Structural and functional characteristics of high alpine soil macro-invertebrate communities

European Journal of Soil Biology 86 (2018) 72–80

Contents lists available at ScienceDirect

European Journal of Soil Biology journal homepage: www.elsevier.com/locate/ejsobi

Structural and functional characteristics of high alpine soil macroinvertebrate communities

T

Michael Steinwandtera,b, Alexander Riefb, Stefan Scheuc,d, Michael Traugottb, Julia Seebera,b,∗ a

Institute for Alpine Environment, Eurac Research, Viale Druso 1, 39100, Bozen/Bolzano, Italy Mountain Agriculture Research Unit, Institute of Ecology, University of Innsbruck, Technikerstrasse 25, 6020, Innsbruck, Austria c J.F. Blumenbach Institute of Zoology and Anthropology, University of Goettingen, Berliner Str. 28, 37073, Goettingen, Germany d Centre of Biodiversity and Sustainable Land Use, University of Göttingen, Von-Siebold-Str. 8, 37075, Göttingen, Germany b

A R T I C LE I N FO

A B S T R A C T

Handling editor: Stefan Schrader

Soil macro-invertebrates play an important role in the formation and functioning of soils, which makes them indispensable for all terrestrial ecosystems, including high alpine soils. However, in the latter, knowledge on species identity, diversity, and functionality of macro-invertebrate soil communities is scarce. Here, we address this knowledge gap by investigating the structural and functional composition of soil macro-invertebrate communities in high alpine sites of the European Alps that differ in sheep grazing intensity (low, medium, and high). Abundance data were combined with the analysis of natural variations in stable isotope ratios (13C/12C, 15 N/14N) of food sources and soil animals, allowing insight into the trophic structure of the decomposer community. The presence of sheep significantly increased the abundance of Nematocera, but reduced the abundance of most other taxa. Diplopoda were found exclusively at the low elevation site with almost no sheep grazing, while Diptera larvae increased in numbers at higher elevation sites. Lumbricidae species were abundant at all sites except the highest site, which was intensively grazed by sheep and also used as a resting place. In contrast, trophic relations were not affected by sheep grazing intensity, four trophic groups were clearly distinguished, pointing to a relatively simple food web structure: (1) primary decomposers, (2) secondary decomposers and dung feeders, (3) root and fungal feeders, and (4) predators. We suggest that the shallow soils with high organic matter do not allow the formation of more complex food webs.

Keywords: Alpine soil fauna Isotopic niche Mixing models Diplopoda Insecta larvae DNA barcoding

1. Introduction Soil macro-invertebrates, including Lumbricidae, Diplopoda, and Insecta larvae, decompose litter material and physically alter the soil structure by mixing organic and mineral matter as well as creating secondary pores, thereby regulating nutrient and carbon cycling, soil aeration, and soil water balance [1]. Thereby, they are indispensable for the functioning of terrestrial ecosystems including those at high altitude. While the macro-invertebrate soil communities of low elevation European mountains (up to ca. 2000 m a.s.l.) have been studied previously [2,3], knowledge on species identity, diversity and functioning of soil macro-invertebrates of high alpine regions is scarce and this also applies to the Central European Alps (but see Ref. [4] for gastropods [5]; for oribatid mites). In the high alpine area of the Central European Alps above 2500 m a.s.l., the vegetation growth period is short, as temperatures are low and the snow cover persists for most of the year [6,7]. Plant communities comprise alpine grasslands (Seslerio-Caricetum sempervirentis,

∗

Caricetum firmae), which contain species that produce a high number of secondary plant compounds such as polyphenols for their protection against radiation and other stressors [8,9], thus producing litter of poor food quality for detritivores [10]. Climate and litter quality shape the soil macro-invertebrate community [11,12], along with a reduced habitat structure in the shallow alpine soils of the European Alps, limiting the leeway of the soil animals [13]. At these elevations, Insecta larvae, especially soil-dwelling larvae of Brachycera and Nematocera, increase in number and biomass, while Lumbricidae and Diplopoda get less abundant as a consequence of unfavourable environmental [14,15] and nutrient conditions [16]. However, sheep might enable macro-invertebrates to extend their upper elevational limit by providing them with a favourable food source in form of dung [17]. Schon et al. (2008) found that the abundance of meso- and macro-invertebrates was promoted by low intensity grazing (i.e. small sheep flocks), until negative effects of intensive pasturing by larger flocks took over (e.g., trampling, soil compaction) [18]. To improve understanding of the soil macro-invertebrate

Corresponding author. Institute for Alpine Environment, Eurac Research, Viale Druso 1, 39100, Bozen/Bolzano, Italy. E-mail addresses: [email protected], [email protected] (J. Seeber).

https://doi.org/10.1016/j.ejsobi.2018.03.006 Received 14 September 2017; Received in revised form 22 March 2018; Accepted 22 March 2018 Available online 09 April 2018 1164-5563/ © 2018 The Authors. Published by Elsevier Masson SAS. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).

European Journal of Soil Biology 86 (2018) 72–80

M. Steinwandter et al.

(fam. Pfurtscheller, farmer and landowner, pers. commun.), whereas the low and high states represented exceptional conditions. We are aware that, with the exception of the medium intensity with two sites serving as true replicates, the soil samples are pseudoreplicated (i.e. the low and high sheep grazing intensity occurred only once), which hampers the generalization of our results beyond this case study. However, since it is difficult in the Central European Alps to find true replicates with the same environmental conditions and because data on alpine soil communities are scarce, we still find value in our study. Soil and environmental properties were measured at each site and on each sampling date. These are: (i) pH of the soil (with a pH Meter, Metrohm, Herisau, Switzerland), (ii) total carbon and nitrogen content as well as C/N ratio of soil and litter material (using an organic elemental analyser, FlashEA 1112, Thermo Fisher Scientific, Waltham, Massachusetts), (iii) soil organic matter (SOM, using the dry combustion method by incinerating the material at 500 °C for 4 h in a muffle furnace, Nabertherm, Lilienthal, Germany), and (iv) inclination of the plots.

community structure and its functioning in high alpine regions in Central Europe, their species composition as well as their trophic structure need to be studied. Functional aspects such as trophic relationships describe what organisms do in an ecosystem [19], and add significantly to the knowledge of the ecology of species [20]. A common method to assess the trophic ecology of animals is the dual stable isotope approach, which provides information on the main food resources (reflected by 13C/12C ratios) and the trophic position (reflected by 15 N/14N ratios), i.e. their position in the food chain [21,22]. Displaying the animals' stable isotope ratios in a bi-plot provides information on the trophic structure of communities, as the heavier isotopes increase in abundance from lower to higher trophic levels [21,23]. Furthermore, the variation in carbon and nitrogen isotope ratios reflect the trophic niche of species or taxa [24], and isotope niches can therefore be used as metrics of the species' trophic ecology [25]. Food web studies using stable isotope analysis have been carried out successfully for several terrestrial as well as for aquatic ecosystems. However, studies on trophic relations of soil animals are still scarce as interpreting isotopic data from soil animals is complicated by a variety of factors: for example, the range of 13C and 15N of the litter material and organic matter, which form the basis of the soil food web, may span a large range of values, changes with time and decomposition status, and increases with soil depth. Most soil animals are omnivorous, feeding on a variety of food sources, making it difficult to pinpoint single sources and to assign them to a distinct trophic level [21]. In addition, microbes, which fractionate nitrogen isotopes, make up part of the nutrition of soil macro-invertebrates, which further complicates interpreting the stable isotope values of consumers. Nevertheless, novel insights into the structure and functioning of soil animal communities has been gained in food web studies (e.g. Refs. [26–28] and studies on trophic relations of individual species (e.g. Refs. [29–31])). In this study, we assessed for the first time structural and functional aspects of a high alpine soil macro-invertebrate community of the Central European Alps. We expected that sheep grazing favours soil macro-invertebrates, especially Lumbricidae due to additional and easily available food sources (i.e. dung), allowing them to colonize high alpine soils. For this purpose we investigated community composition and community-environment relationships using the abundance of invertebrates from three high Alpine soils differing in sheep grazing intensity. In a second step we assessed natural variations in stable isotope ratios of these animals to delineate their trophic positions. For identifying juvenile Lumbricidae species, which lack morphological characters needed for species identification but form a significant proportion of the soil invertebrate community, we employed DNA barcoding [32].

2.2. Soil sampling and identification of soil invertebrates Six soil core samples (i.e. soil with vegetation and litter, diameter 15 cm, soil depth 15 cm if possible, but at least 12 cm) per site and sampling date were taken randomly at least 5 m apart from any other sample. The vegetation was cut to 1 cm on-site and was not used for further analyses. Sampling areas varied between 400 (TH) and 700 m2 (TM). The samples were taken in mid-June 2010, end of July 2010, and end of September 2010 to cover the different activity periods of the soil fauna groups within the vegetation period [13,33]. All soil core samples were transferred to the soil laboratory of the University of Innsbruck on the sampling day. Soil animals were heat-extracted alive for 12 days (damp tissue was placed into the collection bowls) using a modified Kempson apparatus [34], sorted and counted daily, and stored individually at −28 °C until further analyses. Daily collection of live soil animals and freezing them was done to prevent changes in isotope signatures resulting from decay of body tissue and preservation liquids [35]. Insects (adults and larvae) were identified to family level, Lumbricidae (partly by DNA barcoding, see below) and Diplopoda to species level. The frozen soil animals were identified using a dissection stereo-microscope with petri dishes kept constantly on cooling elements. Taxonomic identification of Lumbricidae followed Csuzdi & Zicsi (2003) [36], that of Diplopoda PedroliChristen (1993) [37], and that of all other taxa Schaefer (2009) [38]. 2.3. DNA barcoding

2. Material and methods

Juvenile Lumbricidae were identified to species level using DNA barcoding. DNA was extracted from the first two segments (without gut content) using the peqGOLD Tissue DNA Mini Kit (peqlab, Erlangen, Germany), following manufacturer's instructions. Part of the mitochondrial cytochrome c oxidase subunit one gene (COI) was amplified using the universal invertebrate primers described by Folmer et al. (1994) [39]. PCR was carried out in 10 μl reaction volumes containing 0.2 mM dNTPs (Gencraft), 1 μM of each primer, 1 × buffer (Biotherm), 3 mM MgCl2, 5 μg of bovine serum albumin (BSA), 0.375 U Taq Polymerase (BioTherm), 3.525 μl of PCR-grade RNase-free water and 1.5 μl of DNA extract. Thermal cycling comprised 94 °C for 2 min, 35 cycles at 94 °C for 20 s, 46 °C for 30 s, 72 °C for 45 s and a final elongation at 72 °C for 3 min. PCR products were purified using ExoSAP-IT (USB Corporation) and sequenced in forward direction. The sequences were aligned with BioEdit (Version 7.0.9.0, Hall, 1999) and compared to existing species sequences from the research site [40].

2.1. Study area The study was conducted on the Hoher Burgstall (2470–2600 m a.s.l.), a mountain top above Neustift in the Stubai Valley (Central Alps, Tyrol, Austria, 11.2810° E, 47.1334° N, Fig. A1, see supplementary data). The bedrock is calcareous; the dominating soil type is Rendzina with a high organic matter content and mull as the prevalent humus form. The experimental sites were located in part on strongly inclined slopes exposed to the south or south-west (Table 1). The four sites varied in location (slope S or top T) and sheep grazing intensity (low L, medium M, and H high): (i) SL – site located at the slope with almost no sheep present, (ii) SM – sheep regularly graze the mountain slope, (iii) TM – site at the mountain top which is regularly used by sheep for grazing, and (iv) TH – plane grassland at the mountain top which is, beside for grazing, used by the sheep as resting place during the day and overnight. The medium sheep grazing intensity was considered as the average situation found on a wide range of sites on the study area with approximately 50 sheep roaming the area daily from June to September

2.4. Stable isotope analysis Litter (0 to +1 cm) and soil (−15 to 0 cm) material was air-dried, 73

European Journal of Soil Biology 86 (2018) 72–80

M. Steinwandter et al.

Table 1 Characteristics of the study sites including mean (SD) soil parameters: Ctot soil – total carbon content of the soil, Ntot soil – total nitrogen content of the soil, C/N soil – C-to-N ratio of the soil, SOM – soil organic matter content; superscript letters indicate significant differences between sites with p < 0.05, Tukey's HSD test; SL – alpine grassland with low sheep grazing intensity, SM – alpine grassland with medium sheep grazing intensity, TM – alpine grassland at the mountain top with medium sheep grazing intensity, TM – alpine grassland at the mountain top with high sheep grazing intensity and serving also as sheep resting place. Elevation SL

2470 m

SM

2500 m

TM

2600 m

TH

2600 m

Coordinates 47.1317 11.2833 47.1326 11.2756 47.1340 11.2785 47.1336 11.2796

Inclination

Sheep grazing intensity

pH

Ctot soil [%] a

a

Ntot soil [%]

27%

low

6.0 (0.3)

20.7 (2.7)

1.79

15%

medium

6.2a (0.2)

23.4abc (6.5)

37%

medium

6.3a (0.1)

10%

high

6.2a (0.1)

abc

(0.19)

C/N soil a

SOM [%] abc

11.9 (0.4)

58.6

(5.0)

1.71abc (0.93)

17.2a (8.6)

61.9abc (21.8)

20.5ab (3.5)

1.24b (0.47)

17.6a (3.8)

72.2b (13.2)

29.7c (4.8)

2.62c (0.80)

11.9a (2.3)

47.5c (15.8)

sites, sampling dates, and trophic groups, were evaluated by linear models. To estimate the relative contribution of litter and soil material to the diet of decomposer species as well as the contribution of decomposers to the diet of predators, Baysian mixing models were used as implemented in the SIMMR package in R [44]. For calculating isotope niches of soil animals the R package SIBER was employed [25]. Using δ13C and δ15N values of soil animals, we calculated areas for standard Bayesian ellipses (SEA, comparable to SD in univariate data) and convex hulls (TA). To prevent a bias due to small sample sizes, we calculated SEAc, a corrected metric of SEA. Furthermore, we calculated trophic niche widths, represented by the variance of the mean δ13C (resource niche) and δ15N values (degree of trophic level omnivory) [45]. Differences in niche metrics between sheep grazing intensities and trophic groups were evaluated by linear models. All statistical analysis were performed using R (version 3.4.3 [46]) in RStudio (version 1.1.419 [47]).

milled and along with soil animals (i.e. adult Lumbricidae, juvenile Lumbricidae identified through barcoding, Diplopoda, larvae of Coleoptera and Diptera) prepared for stable isotope analysis (SIA). Additionally, sheep dung material from the sampled site was collected for analyses as potential food source. All the samples (i.e. animals, litter, soil, dung) were dried at 70 °C for at least 24 h, weighed into tin capsules (5 × 9 mm), and stored in a desiccator until measurement. In faunal groups comprising very small individuals (e.g. Chironomidae larvae) several individuals were pooled to ensure the minimum required biomass of the sample for analysis (i.e. 0.2 mg dry weight). Larger animals (most Insecta larvae and Lumbricidae) were analysed on an individual basis, or, when exceeding the maximum sampling weight of 1.2 mg, only a selected body part (i.e. elytra for Coleoptera, the front part of large Lumbricidae). Stable isotope ratios were analysed using an elemental analyser (NA 1500, Carlo Erba, Milan, Italy) and a gas isotope mass spectrometer (MAT 251, Finnigan, Bremen, Germany); for details of the analysis system see Reineking et al. (1993) [41]. Stable isotope natural abundance was expressed using the delta notation with δ (‰) = (Rsample – Rstandard)/Rstandard × 1000, where Rsample and Rstandard represent the ratios of heavier to lighter isotopes (13C/12C and 15 N/14N) of the sample and an international standard (Pee Dee Belemnite for C, atmospheric nitrogen for N).

3. Results 3.1. Abundances of soil invertebrates and community composition The 72 samples inspected contained a total of 1846 soil macro-invertebrates (Fig. 1 and Table A1, see supplementary data). In general, densities of individual taxa per sample were low. An exception were Diptera larvae (Nematocera and Brachycera), which were the most abundant taxon with a total of 1423 individuals (i.e., 77.0% of all individuals), contributing most on the sheep resting place TH (843 individuals). Dominating Nematocera families across all 72 samples were Chironomidae (604 individuals), Cecidomyiidae (509 individuals), and Mycetophilidae that were found exclusively on the two mountain top sites (161 individuals). Brachycera larvae were considerably less abundant with a total of 92 individuals and with Empididae the most abundant family (78 individuals). Overall, a total of 297 Coleoptera were sampled with Staphylinidae (114 adults, 66 larvae) the most abundant of the four identified families (Carabidae, Staphylinidae, Elateridae, Scarabaeidae). Again, total Coleoptera density was highest at TH, but did not differ significantly from the other sites. Altogether 143 Lumbricidae were sampled with the highest abundance at the mountain top site TM with medium sheep grazing intensity. In general, epigeic species were more abundant than endogeic ones with a total of 77 and 53 individuals, respectively. Diplopoda (Enantiulus nanus Latzel, 1884) were generally scarce, but were most abundant at the site with low sheep grazing intensity (SL). Principal components axis 1 and axis 2 explained 18.9% and 13.9% of the variance in species abundances, respectively (Fig. 2a). The first axis was associated with Diptera and Coleoptera larvae as well as Diplopoda, while the second axis reflected Lumbricidae and adult Coleoptera. The site with low sheep grazing intensity (SL) and the sheep resting place at the mountain top (TH) were located on opposing sides

2.5. Statistical analyses Community composition and community-environment relationships were evaluated by ordination analyses using the VEGAN package in R [42]. Taxa with very low abundances of < 20 individuals in all samples of the three dates (n = 72) were excluded from all analyses (17 taxa, from which 11 were Brachycera larvae such as Fanniidae with 6 ind., Stratiomyidae with 8 ind, Muscidae with 4 ind.). Data were log-transformed to account for high abundances of Diptera larvae. High correlation coefficients (Pearson's R) and VIFs (i.e. variation inflation factor as measure of multicollinearity) showed high collinearity between soil parameters (Table 1), which were therefore narrowed down to sheep grazing intensity, pH and soil organic matter content (SOM). As the length of gradient (length of DCA axis 1) was < 3, indicating a linear gradient, data were analysed using principal components analysis (PCA) and redundancy analysis (RDA, including environmental variables) [43]. The significance of terms (environmental parameters) was assessed using permutation tests implemented in the VEGAN package [42]. Effect of sheep grazing intensity on abundance of soil invertebrates was evaluated by linear models with factors “taxon” and “sheep grazing intensity”. To display trophic positions of individual taxa, stable isotope values (δ13C and δ15N) were presented in a bi-plot. To account for variations in stable isotope signatures of the basal resources (litter and soil) between sites, litter 15N data from SL from the first sampling date were used as baseline and all other stable isotope values were calibrated to this value. Differences in isotopic values between sheep grazing intensities, 74

European Journal of Soil Biology 86 (2018) 72–80

M. Steinwandter et al.

Fig. 1. Mean abundances of main taxa in high alpine grassland of the Central European Alps. The abundances are given as individuals m−2 with standard deviations and were calculated as the mean from 18 samples at three sampling dates. The contribution of each taxa to the total abundance at each site are gives as percentage. SL – alpine grassland with low sheep grazing intensity, SM – alpine grassland with medium sheep grazing intensity, TM – alpine grassland at the mountain top with medium sheep grazing intensity, TH – sheep berth at the mountain top.

including larvae of Elateridae (Coleoptera) as well as larvae of Mycetophilidae, Sciaridae and Chironomidae (all Diptera). Group 3 comprises predators including Staphylinidae and Carabidae larvae (both Coleoptera), Chilopoda, and larvae of Rhagionidae and Empididae (both Diptera). The shaded area in Fig. 3 indicates the 95% confidence limits of 15N enrichment for litter feeders (trophic enrichment factor 2.7, lower boundary −1.55, upper boundary −0.86, Table A2, see supplementary data), all Lumbricidae and larvae of Tipulidae come within these limits. Notably, stable isotope signatures of the soil differed markedly from those of litter and animal consumers, (F1,21 = 60.4, p < 0.001). Also, soil δ13C showed a high variability with signatures spanning over 8.6‰.

of PCA axis 1, suggesting that the axis represents a grazing intensity gradient. Redundancy analysis (Fig. 2b) confirmed the strong effect of sheep grazing intensity on the soil macro-invertebrate community structure (p = 0.001). Linear models also showed a highly significant impact of sheep grazing intensity on the abundances of soil macro-invertebrates (F2,441 = 9.8, p < 0.001). 3.2. Soil invertebrate food web 3.2.1. Food web structure Stable isotope signatures of basal resources (litter) and animals did not differ significantly between sampling dates; therefore, samples were treated as if originating from one sampling date. Most soil invertebrate taxa showed high variability in 13C and 15N isotopes (Fig. 3). Within the soil food web four well-separated functional groups were distinguished, with their isotope signatures differing significantly (F2,289 = 141.0, p < 0.001). Group 1 comprises primary decomposers including epigeic Lumbricidae (Lumbricus rubellus Hoffmeister, 1843 and Dendobaena octaedra (Savigny, 1826)) as well as larvae of Tipulidae and Cecidomyiidae (both Diptera). Group 2 comprises the secondary decomposers Octolasion lacteum (Örley, 1885) (Lumbricidae) and E. nanus (Diplopoda), as well as dung-feeding adults and larvae of Scarabaeidae (Aphodius spp., Coleoptera). Group 3 comprises root and fungal feeders

3.2.2. Diet analysis We employed stable isotope mixing models to quantify the proportion of the two main food sources available for decomposer species, i.e. litter and soil material, which confirmed that soil material contributed only little to the nutrition of the decomposers (Fig. 4). By using the same mixing models we obtained estimates for the composition of the diet of predatory taxa (Table 2). In general, estimates of diet proportions varied considerably as indicated by their high standard deviations. However, mixing models still presented a clear picture on the preference of most predatory species: adult and larvae of Staphylinidae 75

European Journal of Soil Biology 86 (2018) 72–80

M. Steinwandter et al.

Fig. 2. Community composition (PCA) (a) and community-environment relationship (RDA) (b) of high alpine soil macro-invertebrates. SL – alpine grassland with low sheep grazing intensity, SM – alpine grassland with medium sheep grazing intensity, TM – alpine grassland at the mountain top with medium sheep grazing intensity, TH – sheep berth at the mountain top, pH – soil pH, sheep – sheep grazing intensity, SOM – soil organic matter content (%). For details on ordinations methods see statistical analysis section.

subalpine sites (except for Nematocera larvae). Some taxa – Myriapoda and some species of Lumbricidae – were scarce or even missing, as they get close to their elevational distribution limit (e.g. E. nanus described to occur from 250 to 2100 m [37]). In our study we compared four sites with three sheep grazing intensities, high abundances of soil macro-invertebrates at the sites with medium (SM, TM) and high sheep grazing intensity (TH) confirmed our expectation that the presence of sheep beneficially affects soil invertebrates. Results of the ordination supported these findings, PCA axis 1 represented a sheep grazing intensity gradient placing the sites with low and high grazing intensity on opposing ends of the axis (Fig. 2a). Also, RDA axis 1 was associated with sheep grazing intensity, indicating a highly significant effect of the presence of sheep on the soil invertebrate community structure (Fig. 2b). High density of soil macroinvertebrates on TH, however, was mainly due to high abundance of Diptera larvae, in particular those of Nematocera, all other taxa decreased in abundance. Considering data from subalpine sites [49], Diptera larvae gradually increase with increasing elevation. In an extensively managed (cattle) pasture at 2000 m at the mountain foot of the “Hoher Burgstall”, Steinwandter et al. (2017) [49] found a mean density in early July 2012 of 14.2 ± 18.4 ind. m−2, reaching 17.7 ± 18.6 ind. m−2 on SL (2470 m), and increasing progressively to 392.6 ± 560.7 ind. m−2 at the mountain top of TH (2600 m). The high standard deviations are due to the aggregated occurrence of Diptera larvae, females lay up to 100 eggs in batches into the soil and the hatching larvae are relatively immobile. Diptera larvae play an important role as decomposers in lowland [53] as well as alpine ecosystems [54,55]. They account for high proportions of the high alpine soil animal community thereby supporting and increasingly substituting the functions of Lumbricidae and Diplopoda in decomposition processes as these groups decrease in density at higher altitude [13,55]. Lumbricidae, contrary to our expectation, were found on all sites, especially abundant on the sites with medium sheep grazing intensity (Table A1, see supplementary data). On TH with high sheep grazing intensity, only 4 individuals of the endogeic O. lacteum were found. This site serves as grazing and resting place for the sheep (as indicated by a high amount of sheep dung), resulting in a high compaction of the

as well as larvae of Empididae and Rhagionidae predominantly fed on invertebrates from the root and fungal feeder group, while Lithobiidae and Carabidae larvae did not have any preference for specific taxa. Lumbricidae contributed only little to the diet of predators, with the exception of O. lacteum, which likely was fed by larvae of Empididae and Rhagionidae. 3.2.3. Isotopic niches We did not find any effect of sheep grazing intensity or trophic group affiliation on any trophic niche metric. The model-estimated ellipse areas (SEAc) varied between taxa ranging from 1.21 (Chilopoda) to 8.86 (Carabidae larvae). Likewise, the δ15N niche widths, representing the degree of trophic level omnivory, covered a wide range with Chironomidae larve having the lowest niche width (0.70) and Sciaridae larvae the largest (8.01). The δ13C niche widths, reflecting the resource niche, were narrower and only ranged between 0.16 (Chilopoda) and 4.40 (D. octaedra). 4. Discussion 4.1. Community composition Our study is one of the few investigating the community composition of soil macro-invertebrates in (high) alpine grassland (e.g. [48,49], both European Alps, and [50,51] for the Himalaya). Mountain regions are less investigated than lowland regions mainly due to more difficult accessibility and harsh environmental conditions, even though almost 25% of the global terrestrial biodiversity occurs in mountain regions (including the alpine belt [52]). The soil community composition of the “Hoher Burgstall” mountain top included similar taxa as at subalpine sites located on 2000 m a.s.l. at the mountain foot with comparable characteristics (e.g. slope, management [49]), and results were in line with the expected composition found in the Central European Alps [13]. However, there were considerable differences regarding the diversity and abundance of the most important soil macro-invertebrate groups. Generally, abundances of most taxonomic groups were much lower compared to the lower 76

European Journal of Soil Biology 86 (2018) 72–80

M. Steinwandter et al.

Fig. 3. Bi-plot illustrating trophic positions of high alpine soil invertebrates. Shaded area depicts 95% confidence limits of enrichment factor for 15N, starting from litter, indicating isotopic area of litter feeding animals. Data were calibrated to litter from SL from the first sampling date (see Methods). Circles indicate functional groups. Car_L – Carabidae larvae, Cec_L – Cecidomyiidae larvae, Chi_L – Chironomidae larvae, D. oct – Dendrobaena octaedra, Ela_L – Elateridae larvae, Emp_L – Empididae larvae, E. nan – Enantiulus nanus, Chilo – Chilopoda, L. rub – Lumbricus rubellus, Myc_L– Mycetophilidae larvae, O. lac – Octolasion lacteum, Rha_L – Rhagionidae larvae, Scara – Scarabidae, Sca_L – Scarabaeidae larvae, Sci_L – Sciaridae larvae, Staph – Staphylinidae, Sta_L – Staphylinidae larvae, Tip_L – Tipulidae larvae.

Fig. 4. Diet proportion of litter for high alpine decomposer taxa. 77

European Journal of Soil Biology 86 (2018) 72–80

M. Steinwandter et al.

Table 2 Diet proportions as estimated by stable isotope mixing models (mean and standard deviation) for predatory taxa. Mean resource use above 15% is marked by bold letters; values below 10% were excluded. Lithobiidae

L. rubellus D. octaedra Cecidomyiidae L. Tipulidae L. O. lacteum Scarabaeidae L. Elateridae L. Sciariadae L. Chironomidae L. Mycetophilidae L.

mean

(SD)

0.111

(0.087)

0.098 0.114

(0.071) (0.091)

0.115 0.099

(0.092) (0.079)

0.103 0.095

(0.091) (0.075)

Carabidae L.

Staphylinidae

Staphylinidae L.

Empididae L.

Rhagionidae L.

mean

(SD)

mean

(SD)

mean

(SD)

mean

(SD)

mean

(SD)

0.098

(0.087)

0.103

(0.070)

0.179

(0.137)

0.181

(0.161)

0.111 0.111 0.114 0.118 0.106

(0.107) (0.105) (0.097) (0.110) (0.104)

0.148 0.095 0.145 0.241

(0.121) (0.097) (0.109) (0.170)

0.106 0.124 0.135 0.242

(0.096) (0.122) (0.099) (0.184)

0.124 0.141

(0.129) (0.110)

0.114 0.163

(0.123) (0.147)

0.208

(0.193)

0.140

(0.141)

already shallow soil, a high degree of disturbance, and the presence of much higher amounts of urea and dung (reflected by high total soil nitrogen, Table 1). The extended distribution of Lumbricidae in high alpine sites in Central Europe might be due to favourable conditions such as the increased availability of dung from large herbivores, an easily available and high quality food source for Lumbricidae [17]. Without grazing and roaming sheep, their natural elevational distribution limit might be well below the sites we have investigated in this study. Harsh environmental conditions, lower amounts of resources as well as reduced and physically unsuitable habitats in unmanaged alpine ecosystems limit their presence in sites above 2000 m a.s.l [16]. However, both intensive grazing and the permanent presence of sheep, a situation found especially on TH, might be unfavourable for epigeic Lumbricidae but evidently not for endogeic species such as O. lacteum. Recently, Bueno & Jiménez (2014) showed that alpine endogeic Lumbricidae respond positively to high densities of livestock and disturbance, making these soils a preferable habitat [48]. Futhermore, in fertilized sheep pastures in north New Zealand, endogeic Lumbricidae were dominant constituting 72–93% of the individuals present [56], confirming the positive impact of livestock on endogeic Lumbricidae.

Table 3 Trophic niche metrics of alpine soil invertebrates: Total area (TA), Baysian Ellipse Area (SEA), ellipse area corrected for small sample size (SEAc), δ13C and δ15N niche widths.

L. rubellus D. octaedra Cecidomyiidae L. Tipulidae L. O. lacteum E. nanus Scarabaeidae Scarabaeidae L. Elateridae L. Sciaridae L. Chironomidae L. Mycetophilidae L. Lithobiidae Carabidae L. Staphylinidae Staphylinidae L. Empididae L. Rhagionidae L.

n

TA

SEA

SEAc

δ13C niche width

δ15N niche width

27 20 24 11 17 21 20 5 10 8 22 9 5 12 30 14 19 7

11.68 13.65 23.87 5.07 7.75 9.70 5.47 2.50 3.78 11.42 3.19 6.61 0.95 14.91 8.75 4.38 8.17 6.10

3.60 4.15 6.43 2.25 3.31 3.10 1.96 2.22 2.05 7.21 1.01 4.47 0.90 8.05 2.40 2.00 3.54 3.97

3.74 4.38 6.72 2.50 3.53 3.26 2.07 2.96 2.31 8.41 1.06 5.11 1.21 8.86 2.48 2.16 3.75 4.77

1.09 4.40 1.52 0.33 0.28 0.59 0.39 0.72 0.23 0.66 0.19 1.28 0.16 2.88 0.43 0.58 1.02 0.23

1.23 0.42 2.80 1.55 4.29 1.73 1.13 0.76 2.34 8.01 0.70 1.58 1.21 2.32 1.40 0.76 1.58 1.70

4.2. Trophic relations [26,62], many insects (such as larvae of Elateridae, Sciaridae, Mycetophilidae, and Chironomidae) feed on a mixture of plant roots, fungi and organic matter (including sheep dung), resulting in highly variable δ15N values. This is reflected in the isotopic and δ15N niches of single taxa (Table 3), e.g. sapro- and mycetophagous Sciaridae larvae ranged over eight δ15N units. However, Chironomidae larvae were an exception as they had narrow niches. Soil dwelling larvae of Orthocladiinae, the only subfamily of Chironomidae with terrestrial representatives, were reported to selectively feed on mosses and lichens [63,64], and occurring in high numbers in soils with wet conditions. However, lichen feeders typically are depleted in 15N [65,66], suggesting that larvae of Chironomidae in our high Alpine soils likely feed on fungi rather than lichens [67]. There are only few predatory invertebrates, mainly Insecta larvae, in high alpine soils. Chilopoda occurred in low densities (only 3 individuals were found each on SL and TM, Table A1, see supplementary data), stable isotope mixing models suggested that they are generalist predators feeding on various non-predatory insect larvae (Table 2). Chilopoda generally have a sit-and-wait hunting strategy [68] and might therefore feed on any potential prey passing by. Notably, according to our mixing models predators mostly prey on species of the root and fungal feeder group and less on species of the secondary decomposer and dung feeder group (Fig. 3). However, they do not feed on primary decomposers, also indicated by their narrow isotopic and δ15N niches (Table 2). An exception to the latter are larvae of Carabidae characterized by a large isotopic niche and variable δ15N values,

To our knowledge, this is the first study to assess the trophic structure of high alpine soil macro-invertebrates. Stable isotope analysis has proven to be a valuable tool to depict food web structures [21], however, delineating distinct trophic levels or groups is challenging ([29,57]). In our study, the dual isotope approach allowed distinguishing four well-separated trophic guilds: (1) primary decomposers, (2) secondary decomposers and dung feeders, (3) root and fungal feeders, and (4) predators (Fig. 3). Using trophic enrichment factors from the literature (Table A2, see supplementary data), we included litter feeders in the narrower sense within the group of decomposers (Fig. 3, shaded area). Epigeic Lumbricidae and larvae of Tipulidae formed part of this group, both taxa are known to feed on decaying organic material [58,59]. Stable isotope mixing models confirmed the high proportion of litter in the diet not only for these taxa, but for all taxa included in the two decomposer groups (Fig. 4). Soil-dwelling larvae of Cecidomyiidae (included in the primary decomposer group) are known to feed on organic substrates and soil [59], however, the high proportion of litter in the diet of the geophagous Lumbricidae O. lacteum is rather surprising. In comparison with all other taxa, this Lumbricidae species has a high δ15N niche width (but not isotopic niche area, Table 3), which might result from feeding on a mixture of mineral and soil organic matter (Table 1). Soil organic matter is abundant in high alpine soils due to slow litter decomposition [60] and shows a high variability in δ15N due to different stages of humification [61]. As known from other soil food webs, omnivory is widespread 78

European Journal of Soil Biology 86 (2018) 72–80

M. Steinwandter et al.

indicating a high level of omnivory. Carabidae are known to feed on both plant and animal material, larvae of some species need to feed on seeds for successful development [69]. This is reflected by their isotopic signals pointing to a mixed diet. Intraguild predation is unlikely to play a major role in the predators studied as their isotopic signatures were very similar and at similar δ15N levels. As mesofauna groups (Collembola, Acari) were not part of this study, trophic connections between macro-invertebrates and micro-arthropods were not assessed. However, we assume that it is very likely that most (small) predators also feed on collembolans and mites inhabiting high alpine soils as well as the nival zone in high species numbers and abundances [13].

altitudinal gradient in the Central Alps, Zootaxa 3626 (2013) 429–454. [6] C. Körner, Alpine plant diversity: a global survey and functional interpretations, in: F.S. Chapin, IIIC. Körner (Eds.), Arctic and Alpine Biodiversity: Patterns, Causes and Ecosystem Consequences, Ecological Studies 113, Springer Verlag, Berlin, 1995, pp. 45–62. [7] C. Körner, G. Nakhutsrishvili, E.M. Spehn, High-elevation land use, biodiversity, and ecosystem functioning, in: E.M. Spehn, M. Liberman, C. Körner (Eds.), Land Use Change and Mountain Biodiversity, CRC Publishers, Boca Raton, 2006, pp. 3–21. [8] M.E. Alonso-Amelot, A. Oliveros, M.P. Calcagno-Pisarelli, Phenolics and condensed tannins in relation to altitude in neotropical Pteridium spp. A field study in the Venezuelan Andes, Biochem. Systemat. Ecol. 32 (2004) 969–981. [9] M.E. Alonso-Amelot, A. Oliveros-Bastidas, M.P. Calcagno-Pisarelli, Phenolics and condensed tannins of high altitude Pteridium arachnoideum in relation to sunlight exposure, elevation, and rain regime, Biochem. Systemat. Ecol. 35 (2007) 1–10. [10] A. Rief, B.A. Knapp, J. Seeber, Palatability of selected alpine plant litters for the decomposer Lumbricus rubellus (Lumbricidae), PLoS One 7 (2012) e45345. [11] M.O. Gessner, C.M. Swan, C.K. Dang, B.G. McKie, R.D. Bardgett, D.H. Wall, S. Hättenschwiler, Diversity meets decomposition, Trends Ecol. Evol. 25 (2010) 372–380. [12] P. García-Palacios, F.T. Maestre, J. Kattge, D.H. Wall, Climate and litter quality differently modulate the effects of soil fauna on litter decomposition across biomes, Ecol. Lett. 16 (2013) 1045–1053. [13] E. Meyer, K. Thaler, Animal diversity at high altitudes in the austrian central Alps, in: F.S. Chapin, IIIC. Körner (Eds.), Arctic and Alpine Biodiversity: Patterns, Causes and Ecosystem Consequences, Ecological Studies 113, Springer Verlag, Berlin, 1995, pp. 95–106. [14] E. Meyer, Ökologische Untersuchungen an Wirbellosen des zentralalpinen Hochgebirges (Obergurgl, Tirol): IV. Aktivitätsdichte, Abundanz und Biomasse der Makrofauna vol. 13, Veröff. Univ. Innsbruck 125-Alpin Biol. Stud, 1980, pp. 1–53. [15] E. Meyer, Bodenbiologische Untersuchungen in den Hohen Tauern 1974–1978. Abundanz und Biomasse von Invertebraten in zentralalpinen Böden (Hohe Tauern, Österreich) 4 Veröff. Österr. MaB-Hochgebirgsprogramm, 1981, pp. 153–178. [16] T. Decaëns, 2010: macroecological patterns in soil communities, Global Ecol. Biogeogr. 19 (2010) 287–302. [17] N.B. Hendriksen, Consumption and utilization of dung by detritivorous and geophagous earthworms in a Danish pasture, Pedobiologia 35 (1991) 65–70. [18] N.L. Schon, A.D. Mackay, M.A. Minor, G.W. Yeates, M.J. Hedley, Soil fauna in grazed New Zealand hill country pastures at two management intensities, Appl. Soil Ecol. 40 (2008) 218–228. [19] O.L. Petchey, K.J. Gaston, Functional diversity: back to basics and looking forward, Ecol. Lett. 9 (2006) 741–758. [20] F.V. Crotty, R.P. Blackshaw, S.M. Adl, R. Inger, P.J. Murray, Divergence of feeding channels within the soil food web determined by ecosystem type, Ecol. Evol. 4 (2014) 1–13. [21] A. Tiunov, Stable isotopes of carbon and nitrogen in soil ecological studies, Biol. Bull. 34 (2007) 395–407. [22] M. Traugott, S. Kamenova, L. Ruess, J. Seeber, M. Plantegenest, Empirically characterising trophic networks: what emerging DNA-based methods, stable isotope and fatty acid analyses can offer, Adv. Ecol. Res. 49 (2013) 177–224. [23] M. Minagawa, E. Wada, Stepwise enrichment of 15N along food chains: further evidence and the relation between δ15N and animal age, Geochem. Cosmochim. Acta 48 (1984) 1135–1140. [24] C.A. Layman, D. Albrey Arrington, C.G. Montaña, D.M. Post, Can stable isotope ratios provide for community-wide measures of trophic structure? Ecology 88 (2007) 42–48. [25] A.L. Jackson, A.C. Parnell, R. Inger, S. Bearhop, Comparing isotopic niche widths among and within communities: SIBER – stable Isotope Bayesian Ellipses in R, J. Anim. Ecol. 80 (2011) 595–602. [26] S. Scheu, M. Falca, The soil food web of two beech forests (Fagus sylvatica) of contrasting humus type: stable isotope analysis of a macro- and a mesofaunadominated community, Oecologia 123 (2000) 285–296. [27] S. Ponsard, R. Arditi, What can stable isotopes (δ15N and δ13C) tell about the food web of soil macro-invertebrates? Ecology 81 (2000) 852–864. [28] B. Klarner, M. Maraun, S. Scheu, Trophic diversity and niche partitioning in a species rich predator guild – natural variations in stable isotope ratios (13C/12C, 15 N/14N) of mesostigmatid mites (Acari, Mesostigmata) from Central European beech forests, Soil Biol. Biochem. 57 (2013) 327–333. [29] K. Oelbermann, S. Scheu, Trophic guilds of generalist feeders in soil animal communities as indicated by stable isotope analysis (15N/14N), Bull. Entomol. Res. 100 (2010) 511–520. [30] E. Corral-Hernández, M. Maraun, J.C. Iturrondobeitia, Trophic structure of oribatid mite communities from six different oak forests (Quercus robur), Soil Biol. Biochem. 83 (2015) 93–99. [31] O. Ferlian, B. Klarner, A.E. Langeneckert, S. Scheu, Trophic niche differentiation and utilisation of food resources in collembolans based on complementary analyses of fatty acids and stable isotopes, Soil Biol. Biochem. 82 (2015) 28–35. [32] T. Decaëns, D. Porco, R. Rougerie, G.G. Brown, S.W. James, Potential of DNA barcoding for earthworm research in taxonomy and ecology, Appl. Soil Ecol. 65 (2013) 35–42. [33] H. Troger, H. Janetschek, E. Meyer, W. Schatz, Abundance of emerging insects (Diptera/Coleoptera/Hymenoptera) in the central high Alps (Tirol: Ötztal), Entomol. Gen. 18 (1994) 241–260. [34] D. Kempson, M. Lloyd, R. Ghelardi, A new extractor for woodland litter, Pedobiologia 3 (1963) 1–21. [35] S. Kaehler, E.A. Pakhomov, Effects of storage and preservation on the δ13C and δ15N signatures of selected marine organisms, Mar. Ecol. Prog. Ser. 219 (2001)

5. Conclusions High alpine soils are characterized by moderate taxa diversity and low density of soil macro-invertebrates presumably due to the harsh environmental conditions. The presence of sheep (and their dung) beneficially affects soil macro-invertebrate communities, allowing some taxa to extend their distributional range to high Alpine soils (especially Lumbricidae). Notably, as indicated by stable isotope analysis, the food web is based almost exclusively on litter (and possibly root) material, even species otherwise known to be geophagous (O. lacteum) prefer – or are constrained – to feed on organic matter (litter or soil organic matter). The macro-invertebrates found in our high Alpine soils may have adapted to the relatively low availability of animal and plant/ litter material. Only few species show narrow isotopic niches indicating specialized feeding, suggesting that they are forced to feed on a wide variety of food resources present in their vicinity. The low mobility in the compact high alpine soils might increase the search time for favourable food resources, resulting in particular in predators to fed on a wider range of prey as in lowland habitats. In contrast to other soil food webs, our high Alpine soil food webs represent relatively simple systems with four well-separated trophic groups. This suggests that the shallow soils with high organic matter predominantly originating from few species of alpine grasses and mosses do not allow the formation of more complex food webs. This is, to our knowledge, the first study to investigate the community and the trophic composition of high alpine soil fauna communities, more studies will be needed to better understand the fragile (high) mountain ecosystems which are increasingly exposed to climate as well as land-use changes. Acknowledgements The study was financially supported by the Austrian Science Fund (project number T441). The authors thank the Department of Innovation, Research and University of the Autonomous Province of Bozen/Bolzano for covering the Open Access publication cost. Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx. doi.org/10.1016/j.ejsobi.2018.03.006. References [1] P. Murray, F. Crotty, N. van Eekeren, Management of grassland systems, soil, and ecosystem services, in: D.H. Wall, R.D. Bardgett, V. Behan-Pelletier, J.E. Herrick, T.H. Jones, K. Ritz, J. Six, D.R. Strong, W. van der Putten (Eds.), Soil Ecology and Ecosystem Services, Oxford University Press, 2012, pp. 282–293. [2] J. Seeber, G.U.H. Seeber, W. Kössler, R. Langel, S. Scheu, E. Meyer, Abundance and trophic structure of macro-decomposers on alpine pastureland (Central Alps, Tyrol): effects of abandonment of pasturing, Pedobiologia 49 (2005) 221–228. [3] A. Rozen, R.W. Myslajek, L. Sobczyk, Altitude versus vegetation as the factors influencing the diversity and abundance of earthworms and other soil macrofauna in montane habitat (Silesian Beskid Mts, Western Carpathians), Pol. J. Ecol. 61 (2013) 145–156. [4] D. Schmera, B. Baur, Gastropod communities in alpine grasslands are characterized by high beta diversity, Community Ecol. 15 (2014) 246–255. [5] B.M. Fischer, H. Schatz, Biodiversity of oribatid mites (Acari: oribatida) along an

79

European Journal of Soil Biology 86 (2018) 72–80

M. Steinwandter et al.

[52] E.M. Spehn, K. Rudmann-Maurer, C. Körner, D. Maselli, Mountain Biodiversity and Global Change, GMBA-DIVERSITAS, Basel, 2010. [53] J. Frouz, Use of soil dwelling Diptera (Insecta, Diptera) as bioindicators: a review of ecological requirements and response to disturbance, Agric. Ecosyst. Environ. 74 (1999) 167–186. [54] J. Seeber, A. Rief, K. Heller, E. Meyer, Emergence rates of dipterans in high alpine soils with special emphasis on the Sciaridae (Insecta: Nematocera), M. D. Gesell. Allg. Ange. Entom. 18 (2012) 367–370. [55] F. Kitz, M. Steinwandter, M. Traugott, J. Seeber, Increased decomposer diversity accelerates and potentially stabilises litter decomposition, Soil Biol. Biochem. 83 (2015) 138–141. [56] N.L. Schon, A.D. Mackay, R.A. Gray, M.A. Minor, Earthworms in New Zealand sheep- and dairy-grazed pastures with focus on anecic Aporrectodea longa, Pedobiologia 54S (2011) S131–S137. [57] T. Eggers, T.H. Jones, You are what you eat… or are you? Trends Ecol. Evol. 15 (2000) 265–266. [58] J.P. Curry, O. Schmidt, The feeding ecology of earthworms – a review, Pedobiologia 50 (2007) 463–477. [59] P. Oosterbroek, The European Families of the Diptera. Identification, Diagnosis, Biology, KNNV Publishing, Utrecht, 2006 204 pp. [60] C. Körner, Alpine Plant Life - Functional Plant Ecology of High Mountain Ecosystems, second ed., Springer Verlag, Heidelberg, 2003. [61] J. Pumpanen, A. Lindén, V.J. Bruckman, F. Berninger, H. Ilvesniemi, M. Oinonen, E. Sonninen, M. Kukumägi, J. Heinonsalo, The effect of roots and easily available carbon on the decomposition of soil organic matter fractions in boreal forest soil, Eur. J. Soil Sci. 68 (2017) 537–546. [62] C. Digel, A. Curtsdotter, J. Riede, B. Klarner, U. Brose, Unravelling the complex structure of forest soil food webs: higher omnivory and more trophic levels, Oikos 123 (2014) 1157–1172. [63] Y.R. Delettre, Larvae of terrestrial Chironomidae (Diptera) colonize the vegetation layer during the rainy season, Pedobiologia 44 (2000) 622–626. [64] J. Frouz, A. Lukesova, Food preference of two species of terrestrial chironomids (Diptera, Chironomidae), Dipterol. Bohemoslov. 7 (1995) 41–46. [65] M. Chahartaghi, R. Langel, S. Scheu, L. Ruess, Feeding guilds in Collembola based on nitrogen stable isotope ratios, Soil Biol. Biochem. 37 (2005) 1718–1725. [66] M. Maraun, G. Erdmann, B.M. Fischer, M.M. Pollierer, R.A. Norton, K. Schneider, S. Scheu, Stable isotopes revisited: their use and limits for oribatid mite trophic ecology, Soil Biol. Biochem. 43 (2011) 877–882. [67] Y.R. Delettre, J. Lagerlöf, Abundance and life history of terrestrial Chironomidae (Dipera) in four Swedish agricultural cropping systems, Pedobiologia 36 (1992) 69–78. [68] B. Günther, B.C. Rall, O. Ferlian, S. Scheu, B. Eitzinger, Variations in prey consumption of centipede predators in forest soils as indicated by molecular gut content analysis, Oikos 123 (2014) 1192–1198. [69] P. Saska, Contrary food requirements of the larvae of two Curtonotus (Coleoptera: Carabidae: Amara) species, Ann. Appl. Biol. 147 (2005) 139–144.

299–304. [36] C. Csuzdi, A. Zicsi, Earthworms of Hungary (Annelida. Oligochaeta, Lumbricidae), Hungarian Natural History Museum, Budapest, 2003. [37] A. Pedroli-Christen, Faunistik der Tausendfüssler der Schweiz (Diplopoda): Documenta Faunistica Helvetica 14, Centre suisse de cartographie de la faune, Neuchatel, 1993 167 pp. [38] M. Schaefer, Brohmer - Fauna von Deutschland: Ein Bestimmungsbuch unserer heimischen Tierwelt, 23th ed., Quelle & Meyer Verlag, Wiebelsheim, 2009. [39] O. Folmer, M. Black, W. Hoeh, R. Lutz, R. Vrijenhoek, DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates, Mol. Mar. Biol. Biotechnol. 3 (1994) 294–299. [40] J. Klarica, A. Kloss-Brandstätter, M. Traugott, A. Juen, Comparing four mitochondrial genes in earthworms – implications for identification, phylogenetics, and discovery of cryptic species, Soil Biol. Biochem. 45 (2012) 23–30. [41] A. Reineking, R. Langel, J. Schikowski, 15-N, 13-C-on-line measurements with an elemental analyser (Carlo Erba, NA 1500), a modified trapping box and a gas isotope mass spectrometer (Finnigan, MAT 251), Isot. Environ. Health Stud. 29 (1993) 169–174. [42] J. Oksanen, F.G. Blanchet, R. Kindt, P. Legendre, P.R. Minchin, R.B. O'Hara, G.L. Simpson, P. Solymos, M.H.H. Stevens, H. Wagner, Vegan: community Ecology Package. R package version 2.4-6, http://CRAN.R-project.org/package=vegan, (2018). [43] J. Lepš, P. Šmilauer, Multivariate Analysis of Ecological Data Using CANOCO, Cambridge University Press, Cambridge, 2003. [44] R. Inger, A. Jackson, A. Parnell, S. Bearhop, SIAR V4 (stable isotope analysis in R): an Ecologist's guide, https://www.tcd.ie/Zoology/research/research/theoretical/ siar/SIAR_For_Ecologists.pdf, (2013). [45] D. Sanders, E. Vogel, E. Knop, Individual and species-specific traits explain niche size and functional role in spiders as generalist predators, J. Anim. Ecol. 84 (2015) 134–142. [46] R Core Team, R: A language and environment for statistical computing, Vienna, R Foundation for Statistical Computing, Vienna, 2017http://www.R-project.org/. [47] RStudio Team, RStudio: Integrated Development for R, RStudio Inc., Boston, 2016http://www.rstudio.com/. [48] C.G. Bueno, J.J. Jiménez, Livestock grazing activities and wild boar rooting affect alpine earthworm communities in the Central Pyrenees (Spain), Appl. Soil Ecol. 83 (2014) 71–78. [49] M. Steinwandter, B.C. Schlick-Steiner, G.U.H. Seeber, F.M. Steiner, J. Seeber, Effects of Alpine land-use changes: soil macrofauna community revisited, Ecol. Evol. 7 (2017) 5389–5399. [50] P. Wu, H. Zhang, Y. Wang, The response of soil macroinvertebrates to alpine meadowdegradation in the Qinghai–Tibetan Plateau, China, Appl. Soil Ecol. 90 (2015) 60–67. [51] X. Yin, L. Qui, Y. Tu, Y. Tao, J. An, W. Xin, Y. Jiang, Characteristics of soilmacrofauna community and CCA analysis between soilmacrofauna and soil properties in the wetlands of the Lhasa River basin, Acta Ecol. Sin. 35 (2015) 210–215.

80