Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
stxb201307271960 Diversity and distribution of ground bryophytes in broadleaved forests in Mabian Dafengding National Nature Reserve, Sichuan, China Yanbin Jianga, Xuehua Liub, Shanshan Songa, Zhong Yuc, Xiaoming Shaoa, a College b of Resources and Environmental Sciences, China Agricultural University, Beijing 100193, China School of Environmental Science and Engineering, Tsinghua University, Beijing 100084, China cMabian Dafengding National Nature Reserve, Leshan 614600, China Abstract: Bryophytes, represented by mosses, liverworts, and hornworts, contribute substantially to forest ecosystems in terms of nutrient cycling, water retention, water availability, plant biomass, and plant community maintenance. Forests provide numerous types of habitat for bryophytes, especially the ground floor. To clarify the ground bryophyte diversity and distribution in broadleaved forests, we used microcoenose sampling to investigate ground bryophytes in 34 sample plots (10 × 10 m) in the Mabian Dafengding National Nature Reserve (MDNR), Sichuan Province. Species diversity and environmental factor relationships were analyzed by using α and β diversity indexes, as well as Pearson’s correlation analysis. Detrended canonical correspondence analysis (DCCA) was applied to analyze the relationships between species distribution and environmental factors. A total of 230 bryophyte species were identified in MDNR. These species include 67 liverwort species belonging to 26 genera of 20 families and 163 moss species belonging to 65 genera of 28 families. Diversity of ground bryophytes was negatively correlated to shrub cover, canopy cover, and tree number, but without significant correlations to altitude, slope, aspect, vegetation type, and herb cover. The DCCA ordination of relationships between ground bryophytes and environmental factors showed that altitude, aspect, vegetation type, and shrub cover were important to the distribution of dominant ground bryophyte species. This study quantitatively related bryophytes diversity and distribution to environmental factors, which is helpful in understanding the ecological niche of various bryophytes. Keywords: ground bryophytes; species diversity; distribution; forest; environment Corresponding author. E-mail address: [email protected] (X. Shao) 1 1. Introduction Bryophytes, with a variety of ecological communities, are widely distributed globally. Ground bryophytes are those that grow in the substrate of floor soil. Ground bryophytes form a dominant ecological group in the forest ecosystems and also serve an important function in forest ecosystems in such processes as the carbon and nitrogen cycles [1–3]. For example, in a boreal forest, ground bryophytes serve as the primary carbon sink over woods and are main contributors to carbon equilibrium [3, 4]. Broadleaved forest is a typical and widely distributed vegetation type in China that provides various suitable environments and substrates for bryophytes. Thus, the distribution of ground bryophytes in the forest is influenced by forest cover (light intensity), stands, composition, and structure [5–9]. Human activities, such as deforestation and reforestation, have changed the forest, consequently influencing the diversity, abundance, and distribution of ground bryophytes [10, 11]. Additional, macroclimatic factors such as rainfall and temperature, site factors such as topography and understory vascular, as well as substrate characteristics such as soil moisture and pH, probably induce certain effects [7–9, 12–15]. Nature reserve, especially with humid environment, is an important area of biodiversity conservation and is a vital ecological region that preserves a large amount of ground bryophytes. The habitat heterogeneity of ground bryophytes is relatively high in broadleaved forests [6, 8]. Thus, ground bryophyte diversity and distribution must be clarified to aid in the conservation of bryophytes and their habitat. In this study, we illustrate the relationships of the diversity and distribution of bryophytes with environmental factors by analyzing ground bryophytes of different environment regimes through correlation and ordination analyses. 2. Methods 2.1. Study area This study was conducted in the Mabian Dafengding National Nature Reserve (MDNR) (28°26′28°47′N, 103°13′-103°25′E). MDNR is located in the southwest of Mabian Yi Autonomous County, Sichuan Province and covers an area of 36 000 hm2, with altitude ranging from 800 m to 4042 m a.s.l.. The altitude difference of more than 3000 m causes the temperatures to be significantly distinct in this region. At 1500 m a.s.l., the annual mean temperature is 14 °C to 15 °C, with 3 °C to 4 °C in the coldest month of January and 22 °C to 24 °C in the hottest month of July. In areas above 3800 m a.s.l. (close to the summit of MDNR), the annual mean temperature is -2 °C to 0 °C , and the minimum extreme temperature is lower than -30 °C. Summer does not occur throughout the year, spring and autumn are short (approximately 60 d), and winter is long (approximately 300 d), thus resulting in a cold temperate climate. The annual precipitation in the study area is 1 739.2 mm, reaching the maximum (more than 2000 mm) between 2000 and 2100 m a.s.l.. Similar to climate, soil types and vegetation are vertically 2 Table 1 Summary of sample plots in MDNR Sample Number of Longitude Latitude Altitude Canopy cover Shrub cover Herb cover Slope Aspect M1 103.3295 28.4924 1553 0.80 0.05 0.92 30 350 3 M2 103.3300 28.4925 1516 0.80 0.10 0.50 40 250 5 M3 103.3314 28.4954 1518 0.40 0.12 0.60 45 295 3 M4 103.3326 28.4958 1554 0.70 0.15 0.64 30 20 4 M5 103.3372 28.4985 1523 0.50 0.45 0.55 25 295 5 M6 103.3370 28.4990 1524 0.70 0.10 1 17 345 4 M7 103.3419 28.5011 1436 0.85 0.20 0.83 35 20 5 M8 103.3410 28.5020 1462 0.80 0.15 0.20 33 30 4 M9 103.3481 28.5745 1434 0.65 0.13 0.45 30 82 4 M10 103.3475 28.5749 1424 0.35 0.12 0.74 28 18 2 M11 103.3460 28.5751 1495 0.60 0.25 0.25 20 270 5 M12 103.3529 28.5755 1339 0.40 0.06 0.70 35 188 2 M13 103.3506 28.5757 1358 0.60 0.100 0.50 32 192 3 M14 103.3495 28.5782 1549 0.55 0.19 0.61 40 250 4 M15 103.3483 28.5790 1464 0.78 0.1 0.28 20 205 2 M16 103.3473 28.5794 1409 0.75 0.14 0.45 28 280 4 M17 103.3462 28.5820 1375 0.30 0.09 0.72 5 252 2 M18 103.3447 28.5841 1464 0.40 0.07 0.50 40 70 4 M19 103.3431 28.5848 1464 0.64 0.14 0.78 45 30 4 M20 103.3387 28.5877 1489 0.42 0.08 0.81 34 180 4 M21 103.3379 28.5914 1619 0.30 0.03 0.71 5 70 2 M22 103.3364 28.5938 1696 0.45 0.18 0.60 35 210 3 M23 103.3783 28.6827 1794 0.60 0.85 0.02 40 165 3 M24 103.3760 28.6831 1924 0.30 0.65 0.38 25 170 1 M25 103.3640 28.6846 1812 0.70 0.16 0.48 5 168 1 M26 103.3684 28.6852 1921 0.12 0.04 0.65 25 110 1 M27 103.3682 28.6859 1951 0.60 0.55 0.30 27.5 195 3 M28 103.3675 28.6884 1979 0.20 0.38 0.47 27 150 1 M29 103.3584 28.6887 2110 0.30 0.95 0.01 45 200 1 M30 103.3598 28.6904 2203 0.80 0.85 0.27 32 246 1 M31 103.3615 28.6915 2113 0.60 0.12 0.70 32 264 2 M32 103.3595 28.6920 2221 0.80 0.90 0.01 45 80 4 M33 103.363 28.6926 2040 0.30 0.60 0.27 20 210 1 M34 103.365 28.6927 2034 0.40 0.01 0.48 22 187 3 plot trees distributed from the bottom to the top. Below 2200 m a.s.l., the soil type is mountainous yellow soil and dominated by tropical evergreen broadleaved forest. Between 1800 and 2400 m a.s.l., the soil type is mountainous yellow brown soil and dominated by evergreen and deciduous broadleaved mixed forest. Between 2400 and 2800 m a.s.l., the soil type is mountainous dark brown soil, and corresponding vegetation is coniferous and broadleaved mixed forest. Between 2800 and 3500 m a.s.l., the soil type is mountainous dark brown coniferous forest soil, vegetated by dark coniferous fir forest. Above 3500 m 3 a.s.l., the soil type is mountain meadow soil, vegetated by subalpine shrub and meadow [16]. 2.2. Field sampling For different broadleaved forest types, several sample sites were surveyed along an altitude gradient from 1300 m to 2300 m in MDNR. Within each site, the community features of trees, shrubs, and herbs, were measured in a 10 × 10 m sample plot. These features included cover, height, and abundance. Thirty-four sample plots were investigated. At the four corners of each sample plot, shrubs and herbs were investigated in quadrates of 3 × 3 m and 1 × 1 m individually. The covers of canopies, shrubs, and herbs were measured by visual estimation, as well as tree height. Other vascular community features, including height and abundance, were recorded by measuring and counting. For ground bryophytes, microcoenose sampling method [17] was employed (sampling with the 50 × 50 cm quadrate at the center of the largest fragment in each of 25 2 × 2m grids) to measure species cover. Unequal quadrates (two to 25) were sampled in the 34 plots. We also recorded the geographic locations and topographic factors of the plots. These parameters include longitude, latitude, altitude, slope, and aspect. Soil pH and moisture were not measured in this study for two reasons. First, previous studies showed that soil pH does not influence the bryophytes diversity and growth of the same vegetation type in one certain region [7]. Second, rainfall is adequate in the study area, and the understory is rather humid. Table 1 shows the detailed environmental information of each sample plot. We collected all bryophyte specimens and identified them according to the literature in the laboratory. Some suspected specimens were identified by native bryologists. All the specimens were stored in BAU. 2.3. Data analysis Frequency = (quadrates of bryophytes presence / total investigated quadrates) / 100% Importance value = (relative cover + relative frequency) / 2 We defined dominant family as a family of five or more bryophyte species and dominant species as species with an importance value higher than 0.5. Dominant species were used to clarify the relationships between distribution and environmental factors. Species diversity involving α-diversity and β-diversity was analyzed. Three α-diversity indexes, namely, Patrick index, Shannon–Wiener index, and Pielous evenness index, as well as two β-diversity indexes, namely, Sφrenson index and modified Morisita–Horn index, were employed. All these indexes were calculated using BIO-DAP software. Patrick index: D = S Shannon–Wiener index: S D = − ∑(Pi lnPi ) i=1 Pielous index: 4 E= H ln(S) Sφrenson index: β=1− 2𝑐 𝑎+𝑏 Modified Morisita–Horn index: β = 2∑ ani ∙ bni (da + db)aN ∙ bN S is the total species richness recorded in the study area. Pi = Ni/N, where Ni is the relative cover of species i, and N is the sum of the relative covers of S species. a and b are the species richness values in the two communities, and c is the common species between the two communities. aN is the species richness of plot A, whereas bN is the species richness of plot B. ani and bni are the abundance values of i species in plots A and B, respectively. da = ∑ an2i /aN 2, and db = ∑ bn2i /bN 2. Pearson’s correlation was used to test the relationships between species diversity and environmental factors in SPSS 19.0 software. Except for dividing sample plots into groups subjectively according to altitude and forest types, we used principal components analysis (PCA) to analyze the inter-correlations of sample plots relating to altitude, slope, aspect, vegetation type, canopy cover, tree numbers, as well as shrub and herb cover of each sample plot. Species distribution and environmental relationships were characterized by detrended canonical correspondence analysis (DCCA). Both PCA and DCCA were executed in software CANOCO for Windows 4.5, and CANODRAW was used to draw the two-dimensional ordination graphs of species and environmental factors. In the PCA graph, the sample plots are scattered or clustered along environments. The first and second axes represent the first and second principal components, respectively, and are strongly related to environmental factors. The distance between a plot and an ordination axis indicates the correlation between the plot and axis or one or several environmental factors. The quadrant in which the sample plot is located can indicate a positive or negative correlation between plot and axis. Sample plots can possibly be divided into groups according to their clusters. In the DCCA graph, the arrows represent environmental factors, the length of arrows indicates the correlations of species distribution and environmental factors, the slope of arrows shows the relationships between environmental factors and ordination axis, and the quadrant in which a sample is plot located illustrates a positive or negative correlation between a plot and an axis. In addition, the distance between a species and a plot represents their relationship; a small distance indicates high relative abundance of the species in the plot. The distances among different species indicate the degrees of distribution divergence. The optimal conditions of different species relative to a quantitative environmental factor can also be determined according to the locations of species projected at the arrow of the environmental factor in the DCCA graph. 5 3. Results 3.1. Diversity of ground bryophytes under broadleaved forest in MDNR We identified 230 ground bryophyte species by conducting survey on 34 sample plots in MDNR. We found 67 liverworts belonging to 20 families and 26 genera, as all as 163 mosses belonging to 28 families and 65 genera. Table 1 displays the dominant bryophyte families. The dominant liverwort families include Lophocoleaceae, Plagiochilaceae, Porellaceae, Calypogeiaceae, Radulaceae, and Lepidoziaceae. The species richness of the six dominant families was 62.7% of the total liverworts, and the coverage was 76.0% of the total population. Among these dominant families, species richness from Lophocoleaceae, Plagiochilaceae, and Porellaceae were relatively greater than that of other families. Species richness of genus Heteroscyphus and Chiloscyphus from Lophocoleaceae were high at 4 and 6, respectively, similar to genus Plagiochila from Plagiochilaceae and genus Porella from Porellaceae, which had species richness of 8 and 7, respectively. Fourteen dominant moss families were found in the 34 sample plots in MDNR: Brachytheciaceae, Mniaceae, Plagiotheciaceae, Hypnaceae, Thuidiaceae, and so on. The species richness of the 14 dominant families of mosses was 86.5% of the total moss species, and coverage was 86.5% of the total. Species richness of these families was more than 10, and coverage was higher than 10%. Table 2 Dominant families of ground bryophytes in MDNR Liverworts No. Mosses Family Genus Species Cover (%) Family Genus Species Cover (%) Lophocoleaceae 2 10 16.3 Brachytheciaceae 6 28 56.3 2 Plagiochilaceae 2 9 10.6 Mniaceae 4 18 38.8 3 Porellacea 1 7 1.6 Plagiotheciaceae 1 14 37.2 4 Lepidoziaceae 2 6 15.2 Fissidentaceae 1 10 8.5 5 Radulaceae 1 5 3.8 Thuidiaceae 4 9 26.9 6 Calypogeiaceae 2 5 2.0 Hypnaceae 4 9 11.4 7 Polytrichaceae 3 8 11.5 8 Neckeraceae 3 7 11.2 9 Bryaceae 3 7 10.8 10 Pottiaceae 6 7 5.1 11 Dicranaceae 3 6 8.0 12 Sematophyllaceae 3 6 8.7 13 Meteoriaceae 6 6 4.8 14 Hylocomiaceae 2 6 2.2 Total 10 42 49.5 Total 49 141 241.4 Percentage (%) 38.5 62.7 76.0 Percentage (%) 75.4 86.5 89.4 6 Tables 2 and 3 show that the mosses are more dominant than the liverworts in terms of species richness and cover. For example, for six of the most dominant families, 88 species were found in the moss families, but only 42 were found in liverworts, and the coverage of these mosses was 179.1%, four times that of the liverworts at 49.5%. For 52 species with an importance value higher than 0.5, only four species were from liverworts, and their total coverage was 19.6%, significantly less than that of mosses at 183.2%. Table 3 also shows that Thuidium cymbifolium, Plagiothecium euryphyllum, Plagiomnium rhynchophorum, and Eurhynchium savatieri were the dominant species in the sampled sites. All these dominant species were creeping growth mosses that occurred in most of the plots with relatively high average covers. The results illustrated that the understory habitats in the broadleaved forest were suitable for these creeping bryophytes. Liverworts were present in most sample plots, but the frequency and cover were relatively lower than those of mosses. The four dominant liverwort species were Heteroscyphus coalitus, Heteroscyphus argutus, Plagiochila ovalifolia, and Conocephalum conicum. 3.2. α- and β-diversity of ground bryophytes in different broadleaved forest types in MDNR The vertical distribution of vegetation in MDNR is apparent, but human disturbance has caused the irregular distribution of some forest types. Thus, to characterize α- and β-diversity of ground bryophytes in different forest types, we divided the 34 sample plots into groups based on altitude and forest types and then compared the diversity indexes. Table 4 shows the six forest types based on the subjective grouping. Through α-diversity analysis, the corresponding value of species diversity differed when using different indexes. The Patrick index showed that species richness in six broadleaved forests in MDNR varied from 46 to 104. The Shannon–Wiener and Pielous indexes consider species coverage, such that the differences in α-diversity among the six forest types were negligible. Shannon–Wiener index values ranged from 3.19 to 3.84, whereas Pielous index values were almost the same at 0.82 or 0.83. The Pielous index revealed that the evenness features of individual ground bryophytes under the six forest types were similar. Within the six forest types, V6 had the highest α-diversity, with 104 species and a Shannon–Wiener index of 3.84. This result may indicate that Davidia involucrate forest is suitable for the growth of various bryophytes. However, the Davidia involucrate forest was the most frequently investigated type (eight sample plots), which may have affected the result. However, this effect was uncertain, because seven sample plots were also investigated for V3, the Patrick value of which was only 71, lower than that for V1 (including five sample plots with Patrick 93). Moreover, V1 and V2 included the same number of sample plots, but the species richness in V2 was 30 less than that in V1 (Figure 1). 7 Table 3 Dominant ground bryophyte species in MDNR No. Species Frequency (%) S1 Thuidium cymbifolium 61.8 Coverage (%) 22.6 Importance value 5.063 S2 Plagiothecium euryphyllum 41.2 17.0 3.662 S3 Plagiomnium rhynchophorum 35.3 12.3 2.795 S4 Eurhynchium savatieri 14.7 12.8 2.314 S5 Eurhynchium laxirete 32.4 6.8 1.904 S6 Thuidium pristocalyx 32.4 6.5 1.851 S7 Homaliodendron crassinervium 11.8 9.9 1.804 S8 Claopodium aciculum 23.5 7.5 1.762 S9 Plagiothecium formosicum 20.6 7.4 1.673 S10 Heteroscyphus coalitus 29.4 5.6 1.643 S11 Leucobryum juniperoideum 38.2 3.0 1.500 S12 Mnium spinosum 5.9 8.9 1.489 S13 Heteroscyphus argutus 17.6 6.5 1.452 S14 Bryhnia trichomitria 11.8 7.4 1.431 S15 Eurhynchium kirishimense 20.6 5.7 1.410 S16 Plagiochila ovalifolia 23.5 5.0 1.386 S17 Dicranum scoparium 23.5 4.9 1.377 S18 Rhizomnium punctatum 14.7 5.7 1.253 S19 Bryum capillare 17.6 4.7 1.181 S20 Hookeria acutifolia 35.3 1.3 1.166 S21 Atrichum undulatum var. gracilisetum 11.8 5.6 1.155 S22 Plagiothecium cavifolium var. fallax 20.6 3.9 1.144 S23 Rhynchostegium contractum 8.8 6.0 1.131 S24 Bryhnia serricuspis 20.6 3.7 1.117 S25 Conocephalum conicum 26.5 2.5 1.098 S26 Hypopterygium flavolimbatum 23.5 3.0 1.092 S27 Mnium lycopodioides 14.7 4.6 1.091 S28 Taxiphyllum subarcuatum 11.8 4.8 1.037 S29 Plagiomnium succulentum 14.7 4.1 1.021 S30 Fissidens anomalus 17.6 3.1 0.944 S31 Atrichum subserratum 20.6 2.4 0.919 S32 Barbella spiculata 11.8 3.9 0.905 S33 Plagiothecium nemorale 20.6 2.2 0.889 S34 Mnium thomsonii 14.7 3.2 0.876 S35 Mnium laevinerve 17.6 2.6 0.869 S36 Brotherella henonii 14.7 2.9 0.833 S37 Aneura pinguis 8.8 3.7 0.794 S38 Heteroscyphus zollingeri 17.6 2.1 0.791 S39 Bryum salakense 14.7 2.5 0.771 S40 Wijkia deflexifolia 8.8 3.4 0.750 S41 Taxiphyllum cuspidifolium 11.8 2.9 0.748 S42 Plagiochila sciophila 11.8 2.8 0.739 S43 Fauriella tenuis 14.7 2.2 0.732 S44 Fissidens involutus 20.6 0.9 0.700 S45 Plagiothecium succulentum 17.6 1.2 0.664 S46 Eurhynchium eustegium 8.8 2.7 0.641 S47 Neodolichomitra yunnanensis 11.8 2.1 0.638 S48 Plagiothecium cavifolium 5.9 3.2 0.635 S49 Duthiella flaccida 5.9 3.1 0.621 S50 Conocephalum japonicum 8.8 2.5 0.614 S51 Trichostomum tenuirostre 14.7 1.3 0.596 S52 Brotherella falcata 14.7 1.1 0.565 8 Table 4 Vegetation types of sampling plots in MDNR No. Vegetation Altitude (m) Sample plots V1 Open evergreen and deciduous broadleaved forest 1300-1450 M9,M10,M12,M13,M17 dominated by Kalopanax septemlobus and Platycarya strobilacea V2 Evergreen broadleaved forest dominated by Fagaceae 1460-1500 M8,M11,M15,M19,M20 V3 Dense evergreen and deciduous broadleaved forest 1400-1550 M1,M2,M4,M7,M14,M16,M18 1520-1700 M3,M5,M6,M21,M22 by 1800-1950 M23,M24,M25,M27 Deciduous broadleaved forest dominated by Davidia 1950-2220 M26,M28,M29,M30,M31,M32,M33, dominated by Aesculus chinensis and Litsea suberosa V4 Mixed evergreen and deciduous broadleaved forest V5 Evergreen broadleaved forest dominated Lithocarpus glabra involucrata M34, 4 Shannon-Wiener Pielous 120 Patrick Shannon-Wiener index Pielous index 3.5 100 3 80 2.5 2 60 1.5 40 Patrick index V6 1 20 0.5 0 0 V1 V2 V3 V4 Vegetation type V5 V6 Figure 1 α-diversity of ground bryophytes in broadleaved forests in MDNR In terms of β diversity, the Sφrenson index (βS) and Morisita–Horn index (βM) measure qualitative and quantitative similarity, respectively. For the qualitative β-diversity index (βS), which only considers species presence or absence, the highest similarity (0.552) was found for the forest types of evergreen broadleaved forest dominated by Fagaceae (V2) and mixed dense evergreen and deciduous broadleaved forest dominated by Aesculus chinensis and Litsea suberosa (V3), whereas the lowest similarity (0.171) as found in the mixed evergreen and deciduous broadleaved forest (V4) and deciduous broadleaved forest dominated by Davidia involucrate (V6). According to the quantitative β-diversity index (βM), which considers both species presence and relative covers, the similarities were somehow different from the above results. The overall similarities were significantly lower at 0.126 to 0.418 (Table 5). The generally low similarities of ground bryophyte communities in the involved broadleaved forest types indicated that species varied significantly with changing forest types, and the species diversity in the study region was very rich. 9 Table 5 β-diversity of ground bryophytes under different broadleaved forests in MDNR βS / βM V1 V2 0.436 / 0.229 V3 0.439 / 0.363 0.552 / 0.398 V4 0.429 / 0.418 0.345 / 0.221 0.390 / 0.361 V5 0.278 / 0.126 0.281 / 0.367 0.295 / 0.145 0.224 / 0.152 V6 0.283 / 0.355 0.262 / 0.136 0.250 / 0.162 0.171 / 0.389 V2 V3 V4 V5 0.410 / 0.325 3.3. Relationships between diversity and environmental factors of ground bryophytes in MDNR In the bryophyte diversity studies, researchers have subjectively divided sample plots into groups based on some rules to compare the differences between environment gradients or regimes. The rules included those used in this study, such as vegetation type. However, the factors that determined species diversity remain unclear. Thus, we conducted a PCA to analyze the relationships among sample plots objectively. Figure 2 shows the PCA result. Thirty-four sample plots were divided into five groups: G1: sample plots 1, 2, 3, 5, 6, 11, 14, 16 and 17; G2: sample plots 12, 13, 15, 20 and 22; G3: sample plots 27, 29, 30, 31, 33 and 34; G4: sample plots 23, 24, 25, 26, 28 and 32; G5: sample plots 4, 7, 8, 9, 10, 18, 19 and 21. The PCA groups were different from the subjective grouping based on vegetation and altitude, thus revealing that the subjective grouping was not a scientific grouping method. Calculating bryophyte diversity per sample plot was efficient in quantifying the relationships between diversity and environmental factors by taking advantage of every measured factor. Thus, we conducted a correlation analysis between Patrick index and environmental factors per sample plot. The correlations revealed that ground bryophyte diversity was significantly negatively correlated to shrub cover (coefficient = -0.339, P = 0.05) and negatively related to canopy cover and number of trees with coefficients of -0.318 and -0.326, respectively. These two factors were significantly autocorrelated (coefficient = 0.533, P = 0.001). Bryophyte diversity was not strongly related to altitude, slope, aspect, vegetation type, and herb cover, with all of their coefficients less than 0.2 (Table 6). Therefore, environmental factors that most affected ground bryophytes in MDNR were vegetation features, including canopy cover and shrub cover. 10 Figure 2 PCA ordination of 34 sample plots in MDNR Table 6 Correlation of Patrick index and environmental factors for ground bryophytes in MDNR Environmental factors Correlation coefficient P Altitude 0.190 0.932 Slope 0.015 0.052 Aspect 0.147 0.406 Vegetation type -0.066 0.713 Canopy cover -0.318 0.067 Number of trees -0.326 0.059 Shrub cover -0.339 0.050 Herb cover 0.180 0.307 3.4. Relationships between distribution and environmental factors of ground bryophytes in MDNR DCCA analysis was employed to analyze the relationships between 52 dominant species and environmental factors. The 52 species were found in the 34 sample plots, and their importance values were higher than 0.5. Owing to the significant autocorrelations between canopy cover and number of trees, the latter was excluded in the DCCA analysis. In Figure 3, primary environmental factors that influenced ground bryophytes, the relationships of bryophytes and environments, and the presence of bryophytes in sample plots were discerned. Considering geographic location, topography, and vegetation, the correlation coefficient between ground bryophytes and environmental factors was 0.933 11 on axis 1 and 0.895 on axis 2. The effects of topographic factors such as altitude and aspect, as well as vegetation factors such as vegetation type and shrub cover, were more important than those of slope, canopy cover, and herb cover to the distribution of ground bryophytes. Figure 3 also indicates that the relationships between different bryophytes and environmental factors varied. These relationships possessed the following features: (1) Altitude, vegetation type, and shrub cover were positive to axis 1 and important to ground bryophyte distribution pattern. Thus, the species with small projected distance to these factors will exhibit increasing abundance as the values of these environment factors increase. The species were Dicranum scoparium (S17), Atrichum undulatum var. gracilisetum (S21), Plagiothecium cavifolium var. fallax (S22), Eurhynchium savatieri (S24, Fissidens anomalus (S30), Atrichum subserratum (S31), Plagiothecium nemorale (S33), Brotherella henonii (S36), Plagiothecium cavifolium (S48), and so on. Canopy cover was negative to axis 1, but its effect on bryophyte distribution was much smaller than that of the abovementioned factors. (2) Aspect was positively related to axis 2, which is another important factor in the distribution of ground bryophytes. The bryophyte species adapted to shady slopes were Plagiomnium rhynchophorum (S3), Plagiochila ovalifolia (S16), Bryum capillare (S19), Conocephalum conicum (S25), Plagiothecium succulentum (S45), and Trichostomum tenuirostre (S51). (3) Eurhynchium laxirete (S5), Claopodium aciculums (S8), Mnium spinosum (S12), Hypopterygium flavolimbatum (S26), Mnium laevinerve (S35), Aneuraceae pinguis (S37), and Taxiphyllum cuspidifolium (S41) were positively correlated to herb cover and slope. (4) The above species were affected by certain factors. However, some species were located in the middle of DCCA graph and surrounded the environmental factors. For example, Thuidium cymbifolium (S1), Plagiothecium euryphyllum(S2), Homaliodendron crassinervium (S7), Leucobryum juniperoideum (S11), Heteroscyphus argutus (S13), Eurhynchium kirishimense (S15), Hookeria acutifolia (S20), Mnium thomsonii (S34), Bryum salakense (S39), Fauriella tenuis (S43), Fissidens involutus (S44), and Eurhynchium eustegium (S46) were scattered and minimally affected by all the factors in this study. These species were present in a few sample plots: M4, M6, M10, M11, M13, M15, M17, M19, and M21. 12 Figure 3 DCCA ordination of 52 dominant ground-bryophytes and environmental factors from MDNR. - 52 bryophyte species, - 34 sample plots 4. Conclusion and discussion The dominant families of ground bryophytes in MDNR included mosses of Brachytheciaceae, Mniaceae, Plagiotheciaceae, Hypnaceae, and Thuidiaceae, as well as liverworts of Lophocoleaceae, Plagiochilaceae and Porellaceae. These families included a number of subtropical genera and species, and their geographic distribution belongs to floras of both Yunnan–Guizhou and Central China [18]. The ground bryophyte species in MDNR were similar to those in Foping Nature Reserve, which may be explained by the fact that the bryophyte flora of Foping Nature Reserve belongs to the transitional zone of North China and Central China [19]. Based on the subjective grouping of bryophyte communities by canopy vegetation types, the resultant similarities were low, and the values of the diversity index were irregular. This finding illustrates that vegetation types strongly affected ground bryophytes, the bryophytes species were significantly changed as the vegetation varied, and the species richness in this study area was high. Moreover, this observation might indicate that the subjective grouping method was unreasonable. To clarify further species diversity and its determinants in the broadleaved forest zone, adequate survey plots and more scientific grouping methods are needed. PCA is one of the effective grouping methods. TWINSPAN, which is a cluster grouping approach that in the light of bryophytes coverage in each sample plot, has also been frequently used [7, 20]. Quantitative studies on bryophyte diversity and environments remain limited [21]. In this study, we 13 set diversity indexes of all sample plots as dependent variables, quantified the correlations between environmental factors and diversity indexes directly, and identified that vegetation, especially upper vegetation such as trees and shrubs, were key factors that influenced ground bryophytes diversity. DCCA ordination of species and environments showed different tendencies: canopy cover was not as important as other factors that affected ground bryophyte distribution, but vegetation type, shrub cover, and altitude limited the distribution range. Acknowledgement This work was supported by the National Natural Science Foundation of China (31300356). Specimen classification was performed with the assistance of Prof. Pengcheng Wu and Meizhi Wang from the Institute of Botany, Chinese Academy of Sciences, as well as Prof. Youfang Wang from East China Normal University. Field works were supported by ShuaRi Luobo and other staff members from the MDNR Administration Bureau. 14 References [1] DeLucia E H, Turnbull M H, Walcroft A S, et al. The contribution of bryophytes to the carbon exchange for a temperate rainforest. Global Change Biology, 2003. 9(8): 1158-1170. [2] Markham J H. Variation in moss-associated nitrogen fixation in boreal forest stands. Oecologia, 2009. 161(2): 353-359. [3] Kolari P, Pumpanen J, Kulmala L, et al. Forest floor vegetation plays an important role in photosynthetic production of boreal forests. Forest Ecology and Management, 2006. 221(1-3): 241-248. [4] Goulden M L, Wofsy S C, Harden J W, et al. Sensitivity of boreal forest carbon balance to soil thaw. Science, 1998. 279(5348): 214-7. [5] Rambo T R, Muir P S. Forest floor bryophytes of Pseudotsuga menziesii Tsuga heterophylla stands in Oregon: Influences of substrate and overstory. Bryologist, 1998. 101(1): 116-130. [6] Marialigeti S, Nemeth B, Tinya F, et al. The effects of stand structure on ground-floor bryophyte assemblages in temperate mixed forests. Biodiversity and Conservation, 2009. 18(8): 2223-2241. [7] Guo S L, Cao T. Distribution patterns of moss species on ground in main ecosystems in Changbai Mountains. Chinese Journal Of Applied Ecology, 1999.10 (3): 270-274. [8] Wu L L,Ji M C,Yan X L. Study on the Relationships between terrestrial bryophytes and environmental factors in Yangjifeng Nature Reserve, China. Journal of Wuhan Botanical Research, 2010, 28( 3) : 324-329 [9] Sun Y, Shao X M, Liu X C, et al. Bryophyte species diversity in main forest vegetations in Dongling Mountain of Beijing. Chinese Journal of Ecology 2007, 26(11): 1725-1731. [10] Shields J M, Webster C R, Glime J M. Bryophyte community response to silvicultural opening size in a managed northern hardwood forest. Forest Ecology and Management, 2007. 252: 222-229. [11] Nelson C R, Halpern C B. Short-term effects of timber harvest and forest edges on ground-layer mosses and liverworts. Canadian Journal of Botany-Revue Canadienne De Botanique, 2005. 83(6): 610-620. [12] Zhang Y M, Cao T, Pan B R. Quantitave classification and ordination analysis on bryophyte vegetation in Bogda Mountain, Xinjiang. Acta Phytoecologica Sinica, 2002. 26(1): 10-16. 15 [13] Bruun H H, Moen J, Virtanen R, et al. Effects of altitude and topography on species richness of vascular plants, bryophytes and lichens in alpine communities. Journal of Vegetation Science, 2006. 17(1): 37-46. [14] Saetersdal M, Gjerde I, Blom H H, et al. Vascular plants as a surrogate species group in complementary site selection for bryophytes, macrolichens, spiders, carabids, staphylinids, snails, and wood living polypore fungi in a northern forest. Biological Conservation, 2004. 115(1): 21-31. [15] Hokkanen P J. Environmental patterns and gradients in the vascular plants and bryophytes of eastern Fennoscandian herb-rich forests. Forest Ecology and Management, 2006. 229(1-3): 73-87. [16] Luo L Q. The first paper in the fourth chapter overview — The vegetation of Mabian Dafengding Nature Reserve. Journal Of Leshan Teachers College, 2003. 18(9): 23-33. [17] Jiang Y B, Liu X H, Tian R X, et al. Field-sampling methods for investigating ground-bryophyte populations in forest vegetation. Polish Journal of Ecology, 2011. 59(2): 317-327. [18] Wu P C, Jia Y. The regionalization and distribution types of the bryophytes in China. Journal of Plant Resources and Environmen, 2006. 15(1): 1-8. [19] Li F X,Wang Y F, Zhan Q F, et al. Species diversity of floor Bryophyte commuities in Foping Nature Reserve. Journal ofPlant Ecology(formerlyActa Phytoecologica Sinica), 2006. 30(6): 919~923. [20] Cao T, Chen Y, Yu J, et al. Distribution patterns of moss species in Shanghai city. Chinese Journal of Applied Ecology,2004,15(10):1785~1791. [21] Raabea S, Müller J, Manthey M, et al. Drivers of bryophyte diversity allow implications for forest management with a focus on climate change. Forest Ecology and Management, 2010. 16