EDP Sciences logo
Submit your paper
Search
Menu
All issues No 418 (2017) Knowl. Manag. Aquat. Ecosyst., 418 (2017) 3 Full HTML
Open Access
Issue
Knowl. Manag. Aquat. Ecosyst.
Number 418, 2017
Article Number 3
Number of page(s) 20
DOI https://doi.org/10.1051/kmae/2016035
Published online 25 January 2017

© V. Pešić et al., Published by EDP Sciences 2017

Licence Creative CommonsThis is an Open Access article distributed under the terms of the Creative Commons Attribution License CC-BY-ND (http://creativecommons.org/licenses/by-nd/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. If you remix, transform, or build upon the material, you may not distribute the modified material.

1 Introduction

Odonates are a significant component of aquatic ecosystems and they are often used as bioindicators of ecosystem health (Oertli, 2008; Dolný et al., 2011). In addition, in aquatic ecosystems where they are the top predators, odonates can influence many other components as they have a wide range of interactions with different organisms (Knight et al., 2005). Odonates have a high dispersal capacity (Conrad et al., 1999) and inhabit a wide range of aquatic habitats including lentic and lotic water bodies. In general, the Odonate fauna of Montenegro has been sufficiently studied and so far 67 species have been recorded for the country (Gligorović et al., 2010; De Knijf et al., 2013; Buczyński et al., 2014).

Springs are considered as groundwater-surface ecotones which host specialized and often endemic or rare taxa (Di Sabatino et al., 2003; Cantonati et al., 2006; Savić et al., in press). The lower and usually stable temperature creates optimal conditions for stenothermal cold-water organisms and thus springs can highly contribute to the regional biodiversity of freshwater ecosystems (Boulton 2005; Pešić et al., 2016). Beside the temperature, the macroinvertebrate composition of springs is also influenced by various environmental factors such as hydrological conditions, physico-chemical parameters and substratum composition (Hahn, 2000; Ilmonen and Paasivirta, 2005; von Fumetti et al., 2006; von Fumetti and Nagel, 2012). The relative importance of different factors in aquatic and terrestrial habitats during larval and adult life stages of odonates, respectively, is poorly understood (Remsburg and Turner, 2009). Despite relatively long duration of the larval stage it seems that the relative distribution of Odonata is not primarily associated with this stage (Harabiš and Dolný, 2010). Generally, water velocity, temperature, shading, disturbance, type of substrate, trophy, aquatic vegetation (its spatial structure and abundance) and predation risk are considered to be the most important factors shaping assemblages of Odonata larvae (Buchwald, 1992; Buss et al., 2004; Johansson et al., 2006; McCauley, 2007; Strange et al., 2007; Buczyński, 2015). The regional distribution of Odonata seems to be mostly affected by dispersal at the adult stage (Hof et al., 2006) while the local distribution is probably mainly affected by interactions at the larva stage (McCauley, 2007).

Little is known about the ecology of odonates in springs, and with exception of some rare papers (e.g., Buczyński, 1999; Buczyński et al., 2003; Borisov, 2015), studies about the ecological factors that determine species distribution in these habitats are still lacking. In this study we postulated that Odonata assemblages in springs might be affected not only by factors acting at the level of individual spring (in this paper “habitat” factors), but also by factors acting in the terrestrial environment (“landscape” factors).

The present study aims to determine which assemblages of Odonata larvae occur in spring habitats in Montenegro. Further, we evaluate the impact of selected environmental and habitat factors on the spatial pattern of these assemblages and check what factors − local habitat factors affecting an individual spring, or landscape factors affecting the broader scale − determine the formation of Odonata larvae assemblages in karstic spring habitats.

2 Material and methods

Odonata larvae were collected from 91 springs (78 rheocrenes, 11 limnocrenes and 2 sublacustrine springs, see Appendix 1) located in the central part of Montenegro. The studied area belongs to the drainage basin of Lake Skadar, the largest lake in the Balkan Peninsula with a surface area that seasonally fluctuates between 370 and 600 km2. There are a number of temporary and permanent karstic springs, most of them rheocrenes and limnocrenes (for spring type definition see e.g., Gerecke and Di Sabatino, 2003). Some of springs are sublacustrine (called cryptodepressions or ‘okos’) and they occur along the shores of Lake Skadar; these springs issue from underwater dolines.

The sampling was done from 2009 to 2015. Odonata larvae were sampled with a small Surber sampler (10 × 10 cm = 0.01 m2, 350 µm mesh width). All samples were immediately preserved in 96% ethanol, and subsequently sorted and identified in the laboratory. The identification and counting of material collected was done on the last-instar larvae and pre-imagines.

At each site water temperature (in winter − Tw and in summer − Ts) and pH were measured with a pH-meter (HI 98127, accuracy 0.1). Three measurements were carried out and the median was used for each for further analysis. The springs were divided into four classes based on their size (SI): 1: <1 m2, 2: 1–5 m2, 3: ≥5–20 m2, 4: >20 m2. Water discharge was determined visually at each site in winter (Diw) and summer (Dis), and the springs were grouped into four classes: 1 (<1 L min−1), 2: (≥1 and <5 L min−1), 3: (≥5 and <25 L min−1), 4: (≤25 L min−1) according to von Fumetti et al. (2006).

The substrate types (anoxic mud − AM, clay − CL, sand − SA, gravel − GR, stones − ST, rocks − RO) present within the sites were categorized into four classes of frequency based on the percentage cover (von Fumetti et al., 2006): 0: 0%; 1: 1–25%; 2: 26–50%; 3: 51–75%; 4: 76–100%. The percentage cover of the aquatic vegetation present within the site (macrophyte − MC, algae − ALG, and mosses − MS) was categorized into four classes: 0: 0%; 1: 1–25%; 2: 26–50%; 3: 51–75%; 4: 76–100%. In total 16 parameters of physicochemical characteristics (discharge, spring size, temperature and pH), substrate composition, aquatic vegetation, spring permanence (CO) and direct anthropogenic impact (AI) on the springs were analysed at the habitat level.

Analysis of the landscape was based on buffer zones marked out as a circle around each sampling site with a radius of 50 m from the waterline of each studied spring. Different land use and land cover types were categorized into four classes of frequency based on the percentage of cover: 0: 0%; 1: 1–25%; 2: 26–50%; 3: 51–75%; 4: 76–100%. The following basic parameters of the landscape were measured and described for each sampling site: (1) altitude (AL), (2) distance to nearby bodies of water (DWB), (3) flooding area (FL) and (4) the surface area of the patches of different types present in the landscape: forest (FO), riparian vegetation (RI), meadows (ME), built-up area (BA), agricultural land (AG) and karst vegetation (KA). In total 9 parameters were analysed at the landscape level.

Statistical analyses were performed using PRIMER 7.0 (Clarke and Gorley, 2015), MVSP v3.21 (Kovach, 2007) and SPSS 19. Dominance (D) and frequency (F) indices were used to evaluate species. The distribution of data was tested with the Kolmogorov–Smirnov test. For cluster analysis based on environmental data, centered and standardized environmental data was classified by the Euclidean distance similarity index. For classification of biotic samples, the Bray–Curtis similarity index on square root transformed data was used. PCA was undertaken on centered and standardized environmental data of the site groups used in the previous cluster analysis. SIMPER analysis was performed to test differences within the faunal composition of clusters A, B, C and D, and I, II and III clusters of sites distinguished by habitat (H) and landscape (L) classification.

Using the SIMPER procedure, dissimilarities between, and similarities within the above-mentioned groups can be explained with individual species and the composition of Odonata assemblages. A repeated measured analysis of variance (ANOVA) and multiple range tests (Fisher's least significant difference (LSD) procedure) were applied to determine the significance of differences in species richness between groups of springs. CCA (ter Braak, 1986) was applied to test the influence of environmental variables on the assemblages investigated. We first performed a forward selection of environmental variables (Legendre et al., 2011). Also, we used the unrestricted Monte Carlo permutation test (ter Braak and Wiertz, 1994) to test the null hypothesis that the selected variables are unrelated to Odonata assemblages in investigated springs.

3 Results

3.1 General characteristics of Odonata fauna

We collected 2979 Odonata larvae during this study. In the material collected were 44 species belonging to 25 genera (Tab. 1). From one to 23 taxa were found per spring. The maximum α-diversity (23 and 22 species respectively) was found in S5 and S6–two large sublacustrine springs at Lake Skadar. In contrast, the lowest α-diversity (single species) was found in S47. The highest frequency was noted for Cordulegaster bidentata (present in 54 springs) and the most abundant species in the material was Cordulegaster bidentata (269 specimens).

The qualitatively richest fauna was found in sublacustrine springs (22.5 ± 0.71 species), followed by limnocrenes (14.8 ± 4.74), while that of the rheocrenes was less rich (6.51 ± 3.89). Using the one way ANOVA, it was confirmed that species richness (F = 39.45, p < 0.001) in different types of springs differed significantly. The LSD analysis revealed significant difference between sublacustrine springs and limnocrenes (p = 0.013) and rheocrenes (p < 0.001) as well between rheocrenes and limnocrenes (p < 0.001).

Table 1

Quantitative occurrence of Odonata in springs in central part of Montenegro. Abbr. − abbreviation, n − number of specimens, D − dominance (%), N − number of springs in which species occurred, F – frequency (%).

3.2 Faunistic similarity between springs

Figure 1 presents a dendrogram grouping springs based on faunistic similarity. The springs were clearly grouped in four clusters. Faunistic similarity between springs ranged from 2.91% to 92.31%. The first cluster includes limnocrenes and small rheocrenes springs at the higher altitude, characterized by their high content of anoxic mud and absence of riparian vegetation. The springs of the second cluster includes small rheocrenes at lower altitudes affected by drought (S11,16,34,38,66,68,82) or anthropogenic disturbance (S33,49,66,68).

Most springs belong to cluster C. This cluster includes most of rheocrenes at the lower and medium altitudes. The springs from this group manifest a wide variation in substrate composition (anoxic mud to limestone) and a greater distance from nearby water bodies. Cluster D consisted of two subclusters. The first one includes sublacustrine springs (S6–7) and limnocrenes (S1–3,7,13,36,44,60,74). The second one includes large lowland karstic springs (called “vrelo”, S4,22–23,41,44,46) and a small rheocrenes (S17,19–21,26–29,42,62) located on the river banks.

ANOVA showed significant difference (F = 13.63, p < 0.001) for the species richness between Odonata assemblages from site clusters A, B, C and D. The highest diversity reveal assemblage type C (12.12 ± 5.81), followed by assemblage type A (6.62 ± 2.92) and type D (6.23± 4.09). The lowest diversity reveal assemblage type B (4.1 ±1.60).

Appendix 2 presents taxa mostly associated with each of the site clusters and dissimilarity in the taxonomic composition between each of the clusters. It can be seen that the community groups separated on the basis of faunistic similarity are better defined, have a higher internal similarity and are more dissimilar to each other than the assemblages in the groups separated by habitat and landscape classification, respectively. Enallagma cyathigerum is a characteristic representative of type A assemblages, while Orthetrum brunneum is characteristic of type B assemblages. Calopteryx splendens and Cordulegaster bidentata are characteristic of springs from group C and D, respectively.

thumbnail Fig. 1

Bray–Curtis similarity of Odonata assemblages within investigated springs.

3.3 Habitat level

Figure 2A presents a dendrogram grouping springs based on habitat factors. Three clusters can be seen. Wilcoxon test (Z = −6.037, p < 0.001) revealed that the clusters of springs based on habitat characteristics were not consistent with those grouped according to faunistic similarity.

To recognize environmental patterns at the habitat level a PCA was undertaken. The first and second PCA axes explain 21.6% and 19.4% variation of the variables analyzed, respectively. The first PCA axis correlated negatively with the discharge (summer: −0.417; winter: −0.426) and the spring size (−0.367). The second PCA axis correlated positively with the percentage of anoxic mud (0.351) and clay (0.35) and correlated negatively with the percentage of stones (−0.353). The spring clusters were much less separated in the PCA plot than the clusters formed on the basis of landscape characteristics (Fig. 3).

Appendix 3 presents taxa mostly associated with each of the site clusters and dissimilarity in the taxonomic composition between each of the clusters. Calopteryx splendens is characteristic representative for the springs from cluster I(H), while Cordulegaster bidentata mostly contributes, to the assemblages in clusters II(H) and III(H).

The results of the CCA analysis summarize the main trends in the relationship between Odonata and habitat factors. The first and second axes explain 30.57% and 24.41% of species variation of the variables analyzed, respectively. The results of CCA analysis revealed that the habitat variables used in ordination explain 37.49% of the total variation in Odonata species. Of the 15 statistically significant parameters shaping the structure of the assemblages, the influence of the parameter “permanence” was the greatest (explaining 4.54% of variation). However, an analysis of the correlations showed that only some species were significantly associated with this factor − Calopteryx virgo correlated positively, while Lestes barbarus and Sympetrum sanguineum correlated negatively with spring permanence. The next parameter having the greatest influence on the formation of Odonata assemblages was “anthropogenic impact” (explaining 4.46% of variation). Figure 4 illustrating the CCA results the species which avoid anthropogenic influence are concentrated in the lower left-hand corner of the diagram. This variable correlated negatively with most species (with the strongest correlations for Calopteryx splendens, Platycnemis pennipes, Onychogomphus forcipatus, and Coenagrion puella), but correlated positively with Orthetrum brunneum, Cordulegaster bidentata and Somatochlora meridionalis.

Another group of species was associated with a set of co-occurring substrate factors − the percentage of the stones, rocks and the gravel (upper left-hand corner of the diagram). The most important factor among them, based on the number of statistically significant correlations between particular species, was the percentage of stones (19 species), while the percentage of rocks (8) and gravel (5) were less important. On other hand, species which showed affinities to springs with a higher content of anoxic mud, clay, algae and macrophyte were concentrated in lower right-hand corner of the diagram. Within this set of co-occurring factors the most important (based on the number of statistically significant correlation between particular species and this parameter) was the percentage of anoxic mud and algae (18 species), followed by the percentage of macrophytes (17), mosses and clay (14).

thumbnail Fig. 2

(A) Similarity distance between sites in clusters I, II, and III reflecting habitat (H) characteristics of investigated springs. (B) Similarity distance between sites in clusters I, II, and III reflecting landscape (L) characteristics of investigated springs.
thumbnail Fig. 3

(A and B) Results of PCA showing habitat (Fig. A) and landscape (Fig. B) characteristics of investigated springs respectively under habitat (H) and landscape (L) classification into clusters I, II and III. (C and D) Results of PCA showing habitat (Fig. C) and landscape (Fig. D) characteristics into faunistic defined clusters A, B, C and D.
thumbnail Fig. 4

CCA bioplot of species by habitat variables based on 91 investigated springs.

3.4 Landscape level

Figure 2C presents a dendrogram grouping springs based on landscape factors. Three clusters can be seen. Outliers encompass one spring (S48) which not belong to any of the clusters. Wilcoxon test (Z = −7.07, p < 0.001) revealed that the clusters of springs formed on the basis of landscape characteristics were not consistent with the grouping based on faunistic similarity.

To recognize the environmental patterns at the landscape level a PCA was undertaken. The first and second PCA axes explain 24.7% and 22.8% variation of the variables analyzed, respectively. The first PCA axis correlated negatively with parameter “meadow” (−0.606). The second PCA axis correlated positively with the percentage of riparian vegetation (0.461) and agricultural land (0.465) and correlated negatively with the altitude (−0.411). The results revealed that spring clusters were clearly separated in the PCA plot. The springs in cluster I(L) were more isolated due to the higher altitude. On other hand, springs from clusters II(L) and III(L) are more scattered over the biplot (more variable) but with a clear tendency for separation, with a stronger preference for riparian vegetation in the latter cluster.

Appendix 3 presents taxa mostly associated with each of the site clusters and dissimilarity in the taxonomic composition between each of the clusters. Cordulegaster bidentata is a characteristic of the assemblages of clusters I(L) and III(L), while Calopteryx splendens is typical of the springs of the cluster II(L).

The results of the CCA analysis summarize the main trend in the relationship between Odonata and the landscape factors. The first and second axes explain 47.34% and 18.92% variation of the variables analyzed, respectively. The results of CCA analysis revealed that the variables used in ordination explain 24.12% of the total variation in Odonata species. Of the 8 statistically significant parameters shaping the structure of the assemblages, the influence of altitude was greatest (explaining 6.2% of variation). Figure 5 illustrating the CCA results there is a large group of species which prefer lower altitude (upper and lower left-hand corner of the diagram).

Another group of species was associated with higher altitude (right-hand side of the diagram). This group includes Cordulia aenea, Aeshna juncea, Pyrrhosoma nymphula, Enallagma cyathigerum and Somatochlora mettalica. All these species showed positive, statistically significant correlations with altitude. In the lower left hand corner of the diagram there is a group of species whose distribution mainly depends on riparian vegetation. Most species correlated positively with this factor, with the strongest correlations for Brachytron pratense, Calopteryx splendens, Erythromma viridulum, Ischnura elegans, Aeshna affinis, Crocothemis erythraea, Platycnemis pennipes, Somatochlora flavomaculata, Trithemis annulata, Libellula fulva, and Orthetrum coerulescens. In the upper left-hand quarter of the diagram there are species whose distribution is determined by the presence of the built-up areas, with the strongest correlation for Cordulegaster bidentata and to a lesser extent for Orthetrum brunneum and Somatochlora meridionalis.

thumbnail Fig. 5

CCA bioplot of species by landscape variables based on 91 investigated springs.

4 Discussion

In 91 springs situated in the central part of Montenegro, a total of 44 species of odonates were recorded. This total is over 60% of the total recorded for Montenegro (De Knijf et al., 2013; Buczyński et al., 2014). The highest numbers of species were caught in two sublacustrine springs. This can be explained by the larger surface area and the standing water body nature of these springs which is primarily induced by the spatial factor, i.e., the location of these water bodies within the lake. Studies on other taxa, such as aquatic Heteroptera (Gligorović et al., 2016) and aquatic gastropods (Pešić and Glöer, 2013) confirm that sublacustrine springs contain the most diverse assemblages. Similarly, limnocrene springs showed considerably higher Odonata diversity when compared to rheocrenes, indicating the importance of lentic habitats in the study area for maintaining the regional biodiversity of Odonata spring populations.

Our research on Odonata larvae assemblages in the karstic springs of the central Montenegro showed that environmental and faunistical classification may not be related. Similar differences in classification of the Dynaric karst springs based on the faunistical and environmental factors affecting the individual springs were pointed out by Płóciennik et al. (2016) and Gligorović et al. (2016).

In the dendrogram of faunistic similarities (Fig. 1) four clusters can be seen: cluster A groups sites from higher altitudes, cluster B includes springs affected by disturbance factors including drought but also a human influence, while cluster C groups aggregates most of the small rheocrenes at lower and medium altitudes. Cluster D is more diverse and encompasses limnocrenes and sublacustrine springs on one side, and a large karstic rheocrenes as well as small riparian springs on the other side. On the other hand, habitat factors divide spring sites into three groups, but no clear trends were observed. The PCA analysis revealed that the spring clusters formed on the basis of habitat parameters were much less differentiated than those clusters formed on the basis of landscape characteristics, suggesting that the impact of habitat factors is blurred by factors acting outside the level of individual springs. The research on karstic springs proves that Odonata communities separated on the basis of faunistic similarity are much better defined and more dissimilar than springs, according, respectively, to their habitat and landscape characters. Faunistic dissimilarity between springs, even on a small spatial scale, was indicated for some groups such as chironomids (Płóciennik et al., 2016) and water bugs (Gligorović et al., 2016).

At the habitat level 19 factors were analyzed. The results of the CCA showed that the greatest influence on Odonata communities have the parameter of “permanence” followed by the “anthropogenic influence”. The latter parameter includes various kinds of anthropogenic modification of the spring habitat for use as drinking water sources, from spring boxes (concrete or wooden boxes placed over the spring to collect and store the water) to piped springs (spring water emerging from an artificial pipe). Both parameters should be considered as disturbance factors, which, as shown in some studies (for example Dmitrović et al., 2016; Płóciennik et al., 2016) may become important factors in shaping spring assemblages especially in karstic springs.

Several studies show that factors outside the aquatic environment have significant impact on spring assemblages. The landscape factors most used in similar studies of macroinvertebrate fauna in springs were altitude, the type and the structure of the landscape, and how it is used, and the proximity of nearby water bodies (Křoupalová et al., 2011; Dumnicka et al., 2007; Martin and Brunke, 2012; Pakulnicka et al., 2016; Stryjecki et al., 2016).

At the landscape level, 9 parameters were analyzed. According to the landscape characteristics, the springs were divided into three groups. The results of the CCA analysis revealed altitude as most important landscape factor. In our study the species most closely associated with the latter factor (the highest correlations with Cordulia aenea, Aeshna juncea, Pyrrhosoma nymphula, Enallagma cyathigerum and Somatochlora mettalica) prefer higher altitude (see De Knijf et al., 2013). According to Harabiš and Dolný (2010) species that prefer higher biotopes are generally scarcer because there is less availability of water biotopes at higher than at lower altitudes. This is in agreement with results of our study which revealed that most species prefer springs at lower altitudes.

The other landscape factors influencing Odonata larvae communities in our study were riparian vegetation and urban environment. Many researchers (e.g., Schindler et al., 2003; Remsburg and Turner, 2009; Buczyński, 2015; Oliveira-Junior et al., 2015) stressed the importance of riparian vegetation on the species richness and distribution of Odonata species. On the other hand, various studies emphasized the negative influence of human activities caused by urbanization and agricultural activities on the distribution and diversity of Odonata, which generally lead to a decrease and homogenization in the richness of Odonata species (Buczyński and Lewandowski, 2011; Willigalla and Fartmann, 2012; Harabiš and Dolný, 2012; Monteiro-Junior et al., 2015). Nevertheless, Goertzen and Suhling (2013) showed that moderately disturbed ruderal and pioneer ponds in residential and agricultural areas increase the number of Odonata species. However, in comparison with the species from springs which are more stenotopic, species from ponds are more eurythermic and thus less susceptible to changes in environment.

The results of the CCA suggest that the presence of Cordulegaster bidentata was related to built-up areas, so this species can be considered an indicator of disturbed habitat. The colonization of this species may be the results of the habitat preferences of this species for small watercourses (Buczyński et al., 2014) and the availability of suitable habitats affected by water regulation and deforestation. These changes, such as the introduction of concrete or wooden spring boxes, results in a reduction in the flow and the formation of large areas of still water, leading to the transformation of these environments into semi-lotic habitats suitable for colonization by the latter species.

Appendices

Appendix 1

General characteristics of the studied springs.

Appendix 2

Results of SIMPER analysis of Odonata assemblages of site groups A, B, C and D.

Appendix 3

Results of SIMPER analysis for Odonata assemblages of site groups I, II and III (habitat classification), and of site groups I, II, III and outlier (landscape classification).

Appendix 4

Physicochemical characteristics (discharge, spring size, temperature and pH), substrate composition and aquatic vegetation of 91 investigated springs.

Appendix 5

Landscape characteristics: distance to nearby bodies of water, flooding area ([+] present and [−] absent) and the surface area of the patches of different types present in the landscape of 91 investigated springs.

References

  • Borisov SN. 2015. Dragonflies (Odonata) of thermal springs in central Asia. Entomol Rev 95: 1203–1211. [CrossRef] [Google Scholar]
  • Boulton AJ. 2005. Chances and challenges in the conservation of groundwaters and their dependent ecosystems. Aquat Conserv 15: 319–323. [CrossRef] [Google Scholar]
  • Buchwald R. 1992. Vegetation and dragonfly fauna – characteristics and examples of biocenological field studies. Vegetatio 101: 99–107. [Google Scholar]
  • Buczyński P. 2015. Dragonflies (Odonata) of anthropogenic waters in middle-eastern Poland. Wydawnictwo Mantis, Olsztyn, 272 p. [Google Scholar]
  • Buczyński P. 1999. Dragonflies (Odonata) of springs areas in Poland − state of research and proposes of further study. In: Biesiadka E, Czachorowski S, eds. Springs in Poland, State of research, monitoring and protection. Wydawnictwo Wyższej Szkoły Pedagogicznej w Olsztynie, Olsztyn, pp. 31–36. [Google Scholar]
  • Buczyński P, Czachorowski S, Moroz M, Stryjecki R. 2003. Odonata, Trichoptera, Coleoptera, and Hydrachnidia of Springs in Kazimierski Landscape Park (Eastern Poland) and factors affecting the characters of these ecosystems. Suppl Acta Hydrobiol 5: 13–29. [Google Scholar]
  • Buczyński P, Lewandowski K. 2011. Dragonfly (Odonata) fauna of Olsztyn (Poland). In: Indykiewicz P, Jerzak L, Böhner J, Kavanagh B, eds. Urban fauna. Studies of animal biology, ecology and conservation in European cities. UTP Bydgoszcz, Bydgoszcz, pp. 107–117. [Google Scholar]
  • Buczyński P, Zawal A, Stępień E, Buczyńska E, Pešić V. 2014. Contribution to the knowledge of dragonflies (Odonata) of Montenegro, with the first record of Ophiogomphus cecilia (Fourcroy, 1785). Ann Univ M Curie-Skłodowska Sec C 68: 57–71. [Google Scholar]
  • Buss DF, Baptista DF, Nessimiam JL, Egler M. 2004. Substrate specificity, environmental degradation and disturbance structuring macroinvertebrate assemblages in neotropical streams. Hydrobiologia 518: 179–188. [CrossRef] [Google Scholar]
  • Cantonati M, Gerecke R, Bertuzzi E. 2006. Springs of the Alps – sensitive ecosystems to environmental change: from biodiversity assessments to long-term studies. Hydrobiologia 562: 59–96. [CrossRef] [Google Scholar]
  • Clarke KR, Gorley RN. 2015. PRIMER v7: User Manual/Tutorial. PRIMER-E, Plymouth, 296 p. [Google Scholar]
  • Conrad KF, Willson KH, Harvey IF, Thomas CJ, Sherratt TN. 1999. Dispersal characteristics of seven odonate species in an agricultural landscape. Ecography 22: 524–531. [CrossRef] [Google Scholar]
  • De Knijf G, Vanappelghem C, Demolder H. 2013. Odonata from Montenegro, with notes on taxonomy, regional diversity and conservation. Odonatologica 42: 1–29. [Google Scholar]
  • Di Sabatino A, Cicolani B, Gerecke R. 2003. Biodiversity and distribution of water mites (Acari, Hydrachnidia) in spring habitats. Freshw Biol 48: 2163–2173. [CrossRef] [Google Scholar]
  • Dmitrović D, Savić D, Pešić V. 2016. Discharge, substrate type and temperature as factors affecting gastropod assemblages in springs in northwestern Bosnia and Herzegovina. Arch Biol Sci Belgrade 68: 613–621. [CrossRef] [Google Scholar]
  • Dolný A, Bárta D, Lhota S, Rusdianto, Drozd P. 2011. Dragonflies (Odonata) in the Bornean rain forest as indicators of changes in biodiversity resulting from forest modification and destruction. Trop Zool 24: 63–86. [Google Scholar]
  • Dumnicka E, Galas J, Koperski P. 2007. Benthic Invertebrates in Karst Springs: Does Substratum or Location Define Communities? Int Rev Hydrobiol 92: 452–464. [CrossRef] [Google Scholar]
  • Gerecke R, Di Sabatino A. 2003. Water mites (Acari, Hydrachnellae) and spring typology in Sicily. Crunoecia 5: 35–41. [Google Scholar]
  • Gligorović B, Pešić V, Zeković A. 2010. Checklist of the Dragonflies of the Skadar Lake Area. Scr Sci Nat Podgor 1: 101–107. [Google Scholar]
  • Gligorović B, Savić A, Protić Lj, Pešić V. 2016. Ecological patterns of water bugs (Hemiptera: Heteroptera) assemblages in karst springs: a case study from central Montenegro. Oceanol Hydrobiol Stud 45: 554–563. [Google Scholar]
  • Goertzen D, Suhling F. 2013. Promoting dragonfly diversity in cities: major determinants and implications for urban pond design. J Insect Conserv 17: 399–409. [Google Scholar]
  • Hahn HJ. 2000. Studies on classifying of undisturbed springs in southwestern Germany by macrobenthic communities. Limnologica 30: 247–259. [CrossRef] [Google Scholar]
  • Harabiš F, Dolný A. 2010. Ecological factors determining the density-distribution of Central European dragonflies (Odonata). Eur J Entomol 107: 571–577. [CrossRef] [Google Scholar]
  • Harabiš F, Dolný A. 2012. Human altered ecosystems: suitable habitats as well as ecological traps for dragonflies (Odonata): the matter of scale. J Insect Conserv 16: 121–130. [CrossRef] [Google Scholar]
  • Hof C, Brändle M, Brandl R. 2006. Lentic odonates have larger and more northern ranges than lotic species. J Biogeogr 33: 63–70. [CrossRef] [Google Scholar]
  • Ilmonen J, Paasivirta L. 2005. Benthic macrocrustacean and insect assemblages in relation to spring habitat characteristics: patterns in abundance and diversity. Hydrobiologia 533: 99–113. [CrossRef] [Google Scholar]
  • Johansson F, Englund G, Brodin T, Gardfjell H. 2006. Species abundance models and patterns in dragonfly communities: effects of fish predators. Oikos 114: 27–36. [CrossRef] [Google Scholar]
  • Knight TM, McCoy MW, Chase JM, McCoy KA, Holt RD. 2005. Trophic cascades across ecosystems. Nature 437: 880–883. [CrossRef] [PubMed] [Google Scholar]
  • Kovach WL. 2007. MVSP – A Multivariate Statistical Package for Windows, ver. 3.21. Kovach Computing Services, Pentraeth, Wales, UK. [Google Scholar]
  • Křoupalová V, Bojková J, Schenková J, Pařil P, Horsák M. 2011. Small-scale distribution of aquatic macroinvertebrates in two spring fens with different groundwater chemistry. Int Rev Hydrobiol 96: 235–256. [CrossRef] [Google Scholar]
  • Legendre P, Oksanen J, ter Braak CJF. 2011. Testing the significance of canonical axes in redundancy analysis. Methods Ecol Evol 2: 269–277. [CrossRef] [Google Scholar]
  • Martin P, Brunke M. 2012. Faunal typology of lowland springs in northern Germany. Freshw Sci 31: 542–562. [CrossRef] [Google Scholar]
  • McCauley SJ. 2007. The role of local and regional processes in structuring larval dragonfly distributions across habitat gradients. Oikos 116: 121–133. [CrossRef] [Google Scholar]
  • Monteiro-Junior CS, Juen L, Hamada N. 2015. Analysis of urban impacts on aquatic habitats in the central Amazon basin: adult odonates as bioindicators of environmental quality. Ecol Indic 48: 303–311. [CrossRef] [Google Scholar]
  • Oertli B. 2008. The use of dragonflies in the assessment and monitoring of aquatic habitats. In: Córdoba-Aguilar A, ed. Dragonflies and damselflies: model organisms for ecological and evolutionary research. Oxford University Press, Oxford, pp. 79–95. [Google Scholar]
  • Oliveira-Junior JMB, Shimano Y, Gardner TA, Hughes RM, De Marco PJ, Juen L. 2015. Neotropical dragonflies (Insecta: Odonata) as indicators of ecological condition of small streams in the eastern Amazon. Austral Ecol 40: 733–744. [CrossRef] [Google Scholar]
  • Pakulnicka J, Buczyński P, Dąbkowski P, et al. 2016. Aquatic beetles (Coleoptera) in springs of a small lowland river: habitat factors vs. landscape factors. Knowl Manag Aquat Ecosyst 417: 29. [CrossRef] [EDP Sciences] [Google Scholar]
  • Pešić V, Glöer P. 2013. A new freshwater snail genus (Hydrobiidae, Gastropoda) from Montenegro, with a discussion on gastropod diversity and endemism in Skadar Lake. ZooKeys 281: 69–90. [CrossRef] [Google Scholar]
  • Pešić V, Dmitrović D, Savić A, von Fumetti S. 2016. Studies on eucrenal-hypocrenal zonation of springs along the river mainstream: a case study of a karst canyon in Bosnia and Herzegovina. Biologia 71: 809–817. [Google Scholar]
  • Płóciennik M, Dmitrović D, Pešić V, Gadawski P. 2016. Ecological patterns of Chironomidae assemblages in Dynaric karst springs. Knowl Manag Aquat Ecosyst 411: 11. [Google Scholar]
  • Remsburg AJ, Turner MG. 2009. Aquatic and terrestrial drives of dragonfly (Odonata) assemblages within and among north-temperate lakes. J North Am Benthol Soc 28: 44–56. [CrossRef] [Google Scholar]
  • Savić A, Dmitrović D, Pešić V. in press. Ephemeroptera, Plecoptera and Trichoptera assemblages of karst springs in relation to environmental factors: a case study in central Bosnia and Hercegovina. Turk J Zool. [Google Scholar]
  • Schindler M, Fesl C, Chovanec A. 2003. Dragonfly associations (Insecta: Odonata) in relation to habitat variables: a multivariate approach. Hydrobiologia 497: 169–180. [CrossRef] [Google Scholar]
  • Strange AM, Griffiths GH, Hine S, Young K, Holloway GJ. 2007. Habitat associations of the Small Red Damselfly (Ceriagrion tenellum) (De Villiers) in heathland in southern England (Zygoptera: Coenagrionidae). J Insect Conserv 11: 241–249. [CrossRef] [Google Scholar]
  • Stryjecki R, Zawal A, Stępień E, et al. 2016. Water mites (Acari, Hydrachnidia) of water bodies of the Krąpiel River valley: interactions in the spatial arrangement of a river valley. Limnology 17: 247–261. [CrossRef] [Google Scholar]
  • ter Braak CJF. 1986. Canonical Correspondence Analysis: a new eigenvector technique for multivariate direct gradient analysis. Ecology 67: 1167–1179. [CrossRef] [Google Scholar]
  • ter Braak CJF, Wiertz J. 1994. On the statistical analysis of vegetation change: a wetland affected by water extraction and soil acidification. J Veg Sci 5: 361–372. [CrossRef] [Google Scholar]
  • von Fumetti S, Nagel P. 2012. Discharge variability and its effect on faunistic assemblages in springs. Freshw Sci 31: 647–656. [CrossRef] [Google Scholar]
  • von Fumetti S, Nagel P, Scheifhacken N, Baltes B. 2006. Factors governing macrozoobenthic assemblages in perennial springs in north-western Switzerland. Hydrobiologia 568: 467–475. [CrossRef] [Google Scholar]
  • Willigalla C, Fartmann T. 2012. Patterns in the diversity of dragonflies (Odonata) in cities across Central Europe. Eur J Entomol 109: 235–245. [CrossRef] [Google Scholar]

Cite this article as: Pešić V, Gligorović B, Savić A, Buczyński P. 2017. Ecological patterns of Odonata assemblages in karst springs in central Montenegro. Knowl. Manag. Aquat. Ecosyst., 418, 3.

All Tables

Table 1

Quantitative occurrence of Odonata in springs in central part of Montenegro. Abbr. − abbreviation, n − number of specimens, D − dominance (%), N − number of springs in which species occurred, F – frequency (%).

In the text
Appendix 1

General characteristics of the studied springs.

In the text
Appendix 2

Results of SIMPER analysis of Odonata assemblages of site groups A, B, C and D.

In the text
Appendix 3

Results of SIMPER analysis for Odonata assemblages of site groups I, II and III (habitat classification), and of site groups I, II, III and outlier (landscape classification).

In the text
Appendix 4

Physicochemical characteristics (discharge, spring size, temperature and pH), substrate composition and aquatic vegetation of 91 investigated springs.

In the text
Appendix 5

Landscape characteristics: distance to nearby bodies of water, flooding area ([+] present and [−] absent) and the surface area of the patches of different types present in the landscape of 91 investigated springs.

In the text

All Figures

thumbnail Fig. 1

Bray–Curtis similarity of Odonata assemblages within investigated springs.
In the text
thumbnail Fig. 2

(A) Similarity distance between sites in clusters I, II, and III reflecting habitat (H) characteristics of investigated springs. (B) Similarity distance between sites in clusters I, II, and III reflecting landscape (L) characteristics of investigated springs.
In the text
thumbnail Fig. 3

(A and B) Results of PCA showing habitat (Fig. A) and landscape (Fig. B) characteristics of investigated springs respectively under habitat (H) and landscape (L) classification into clusters I, II and III. (C and D) Results of PCA showing habitat (Fig. C) and landscape (Fig. D) characteristics into faunistic defined clusters A, B, C and D.
In the text
thumbnail Fig. 4

CCA bioplot of species by habitat variables based on 91 investigated springs.
In the text
thumbnail Fig. 5

CCA bioplot of species by landscape variables based on 91 investigated springs.
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.