Introduction

Oribatid mites are mainly saprophages that play an important role in the decomposition of the soil organic matter (Schuster 1956; Seniczak 1978), including in the Arctic regions. They consume the litter, reducing it to pellets or smaller fragments and hence increasing the surface area available for further microbial decomposition (Schuster 1956; Seniczak 1978). Microbial activity is particularly intensive in the gut tracts of mites (Seniczak and Stefaniak 1978; Stefaniak and Seniczak 1976, 1981), releasing the elements for plants growth. The oribatid mites are usually abundant and rich in species, but these population parameters depend on the type and extent of the vegetation cover and quantity of organic matter (Coulson et al. 2003; Seniczak 1978; Seniczak and Plichta 1978; Seniczak et al. 2014a). Bayartogtokh et al. (2011) listed 81 species of these mites from the Svalbard archipelago, which comprises only 0.8% of the total 10,000 species of recognized oribatid mite species (Behan-Pelletier and Schatz 2010). However, within the Svalbard inventory there is some taxonomic uncertainty regarding the identification of some species (Coulson et al. 2014). Nonetheless, population parameters, such as distribution, density, stage and sex structure of the Svalbard oribatid communities are unknown, except for the four species investigated by Seniczak et al. (2014a, 2015a, b, 2016).

Diapterobates notatus (Thorell, 1871) is common and abundant in the Svalbard archipelago. Seniczak and Plichta (1978) found it to be dominant in moss–lichen tundra at Hornsund (Saxifraga oppositifolia, Cetraria delisei, C. delisei + Salix polaris and Drepanocladus uncinatus) and juveniles greatly outnumbered adults. In the moss–grass tundra at Petuniabukta and Vestpynten (patches of moss, grasses, moss mixed with grasses, Cassiope tetragona, S. polaris and bare soil), D. notatus was not so abundant (Seniczak et al. 2014a), but the juveniles also dominated the adults. In patches of six vascular plant species (Luzula confusa, Dryas octopetala, C. tetragona, S. polaris, Silene acaulis and S. oppositifolia) investigated by Coulson et al. (2003), the density of D. notatus was also lower, but these authors counted only the adults and omitted juveniles. In these three papers, S. paludicola Thor, 1930 was absent.

Here we focus on D. notatus and S. paludicola, which are closely related ceratozetid species occurring in the Svalbard archipelago, but which have dissimilar morphology, ecology and distributions. The former species prefers humid to semi-dry soils covered with mosses, grasses, heath-like vegetation and sometimes willow or birch scrub (Haarlov 1967), with a wide variety of mesic habitats, such as bog, shrub, dry tundra and littoral areas (Behan-Pelletier 1997), whereas the latter species prefers moist to wet habitats (Behan-Pelletier 1985).

We hypothesize that the population density and certain other population parameters of D. notatus and S. paludicola are related to the type and composition of the vegetation cover, in this paper referred to as “vegetation classes”. The vegetation cover creates the microhabitats for the mite community, integrating various important environmental factors that control the population parameters of these species or can explain absence. A limited number of studies have investigated the distribution of oribatid mites in the Svalbard archipelago (Bayartogtokh et al. 2011; Coulson et al. 2003, 2014; Seniczak et al. 2014a, 2015a, b), but these concerned single regions or omitted the demographically important juvenile stages. Our aim is to compare the density, stage and sex structure, number of gravid females, and the body size of D. notatus and S. paludicola from diverse vegetation classes and locations in Svalbard.

Material, methods and terminology

Sampling and mite extraction, determination and analyses

We investigated the distribution of D. notatus and S. paludicola in the Svalbard archipelago (Fig. 1) using 172 samples from 33 locations (Fig. 2). These samples were collected mainly in the rocky coastal zone (1–5 m a. s. l.) except for the vegetated slope in Ringhorndalen (about 100 m a. s. l.), during the polar summer (a period of around 6 weeks from late June to early August) between 2009 and 2014. The number of samples from each location varied between 1 and 12. Soil samples were collected from the typical vegetation communities at each location. Each sample had dimensions of 10 × 10 cm by c. 5 cm deep (the usual maximum depth of the organic soil). The samples (soil and plant cover) were kept cool until being extracted in Tullgren funnels (Burkard Scientific Ltd., Uxbridge, UK) at the University Centre in Svalbard (UNIS) within 5 days of collection. Extracted animals were preserved in 96% ethanol. The density, stage structure and sex ratio, number of gravid females and the body length in very abundant samples on the basis of 30 individuals selected at random was determined for both species.

Fig. 1
figure 1

Location of the Svalbard archipelago (box) in the European High Arctic

Full size image
Fig. 2
figure 2

Number of vegetation classes and presence of Diapterobates notatus (D) and Svalbardia paludicola (S) at sampling locations in the Svalbard archipelago; definitions of the vegetation classes are given in “Material, methods and terminology”

Full size image

We identified the juveniles of D. notatus and S. paludicola according to Seniczak (1980, 1993) and Behan-Pelletier (1985). The larva of D. notatus has the posterior gastronotal setae (dp, lp and h 1) longer than that of S. paludicola, whereas the nymphs of D. notatus have the gastronotal setae c 2, da, dm, la, lm and h 2 distinctly longer than those of S. paludicola (Figs. 3, 4). Moreover, the juveniles of S. paludicola have a humeral organ (oh) present, whereas those of D. notatus do not possess this organ. Species nomenclature follows Subías (2004). Identified material is deposited as alcohol samples in the Department of Evolutionary Biology, Kazimierz Wielki University in Bydgoszcz (Poland).

Fig. 3
figure 3

Diapterobates notatus, lateral aspect, legs partially drawn, scale bar 50 μm; a larva, b tritonymph

Full size image
Fig. 4
figure 4

Svalbardia paludicola, lateral aspect, legs partially drawn, scale bar 50 μm; a larva, b tritonymph

Full size image

Statistical analysis of the density of D. notatus was performed only for locations with five or more replicates. We compared the density and stage structure of D. notatus between vegetation classes using Euclidean distances. The resemblance matrices were produced using a cluster analysis and employing an unweighted pair-group method with arithmetic averaging (UPGMA) (Jongman at al. 1987; Piernik 2008). Normality was examined by the Kolmogorov–Smirnov test and homogeneity of variance with the Levene’s test. As the assumptions of variance analysis were not met, nonparametric tests were used. Kruskal–Wallis was to inspect the raw data and in the case of significant differences between means, a multiple comparison test on ranks was employed (Stanisz 2006). The level of significance for all statistical tests was accepted at α = 0.05. Data were analyzed using MS Excel 2007, Statistica 10.0 and MVSP 3.2 (Kovach Computing Services 2010).

Vegetation and land cover types

The distribution of D. notatus and S. paludicola was determined in revised vegetation classes of the Svalbard vegetation map (Johansen et al. 2012) and given in Fig. 2:

  1. 1.

    Pioneer vegetation on wet vegetated flats, beaches, slopes and river fans (class 8), with some exposed Dryas dominated vegetation and graminoids Luzula sp.

  2. 2.

    Moderate snowbed and snowflush communities (class 9), with snow protection during winter, plant composition from moss-rich communities to tussock flats, and Deschampsia alpina, Poa alpina, S. polaris, Equisetum arvense and moss.

  3. 3.

    Swamps, hygrophilous vegetation and wet moss tundra (class 10), including bird cliffs or bird manured areas, with moss and Arctophila fulva, Carex subsphatacea, Eriophorum triste, Dupontia psilosantha and Ranunculus hyperboreus.

  4. 4.

    Moss tundra (class 11), occurs in areas with moderate snow cover during winter.

  5. 5.

    Rich moss tundra and bird cliff vegetation (class 12), with moist moss Tomentypnum nitens, grasses (Alopecurus ovatus, Dupontia fisheri, Dupontia pelligera and Eriophorum scheuchzeri) and sedges (Carex parallela and Carex saxatilis).

  6. 6.

    Arctic meadows—rich vegetation (class 13), with grasses and sedges and in the drier parts of Dupontia meadows. This class includes bird cliff vegetation communities, with grasses (P. alpina, Festuca rubra richardsonii, Festuca baffinensis and Trisetum spicatum) and forbs (Taraxacum brachyceras and Cochlearia groenlandica).

  7. 7.

    Open dry-grass communities (class 14), comprise open, extremely dry communities characterized by grass, sedge and rush species and lichens, High Arctic steppe vegetation.

  8. 8.

    Open Dryas/Carex rupestris communities (class 15), with Carex rupestris, Carex nardina, S. oppositifolia and scattered stands of D. octopetala.

  9. 9.

    Established dense Dryas heaths (class 16), with a significant winter snow cover and is dominated by species such as D. octopetala, C. tetragona, S. oppositifolia, and Papaver dahlianum.

  10. 10.

    Luzula vegetation—sparse graminoid vegetation (class 17), is composed of Luzula sp., moss Racomitrium lanuginosum and lichens (Flavocetraria nivalis, Cladonia mitis and C. delisei).

  11. 11.

    Gravel barren communities (class 18), with barren gravel communities constitute none-to-sparsely vegetated gravel-boulders, block fields and large periglacial polygons.

Results

Distribution, density and stage structure of species of Diapterobates notatus and Svalbardia paludicola

Diapterobates notatus was present in 133 samples (Table 1) from a total 172 total samples investigated, so the constancy of species was high (77% of all samples). Only at Alkepynten (moss tundra), Hornsund (rich moss tundra), Blåbukta and Hambergbukta (Luzula vegetation) was this species absent. Diapterobates notatus occurred in all 11 vegetation classes, but the highest mean density (5759 ± 7155.4 ind m−2) was achieved in open Dryas/Carex rupestris communities (Table 1). A relatively great mean density occurred in moderate snowbed communities, 4450 ± 6010.4 ind m−2) but lower densities (150–312 ind m−2) were noted in pioneer vegetation, moss tundra, open dry-grass communities and gravel barren communities. In other vegetation classes the mean densities were 1836–2856 ind m−2.

Table 1 Mean density of Diapterobates notatus (ind dm−2, ±SD and range) in vegetation classes in Svalbard
Full size table

The stage structure of the D. notatus population was determined from a total of 1962 juveniles and 1275 adults. In most vegetation classes the juveniles dominated the adults. Juveniles comprised between 51.3% (rich moss tundra) and 100% (gravel barren communities) of the population, except for pioneer vegetation where the proportion declined to 33.3% (Table 1). In most vegetation classes the larvae or protonymphs dominated the other juvenile stages and the proportion of subsequent nymphs (protonymph to tritonymph), usually decreased. The density of juveniles and adults of D. notatus was the highest in open Dryas/Carex rupestris communities, as compared to that in moss tundra and open dry-grass communities (Fig. 5).

Fig. 5
figure 5

Structure of the mean density of adults (AD) and juveniles (JUV) of Diapterobates notatus in 11 vegetation classes (the names of vegetation classes are given in “Material, methods and terminology”)

Full size image

Diapterobates notatus displayed significant variations in density between vegetation classes (Table 2). For example, in Arctic meadows the mean density of this species was significantly greater for larvae, juveniles, adults and all life stages than in moss tundra. In the former vegetation class, the mean density of D. notatus was also significantly greater than in open dry-grass communities but this was due to the large numbers of both juveniles and adults. In open Dryas/Carex rupestris communities, the mean density of D. notatus was significantly greater than in moss tundra and open dry-grass communities due to large numbers of larvae, deutonymphs, all juveniles, adults and all life stages (moss tundra) or adults and all instars (open dry – grass communities). In established Dryas heaths, the mean density of D. notatus was significantly greater than in moss tundra and open dry-grass communities due to the large numbers of adults and all instars (moss tundra) and tritonymphs, adults and all instars (open dry-grass communities).

Table 2 Significant differences in densities of D. notatus at P ≤ 0.05 between vegetation classes in Svalbard
Full size table

The preferences of D. notatus to particular vegetation classes are clear from the cluster analysis (Fig. 6), which separates open Dryas/Carex rupestris communities, with the greatest density of this species (Table 2), from the other plant classes. Moss tundra and open dry-grass communities cluster together due to their low densities.

Fig. 6
figure 6

Cluster analysis of the density of Diapterobates notatus in the vegetation classes (Euclidean distances) (definitions of the vegetation classes are given in “Material, methods and terminology”)

Full size image

Svalbardia palludicola was noted in only two vegetation classes, but only in Arctic meadows was this species abundant (Table 3) and the juveniles were more abundant than the adults. In one sample from Reinsdyrflya this species successfully competed with D. notatus and was dominant.

Table 3 Stage structure and sex ratio of Svalbardia palludicola in Svalbard
Full size table

Sex structure, number of gravid females and body length of Diapterobates notatus and Svalbardia paludicola

The mean sex ratio (females to males) of D. notatus was 1:0.7, but this varied between vegetation classes from 1:0.4 to 1:1.0 (Table 4) (no statistically significant differences, P > 0.05). In all vegetation classes, the sex ratio varied greatly between samples, especially in established dense Dryas heaths (1:0.1–1:7.5). For S. palludicola the males were more abundant than females (sex ratio 1:1.2), but the estimate was based from only one sample. The occurrence of gravid females of D. notatus was generally high (58–95%) and the differences between vegetation classes were insignificant (Table 4). The gravid females carried 6–7 large eggs, rarely eight eggs, in all vegetation classes.

Table 4 Sex ratio of Diapterobates notatus and number of gravid females in chosen vegetation classes in Svalbard
Full size table

Females of D. notatus were usually larger than the males (Table 5), but the body length of the mites varied between vegetation classes. However, in Arctic meadows females were significantly larger than those in swamps, hygrophilous vegetation, rich moss tundra Arctic meadows and established Dryas heaths. In all vegetation classes, the range of both female and male body length was large and the values overlapped.

Table 5 Comparison of average body length of females and males of D. notatus in some vegetation classes in Svalbard
Full size table

New finding for Svalbard

We found two females of Scutozetes clavatosensillus Ermilov, Martens & Tolstikov, 2013 in Mosselbukta (north Spitsbergen, established Dryas heaths). This is a first record for Svalbard.

Discussion

This study demonstrates that D. notatus and S. paludicola are typical components of the oribatid mite fauna of the Svalbard archipelago. The former species is distinctly more common than the latter species. In Svalbard it prefers the open Dryas/Carex rupestris communities, with S. oppositifolia, in which the density of D. notatus is usually high (Coulson et al. 2003; Seniczak and Plichta 1978). However, in this study the mean density of D. notatus is relatively low (5759 ind m−2) compared to previous reports. For example, in patches of S. oppositifolia at Hornsund this species achieved a density of 64,410 ind m−2 and comprised 82.5% of all oribatid mites (Seniczak and Plichta 1978). In other plant patches with C. delisei, C. delisei + S. polaris, and D. uncinatus, the density was lower (9635–17,434 ind m−2), but D. notatus still dominated, forming 65–91% of the oribatid mite community. In all plant patches, the larvae and protonymphs comprised more than 50% of all individuals of this species. Intriguingly, the lowest percent of D. notatus was observed in patches of D. uncinatus, where the density of Mesostigmata was the greatest. The latter group of mites predates oribatid mites (Karg 1993; Ruf and Beck 2005) and reduces their density. Similarly, in patches of moss + grasses at Vestpynten (Longyearbyen, Adventfjord), D. notatus was also abundant (37,600 ind m−2), but was not a dominant (Seniczak et al. 2014a). In five vascular plant species investigated by Coulson et al. (2003), the mean density of adult D. notatus in Dryas patches was clearly lower (4.4 ind m−2) than in Saxifraga pathes (35.8 ind m−2). Relatively low mean density of D. notatus in the open Dryas/Carex rupestris communities can be explained by sampling the most representative areas of this vegetation class, not in plant patches of S. oppositifolia.

In other vegetation classes, the density of D. notatus was lower or low, and significant differences in mean values were noted between vegetation classes, but there are no data in the literature on how vegetation may influence the population density of D. notatus. A low density of D. notatus in pioneer vegetation can be explained by the type of vegetation, that of open dry-grass communities, by the extremely arid communities, and that of in gravel barren communities, by the sparsely vegetated and mobile gravel, all of which provide poor habitats for oribatid mites. Behan-Pelletier (1997) also found D. notatus to be rare in barren habitats. Although a more open vegetation may allow the ground to achieve greater summer temperatures (Coulson et al. 1993) and the early melt of the thin snow that accumulates along ridge crests results in a longer snow free summer period (Ávila-Jiménez and Coulson 2011), these environmental benefits of pioneer and open vegetation may not offset the reduced development of organic soils and lack of suitable microhabitats. However, in vegetation classes with low densities, the juveniles of D. notatus were present suggesting the development of this species here despite the potentially extremely demanding living conditions. The wide variance of D. notatus densities in the various vegetation classes could be due to the mosaic pattern of distribution of plant species (Coulson et al. 2003; Jónsdóttir 2005; Seniczak and Plichta 1978; Seniczak et al. 2014a) or the heterogenous distribution of the oribatid mites that often occur in clusters.

The common occurrence of D. notatus in Svalbard appears to be mainly due to the wide ecological tolerance of this species, its high reproductive potential, the large proportion of juveniles, and the presence of long setae in the juvenile stages. This species is a generalist, but prefers vegetation communities with a low diversity of plant species, a characteristic of most vegetation classes in Svalbard (Coulson et al. 2003; Seniczak and Plichta 1978; Seniczak et al. 2014a). Female D. notatus were more abundant than the males and the majority were gravid resulting in a high proportion of juveniles in the populations. A closely related species, Diapterobates humeralis (Hermann, 1804), is abundant and common on the lichen-covered bark of young Scots pine (Pinus sylvestris) and larch (Larix decidua) in Poland. In the spring and early summer, the juveniles dominate the adults, achieving proportions of 79.6% and 74.3% of all individuals, respectively (Seniczak 1974), and the juveniles also have long setae, as those of D. notatus.

Relatively high density of D. notatus in the open Dryas/Carex rupestris communities was achieved partly due to the juveniles, which were the most abundant there, especially the protonymphs and deutonymphs. However, the percent of juveniles in the population was not high and may suggest the stability of the population, compared to the less abundant populations of D. notatus with higher proportions of juveniles which often indicates the growth phase of a population. Generally, a large proportion of juveniles in the population of D. notatus may enhance dispersal and colonization success in the Arctic ecosystem since the juveniles are poorly sclerotized, weigh less than the adults, and have long prodorsal and gastronotal setae that enlarge their body dimensions and enhance passive dispersal by the wind. Transport of juveniles and adults of D. notatus by birds (Krivolutsky and Lebedeva 2004) and insects (Coulson 2009) may also be important for in potential long-distance dispersal distribution of this species in Svalbard. However, the juveniles are believed to be more sensitive to adverse environmental conditions than adults (Seniczak 1978; Seniczak and Plichta 1978; Seniczak et al. 2014a). Mite mortality is also a result of predation; for example, from gamasid mites (Karg 1993; Ruf and Beck 2005), especially on the juvenile stages. These predators were present in the samples, and while they were not investigated here, may typically occur at approximately 1000 ind m−2 (Gwiazdowicz and Coulson 2011).

The sex ratio of oribatid mites varies between the species and the number of eggs carried by gravid females generally depends on body size and season. For example, the mean sex ratio of D. notatus was 1:0.7 and in the polar summer the medium-sized gravid females (about 675 μm) usually carried 6–8 eggs, suggesting the high developmental potential of this species. In a slightly smaller species, for example, Mycobates sarekensis (Trägårdh, 1910), males usually dominated (mean sex ratio 1:1.3) and gravid females usually carried 6–7 eggs (Seniczak et al. 2015b). In abundant populations of the distinctly smaller Arctic species, Oribatella arctica (Thor 1930), the males also dominated (sex ratio 1:0.9–1:1.8), but the proportion of gravid females was low (14%), and these (body length about 440 μm) usually carried only one or two large eggs (Seniczak et al. 2015a).

By contrast, S. paludicola was not common, which is consistent with the observations of Thor (1930). This species was the most abundant in the Arctic meadows at Reinsdyrflya which is rich in plant species, including Dupontia sp., and which is consistent with Behan-Pelletier (1985). The reduced dispersal of S. paludicola to other Arctic meadows and other vegetation classes than D. notatus may be partly the result of the morphology of its juveniles; they have shorter gastronotal setae and are less suitable for anemochory than D. notatus. Interestingly, S. paludicola successfully competed with D. notatus in one sample from Reinsdyrflya. This suggests that more studies on the population biology of this species are required to determine the factors restricting abundances in the other samples and vegetation classes.

In High Arctic Svalbard, the body size of D. notatus depends on the vegetation class being the largest in rich vegetation of Arctic meadows, and the smallest in the poorer vegetation of hygrophilous vegetation and wet moss tundra. In other species of Oribatida, the body size also depends on microhabitat. For example, on Olympus Mtn. (Greece) the largest adults of Eueremaeus oblongus (C. L. Koch, 1835) were recorded from moss patches and the smallest from dead wood (Seniczak et al. 2014b). In the Mediterranean region, the largest adults of Oppia denticulata (R. & G. Canestrini, 1882) were found in pine litter in Vodice (Croatia), and the smallest in pine litter in Bilbao (northern Spain) (Seniczak et al. 2013).

The question arises about the population consequences of the small body size of D. notatus. Smaller gravid females carry a lower number of eggs, but it does not appear to limit the density of a species. The small Arctic species O. arctica achieves distinctly greater densities in Svalbard (Seniczak et al. 2015a) than D. notatus. The smaller body size may also influence the vertical distribution in the soil profile, which partly depends on the soil pores and interstitial spaces that get smaller with the soil depth (Seniczak 1978). The lack of suitable soil pores and cavities for oribatid mites to inhabits was suggested by Coulson et al. (2015) as a possible explanation for the lack of oribatid mites in the imported soils at the abandoned mining settlement of Pyramided in Svalbard. Smaller individuals of D. notatus may penetrate deeper soil horizons and obtain access to additional resources, but this distribution limits their passive dispersal by the wind to surrounding areas. These examples indicate the complicated consequences of the smaller body size on the interactions between environment and mite.

In this project, two females of S. clavatosensillus were obtained from established Dryas heaths at Mosselbukta (north Spitsbergen). This is a new species record for Svalbard, and is especially interesting as it was recently described from Nepal (Ermilov et al. 2013) and its wider distribution is unknown. It is, therefore, difficult to explain the history of this species in Svalbard.