperate ombrotrophic peatlands over a 35 year period. tests for homogeneity in multivariate dispersion. We found a significant increase in species PLoS ONE 15(2): e0229146. https://doi.org/ richness and beta diversity over the last 35 years associated with major species turnovers, 10.1371/journal.pone.0229146 indicating a biotic differentiation of the Sphagnum-bog plant communities. These changes Editor: RunGuo Zang, Chinese Academy of were mostly associated with an increase in the abundance and frequency of shade-tolerant Forestry, CHINA and facultative species, suggesting a global phenomenon of woody encroachment. Received: September 11, 2019 Because the observed changes occurred in a few decades on sites free of in situ human dis- Accepted: January 30, 2020 turbances, we suggest that they were likely induced by the synergic effect of the agricultural drainage occurring in the surrounding mineral soils, climate warming, and nitrogen atmo- Published: February 13, 2020 spheric depositions. We also believe that further changes are to be expected, as the trigger- Peer Review History: PLOS recognizes the ing factors persist. Finally, our results highlight the need for increased bog conservation or benefits of transparency in the peer review process; therefore, we enable the publication of restauration efforts. Indeed, a rise in beta diversity due to the introduction of nearby terres- all of the content of peer review and author trial species could induce biotic homogenization of the bog flora with that of surrounding responses alongside final, published articles. The habitats and ultimately impoverish the regional species pool. editorial history of this article is available here: https://doi.org/10.1371/journal.pone.0229146 Copyright:© 2020 Pinceloup et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and Introduction reproduction in any medium, provided the original Global changes in climate and land use are occurring at an unprecedented rate [1,2], often trig- author and source are credited. gering drastic shifts in plant communities [3,4]. The overriding influence of human activities Data Availability Statement: 1982 data are on plant communities has been shown to induce modification in species abundance [5,6] and available in the Examine database of the Ministère ´ distribution [7,8], and to increase the risk of population extinction [9,10]. At a global scale, Energie et Ressources naturelles of Quebec, (http:// sigeom.mines.gouv.qc.ca/signet/classes/I1102_ such changes have most often been associated with biodiversity loss through the simplification PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 1 / 15 Temporal vegetation changes in bogs examine?l=F). Relevant data are within the of communities (decrease in beta diversity), a process called biotic homogenization [11,12]. manuscript and its Supporting Information files. Biotic homogenization usually occurs when exotic or ruderal native species replace specialist native species [11,13,14]. Nevertheless, local species invasions and extinctions are not necessar- Funding: This research received financial support from the Natural Sciences and Engineering ily causally related and can follow complex pathways that can also lead to biotic differentiation Research Council of Canada (Discovery grants to (increase in beta diversity) [15–17] or even no change in beta diversity [18]. Biotic differentia- Ste ´phanie Pellerin Pellerin: RGPIN-2014-05367 tion has mostly been observed at local scales when the introduction of numerous exotic species and Monuique Poulin RGPIN-2014-05663). The outweighs the loss of natives or when disturbances increase habitat heterogeneity thus promot- funders had no role in study design, data collection ing the introduction of various different species [15,19–21]. and analysis, decision to publish, or preparation of the manuscript. Wetland flora are usually more homogeneous than terrestrial flora at local and continental scales [22,23]. This naturally low level of beta diversity within and among sites is likely related, Competing interests: The authors have declared among other reasons, to the uniformity of stressful environmental conditions found in wet- that no competing interests exist. lands that can be tolerated by only a small set of specialized species [22,24,25]. Despite wet- lands being among the most disturbed ecosystems in the world [26], few studies have evaluated whether direct and indirect human disturbances induce further biotic homogeniza- tion in their plant communities (e.g., [17,27–29]). Among them, Price et al. [28] found a subtle process of biotic homogenization over time in the herbaceous emergent communities of palus- trine wetlands of Illinois (USA). This process was associated with an increase in the presence and abundance of the exotic Phalaris arundinacea from the Poaceae family, and the decline of numerous other native and exotic species. On the other hand, Brice et al. [17] observed biotic differentiation in the herbaceous communities of urban riparian swamps due, in part, to the increase in habitat heterogeneity induced by urbanization. Similarly, Zhang et al. [29] showed taxonomic differentiation of macrophyte communities on a river floodplain exposed to human-induced disturbances. In fact, because many types of wetlands are naturally species- poor, they are potentially more susceptible to biotic differentiation than to homogenization following disturbances [30]. Ombrotrophic peatlands (bogs) are peat-forming naturally species-poor wetlands [31,32] that are mostly viewed as resilient ecosystems with important compositional changes usually occurring over centuries to millennia [33]. Still, several studies have shown recent rapid and drastic compositional changes in response to human-induced disturbances (drainage, eutro- phication, climate warming; land-use modifications), through their influence on water and nutrient availability (e.g., [34–40]). For instance, drainage and contemporary climate change (warmer and/or drier climatic conditions) have been shown to enhance shrub and tree encroachment, hamper Sphagnum growth and facilitate the establishment of generalist and exotic species [33, 35,37,41]. This shift from open bog toward forested communities is also associated with a decrease in the cover and richness of other specialist bog plant species (often heliophilic species) and a concomitant increase of shade or drought-tolerant species [33,37,41–44]. More recently, a study investigating the effects of woody encroachment in bog flora, using a space for time substitution, found within-site biotic differentiation associated with the introduction of terrestrial shade-tolerant species [45]. Although vegetation changes in bogs have been the subject of numerous studies (e.g., [33;38,44]), temporal changes in plant communities at a regional scale, and how these changes affect beta diversity, remain to be explored. In this study, we investigated the floristic changes that occurred over 35 years in the vascu- lar plant communities of temperate bogs of southern Que ´bec (Canada). These bogs are mostly dominated by ericaceous shrubs (e.g., Kalmia angustifolia, Rhododendron groenlandicum, Cha- maedaphne calyculata), Sphagnum mosses (e.g., Sphagnum capillifolium, S. magellanicum., S. rubellum) and scattered Picea mariana or Larix laricina thickets. We revisited 76 plots origi- nally surveyed in 1982 and located in 16 different bogs spread over the St. Lawrence Lowlands region. These bogs have been isolated in an agricultural landscape for more than 50 years, PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 2 / 15 Temporal vegetation changes in bogs which has induced a drying of the peat surface on many sites (e.g., [41,46]). We specifically aimed to 1) determine changes in taxa richness, composition and beta diversity and 2) assess whether biotic homogenization or differentiation occurred over time. We hypothesized that the studied sites would be characterized by biotic differentiation in response to species enrich- ment by terrestrial and shade-tolerant plants. We also expected to observe a woody encroach- ment phenomenon. Finally, we acknowledge that focusing on vascular plant communities may underestimate observed changes as mosses may represent a high proportion of the diver- sity in bogs, but mosses were rarely identified at the species level in 1982. Methodology Study area The study area is in the St. Lawrence Lowlands, southern Que ´bec (Canada), and extends from 45˚01’N to 46˚50’N latitude and from 70˚55’W to 74˚17’W longitude. The area is enclosed by the Canadian Shield to the north and the Appalachian Mountains to the south. The Lowlands have a flat topography and are characterized by deep arable soils derived from glacial and marine deposits. The landscape is composed of about 50% agricultural fields, 30% woodlands, 10% urbanized areas and 10% wetlands [47], of which 35% are bogs [48]. Over the last few decades, 19% of the wetland areas have been lost or disturbed by human activities, mainly agri- culture and forestry, but also urbanization and peat mining [48]. Tree encroachment by Picea mariana, Pinus banksiana, Larix laricina, Betula populifolia, and Acer rubrum has been reported in many of the remaining bogs [35,41,45,46,49]. The mean annual temperature of the study area fluctuates from 4.1˚C in the northeast to 6.7˚C in the southwest [50]. Average precipitation ranges from a high of 1120 mm in the northeast to a low of 965 mm in the southwest, of which 23% and 17% fall as snow, respectively [50]. Original sampling During the 1980s, a large inventory campaign was conducted by the Que ´bec Ministry of Energy and Natural Resources to evaluate the peat resource in southern Que ´bec peatlands greater than 40 ha and with a peat deposit thicker than 30 cm [51]. During this large-scale inventory, hundreds of vegetation surveys were conducted. All plots in the study area were sampled in 1982. All plots (20 m × 20 m) were delineated in the center of a seemingly homoge- nous plant community. The percent cover of all vascular taxa and strata (tree, shrub, herbs, mosses) was visually estimated according to six classes: <1%, 1-5%, 6–25%, 26–50%, 51–75%, >75%. Mosses were rarely identified at the species level in original surveys and thus not inves- tigated in the present study. Other information available in the field sheets includes geographic coordinates of the plot, peat thickness, soil pH, biotope relative dominance (pool, hummock, hollow, etc.), vegetation structure (woody, shrubby, herbaceous, etc.) and general information on the location of the plot within the bog (margins, expanse, near a drainage ditch, etc.). It should be noted that vegetation data from this original survey has never been analyzed. Resampling: 2015 and 2017 Historical sampling plots were not permanently marked. However, as mentioned above geo- graphic coordinates were available from the original field sheets. Before field sampling, we dis- carded all original plots established on minerotrophic peatland habitats or disturbed areas. We also rejected all plots located in areas that had been affected by land conversion or subjected to woodcutting since earlier surveys. We identified these areas using 1980–2015 Google Earth PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 3 / 15 Temporal vegetation changes in bogs Digital Globe satellite imagery. During the summer of 2015 and 2017 (hereafter referred to as 2017 for simplification), 76 of the original plots spread over 16 bogs were relocated using a GPS (S1 Table). All plots were on private lands or on lands protected by non-governmental agencies. We obtained a verbal permission from the owner of each site before sampling. In addition to geographic coordinates, we drew upon all environmental information available from original studies in this endeavor. We surveyed the vegetation following the same method- ology used in 1982. Due to the abundance of visual landmarks, we estimate that we were able to relocate all plots within 30–40 m from their original position. Although a relocation error could introduce a source of variance in the dataset, systematic biases favoring vegetation change in one direction are unlikely to occur, especially when environmental variables are used in combination with geographic location to relocate unmarked plots [13,52]. Data analysis Before analyses, taxonomic nomenclature was standardized according to the Database of Vas- cular Plants of Canada (VASCAN) [53]. We lumped all sedge taxa (Carex) for both years at the genus level, as very few individuals were identified to the species level in 1982. We also merged Vaccinium myrtilloides and Vaccinium angustifolium under Vaccinum cf. angustifolium because they were often undistinguished in 1982. Taxa richness. Change in taxa richness at the plot level (alpha diversity) between 1982 and 2017 was evaluated using a Wilcoxon signed rank test. We also considered each taxa’s habi- tat preferences by comparing richness of taxa sorted by their wetland indicator status [54]. Accordingly, taxa that almost always occur in wetlands were classified as “obligate”, those that usually occur in wetlands, but may occur in non-wetland habitats as “facultative”, and those that occur equally in wetlands and non-wetland habitats or that usually occur in non-wetland habi- tats as “non-wetland” (S2 Table). No taxa that almost always occur in non-wetland habitats (upland) were found. The wetland indicator status of each taxon was retrieved from the National Wetland Plant List [55] and Lapointe et al. [56]. Generalized Poisson mixed models were used to compare the number of taxa between groups of habitat preferences and between years as well as the interaction between both factors. Plots were treated as a random factor, allowing us to con- trol for pairing between plots and dependence between the numbers of taxa of different prefer- ences in the same plot. Because a significant interaction was found between year and habitat preference, we conducted post-hoc Tukey tests between habitats independently for each year. We verified that model residuals followed a normal distribution of homogeneous variance. We also compared richness of taxa according to their tolerance to shade [57] (S2 Table). We classified taxa into three broad categories of shade tolerance: “tolerant” for taxa that grow well under a limited amount of light, “mid-tolerant” for taxa that can tolerate shade during some phases of their life cycle, and “intolerant” for taxa that only grow with a high amount of available light. Shade tolerance information was retrieved from the online databases TRY [58] and PLANTS [59], as well as from Humbert et al. [60]. We used the same linear mixed-effect model procedure described above. Taxa frequency and abundance. Taxa with the greatest changes in terms of frequency of occurrence were identified by comparing the number of plots occupied by each taxon for each year using Chi-square goodness-of-fit tests (taxa were tested one by one). A Yates continuity correction was applied [61]. Finally, we evaluated whether the abundance of some taxa changed over time using plant cover. The cover data were transformed into percentages using the midpoint of each class. Only species with a mean of� 5% coverage for at least one of the sampling periods were used (12 species analyzed). Differences between 1982 and 2017 were tested using a two-tailed paired t-test (species were tested one by one). PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 4 / 15 Temporal vegetation changes in bogs Beta diversity. Differences in beta diversity between 1982 and 2017 were analyzed using a distance-based test for homogeneity of multivariate dispersions (PERMDISP) [62]. PERM- DISP calculates the distance of each site (here, plot) to the centroid in an ordination space (principal coordinate analysis) and then tests whether these distances are different between groups (here, sampling years) through permutation tests. More precisely, two site-by-species matrices were first computed; one using covers (median of cover classes) and the other using presence-absence data. Then, for each matrix, a site-by-site distance matrix was built using the Hellinger distance [63]. This distance matrix was used to compute the centroid of each group of sites. The distance of each site to its associated group centroid was calculated, and the dis- persion of these distances (within-group variance) was used as an estimate of beta diversity (the greater the within-group variance, the higher the beta diversity). The site distances to cen- troid were subjected to a paired t-test with 9999 permutations to determine whether disper- sions differed between groups. To detect a shift in species composition between year (i.e., turnover), we tested for location differences between centroids using PERMANOVA with pseudo-F ratios (9999 permutations) [64]. Because this test is sensitive to differences in multi- variate dispersions [65], data visualization was used to support the interpretation of the statisti- cal test. The differences in multivariate dispersion and composition were illustrated in principal coordinates analysis ordinations (PCoA). Finally, we decomposed beta diversity into replacement (turnover) and richness difference (species gain/loss) using Jaccard dissimilarity indices and presence-absence data [66]. We decomposed beta diversity for each year as well as between years. All statistical analyses were performed using R 3.3.2 [67]. Wilcoxon signed rank tests were performed using the wilcox.test, chi-square tests using chisq.test, paired t-test using t.test, and generalized Poisson mixed models using glm(), all from the stats package [67]. Tukey tests were done with the lsmeans() function in the lsmeans library [68]. Hellinger transformations were done using decostand(), multivariate dispersion analyses were performed using betadis- per(), centroid locations were tested using adonis2(), all from the vegan package [69]. Permuta- tion t-tests were done using perm.test(), from the broman package [70]. Beta diversity partitioning was conducted using the beta.div.comp function [66]. Results Taxa richness A total of 37 vascular taxa were recorded in 1982 and 57 in 2017 (total = 67), and only one exotic (Frangula alnus) was found, in 2017 (S2 Table). Thirty new taxa were detected in 2017, while ten were lost. Most of the taxa gained were found in <10 plots, except Aronia melano- carpa (33 plots), Gaylussacia baccata (18 plots), Vaccinium corymbosum (18 plots), Erio- phorum virginicum (17 plots), Platanthera blephariglottis var. blephariglottis (14 plots) and Coptis trifolia (11 plots). Furthermore, few of the taxa gained were abundant in the plots where they were found (mean cover >5%): Rubus allegheniensis (39% mean cover; 4 plots), V. corym- bosum (24%; 18 plots), Frangula alnus (22%; 3 plots), G. baccata (12%; 18 plots) and Osmun- dastrum cinnamomeum (8%; 5 plots). All taxa lost were present in five plots or less in 1982, except Alnus incana subsp. rugosa (8 plots). Taxa richness per plot ranged from 3 to 15 in 1982 and from 7 to 21 in 2017 (Fig 1). The number of taxa increased in 61 plots, remained similar in five plots and decreased in ten plots. Overall, mean richness per plot significantly increased over time, from an average of 10 to 14 taxa (V = 165; p� 0.005). The number of “facultative” and “non-wetland” taxa per plot significantly increased with time, while the number of “obligate” taxa remained similar (Fig 2A). The number of “obligate” PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 5 / 15 Temporal vegetation changes in bogs Fig 1. Vascular taxa richness in temperate bogs of southern Quebec (Canada) in 1982 and 2017. Plots sampled in 1982 are compared to the same plots re-sampled in 2017. The plain diagonal line is the 1:1 line (no change in richness between the two years) and the dashed line represents the mean richness increase. Grey dots represent at least two plots with the exact same number of species in 1982 and 2017. https://doi.org/10.1371/journal.pone.0229146.g001 Fig 2. Richness of a) obligate, facultative and non-wetland taxa and b) shade tolerant, mid-tolerant and intolerant taxa in 1982 and 2017 in temperate bogs of southern Que ´bec (Canada). Provided are mean taxa richness (black dot), median (line), 25–75% quartiles (boxes) and ranges (whiskers). Different lower-case letters indicate a significant difference (α = 0.05; Generalized Poisson mixed models) between years, whereas different upper-case letters indicate a significant difference (α = 0.05; Tukey’s test) within the year. https://doi.org/10.1371/journal.pone.0229146.g002 PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 6 / 15 Temporal vegetation changes in bogs Fig 3. Plant taxa with significant changes (p� 0.05) in the frequency of occurrence (number of plots where species occurs) between 1982 and 2017 in temperate bogs of southern Que ´bec (Canada). Taxa were tested one by one (chi-square goodness-of-fit tests with Yates correction). On the left side are taxa with lower frequency in 2017 than in 1982, and on the right side, taxa with higher frequency in 2017 than in 1982. https://doi.org/10.1371/journal.pone.0229146.g003 taxa was about two times greater than that of “facultative” and “non-wetland” taxa in 1982 and about 1.5 times greater in than that of “facultative” taxa in 2017 (Fig 2A). The amount of shade “tolerant” and “mid-tolerant” taxa per plot significantly increased over time, while the richness of shade “intolerant” taxa remained similar for both years (Fig 2B). It follows that the number of “shade-intolerant” taxa was greater than that of the two other groups of taxa for both years. Species frequency and abundance Of the 67 taxa tested, 14 significantly changed their frequency of occurrence over time (Fig 3) and six experienced a significant change in their abundance (Fig 4). More precisely, 11 taxa significantly increased in frequency, among which Acer rubrum and Vaccinium corymbosum also increased in abundance. The increase in frequency was particularly marked (�20 plots) for Cypripedium acaule, Aronia melanocarpa, Gaultheria hispidula, Ilex mucronata and Acer rubrum. Taxa that became more frequent or abundant over time were mostly “facultative” or “non-wetland” taxa. Inversely, two taxa significantly decreased in frequency (Fig 3) and four decreased in abundance (Fig 4). Only Eriophorum vaginatum subsp. spissum decreased both in frequency and abundance. While E. vaginatum was still found in 43 plots in 2017 (62 plots in 1982), Betula pumila was found in only one plot the same year (22 plots in 1982). Taxa decreas- ing in frequency or abundance were all wetland “obligate” taxa. Beta diversity Beta diversity increased between 1982 and 2017 (presence-absence: T = 98.79; p< 0.0001; abundance data: T = 109.28; p< 0.0001), meaning that vascular plant communities of bogs dif- fered more markedly from one site to another in 2017 than in 1982 (Fig 5). This increase in beta diversity was associated with composition turnover as a significant difference in centroid position between years was also found for both presence-absence (F = 9.085; p� 0.0001) and abundance data (F = 9.514; p� 0.0001). PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 7 / 15 Temporal vegetation changes in bogs Fig 4. Plant taxa mean cover in 2017 vs 1982 in temperate bogs of southern Que ´bec (Canada). Only taxa with significant change in mean cover (two-tailed paired t-test; p� 0.05) are presented. The plain line is the 1:1 line. Carex spp. is plotted with a different symbol, as all species were lumped together and, therefore, not associated to a particular wetland status. https://doi.org/10.1371/journal.pone.0229146.g004 Finally, beta diversity in both 1982 and 2017, as well as beta diversity changes between 1982 and 2017, resulted mostly from taxa replacement rather than richness differences (Fig 6). For instance, when comparing all pairs of plots between 1982 and 2017, replacement accounted for 48.5% of the beta diversity decomposition while richness difference accounted only for 12.6%. Discussion In this study, we hypothesized that the studied sites would be characterized by a temporal biotic differentiation of the vascular flora in temperate bogs in response to species enrichment by terrestrial shade-tolerant plants. Our results showed that the richness of the studied bogs increased from 1982 to 2017, mostly due to the establishment of facultative and shade-tolerant plants. As expected, these changes induced a biotic differentiation, since we found higher beta diversity in 2017 than 35 years prior. This decrease in similarity with time mostly represented the outcome of species turnover. This indicates that newly established taxa are not similar across the study area and/or that various taxa are being replaced over time. Changes in plant environmental preferences The analyses of plant richness according to individual habitat preference indicated an enrich- ment of taxa associated to drier habitats as we found more facultative and non-wetland plants in recent surveys than in the older ones. For example, Acer rubrum, a species that usually occur in non-wetland habitats, which was present in only five plots in 1982, is more recently present in 25 plots and has experienced a significant increase in cover over time (Figs 3 and 4). Acer rubrum introduction and increase in density, often following drainage, have been recently documented in many North American wetlands including bogs [50,71,72]. On the other hand, PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 8 / 15 Temporal vegetation changes in bogs Fig 5. Effect of time on beta diversity for vascular plant taxa in bogs of southern Quebec (Canada), using presence-absence data. Beta diversity was measured as the distance of sites to their group centroid, here represented on the first two axes of the PCoA and using the boxplot of the sites-to-centroid distance. Provided are median (line), 25–75% quartiles (boxes) and ranges (whiskers). Different letters indicate a significant difference (p<0.05) determined by t-test. Circles are ellipses of standard deviation. https://doi.org/10.1371/journal.pone.0229146.g005 Fig 6. Triangular plot of the relationship among a) the 76 plots in 1982 and 2017, and b) for all pairs of plots between 1982 vs 2017, illustrating the contributions of three components of beta diversity over time: similarity of species composition, richness difference and replacement of species. The large dots inside the triangle represent the centroid, and the smaller dots represent the mean values of the three components. Values are indicated as raw percentages (i.e. 0.168 equals 16.8% of total beta diversity). https://doi.org/10.1371/journal.pone.0229146.g006 PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 9 / 15 Temporal vegetation changes in bogs although the richness of obligate wetland plants did not significantly change over time, some of these plants experienced a noteworthy decrease in frequency and abundance. This was the case for Eriophorum vaginatum subsp. spissum, a deep rooting wetland obligate typical of open bogs [73]. Despite the strong capacity of E. vaginatum to persist in peatlands, notably due to its typical tussock growth with basal meristematic tissue, surveys in vacuum mined peatlands have revealed its sensitivity to water-table drawdown [74]. Our results suggest that the studied bogs experienced an increase in shading level over time, as indicated by the gain of shade-tolerant and mid-tolerant plants. This increased level of shading is likely the result of shrub and tree encroachment in bogs, as suggested, among other reasons, by the increase in frequency and abundance of Acer rubrum and Vaccinium corymbo- sum. In fact, a quarter of the plots investigated presented an increase in tree cover (mean increase of 4.7%) and, more importantly, three quarter of the plots showed an increase in shrub (mean increase of 37.2%) cover (S3 Table). While our results indicate that shaded habi- tats are becoming more widespread in the studied bogs, we also found no significant change in the richness of shade-intolerant species, although the variability of the data was higher in 2017 than in 1982 (Fig 1). Furthermore, most of the species that significantly decreased in occur- rence and/or abundance (Betula pumila, Chamaedaphne calyculata, E. vaginatum) were shade-intolerant. The lack of significant change in the richness of the shade-intolerant species at the plot level may be due to the time-lag after disturbance, which implies that changes may still occur in the coming decades [75]. Overall, a more variable hydrologic environment and structure of plant strata likely allowed the co-occurrence of heliophilic bog specialists and shade and drought-tolerant plants more typical of forest environments. These results concur with the phenomenon of woody encroachment increasingly observed in bogs worldwide (e.g., [35,40,41,50]). Beta diversity As expected, we found a biotic differentiation (increase in beta diversity) over a period of 35 years in temperate Sphagnum-dominated bogs of the southern Quebec. Changes in beta diver- sity are led by two primary components: species turnover and richness differences [30,66]. Although we found an increase in plant richness over time in nearly all the sampling plots, our results indicated that the observed changes in beta diversity were mainly due to species replacement. For instance, our beta diversity partitioning analyses indicated that the influence of species replacement was four times greater than that of richness difference in beta diversity changes between 1982 and 2017. A recent study using a time-for-space substitution, conducted in the same study area but in different bogs, also found a higher beta diversity on sites recently subjected to woody encroachment than in open bog habitats, and this increase was also mostly driven by species turnover [45]. Significant species turnover with time, mainly driven by the replacement of typical bog species by trees, shrubs and other widespread forest species, was also found in Sweden [33,37], although the effect on beta diversity was not analysed. Potential causes of the observed changes Rapid changes in peatland plant communities have been mostly attributed to human-induced disturbances such as increasing temperature, atmospheric nitrogen deposition and drainage [34–40, 76]. In the studied bogs, the observed changes probably resulted from the interaction among all these factors, of which those influencing the hydrology of the sites likely predomi- nated. Because the study sites are located in seemingly undisturbed and undrained bogs, the gradual drying of the peat deposits likely resulted from the extensive drainage of mineral soils in the surrounding agricultural catchments [41,46,50]. For instance, using paleohydrological PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 10 / 15 Temporal vegetation changes in bogs reconstruction (testate amoebae), it was recently showed that the water level drawdown in two bogs of the study area matches the peak of agricultural activities and ditch-digging mechaniza- tion in their surrounding mineral catchment [46]. On the other hand, since 1960, southern Que ´bec has experienced an average temperature increase of 1.2˚C [77], which may also have contributed to peat drying. Indeed, according to a modeling study [78], even a 1˚C tempera- ture increase can dry the peat surface enough to shift open bogs into tree-dominated ecosys- tems. In conjunction with regional drainage and climate warming, nutrient enrichment may have facilitated the observed changes. Indeed, six of the 16 bogs studied are located in a region with some of the highest levels of wet nitrogen deposition in eastern North America [79]. All the bogs might also have been enriched by dry mineral dust from adjacent agricultural lands [39,46]. Finally, we could not rule out that some of the changes may also be associated with intrinsic ecosystem dynamics such as peat accumulation and mostly competition among plants (e.g., [80–82]). Conclusion In conclusion, our results reveal that rather rapid compositional changes in the plant commu- nities of temperate Sphagnum-dominated bogs isolated in an agricultural landscape induced an overall biotic differentiation. Because these changes occurred in a few decades within sites free of in situ human disturbances, we suggest that they were likely induced by the synergy effect of drainage in the surrounding mineral agricultural soils, climate warming, and nitrogen atmospheric depositions, as observed in several other bogs. We also believe that further changes are to be expected, since the triggering factors persist and obligate and shade-intoler- ant species remain abundant, unlike other bogs, where woody encroachment is more pro- nounced (e.g., [35]). Finally, our results also highlight the need for increased conservation efforts in such habi- tats, as bogs face two challenges: their surface is being lost to agriculture and urbanization [48] and their floristic distinctiveness is disappearing. Indeed, peatlands are usually species poor ecosystems with a highly specialized flora. An increase in beta diversity of their plant commu- nity due to the introduction of nearby species could increase their similarity with surrounding mesic habitats and ultimately impoverish the regional species pool. Supporting information S1 Table. Geographical coordinates (degree decimal) of the plots. (DOCX) S2 Table. List of plant taxa sampled. Taxa name, shade-tolerance (T = tolerant; M = mid-tol- erant; I = intolerant), wetland indicator status (Obl = obligate; Fac = facultative; NW = non- wetland), frequency of occurrence for both years (number of plots) and in brackets mean abundance in the plots where the species is present are indicated. NA indicates inconsistent or unavailable information. (DOCX) S3 Table. Tree and shrub cover (in %) in sampled plots for both survey periods. (DOCX) Acknowledgments We are grateful to the landowners who allowed us to work on their lands, to field assistants, to S. Daigle and P. Legendre for statistical advice and K. Grislis for linguistic revision. PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 11 / 15 Temporal vegetation changes in bogs Author Contributions Conceptualization: Nicolas Pinceloup, Monique Poulin, Ste ´phanie Pellerin. Data curation: Nicolas Pinceloup, Marie-He ´lène Brice, Ste ´phanie Pellerin. Formal analysis: Nicolas Pinceloup, Marie-He ´lène Brice. Funding acquisition: Monique Poulin, Ste ´phanie Pellerin. Methodology: Nicolas Pinceloup, Monique Poulin, Marie-He ´lène Brice, Ste ´phanie Pellerin. Resources: Monique Poulin, Ste ´phanie Pellerin. Validation: Ste ´phanie Pellerin. Writing – original draft: Nicolas Pinceloup, Monique Poulin, Marie-He ´lène Brice, Ste ´phanie Pellerin. Writing – review & editing: Nicolas Pinceloup, Monique Poulin, Ste ´phanie Pellerin. References 1. Lindner M, Maroschek M, Netherer S, Kremer A, Barbati A, Garcia-Gonzalo J, et al. Climate change impacts, adaptive capacity, and vulnerability of European forest ecosystems. Forest Ecol Manage. 2010; 259: 698–709. 2. Song XP, Hansen MC, Stehman SV, Potapov PV, Tyukavina A, Vermote EF, Townshend R. Global land change from 1982 to 2016. Nature. 2018; 560: 639–643. https://doi.org/10.1038/s41586-018- 0411-9 PMID: 30089903 3. Walker MD, Wahren CH, Hollister RD, Henry GHR, Ahlquist LE, Alatalo JM, et al. Plant community responses to experimental warming across the tundra biome. Proc Natl Acad Sci U S A. 2006; 103: 1342–1346. https://doi.org/10.1073/pnas.0503198103 PMID: 16428292 4. Eppinga MB, Rietkerk M, Wassen MJ, De Ruiter PC. Linking habitat modification to catastrophic shifts and vegetation patterns in bogs. Plant Ecol. 2009; 200: 53–68. 5. Smart SM, Bunce RGH, Marrs R, LeDuc M, Firbank LG, Maskell LC, et al. Large-scale changes in the abundance of common higher plant species across Britain between 1978, 1990 and 1998 as a conse- quence of human activity: tests of hypothesised changes in trait representation. Biol Conserv. 2005; 124: 355–371. 6. McGill BJ, Dornelas M, Gotelli NJ, Magurran AE. Fifteen forms of biodiversity trend in the anthropocene. Trends Ecol Evol. 2015; 30: 104–113. https://doi.org/10.1016/j.tree.2014.11.006 PMID: 25542312 7. Parmesan C, Ryrholm N, Stefanescu C, Hill JK, Thomas CD, Descimon H, et al. Poleward shifts in geo- graphical ranges of butterfly species associated with regional warming. Nature. 1999; 399: 579–583. 8. Dyer EE, Cassey P, Redding DW, Collen B, Franks V, Gastion KJ, et al. The global distribution and driv- ers of alien bird species richness. PloS Biol. 2017 Jan 12. https://doi.org/10.1371-journal.pbio.2000942 9. Sax DF, Gaines SD. Species invasions and extinction: the future of native biodiversity on islands. Proc Natl Acad Sci U S A. 2008; 105: 11490–11497. https://doi.org/10.1073/pnas.0802290105 10. Ceballos G, Ehrlich PR, Barnosky AD, Garcıa A, Pringle RM, Palmer TM. Accelerated modern human- induced species losses: entering the sixth mass extinction. Sci Adv. 2015; https://doi.org/10.1126/ sciadv.1400253 PMID: 26601195 11. McKinney ML, Lockwood JL. Biotic homogenization: a few winners replacing many losers in the next mass extinction. Trends Ecol Evol. 1999; 14: 450–453. https://doi.org/10.1016/s0169-5347(99)01679- 1 PMID: 10511724 12. Olden JD, Rooney TP. On defining and quantifying biotic homogenization. Global Ecol Biogeogr. 2006; 15: 113–120. 13. McCune JL, Vellend M. Gains in native species promote biotic homogenization over four decades in a human-dominated landscape. J Ecol. 2013; 101: 1542–1551. 14. Zeeman BJ, McDonnell MJ, Kendal D, Morgan JW. Biotic homogenization in an increasingly urbanized temperate grassland ecosystem. J Veg Sci. 2017; 28: 550–561. 15. Gong C, Chen J, Yu S. Biotic homogenization and differentiation of the flora in artificial and near-natural habitats across urban green spaces. Landsc Urban Plan. 2013; 120: 158–169. PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 12 / 15 Temporal vegetation changes in bogs 16. Aronson MF, Handel SN, La Puma IP, Clemants SE. Urbanization promotes non-native woody species and diverse plant assemblages in the New York metropolitan region. Urban Ecosyst. 2015; 18: 31–45. 17. Brice MH, Pellerin S, Poulin M. Does urbanization lead to taxonomic and functional homogenization in riparian forests? Divers Distrib. 2017; 23: 828–840. 18. Magura T, Lovei GL, Tothmeresz B Does urbanization decrease diversity in ground beetle (Carabidae) assemblages? Global Ecol Biogeogr 2010; 19: 16–26. 19. Olden JD, Poff N. Ecological processes driving biotic homogenization: testing a mechanistic model using fish faunas. Ecology. 2004; 85: 1867–1875. 20. Qian H, McKinney ML, Ku ¨ hn I. Effects of introduced species on floristic similarity: comparing two US states. Basic Appl Ecol. 2008; 9: 617–625. 21. Petsch DK. Causes and consequences of biotic homogenization in freshwater ecosystems. Int Rev Hydrobiol. 2016; 101: 113–122. 22. Santamarı ´a L. Why are most aquatic plants widely distributed? Dispersal, clonal growth and small-scale heterogeneity in a stressful environment. Acta Oecol. 2002; 23: 137–154. 23. Qian H, Guo Q. Linking biotic homogenization to habitat type, invasiveness and growth form of natural- ized alien plants in North America. Divers Distrib. 2010; 16: 119–125. 24. Barrett SCH, Echert CG, Husband BC. Evolutionary processes in aquatic plant populations. Aquat Bot. 1993; 44: 105–145. 25. Colmer TD, Voesenek LACJ. Flooding tolerance: suites of plant traits in variable environments. Funct Plant Biol. 2009; 36: 665–681. 26. Zedler JB, Kercher S. Wetland resources: status, trends, ecosystem services, and restorability. Annu Rev Environ Resour. 2005; 30: 39–74. 27. Lougheed VL, Mcintosh MD, Parker CA, Stevenson RJ. Wetland degradation leads to homogenization of the biota at local and landscape scales. Freshwater Biol. 2008; 53: 2402–2413. 28. Price EPF, Spyreas G, Matthews JW. Biotic homogenization of regional wetland plant communities within short timescales in the presence of an aggressive invader. J. Ecol. 2018; 106: 1180–1190. 29. Zhang M, Garcia Molinos J, Zhang X, Xu J. Functional and taxonomic differentiation of macrophyte assemblages across the Yangtze river floodplain under human impacts. Front Plant Sci. 2018; https:// doi.org/10.3389/fpls.2018.00387 PMID: 29636763 30. Olden JD, Poff N. Toward a mechanistic understanding and prediction of biotic homogenization. Am Nat. 2003; 162: 442–460. https://doi.org/10.1086/378212 PMID: 14582007 31. Vitt DH, Li Y, Belland RJ. Patterns of bryophyte diversity in peatlands of continental western Canada. Bryologist. 1995; 98: 218–227. 32. Warner BG, Asada T. Biological diversity of peatlands in Canada. Aquat Sci. 2006; 68: 240–253. 33. Gunnarsson U, Maimer N, Rydin H. Dynamics or constancy in Sphagnum dominated mire ecosystems? A 40-year study. Ecography. 2002; 25: 685–704. 34. Berendse F, Van Breemen N, Rydin H, Buttler A, Heijmans M., Hoosbeek MR, et al. Raised atmo- spheric CO2 levels and increased N deposition cause shifts in plant species composition and production in Sphagnum bogs. Glob Change Biol. 2001; 7: 571–598. 35. Pellerin S, Lavoie C. Reconstructing the recent dynamics of mires using a multitechnique approach. J Ecol. 2003; 91: 1008–1021. 36. Linderholm HW, Leine M. An assessment of twentieth century tree-cover changes on a Southern Swed- ish peatland combining dendrochronology and aerial photograph analysis. Wetlands. 2004; 24: 357– 37. Kapfer J, Grytnes JA, Gunnarsson U, Birks HJB. Fine-scale changes in vegetation composition in a boreal mire over 50 years. J Ecol. 2011; 99: 1179–1189. 38. Ha ´ jkova ´ P, Ha ´ jek M, Rybnı ´ček K, Jirous ˇ ek M, Tichy ´ L, Kra ´ lova ´ S, Mikula ´ s ˇ kova ´ E. Long–term vegetation changes in bogs exposed to high atmospheric deposition, aerial liming and climate fluctuations. J Veg Sci. 2011; 22: 891–904. 39. Ireland AW, Booth RK. Upland deforestation triggered an ecosystem state-shift in a kettle peatland. J Ecol. 2012; 100: 586–596. 40. Talbot J, Roulet NT, Sonnentag O, Moore TR. Increases in aboveground biomass and leaf area 85 years after drainage in a bog. Botany. 2014; 92: 713–721. 41. Pasquet S, Pellerin S, Poulin M. Three decades of vegetation changes in peatlands isolated in an agri- cultural landscape. Appl Veg Sci. 2015; 18: 220–229. 42. Laine J, Vasander H, Sallantaus T. Ecological effects of peatland drainage for forestry. Environ Rev. 1995; 3: 286–303. PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 13 / 15 Temporal vegetation changes in bogs 43. Lachance D, Lavoie C, Desrochers A. The impact of peatland afforestation on plant and bird diversity in southeastern Que ´ bec. Ecoscience. 2005; 12: 161–171. 44. Pellerin S, Mercure M, Desaulniers AS, Lavoie C. Changes in plant communities over three decades on two disturbed bogs in southeastern Que ´ bec. Appl Veg Sci. 2009; 12: 107–118. 45. Favreau M, Pellerin S, Poulin M. Tree encroachment induces biotic differentiation in Sphagnum-domi- nated bogs. Wetlands. 2019; 39: 841–852. 46. Pellerin S, Lavoie M, Boucheny A, Larocque M, Garneau M. Recent vegetation dynamics and hydrolog- ical changes in bogs located in an agricultural landscape. Wetlands. 2016; 36: 159–168. 47. Statistics Canada. St-Laurent Lowlands ecoregion. 2017. Available from: http://www.statcan.gc.ca/pub/ 16-002-x/2010002/tbl/11285/tbl001-eng.htm. 48. Poulin M, Pellerin S, Cimon-Morin J, Lavalle ´ e S, Courchesne G, Tendland Y. Inefficacy of wetland legis- lation for conserving Quebec wetlands as revealed by mapping of recent disturbances. Wetl Ecol Manag. 2016; 24: 651–665. 49. Beauregard P, Lavoie M, Pellerin S. Recent gray birch (Betula populifolia) encroachment in temperate peatlands of eastern North America. Wetlands. 2019; https://doi.org/10.1007/s13157-019-01186-3 50. Environment Canada. Canadian Climate Normals 1981–2010. 2011. Available from: http://climate. weather.gc.ca/climate_normals/ 51. Buteau P. Atlas des tourbières du Que ´ bec me ´ ridional. Que ´ bec: Direction ge ´ ne ´ rale de l’exploration ge ´ o- logique et mine ´ rale, Direction de la recherche ge ´ ologique, Service ge ´ ologique de Que ´ bec; 1989. 52. Kopecky ´ M, Macek M. Vegetation resurvey is robust to plot location uncertainty. Divers Distrib. 2015; 21: 322–330. https://doi.org/10.1111/ddi.12299 PMID: 28503083 53. Brouillet L, Coursol F, Meades SJ, Favreau M, Anions M, B.lisle P, Desmet P. VASCAN, the Database of Vascular Plants of Canada. 2017 Jul 10. Available from: http://data.canadensys.net/vascan/ 54. Reed PB. National list of plant species that occur in wetlands: national summary [report no. 88–24]. Washington DC: U.S. Department of the Interior, Fish and Wildlife Service; 1988. 55. Lichvar RW, Butterwick M, Melvin NC, Kirchner WN. The national wetland plant list: 2014 update of wet- land ratings. Phytoneuron. 2014; 41: 1–42. 56. Lapointe M, Lavoie A, Leboeuf M. Plantes de milieux humides et de bord de mer du Que ´ bec et des Mar- itimes. Que ´ bec: Editions Michel Quintin; 2014. 57. Kobe RK, Pacala SW, Silander JA, Canham CD. Juvenile tree survivorship as a component of shade tolerance. Ecol Appl. 1995; 5: 517–532. 58. Kattge J, Dı ´az S, Lavorel S, Prentice IC, Leadley P, Bo ¨ nisch G, et al. TRY—a global database of plant traits. Glob Change Biol. 2011; 17: 2905–2935. 59. USDA, NRCS. The PLANTS Database. 2015 May 6. Available from http://plants.sc.egov.usda.gov 60. Humbert L, Gagnon D, Kneeshaw D, Messier C. A shade tolerance index for common understory spe- cies of northeastern North America. Ecol Indic. 2007; 7: 195–207. 61. Camilli G, Hopkins KD. Applicability of chi-square to 2x2 contingency tables with small expected cell fre- quencies. Psychol Bull. 1978; 85: 163–167. 62. Anderson MJ. Distance-based tests for homogeneity of multivariate dispersions. Biometrics. 2006; 62: 245–253. https://doi.org/10.1111/j.1541-0420.2005.00440.x PMID: 16542252 63. Legendre P, De Ca ´ ceres M. Beta diversity as the variance of community data: dissimilarity coefficients and partitioning. Ecol Lett. 2013; 16: 951–963. https://doi.org/10.1111/ele.12141 PMID: 23809147 64. Anderson MJ. Permutation tests for univariate or multivariate analysis of variance and regression. Can J Fish Aquat Sci. 2001; 58: 626–639. 65. Anderson MJ, Walsh DCI. PERMANOVA, ANOSIM, and the Mantel test in the face of heterogeneous dispersions: What null hypothesis are you testing? Ecolo Monogr. 2013; 83: 557–574. 66. Legendre P. Interpreting the replacement and richness difference components of beta diversity. Global Ecol Biogeogr. 2014; 23: 1324–1334. 67. R Core Team. R: A Language and Environment for Statistical Computing. Vienna: R Foundation for Statistical Computing; 2016. 68. Lenth RV. Least-Squares Means: the package. J Stat Softw. 2016; 69: 1–33. 69. Oksanen J, Kindt R, Legendre P, O’Hara B, Stevens MHH, Oksanen MJ, Suggests MASS. Vegan: Community Ecology Package. R Package Version 2.4–2; 2017. 70. Broman KW, Broman AT. broman: Karl Broman’s R Code. R Package Version 0.67–4; 2017. 71. Swinehart AL, Parker GR. Palaeoecology and development of peatlands in Indiana. Am Midl Nat. 2006; 143: 267–297. PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 14 / 15 Temporal vegetation changes in bogs 72. Warren RJ II, Rossell IM, Moorhead KK. Colonization and establishment of red maple (Acer rubrum) in a southern Appalachian wetland. Wetlands. 2004; 24: 364–374. 73. Wein RW. Biological flora of the British Islands. Eriophorum vaginatum L. J Ecol. 1973; 61: 601–615. 74. Lavoie C, Marcoux K, Saint-Louis A, Price JS. The dynamics of a cotton-grass (Eriophorum vaginatum L.) cover expansion in a vacuum-mined peatland, southern Que ´ bec, Canada. Wetlands, 2005; 25: 64– 75. Helm A, Hanski I, Pa ¨ rtel M. Slow response of plant species richness to habitat loss and fragmentation. Ecol Lett. 2006; 9: 72–77. https://doi.org/10.1111/j.1461-0248.2005.00841.x PMID: 16958870 76. Frele ´ choux F, Buttler A, Schweingruber FH, Gobat JM. Stand structure, invasion, and growth dynamics of bog pine (Pinus uncinata var. rotundata) in relation to peat cutting and drainage in the Jura Moun- tains, Switzerland. Can J For Res. 2000; 30: 1114–1126. 77. Yagouti A, Boulet G, Vincent L, Vescovi L, Mekis E. Observed changes in daily temperature and precipi- tation indices for southern Que ´ bec, 1960–2005. Atmos Ocean. 2008; 46: 243–356. 78. Heijmans MMPD, Mauquoy D, van Geel B, Berendse F. Long-term effects of climate change on vegeta- tion and carbon dynamics in peat bogs. J Veg Sci. 2008; 19: 307–320. 79. Turunen J, Roulet NT, Moore TR. Nitrogen deposition and increased carbon accumulation in ombro- trophic peatlands in eastern Canada. Global Biogeochem Cycles 2004; 18: https://doi.org/10.1029/ 2003GB002154 80. Payette S. Late-Holocene development of subarctic ombrotrophic peatlands: allogenic and autogenic succession. Ecology 1988; 69: 516–531. 81. Loiselle J, Yu Z. Surface vegetation patterning controls carbon accumulation in peatlands. Geophys Res Lett 2013; 40: 5508–5513. 82. Gałka N, Szal M, Broder T, Loiselle J, Knorr KH. Peatbog resilience to pollution and climate change over the past 2700 years in the Harz Mountains, Germany. Ecol Indic 2019; 97: 183–193. PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 15 / 15 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png PLoS ONE Public Library of Science (PLoS) Journal

Vegetation changes in temperate ombrotrophic peatlands over a 35 year period

PLoS ONE, Volume 15 (2) – Feb 13, 2020

Loading next page...
 
/lp/public-library-of-science-plos-journal/vegetation-changes-in-temperate-ombrotrophic-peatlands-over-a-35-year-l1SEtVE0VK
Publisher
Public Library of Science (PLoS) Journal
Copyright
Copyright: © 2020 Pinceloup et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: 1982 data are available in the Examine database of the Ministère Énergie et Ressources naturelles of Québec, (http://sigeom.mines.gouv.qc.ca/signet/classes/I1102_examine?l=F). Relevant data are within the manuscript and its Supporting Information files. Funding: This research received financial support from the Natural Sciences and Engineering Research Council of Canada (Discovery grants to Stéphanie Pellerin Pellerin: RGPIN-2014-05367 and Monuique Poulin RGPIN-2014-05663). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist.
eISSN
1932-6203
DOI
10.1371/journal.pone.0229146
Publisher site
See Article on Publisher Site

Abstract

a1111111111 a1111111111 Global changes in climate and land use are occurring at an unprecedented rate, often trig- a1111111111 gering drastic shifts in plant communities. This study aims to reconstruct the changes that a1111111111 occurred over 35 years in the plant communities of temperate bogs subjected to indirect human-induced disturbances. In 2015–17, we resurveyed the vascular flora of 76 plots located in 16 bogs of southern Que ´ bec (Canada) first sampled in 1982. We evaluated changes in species richness, frequency of occurrence and abundance, while considering OPENACCESS species shade-tolerance and preferential habitat. We calculated beta diversity as between- Citation: Pinceloup N, Poulin M, Brice M-H, site similarities in composition, and evaluated differences between the two surveys using Pellerin S (2020) Vegetation changes in temperate ombrotrophic peatlands over a 35 year period. tests for homogeneity in multivariate dispersion. We found a significant increase in species PLoS ONE 15(2): e0229146. https://doi.org/ richness and beta diversity over the last 35 years associated with major species turnovers, 10.1371/journal.pone.0229146 indicating a biotic differentiation of the Sphagnum-bog plant communities. These changes Editor: RunGuo Zang, Chinese Academy of were mostly associated with an increase in the abundance and frequency of shade-tolerant Forestry, CHINA and facultative species, suggesting a global phenomenon of woody encroachment. Received: September 11, 2019 Because the observed changes occurred in a few decades on sites free of in situ human dis- Accepted: January 30, 2020 turbances, we suggest that they were likely induced by the synergic effect of the agricultural drainage occurring in the surrounding mineral soils, climate warming, and nitrogen atmo- Published: February 13, 2020 spheric depositions. We also believe that further changes are to be expected, as the trigger- Peer Review History: PLOS recognizes the ing factors persist. Finally, our results highlight the need for increased bog conservation or benefits of transparency in the peer review process; therefore, we enable the publication of restauration efforts. Indeed, a rise in beta diversity due to the introduction of nearby terres- all of the content of peer review and author trial species could induce biotic homogenization of the bog flora with that of surrounding responses alongside final, published articles. The habitats and ultimately impoverish the regional species pool. editorial history of this article is available here: https://doi.org/10.1371/journal.pone.0229146 Copyright:© 2020 Pinceloup et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and Introduction reproduction in any medium, provided the original Global changes in climate and land use are occurring at an unprecedented rate [1,2], often trig- author and source are credited. gering drastic shifts in plant communities [3,4]. The overriding influence of human activities Data Availability Statement: 1982 data are on plant communities has been shown to induce modification in species abundance [5,6] and available in the Examine database of the Ministère ´ distribution [7,8], and to increase the risk of population extinction [9,10]. At a global scale, Energie et Ressources naturelles of Quebec, (http:// sigeom.mines.gouv.qc.ca/signet/classes/I1102_ such changes have most often been associated with biodiversity loss through the simplification PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 1 / 15 Temporal vegetation changes in bogs examine?l=F). Relevant data are within the of communities (decrease in beta diversity), a process called biotic homogenization [11,12]. manuscript and its Supporting Information files. Biotic homogenization usually occurs when exotic or ruderal native species replace specialist native species [11,13,14]. Nevertheless, local species invasions and extinctions are not necessar- Funding: This research received financial support from the Natural Sciences and Engineering ily causally related and can follow complex pathways that can also lead to biotic differentiation Research Council of Canada (Discovery grants to (increase in beta diversity) [15–17] or even no change in beta diversity [18]. Biotic differentia- Ste ´phanie Pellerin Pellerin: RGPIN-2014-05367 tion has mostly been observed at local scales when the introduction of numerous exotic species and Monuique Poulin RGPIN-2014-05663). The outweighs the loss of natives or when disturbances increase habitat heterogeneity thus promot- funders had no role in study design, data collection ing the introduction of various different species [15,19–21]. and analysis, decision to publish, or preparation of the manuscript. Wetland flora are usually more homogeneous than terrestrial flora at local and continental scales [22,23]. This naturally low level of beta diversity within and among sites is likely related, Competing interests: The authors have declared among other reasons, to the uniformity of stressful environmental conditions found in wet- that no competing interests exist. lands that can be tolerated by only a small set of specialized species [22,24,25]. Despite wet- lands being among the most disturbed ecosystems in the world [26], few studies have evaluated whether direct and indirect human disturbances induce further biotic homogeniza- tion in their plant communities (e.g., [17,27–29]). Among them, Price et al. [28] found a subtle process of biotic homogenization over time in the herbaceous emergent communities of palus- trine wetlands of Illinois (USA). This process was associated with an increase in the presence and abundance of the exotic Phalaris arundinacea from the Poaceae family, and the decline of numerous other native and exotic species. On the other hand, Brice et al. [17] observed biotic differentiation in the herbaceous communities of urban riparian swamps due, in part, to the increase in habitat heterogeneity induced by urbanization. Similarly, Zhang et al. [29] showed taxonomic differentiation of macrophyte communities on a river floodplain exposed to human-induced disturbances. In fact, because many types of wetlands are naturally species- poor, they are potentially more susceptible to biotic differentiation than to homogenization following disturbances [30]. Ombrotrophic peatlands (bogs) are peat-forming naturally species-poor wetlands [31,32] that are mostly viewed as resilient ecosystems with important compositional changes usually occurring over centuries to millennia [33]. Still, several studies have shown recent rapid and drastic compositional changes in response to human-induced disturbances (drainage, eutro- phication, climate warming; land-use modifications), through their influence on water and nutrient availability (e.g., [34–40]). For instance, drainage and contemporary climate change (warmer and/or drier climatic conditions) have been shown to enhance shrub and tree encroachment, hamper Sphagnum growth and facilitate the establishment of generalist and exotic species [33, 35,37,41]. This shift from open bog toward forested communities is also associated with a decrease in the cover and richness of other specialist bog plant species (often heliophilic species) and a concomitant increase of shade or drought-tolerant species [33,37,41–44]. More recently, a study investigating the effects of woody encroachment in bog flora, using a space for time substitution, found within-site biotic differentiation associated with the introduction of terrestrial shade-tolerant species [45]. Although vegetation changes in bogs have been the subject of numerous studies (e.g., [33;38,44]), temporal changes in plant communities at a regional scale, and how these changes affect beta diversity, remain to be explored. In this study, we investigated the floristic changes that occurred over 35 years in the vascu- lar plant communities of temperate bogs of southern Que ´bec (Canada). These bogs are mostly dominated by ericaceous shrubs (e.g., Kalmia angustifolia, Rhododendron groenlandicum, Cha- maedaphne calyculata), Sphagnum mosses (e.g., Sphagnum capillifolium, S. magellanicum., S. rubellum) and scattered Picea mariana or Larix laricina thickets. We revisited 76 plots origi- nally surveyed in 1982 and located in 16 different bogs spread over the St. Lawrence Lowlands region. These bogs have been isolated in an agricultural landscape for more than 50 years, PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 2 / 15 Temporal vegetation changes in bogs which has induced a drying of the peat surface on many sites (e.g., [41,46]). We specifically aimed to 1) determine changes in taxa richness, composition and beta diversity and 2) assess whether biotic homogenization or differentiation occurred over time. We hypothesized that the studied sites would be characterized by biotic differentiation in response to species enrich- ment by terrestrial and shade-tolerant plants. We also expected to observe a woody encroach- ment phenomenon. Finally, we acknowledge that focusing on vascular plant communities may underestimate observed changes as mosses may represent a high proportion of the diver- sity in bogs, but mosses were rarely identified at the species level in 1982. Methodology Study area The study area is in the St. Lawrence Lowlands, southern Que ´bec (Canada), and extends from 45˚01’N to 46˚50’N latitude and from 70˚55’W to 74˚17’W longitude. The area is enclosed by the Canadian Shield to the north and the Appalachian Mountains to the south. The Lowlands have a flat topography and are characterized by deep arable soils derived from glacial and marine deposits. The landscape is composed of about 50% agricultural fields, 30% woodlands, 10% urbanized areas and 10% wetlands [47], of which 35% are bogs [48]. Over the last few decades, 19% of the wetland areas have been lost or disturbed by human activities, mainly agri- culture and forestry, but also urbanization and peat mining [48]. Tree encroachment by Picea mariana, Pinus banksiana, Larix laricina, Betula populifolia, and Acer rubrum has been reported in many of the remaining bogs [35,41,45,46,49]. The mean annual temperature of the study area fluctuates from 4.1˚C in the northeast to 6.7˚C in the southwest [50]. Average precipitation ranges from a high of 1120 mm in the northeast to a low of 965 mm in the southwest, of which 23% and 17% fall as snow, respectively [50]. Original sampling During the 1980s, a large inventory campaign was conducted by the Que ´bec Ministry of Energy and Natural Resources to evaluate the peat resource in southern Que ´bec peatlands greater than 40 ha and with a peat deposit thicker than 30 cm [51]. During this large-scale inventory, hundreds of vegetation surveys were conducted. All plots in the study area were sampled in 1982. All plots (20 m × 20 m) were delineated in the center of a seemingly homoge- nous plant community. The percent cover of all vascular taxa and strata (tree, shrub, herbs, mosses) was visually estimated according to six classes: <1%, 1-5%, 6–25%, 26–50%, 51–75%, >75%. Mosses were rarely identified at the species level in original surveys and thus not inves- tigated in the present study. Other information available in the field sheets includes geographic coordinates of the plot, peat thickness, soil pH, biotope relative dominance (pool, hummock, hollow, etc.), vegetation structure (woody, shrubby, herbaceous, etc.) and general information on the location of the plot within the bog (margins, expanse, near a drainage ditch, etc.). It should be noted that vegetation data from this original survey has never been analyzed. Resampling: 2015 and 2017 Historical sampling plots were not permanently marked. However, as mentioned above geo- graphic coordinates were available from the original field sheets. Before field sampling, we dis- carded all original plots established on minerotrophic peatland habitats or disturbed areas. We also rejected all plots located in areas that had been affected by land conversion or subjected to woodcutting since earlier surveys. We identified these areas using 1980–2015 Google Earth PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 3 / 15 Temporal vegetation changes in bogs Digital Globe satellite imagery. During the summer of 2015 and 2017 (hereafter referred to as 2017 for simplification), 76 of the original plots spread over 16 bogs were relocated using a GPS (S1 Table). All plots were on private lands or on lands protected by non-governmental agencies. We obtained a verbal permission from the owner of each site before sampling. In addition to geographic coordinates, we drew upon all environmental information available from original studies in this endeavor. We surveyed the vegetation following the same method- ology used in 1982. Due to the abundance of visual landmarks, we estimate that we were able to relocate all plots within 30–40 m from their original position. Although a relocation error could introduce a source of variance in the dataset, systematic biases favoring vegetation change in one direction are unlikely to occur, especially when environmental variables are used in combination with geographic location to relocate unmarked plots [13,52]. Data analysis Before analyses, taxonomic nomenclature was standardized according to the Database of Vas- cular Plants of Canada (VASCAN) [53]. We lumped all sedge taxa (Carex) for both years at the genus level, as very few individuals were identified to the species level in 1982. We also merged Vaccinium myrtilloides and Vaccinium angustifolium under Vaccinum cf. angustifolium because they were often undistinguished in 1982. Taxa richness. Change in taxa richness at the plot level (alpha diversity) between 1982 and 2017 was evaluated using a Wilcoxon signed rank test. We also considered each taxa’s habi- tat preferences by comparing richness of taxa sorted by their wetland indicator status [54]. Accordingly, taxa that almost always occur in wetlands were classified as “obligate”, those that usually occur in wetlands, but may occur in non-wetland habitats as “facultative”, and those that occur equally in wetlands and non-wetland habitats or that usually occur in non-wetland habi- tats as “non-wetland” (S2 Table). No taxa that almost always occur in non-wetland habitats (upland) were found. The wetland indicator status of each taxon was retrieved from the National Wetland Plant List [55] and Lapointe et al. [56]. Generalized Poisson mixed models were used to compare the number of taxa between groups of habitat preferences and between years as well as the interaction between both factors. Plots were treated as a random factor, allowing us to con- trol for pairing between plots and dependence between the numbers of taxa of different prefer- ences in the same plot. Because a significant interaction was found between year and habitat preference, we conducted post-hoc Tukey tests between habitats independently for each year. We verified that model residuals followed a normal distribution of homogeneous variance. We also compared richness of taxa according to their tolerance to shade [57] (S2 Table). We classified taxa into three broad categories of shade tolerance: “tolerant” for taxa that grow well under a limited amount of light, “mid-tolerant” for taxa that can tolerate shade during some phases of their life cycle, and “intolerant” for taxa that only grow with a high amount of available light. Shade tolerance information was retrieved from the online databases TRY [58] and PLANTS [59], as well as from Humbert et al. [60]. We used the same linear mixed-effect model procedure described above. Taxa frequency and abundance. Taxa with the greatest changes in terms of frequency of occurrence were identified by comparing the number of plots occupied by each taxon for each year using Chi-square goodness-of-fit tests (taxa were tested one by one). A Yates continuity correction was applied [61]. Finally, we evaluated whether the abundance of some taxa changed over time using plant cover. The cover data were transformed into percentages using the midpoint of each class. Only species with a mean of� 5% coverage for at least one of the sampling periods were used (12 species analyzed). Differences between 1982 and 2017 were tested using a two-tailed paired t-test (species were tested one by one). PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 4 / 15 Temporal vegetation changes in bogs Beta diversity. Differences in beta diversity between 1982 and 2017 were analyzed using a distance-based test for homogeneity of multivariate dispersions (PERMDISP) [62]. PERM- DISP calculates the distance of each site (here, plot) to the centroid in an ordination space (principal coordinate analysis) and then tests whether these distances are different between groups (here, sampling years) through permutation tests. More precisely, two site-by-species matrices were first computed; one using covers (median of cover classes) and the other using presence-absence data. Then, for each matrix, a site-by-site distance matrix was built using the Hellinger distance [63]. This distance matrix was used to compute the centroid of each group of sites. The distance of each site to its associated group centroid was calculated, and the dis- persion of these distances (within-group variance) was used as an estimate of beta diversity (the greater the within-group variance, the higher the beta diversity). The site distances to cen- troid were subjected to a paired t-test with 9999 permutations to determine whether disper- sions differed between groups. To detect a shift in species composition between year (i.e., turnover), we tested for location differences between centroids using PERMANOVA with pseudo-F ratios (9999 permutations) [64]. Because this test is sensitive to differences in multi- variate dispersions [65], data visualization was used to support the interpretation of the statisti- cal test. The differences in multivariate dispersion and composition were illustrated in principal coordinates analysis ordinations (PCoA). Finally, we decomposed beta diversity into replacement (turnover) and richness difference (species gain/loss) using Jaccard dissimilarity indices and presence-absence data [66]. We decomposed beta diversity for each year as well as between years. All statistical analyses were performed using R 3.3.2 [67]. Wilcoxon signed rank tests were performed using the wilcox.test, chi-square tests using chisq.test, paired t-test using t.test, and generalized Poisson mixed models using glm(), all from the stats package [67]. Tukey tests were done with the lsmeans() function in the lsmeans library [68]. Hellinger transformations were done using decostand(), multivariate dispersion analyses were performed using betadis- per(), centroid locations were tested using adonis2(), all from the vegan package [69]. Permuta- tion t-tests were done using perm.test(), from the broman package [70]. Beta diversity partitioning was conducted using the beta.div.comp function [66]. Results Taxa richness A total of 37 vascular taxa were recorded in 1982 and 57 in 2017 (total = 67), and only one exotic (Frangula alnus) was found, in 2017 (S2 Table). Thirty new taxa were detected in 2017, while ten were lost. Most of the taxa gained were found in <10 plots, except Aronia melano- carpa (33 plots), Gaylussacia baccata (18 plots), Vaccinium corymbosum (18 plots), Erio- phorum virginicum (17 plots), Platanthera blephariglottis var. blephariglottis (14 plots) and Coptis trifolia (11 plots). Furthermore, few of the taxa gained were abundant in the plots where they were found (mean cover >5%): Rubus allegheniensis (39% mean cover; 4 plots), V. corym- bosum (24%; 18 plots), Frangula alnus (22%; 3 plots), G. baccata (12%; 18 plots) and Osmun- dastrum cinnamomeum (8%; 5 plots). All taxa lost were present in five plots or less in 1982, except Alnus incana subsp. rugosa (8 plots). Taxa richness per plot ranged from 3 to 15 in 1982 and from 7 to 21 in 2017 (Fig 1). The number of taxa increased in 61 plots, remained similar in five plots and decreased in ten plots. Overall, mean richness per plot significantly increased over time, from an average of 10 to 14 taxa (V = 165; p� 0.005). The number of “facultative” and “non-wetland” taxa per plot significantly increased with time, while the number of “obligate” taxa remained similar (Fig 2A). The number of “obligate” PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 5 / 15 Temporal vegetation changes in bogs Fig 1. Vascular taxa richness in temperate bogs of southern Quebec (Canada) in 1982 and 2017. Plots sampled in 1982 are compared to the same plots re-sampled in 2017. The plain diagonal line is the 1:1 line (no change in richness between the two years) and the dashed line represents the mean richness increase. Grey dots represent at least two plots with the exact same number of species in 1982 and 2017. https://doi.org/10.1371/journal.pone.0229146.g001 Fig 2. Richness of a) obligate, facultative and non-wetland taxa and b) shade tolerant, mid-tolerant and intolerant taxa in 1982 and 2017 in temperate bogs of southern Que ´bec (Canada). Provided are mean taxa richness (black dot), median (line), 25–75% quartiles (boxes) and ranges (whiskers). Different lower-case letters indicate a significant difference (α = 0.05; Generalized Poisson mixed models) between years, whereas different upper-case letters indicate a significant difference (α = 0.05; Tukey’s test) within the year. https://doi.org/10.1371/journal.pone.0229146.g002 PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 6 / 15 Temporal vegetation changes in bogs Fig 3. Plant taxa with significant changes (p� 0.05) in the frequency of occurrence (number of plots where species occurs) between 1982 and 2017 in temperate bogs of southern Que ´bec (Canada). Taxa were tested one by one (chi-square goodness-of-fit tests with Yates correction). On the left side are taxa with lower frequency in 2017 than in 1982, and on the right side, taxa with higher frequency in 2017 than in 1982. https://doi.org/10.1371/journal.pone.0229146.g003 taxa was about two times greater than that of “facultative” and “non-wetland” taxa in 1982 and about 1.5 times greater in than that of “facultative” taxa in 2017 (Fig 2A). The amount of shade “tolerant” and “mid-tolerant” taxa per plot significantly increased over time, while the richness of shade “intolerant” taxa remained similar for both years (Fig 2B). It follows that the number of “shade-intolerant” taxa was greater than that of the two other groups of taxa for both years. Species frequency and abundance Of the 67 taxa tested, 14 significantly changed their frequency of occurrence over time (Fig 3) and six experienced a significant change in their abundance (Fig 4). More precisely, 11 taxa significantly increased in frequency, among which Acer rubrum and Vaccinium corymbosum also increased in abundance. The increase in frequency was particularly marked (�20 plots) for Cypripedium acaule, Aronia melanocarpa, Gaultheria hispidula, Ilex mucronata and Acer rubrum. Taxa that became more frequent or abundant over time were mostly “facultative” or “non-wetland” taxa. Inversely, two taxa significantly decreased in frequency (Fig 3) and four decreased in abundance (Fig 4). Only Eriophorum vaginatum subsp. spissum decreased both in frequency and abundance. While E. vaginatum was still found in 43 plots in 2017 (62 plots in 1982), Betula pumila was found in only one plot the same year (22 plots in 1982). Taxa decreas- ing in frequency or abundance were all wetland “obligate” taxa. Beta diversity Beta diversity increased between 1982 and 2017 (presence-absence: T = 98.79; p< 0.0001; abundance data: T = 109.28; p< 0.0001), meaning that vascular plant communities of bogs dif- fered more markedly from one site to another in 2017 than in 1982 (Fig 5). This increase in beta diversity was associated with composition turnover as a significant difference in centroid position between years was also found for both presence-absence (F = 9.085; p� 0.0001) and abundance data (F = 9.514; p� 0.0001). PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 7 / 15 Temporal vegetation changes in bogs Fig 4. Plant taxa mean cover in 2017 vs 1982 in temperate bogs of southern Que ´bec (Canada). Only taxa with significant change in mean cover (two-tailed paired t-test; p� 0.05) are presented. The plain line is the 1:1 line. Carex spp. is plotted with a different symbol, as all species were lumped together and, therefore, not associated to a particular wetland status. https://doi.org/10.1371/journal.pone.0229146.g004 Finally, beta diversity in both 1982 and 2017, as well as beta diversity changes between 1982 and 2017, resulted mostly from taxa replacement rather than richness differences (Fig 6). For instance, when comparing all pairs of plots between 1982 and 2017, replacement accounted for 48.5% of the beta diversity decomposition while richness difference accounted only for 12.6%. Discussion In this study, we hypothesized that the studied sites would be characterized by a temporal biotic differentiation of the vascular flora in temperate bogs in response to species enrichment by terrestrial shade-tolerant plants. Our results showed that the richness of the studied bogs increased from 1982 to 2017, mostly due to the establishment of facultative and shade-tolerant plants. As expected, these changes induced a biotic differentiation, since we found higher beta diversity in 2017 than 35 years prior. This decrease in similarity with time mostly represented the outcome of species turnover. This indicates that newly established taxa are not similar across the study area and/or that various taxa are being replaced over time. Changes in plant environmental preferences The analyses of plant richness according to individual habitat preference indicated an enrich- ment of taxa associated to drier habitats as we found more facultative and non-wetland plants in recent surveys than in the older ones. For example, Acer rubrum, a species that usually occur in non-wetland habitats, which was present in only five plots in 1982, is more recently present in 25 plots and has experienced a significant increase in cover over time (Figs 3 and 4). Acer rubrum introduction and increase in density, often following drainage, have been recently documented in many North American wetlands including bogs [50,71,72]. On the other hand, PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 8 / 15 Temporal vegetation changes in bogs Fig 5. Effect of time on beta diversity for vascular plant taxa in bogs of southern Quebec (Canada), using presence-absence data. Beta diversity was measured as the distance of sites to their group centroid, here represented on the first two axes of the PCoA and using the boxplot of the sites-to-centroid distance. Provided are median (line), 25–75% quartiles (boxes) and ranges (whiskers). Different letters indicate a significant difference (p<0.05) determined by t-test. Circles are ellipses of standard deviation. https://doi.org/10.1371/journal.pone.0229146.g005 Fig 6. Triangular plot of the relationship among a) the 76 plots in 1982 and 2017, and b) for all pairs of plots between 1982 vs 2017, illustrating the contributions of three components of beta diversity over time: similarity of species composition, richness difference and replacement of species. The large dots inside the triangle represent the centroid, and the smaller dots represent the mean values of the three components. Values are indicated as raw percentages (i.e. 0.168 equals 16.8% of total beta diversity). https://doi.org/10.1371/journal.pone.0229146.g006 PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 9 / 15 Temporal vegetation changes in bogs although the richness of obligate wetland plants did not significantly change over time, some of these plants experienced a noteworthy decrease in frequency and abundance. This was the case for Eriophorum vaginatum subsp. spissum, a deep rooting wetland obligate typical of open bogs [73]. Despite the strong capacity of E. vaginatum to persist in peatlands, notably due to its typical tussock growth with basal meristematic tissue, surveys in vacuum mined peatlands have revealed its sensitivity to water-table drawdown [74]. Our results suggest that the studied bogs experienced an increase in shading level over time, as indicated by the gain of shade-tolerant and mid-tolerant plants. This increased level of shading is likely the result of shrub and tree encroachment in bogs, as suggested, among other reasons, by the increase in frequency and abundance of Acer rubrum and Vaccinium corymbo- sum. In fact, a quarter of the plots investigated presented an increase in tree cover (mean increase of 4.7%) and, more importantly, three quarter of the plots showed an increase in shrub (mean increase of 37.2%) cover (S3 Table). While our results indicate that shaded habi- tats are becoming more widespread in the studied bogs, we also found no significant change in the richness of shade-intolerant species, although the variability of the data was higher in 2017 than in 1982 (Fig 1). Furthermore, most of the species that significantly decreased in occur- rence and/or abundance (Betula pumila, Chamaedaphne calyculata, E. vaginatum) were shade-intolerant. The lack of significant change in the richness of the shade-intolerant species at the plot level may be due to the time-lag after disturbance, which implies that changes may still occur in the coming decades [75]. Overall, a more variable hydrologic environment and structure of plant strata likely allowed the co-occurrence of heliophilic bog specialists and shade and drought-tolerant plants more typical of forest environments. These results concur with the phenomenon of woody encroachment increasingly observed in bogs worldwide (e.g., [35,40,41,50]). Beta diversity As expected, we found a biotic differentiation (increase in beta diversity) over a period of 35 years in temperate Sphagnum-dominated bogs of the southern Quebec. Changes in beta diver- sity are led by two primary components: species turnover and richness differences [30,66]. Although we found an increase in plant richness over time in nearly all the sampling plots, our results indicated that the observed changes in beta diversity were mainly due to species replacement. For instance, our beta diversity partitioning analyses indicated that the influence of species replacement was four times greater than that of richness difference in beta diversity changes between 1982 and 2017. A recent study using a time-for-space substitution, conducted in the same study area but in different bogs, also found a higher beta diversity on sites recently subjected to woody encroachment than in open bog habitats, and this increase was also mostly driven by species turnover [45]. Significant species turnover with time, mainly driven by the replacement of typical bog species by trees, shrubs and other widespread forest species, was also found in Sweden [33,37], although the effect on beta diversity was not analysed. Potential causes of the observed changes Rapid changes in peatland plant communities have been mostly attributed to human-induced disturbances such as increasing temperature, atmospheric nitrogen deposition and drainage [34–40, 76]. In the studied bogs, the observed changes probably resulted from the interaction among all these factors, of which those influencing the hydrology of the sites likely predomi- nated. Because the study sites are located in seemingly undisturbed and undrained bogs, the gradual drying of the peat deposits likely resulted from the extensive drainage of mineral soils in the surrounding agricultural catchments [41,46,50]. For instance, using paleohydrological PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 10 / 15 Temporal vegetation changes in bogs reconstruction (testate amoebae), it was recently showed that the water level drawdown in two bogs of the study area matches the peak of agricultural activities and ditch-digging mechaniza- tion in their surrounding mineral catchment [46]. On the other hand, since 1960, southern Que ´bec has experienced an average temperature increase of 1.2˚C [77], which may also have contributed to peat drying. Indeed, according to a modeling study [78], even a 1˚C tempera- ture increase can dry the peat surface enough to shift open bogs into tree-dominated ecosys- tems. In conjunction with regional drainage and climate warming, nutrient enrichment may have facilitated the observed changes. Indeed, six of the 16 bogs studied are located in a region with some of the highest levels of wet nitrogen deposition in eastern North America [79]. All the bogs might also have been enriched by dry mineral dust from adjacent agricultural lands [39,46]. Finally, we could not rule out that some of the changes may also be associated with intrinsic ecosystem dynamics such as peat accumulation and mostly competition among plants (e.g., [80–82]). Conclusion In conclusion, our results reveal that rather rapid compositional changes in the plant commu- nities of temperate Sphagnum-dominated bogs isolated in an agricultural landscape induced an overall biotic differentiation. Because these changes occurred in a few decades within sites free of in situ human disturbances, we suggest that they were likely induced by the synergy effect of drainage in the surrounding mineral agricultural soils, climate warming, and nitrogen atmospheric depositions, as observed in several other bogs. We also believe that further changes are to be expected, since the triggering factors persist and obligate and shade-intoler- ant species remain abundant, unlike other bogs, where woody encroachment is more pro- nounced (e.g., [35]). Finally, our results also highlight the need for increased conservation efforts in such habi- tats, as bogs face two challenges: their surface is being lost to agriculture and urbanization [48] and their floristic distinctiveness is disappearing. Indeed, peatlands are usually species poor ecosystems with a highly specialized flora. An increase in beta diversity of their plant commu- nity due to the introduction of nearby species could increase their similarity with surrounding mesic habitats and ultimately impoverish the regional species pool. Supporting information S1 Table. Geographical coordinates (degree decimal) of the plots. (DOCX) S2 Table. List of plant taxa sampled. Taxa name, shade-tolerance (T = tolerant; M = mid-tol- erant; I = intolerant), wetland indicator status (Obl = obligate; Fac = facultative; NW = non- wetland), frequency of occurrence for both years (number of plots) and in brackets mean abundance in the plots where the species is present are indicated. NA indicates inconsistent or unavailable information. (DOCX) S3 Table. Tree and shrub cover (in %) in sampled plots for both survey periods. (DOCX) Acknowledgments We are grateful to the landowners who allowed us to work on their lands, to field assistants, to S. Daigle and P. Legendre for statistical advice and K. Grislis for linguistic revision. PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 11 / 15 Temporal vegetation changes in bogs Author Contributions Conceptualization: Nicolas Pinceloup, Monique Poulin, Ste ´phanie Pellerin. Data curation: Nicolas Pinceloup, Marie-He ´lène Brice, Ste ´phanie Pellerin. Formal analysis: Nicolas Pinceloup, Marie-He ´lène Brice. Funding acquisition: Monique Poulin, Ste ´phanie Pellerin. Methodology: Nicolas Pinceloup, Monique Poulin, Marie-He ´lène Brice, Ste ´phanie Pellerin. Resources: Monique Poulin, Ste ´phanie Pellerin. Validation: Ste ´phanie Pellerin. Writing – original draft: Nicolas Pinceloup, Monique Poulin, Marie-He ´lène Brice, Ste ´phanie Pellerin. Writing – review & editing: Nicolas Pinceloup, Monique Poulin, Ste ´phanie Pellerin. References 1. Lindner M, Maroschek M, Netherer S, Kremer A, Barbati A, Garcia-Gonzalo J, et al. Climate change impacts, adaptive capacity, and vulnerability of European forest ecosystems. Forest Ecol Manage. 2010; 259: 698–709. 2. Song XP, Hansen MC, Stehman SV, Potapov PV, Tyukavina A, Vermote EF, Townshend R. Global land change from 1982 to 2016. Nature. 2018; 560: 639–643. https://doi.org/10.1038/s41586-018- 0411-9 PMID: 30089903 3. Walker MD, Wahren CH, Hollister RD, Henry GHR, Ahlquist LE, Alatalo JM, et al. Plant community responses to experimental warming across the tundra biome. Proc Natl Acad Sci U S A. 2006; 103: 1342–1346. https://doi.org/10.1073/pnas.0503198103 PMID: 16428292 4. Eppinga MB, Rietkerk M, Wassen MJ, De Ruiter PC. Linking habitat modification to catastrophic shifts and vegetation patterns in bogs. Plant Ecol. 2009; 200: 53–68. 5. Smart SM, Bunce RGH, Marrs R, LeDuc M, Firbank LG, Maskell LC, et al. Large-scale changes in the abundance of common higher plant species across Britain between 1978, 1990 and 1998 as a conse- quence of human activity: tests of hypothesised changes in trait representation. Biol Conserv. 2005; 124: 355–371. 6. McGill BJ, Dornelas M, Gotelli NJ, Magurran AE. Fifteen forms of biodiversity trend in the anthropocene. Trends Ecol Evol. 2015; 30: 104–113. https://doi.org/10.1016/j.tree.2014.11.006 PMID: 25542312 7. Parmesan C, Ryrholm N, Stefanescu C, Hill JK, Thomas CD, Descimon H, et al. Poleward shifts in geo- graphical ranges of butterfly species associated with regional warming. Nature. 1999; 399: 579–583. 8. Dyer EE, Cassey P, Redding DW, Collen B, Franks V, Gastion KJ, et al. The global distribution and driv- ers of alien bird species richness. PloS Biol. 2017 Jan 12. https://doi.org/10.1371-journal.pbio.2000942 9. Sax DF, Gaines SD. Species invasions and extinction: the future of native biodiversity on islands. Proc Natl Acad Sci U S A. 2008; 105: 11490–11497. https://doi.org/10.1073/pnas.0802290105 10. Ceballos G, Ehrlich PR, Barnosky AD, Garcıa A, Pringle RM, Palmer TM. Accelerated modern human- induced species losses: entering the sixth mass extinction. Sci Adv. 2015; https://doi.org/10.1126/ sciadv.1400253 PMID: 26601195 11. McKinney ML, Lockwood JL. Biotic homogenization: a few winners replacing many losers in the next mass extinction. Trends Ecol Evol. 1999; 14: 450–453. https://doi.org/10.1016/s0169-5347(99)01679- 1 PMID: 10511724 12. Olden JD, Rooney TP. On defining and quantifying biotic homogenization. Global Ecol Biogeogr. 2006; 15: 113–120. 13. McCune JL, Vellend M. Gains in native species promote biotic homogenization over four decades in a human-dominated landscape. J Ecol. 2013; 101: 1542–1551. 14. Zeeman BJ, McDonnell MJ, Kendal D, Morgan JW. Biotic homogenization in an increasingly urbanized temperate grassland ecosystem. J Veg Sci. 2017; 28: 550–561. 15. Gong C, Chen J, Yu S. Biotic homogenization and differentiation of the flora in artificial and near-natural habitats across urban green spaces. Landsc Urban Plan. 2013; 120: 158–169. PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 12 / 15 Temporal vegetation changes in bogs 16. Aronson MF, Handel SN, La Puma IP, Clemants SE. Urbanization promotes non-native woody species and diverse plant assemblages in the New York metropolitan region. Urban Ecosyst. 2015; 18: 31–45. 17. Brice MH, Pellerin S, Poulin M. Does urbanization lead to taxonomic and functional homogenization in riparian forests? Divers Distrib. 2017; 23: 828–840. 18. Magura T, Lovei GL, Tothmeresz B Does urbanization decrease diversity in ground beetle (Carabidae) assemblages? Global Ecol Biogeogr 2010; 19: 16–26. 19. Olden JD, Poff N. Ecological processes driving biotic homogenization: testing a mechanistic model using fish faunas. Ecology. 2004; 85: 1867–1875. 20. Qian H, McKinney ML, Ku ¨ hn I. Effects of introduced species on floristic similarity: comparing two US states. Basic Appl Ecol. 2008; 9: 617–625. 21. Petsch DK. Causes and consequences of biotic homogenization in freshwater ecosystems. Int Rev Hydrobiol. 2016; 101: 113–122. 22. Santamarı ´a L. Why are most aquatic plants widely distributed? Dispersal, clonal growth and small-scale heterogeneity in a stressful environment. Acta Oecol. 2002; 23: 137–154. 23. Qian H, Guo Q. Linking biotic homogenization to habitat type, invasiveness and growth form of natural- ized alien plants in North America. Divers Distrib. 2010; 16: 119–125. 24. Barrett SCH, Echert CG, Husband BC. Evolutionary processes in aquatic plant populations. Aquat Bot. 1993; 44: 105–145. 25. Colmer TD, Voesenek LACJ. Flooding tolerance: suites of plant traits in variable environments. Funct Plant Biol. 2009; 36: 665–681. 26. Zedler JB, Kercher S. Wetland resources: status, trends, ecosystem services, and restorability. Annu Rev Environ Resour. 2005; 30: 39–74. 27. Lougheed VL, Mcintosh MD, Parker CA, Stevenson RJ. Wetland degradation leads to homogenization of the biota at local and landscape scales. Freshwater Biol. 2008; 53: 2402–2413. 28. Price EPF, Spyreas G, Matthews JW. Biotic homogenization of regional wetland plant communities within short timescales in the presence of an aggressive invader. J. Ecol. 2018; 106: 1180–1190. 29. Zhang M, Garcia Molinos J, Zhang X, Xu J. Functional and taxonomic differentiation of macrophyte assemblages across the Yangtze river floodplain under human impacts. Front Plant Sci. 2018; https:// doi.org/10.3389/fpls.2018.00387 PMID: 29636763 30. Olden JD, Poff N. Toward a mechanistic understanding and prediction of biotic homogenization. Am Nat. 2003; 162: 442–460. https://doi.org/10.1086/378212 PMID: 14582007 31. Vitt DH, Li Y, Belland RJ. Patterns of bryophyte diversity in peatlands of continental western Canada. Bryologist. 1995; 98: 218–227. 32. Warner BG, Asada T. Biological diversity of peatlands in Canada. Aquat Sci. 2006; 68: 240–253. 33. Gunnarsson U, Maimer N, Rydin H. Dynamics or constancy in Sphagnum dominated mire ecosystems? A 40-year study. Ecography. 2002; 25: 685–704. 34. Berendse F, Van Breemen N, Rydin H, Buttler A, Heijmans M., Hoosbeek MR, et al. Raised atmo- spheric CO2 levels and increased N deposition cause shifts in plant species composition and production in Sphagnum bogs. Glob Change Biol. 2001; 7: 571–598. 35. Pellerin S, Lavoie C. Reconstructing the recent dynamics of mires using a multitechnique approach. J Ecol. 2003; 91: 1008–1021. 36. Linderholm HW, Leine M. An assessment of twentieth century tree-cover changes on a Southern Swed- ish peatland combining dendrochronology and aerial photograph analysis. Wetlands. 2004; 24: 357– 37. Kapfer J, Grytnes JA, Gunnarsson U, Birks HJB. Fine-scale changes in vegetation composition in a boreal mire over 50 years. J Ecol. 2011; 99: 1179–1189. 38. Ha ´ jkova ´ P, Ha ´ jek M, Rybnı ´ček K, Jirous ˇ ek M, Tichy ´ L, Kra ´ lova ´ S, Mikula ´ s ˇ kova ´ E. Long–term vegetation changes in bogs exposed to high atmospheric deposition, aerial liming and climate fluctuations. J Veg Sci. 2011; 22: 891–904. 39. Ireland AW, Booth RK. Upland deforestation triggered an ecosystem state-shift in a kettle peatland. J Ecol. 2012; 100: 586–596. 40. Talbot J, Roulet NT, Sonnentag O, Moore TR. Increases in aboveground biomass and leaf area 85 years after drainage in a bog. Botany. 2014; 92: 713–721. 41. Pasquet S, Pellerin S, Poulin M. Three decades of vegetation changes in peatlands isolated in an agri- cultural landscape. Appl Veg Sci. 2015; 18: 220–229. 42. Laine J, Vasander H, Sallantaus T. Ecological effects of peatland drainage for forestry. Environ Rev. 1995; 3: 286–303. PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 13 / 15 Temporal vegetation changes in bogs 43. Lachance D, Lavoie C, Desrochers A. The impact of peatland afforestation on plant and bird diversity in southeastern Que ´ bec. Ecoscience. 2005; 12: 161–171. 44. Pellerin S, Mercure M, Desaulniers AS, Lavoie C. Changes in plant communities over three decades on two disturbed bogs in southeastern Que ´ bec. Appl Veg Sci. 2009; 12: 107–118. 45. Favreau M, Pellerin S, Poulin M. Tree encroachment induces biotic differentiation in Sphagnum-domi- nated bogs. Wetlands. 2019; 39: 841–852. 46. Pellerin S, Lavoie M, Boucheny A, Larocque M, Garneau M. Recent vegetation dynamics and hydrolog- ical changes in bogs located in an agricultural landscape. Wetlands. 2016; 36: 159–168. 47. Statistics Canada. St-Laurent Lowlands ecoregion. 2017. Available from: http://www.statcan.gc.ca/pub/ 16-002-x/2010002/tbl/11285/tbl001-eng.htm. 48. Poulin M, Pellerin S, Cimon-Morin J, Lavalle ´ e S, Courchesne G, Tendland Y. Inefficacy of wetland legis- lation for conserving Quebec wetlands as revealed by mapping of recent disturbances. Wetl Ecol Manag. 2016; 24: 651–665. 49. Beauregard P, Lavoie M, Pellerin S. Recent gray birch (Betula populifolia) encroachment in temperate peatlands of eastern North America. Wetlands. 2019; https://doi.org/10.1007/s13157-019-01186-3 50. Environment Canada. Canadian Climate Normals 1981–2010. 2011. Available from: http://climate. weather.gc.ca/climate_normals/ 51. Buteau P. Atlas des tourbières du Que ´ bec me ´ ridional. Que ´ bec: Direction ge ´ ne ´ rale de l’exploration ge ´ o- logique et mine ´ rale, Direction de la recherche ge ´ ologique, Service ge ´ ologique de Que ´ bec; 1989. 52. Kopecky ´ M, Macek M. Vegetation resurvey is robust to plot location uncertainty. Divers Distrib. 2015; 21: 322–330. https://doi.org/10.1111/ddi.12299 PMID: 28503083 53. Brouillet L, Coursol F, Meades SJ, Favreau M, Anions M, B.lisle P, Desmet P. VASCAN, the Database of Vascular Plants of Canada. 2017 Jul 10. Available from: http://data.canadensys.net/vascan/ 54. Reed PB. National list of plant species that occur in wetlands: national summary [report no. 88–24]. Washington DC: U.S. Department of the Interior, Fish and Wildlife Service; 1988. 55. Lichvar RW, Butterwick M, Melvin NC, Kirchner WN. The national wetland plant list: 2014 update of wet- land ratings. Phytoneuron. 2014; 41: 1–42. 56. Lapointe M, Lavoie A, Leboeuf M. Plantes de milieux humides et de bord de mer du Que ´ bec et des Mar- itimes. Que ´ bec: Editions Michel Quintin; 2014. 57. Kobe RK, Pacala SW, Silander JA, Canham CD. Juvenile tree survivorship as a component of shade tolerance. Ecol Appl. 1995; 5: 517–532. 58. Kattge J, Dı ´az S, Lavorel S, Prentice IC, Leadley P, Bo ¨ nisch G, et al. TRY—a global database of plant traits. Glob Change Biol. 2011; 17: 2905–2935. 59. USDA, NRCS. The PLANTS Database. 2015 May 6. Available from http://plants.sc.egov.usda.gov 60. Humbert L, Gagnon D, Kneeshaw D, Messier C. A shade tolerance index for common understory spe- cies of northeastern North America. Ecol Indic. 2007; 7: 195–207. 61. Camilli G, Hopkins KD. Applicability of chi-square to 2x2 contingency tables with small expected cell fre- quencies. Psychol Bull. 1978; 85: 163–167. 62. Anderson MJ. Distance-based tests for homogeneity of multivariate dispersions. Biometrics. 2006; 62: 245–253. https://doi.org/10.1111/j.1541-0420.2005.00440.x PMID: 16542252 63. Legendre P, De Ca ´ ceres M. Beta diversity as the variance of community data: dissimilarity coefficients and partitioning. Ecol Lett. 2013; 16: 951–963. https://doi.org/10.1111/ele.12141 PMID: 23809147 64. Anderson MJ. Permutation tests for univariate or multivariate analysis of variance and regression. Can J Fish Aquat Sci. 2001; 58: 626–639. 65. Anderson MJ, Walsh DCI. PERMANOVA, ANOSIM, and the Mantel test in the face of heterogeneous dispersions: What null hypothesis are you testing? Ecolo Monogr. 2013; 83: 557–574. 66. Legendre P. Interpreting the replacement and richness difference components of beta diversity. Global Ecol Biogeogr. 2014; 23: 1324–1334. 67. R Core Team. R: A Language and Environment for Statistical Computing. Vienna: R Foundation for Statistical Computing; 2016. 68. Lenth RV. Least-Squares Means: the package. J Stat Softw. 2016; 69: 1–33. 69. Oksanen J, Kindt R, Legendre P, O’Hara B, Stevens MHH, Oksanen MJ, Suggests MASS. Vegan: Community Ecology Package. R Package Version 2.4–2; 2017. 70. Broman KW, Broman AT. broman: Karl Broman’s R Code. R Package Version 0.67–4; 2017. 71. Swinehart AL, Parker GR. Palaeoecology and development of peatlands in Indiana. Am Midl Nat. 2006; 143: 267–297. PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 14 / 15 Temporal vegetation changes in bogs 72. Warren RJ II, Rossell IM, Moorhead KK. Colonization and establishment of red maple (Acer rubrum) in a southern Appalachian wetland. Wetlands. 2004; 24: 364–374. 73. Wein RW. Biological flora of the British Islands. Eriophorum vaginatum L. J Ecol. 1973; 61: 601–615. 74. Lavoie C, Marcoux K, Saint-Louis A, Price JS. The dynamics of a cotton-grass (Eriophorum vaginatum L.) cover expansion in a vacuum-mined peatland, southern Que ´ bec, Canada. Wetlands, 2005; 25: 64– 75. Helm A, Hanski I, Pa ¨ rtel M. Slow response of plant species richness to habitat loss and fragmentation. Ecol Lett. 2006; 9: 72–77. https://doi.org/10.1111/j.1461-0248.2005.00841.x PMID: 16958870 76. Frele ´ choux F, Buttler A, Schweingruber FH, Gobat JM. Stand structure, invasion, and growth dynamics of bog pine (Pinus uncinata var. rotundata) in relation to peat cutting and drainage in the Jura Moun- tains, Switzerland. Can J For Res. 2000; 30: 1114–1126. 77. Yagouti A, Boulet G, Vincent L, Vescovi L, Mekis E. Observed changes in daily temperature and precipi- tation indices for southern Que ´ bec, 1960–2005. Atmos Ocean. 2008; 46: 243–356. 78. Heijmans MMPD, Mauquoy D, van Geel B, Berendse F. Long-term effects of climate change on vegeta- tion and carbon dynamics in peat bogs. J Veg Sci. 2008; 19: 307–320. 79. Turunen J, Roulet NT, Moore TR. Nitrogen deposition and increased carbon accumulation in ombro- trophic peatlands in eastern Canada. Global Biogeochem Cycles 2004; 18: https://doi.org/10.1029/ 2003GB002154 80. Payette S. Late-Holocene development of subarctic ombrotrophic peatlands: allogenic and autogenic succession. Ecology 1988; 69: 516–531. 81. Loiselle J, Yu Z. Surface vegetation patterning controls carbon accumulation in peatlands. Geophys Res Lett 2013; 40: 5508–5513. 82. Gałka N, Szal M, Broder T, Loiselle J, Knorr KH. Peatbog resilience to pollution and climate change over the past 2700 years in the Harz Mountains, Germany. Ecol Indic 2019; 97: 183–193. PLOS ONE | https://doi.org/10.1371/journal.pone.0229146 February 13, 2020 15 / 15

Journal

PLoS ONEPublic Library of Science (PLoS) Journal

Published: Feb 13, 2020

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create folders to
organize your research

Export folders, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off