Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Does littoral substrate affect macroinvertebrate assemblages in Mediterranean lakes?

Does littoral substrate affect macroinvertebrate assemblages in Mediterranean lakes? Aquat Ecol https://doi.org/10.1007/s10452-023-10037-7 Does littoral substrate affect macroinvertebrate assemblages in Mediterranean lakes? Efpraxia Mavromati  · Dimitra Kemitzoglou · Vasiliki Tsiaoussi Received: 3 February 2023 / Accepted: 25 May 2023 © The Author(s) 2023 Abstract The objective of this study was to inves- Keywords Lakes · Macroinvertebrates · Substrate tigate the effects of substrate type in macroinverte- type · Species richness · Community composition brate assemblages in Mediterranean lakes. Samplings have taken place in the littoral zone of 21 lakes in Greece, between 2015 and 2018. We compared ben- Introduction thic macroinvertebrate assemblages among three substrate types of their littoral zones; sandy, covered Freshwater biodiversity is worldwide declining at with macrophytes and stony substrate. Benthic mac- unprecedented rates (IPBES 2019). Factors under- roinvertebrate assemblages at sites with extended lying this biodiversity loss have been thoroughly macrophyte cover differed only slightly in composi- studied over the years and can be categorized in six tion and abundance from the ones found in stony and groups: hydrological alterations, habitat degradation sandy substrates. Coenagrionidae were indicative of and loss, pollution, overexploitation, invasive spe- sites covered with macrophytes and Oligochaeta and cies and climate change (Dudgeon et  al. 2006; Reid Erpobdellidae were representative of stony substrates. et  al. 2019; Arthington 2021). Most of these factors The type of substrate proved to be a statistically sig- are linked to human activities, which resulted in many nificant factor influencing the number of benthic restoration activities being carried out, focusing on macroinvertebrate taxa, the relative abundance of mitigating these anthropogenic impacts (Gething Oligochaeta and the relative abundance of Odonata. et al. 2020). However, in many cases there are knowl- In the context of designing site-adapted management edge gaps that hamper restoration efforts. For exam- measures, priority could be given to the conservation ple, although it is well known that lakes are subject and restoration of aquatic vegetation in lake litto- to morphological and hydrological modifications, the ral zones, which host rich macroinvertebrate assem- effects of these changes have  been less understood blages with abundant taxa of Odonata. (Poikane et al. 2020a). In Europe, we have increased our knowledge of freshwater ecosystems structure and function over the years, owing mainly to the implementation of Communicated by Télesphore Sime-Ngando. the Water Framework Directive (WFD). In this con- E. Mavromati (*) · D. Kemitzoglou · V. Tsiaoussi  text, benthic macroinvertebrates have been  among The Goulandris Natural History Museum, Greek Biotope/ the most widely used biological quality elements for Wetland Centre, 14th km Thessaloniki - Mihaniona, ecological assessments purposes especially in rivers 57001 Thermi, Greece and lakes, as reported in the 2018 EC Intercalibration e-mail: emavromati@ekby.gr Vol.: (0123456789) 1 3 Aquat Ecol Decision (European Commission 2018). Recently, their confounding effects (Solimini et  al. 2006). As they have become more popular in bio-assessment a rule of thumb, waterbodies in good ecological sta- methods as they have been proved sensitive not only tus could support a variety of species-indicators for to eutrophication and acidification but also to mor - undisturbed waters with high abundance values. phological changes and general degradation (Urbanič Except from their response to human stressors, et  al. 2012; Poikane et  al. 2016, 2020b; Mavromati freshwater macroinvertebrate taxa vary in response et al. 2021). Ntislidou et al. (2021) characterize them to habitat heterogeneity and substrate type. Gener- as “ecosystem engineers” playing an important role to ally, high habitat heterogeneity and coarser substrates aquatic ecosystem services as they participate in vari- present great invertebrate diversity and abundance ous biogeochemical processes. In the Mediterranean (Zenker and Baier 2009). According to Graça et  al. region, one such WFD-compliant assessment system (2015), coarse substrates are more complex than fine based on littoral benthic macroinvertebrates of Greek ones, as they retain coarser particulate organic matter lakes has recently been developed (HeLLBI; Mavro- and abundant with microalgae with higher biomass, mati et al. 2021). HeLLBI was designed and applied which can be used as food resources. Most studies to respond to anthropogenic impacts, and in particular accept the hypothesis that stony substrates present eutrophication and morphological changes (artificial higher richness and abundance of benthic taxa and in shoreline). particular abundance of collector–gatherers and filter- The distribution of macroinvertebrate communi- collectors (Pereira et al. 2017). ties in freshwater ecosystems is affected by both natu- Additionally, the presence of macrophytes in the ral and human factors (Dou et al. 2022). The natural littoral zone results in higher habitat heterogene- factors include all these environmental parameters ity. Dense macrophyte stands could alter inverte- that characterize a single site or water body, such as brate communities in numerous ways. Not only do temperature, water depth, dissolved oxygen, pΗ, con- they serve as food resources directly and indirectly ductivity, the spatial heterogeneity of habitats (Free through the growth of periphyton, but they can also et  al. 2009; Dou et al. 2022). Human activities affect access and deplete sediment nutrient loads (Waters macroinvertebrate communities either by nutrient and Giovanni 2002; Tolonen et al. 2003; Zenker and enrichment caused by land use activities or due to Baier 2009; Salmon et al. 2022). shoreline modification (Ntislidou et al. 2018; Bartels The overall objective of the study is to investigate et  al. 2021; Mavromati et  al. 2021). Differences in the effects of substrate types in benthic macroinver - biological communities across various environmental tebrate assemblages in Mediterranean lakes differing parameters have been extensively studied in streams in trophic status and other environmental parameters. and rivers but in lakes, especially in the Mediterra- In particular, we want to investigate the differences nean region, these studies are limited (Vinson and in benthic macroinvertebrate assemblages among Hawkins 1998). the sandy substrates, substrates covered with macro- Biological diversity can be studied several ways; phytes and stony substrates in Mediterranean natural it can be characterized by the diversity of species lakes, identify the indicator taxa characterizing each within sites and quantified by measuring the number substrate type and explore the role of macrophytes in of taxa or by diversity indices (Costa and Melo 2008). shaping these macroinvertebrate assemblages. Another way of measuring diversity is by describing the dissimilarities of biological assemblages between different environments (Costa and Melo 2008). Materials and methods As bio-indicators, freshwater macroinvertebrate taxa respond differently to stressors and they have Study area and sampling procedure the ability to incorporate the effects of the stressors they are exposed to, in combination and over time. As Benthic macroinvertebrates data were collected from they show spatial variation along different lake zones 21 lakes (Fig.  1), including two transboundary lakes, and depths, habitats and even lakes, finding out which Megali and Mikri Prespa, all belonging to the Greek natural factors affect their  assemblages is essential National Water Monitoring Network (Mavromati in order to separate them from human pressures and et al. 2021). Vol:. (1234567890) 1 3 Aquat Ecol Fig. 1 Twenty-one studied lakes of the Greek National Moni- votida, 13: Paralimni, 14: Petron, 15: Stymfalia, 16: Tricho- toring Network. 1: Amvrakia, 2: Cheimaditida, 3: Dystos, 4: nida, 17: Vegoritida, 18: Volvi, 19: Voulkaria, 20: Yliki, 21: Ismarida, 5: Kastoria, 6: Koroneia, 7: Kourna, 8: Lysimacheia, Zazari 9: Megali Prespa, 10: Mikri Prespa, 11: Ozeros, 12: Pam- The lakes included in this study belong to three In total, 97 littoral sampling sites have been sur- different natural lake types, according to the mixing veyed in the spring season during the 2015–2018 regime and depth gradient (Kagalou et  al. 2021). sampling campaign. Most lakes were sampled As most Mediterranean lakes, they face multiple once. Six lakes were sampled for two years and pressures including nutrient loading from point and one was sampled for three years. A nonparametric non-point sources, water abstraction and morpho- Kruskal–Wallis test was applied in this dataset and logical changes (Latinopoulos et  al. 2016). Moreo- showed that there were no differences among the ver, they seem to be affected by regional landscape sampling years (Mavromati et  al. 2021). The num- characteristics, in comparison with most cold tem- ber and location of the sampling sites for each lake perate and tropical lakes, by groundwater hydrology were selected according to lake size, habitat maps, and by the Mediterranean climate (Alvarez Cobelas and land use data of the lakes and their catchment et al. 2005; Mavromati et al. 2018). areas. Vol.: (0123456789) 1 3 Aquat Ecol Samples were collected using a semi-quantitative Abundance of Ephemeroptera, Plecoptera, Trichop- approach, which consists of a three-minute kick/ tera, Coleoptera, Bivalvia and Odonata (%EPTCBO), sweep with a standard hand net (500 μm mesh size), Simpson’s Diversity Index (Simpson) and Shannon at the littoral zone of each lake (up to 1.2 m depth of Diversity Index (Shannon). water). This particular approach was selected to cover potential effects of lakeshore modifications and it was conducted during the development of the littoral mac- Statistical analysis roinvertebrate assessment method (Mavromati et  al. 2021). At each sampling site, the cover with aquatic The  nonparametric Kruskal–Wallis  test was applied macrophytes (e.g. Phragmites australis, Potamogeton to examine statistically significant differences of mac- sp.) was recorded as a percentage. Furthermore, vis- roinvertebrate community indicators between differ - ual assessments of substrate composition were made ent substrate types (stony substrate, macrophytes and based on the predominant substratum size using the sandy substrate), followed by Wilcoxon signed rank size categories given in the Wentworth scale (Went- pairwise tests for the statistically significant indica- worth 1922). The substrate composition was further tors. Nonparametric tests were chosen, as the data grouped into two categories: sandy substrate (< 2 mm did not meet the assumptions of parametric tests. diameter) and stony substrate (> 2  mm diameter). Boxplots were used to show distributions of numeric As a result, each site sampling site was categorised values of statistically significant biological commu- into one of the three substrate types: stony substrate nity indicators. Boxplots were prepared in R using the (> 2  mm diameter), macrophyte and sandy substrate ggplot2 function (R Core Team 2018). (< 2 mm diameter). Non-metric Multidimensional Scaling (NMDS) Sieving was carried out on site; sorting, identifica- is an ordination technique, which enables complex tion and counting were carried out in the laboratory, multivariate data to be visualised in two dimensions. and the samples were preserved in vials containing This technique was employed to check for differences 70% ethanol. The littoral invertebrate fauna was iden- in benthic macroinvertebrate assemblages with sam- tified to family level, except oligochaetes, which were ples being a priori grouped by substrate type. NMDS identified as a subclass. Finally, the total number of centroids were calculated as the centre points of all individuals of each taxon was recorded and their rela- replicates for each sampling method in multidimen- tive abundance was calculated. sional space, as Gething et  al. (2020) suggested, using Bray–Curtis similarity coefficients on fourth Indicators of benthic macroinvertebrate assemblages root-transformed data. According to Anderson et  al. (2008), data transformation has been recommended Several indicators of benthic macroinvertebrate as a way to reduce the contribution of highly abun- assemblages (16 in total) were calculated from the dant species in relation to less abundant ones in the whole dataset: Number of Taxa (Taxa), Number of calculation of Bray–Curtis measure; when the trans- EPT taxa (EPT), Relative Abundance of Mollusca formation is severe (e.g. fourth root), rare species will (%Mollusca), Relative Abundance of Chironomi- have higher contribution to the analysis (Clarke et al. dae (%Chironomidae), Relative Abundance of Oli- 2014). This test statistic calculates a pseudo-F value, gochaeta (%Oligochaeta), Relative Abundance of similar to the F value in ANOVA. Larger pseudo-F Odonata (%Odonata), Relative Abundance of Bival- values indicate more pronounced group separation; via (%Bivalvia), Relative Abundance of Gastropoda however, its significance is usually of more interest (%Gastropoda), Relative Abundance of Ephemer- than its magnitude. The ordination diagram was also optera, Plecoptera and Trichoptera (%EPT), Rela- enhanced with the convex polygons around each cen- tive Abundance of Ephemeroptera, Coleoptera and troid (Costa and Melo 2008). Differences in the ben- Odonata (%ECO), Relative Abundance of Ephemer- thic fauna between the three different substrate types optera, Trichoptera and Odonata (%ETO), Relative were statistically tested via a permutational multi- Abundance of Ephemeroptera, Plecoptera and Odo- variate analysis of variance (PERMANOVA) using nata (%EPO), Relative Abundance of Ephemerop- the adonis2 function in vegan (Oksanen et  al. 2012; tera, Plecoptera and Coleoptera (%EPC), Relative Arbizu 2020; Gething et al. 2020). Vol:. (1234567890) 1 3 Aquat Ecol In order to analyse the multivariate homogeneity Results of group dispersions (variances) of macroinverte- brates assemblages and characterise their variability, The dataset used in this study included 77 taxa the betadisper function in vegan package was calcu- (together with Oligochaete subclass). The taxonomic lated based on Bray–Curtis distances (Oksanen et al. groups with higher numbers of taxa were Diptera 2012). Ordination plots were prepared and statistical (13), Gastropoda (10), Coleoptera (10) and Odonata analyses were performed in R version 3.5.3 (R Core (7). The highest number of taxa were observed in Team 2018). Lakes Lysimacheia and Vegoritida (23 and 21 taxa, The Similarity Percentages Analysis (SIMPER) respectively) and the lowest in Lake Koroneia (one was performed in Primer v7 software, to examine taxon). The lake with the highest number of EPT which taxa contributed most to the average similar- taxa was Vegoritida (6 taxa); on the other hand, Lake ity of sampling sites within each substrate type and Koroneia had no taxa belonging to orders Ephemer- the average dissimilarity between different types optera, Plecoptera and Trichoptera. of substrate (Clarke and Gorley 2015). Το identify According to SIMPER, the benthic fauna of our dominant taxa within each macroinvertebrate assem- littoral sites was dominated by five taxa: Chirono- blage for each substrate type, the indicator value midae, Oligochaeta, Corixidae, Gammaridae and index was applied using the indval function within Caenidae. The results of the analysis describing the the ladsv package and the indicators function within macroinvertebrates assemblages of the 97 sampling the indicspecies package (De Caceres et  al. 2016). sites are shown in Table 1. Chironomidae contributed The former is performed by a permutation test to most to the average similarity of all sites; especially assess the statistical significance of the association in sites with sandy substrate they almost represented between species and site groups, yielding a percent- 40% of all taxa. Benthic macroinvertebrate commu- age indicator value for each species (De Caceres and nities which occurred at sites with extended macro- Legendre 2009; Legendre and Legendre 2012). An phyte cover differed only slightly from the ones found indicator value of 0.25 was accepted as being ecologi- in stony and sandy substrates, in composition and cally relevant (Dufrene and Legendre 1997), and all abundance. Caenidae were mostly found in substrates significant indicators with a fidelity value below 0.25 dominated by macrophytes. The average dissimilar- were removed to exclude rare taxa (De Caceres et al. ity between stony substrates and sites covered with 2012). All analyses were performed with the use of macrophytes was 58.08%, between stony and sandy ladsv (Dufrene and Legendre 1997), indicspecies (De substrates was 58.55% and between sandy substrates Caceres and Legendre 2009), and tidyverse (Wickam and sites covered with macrophytes was 60.20% 2017) R packages in R environment version 3.5.3 (R (Table  2). The taxa that contributed to the average Core Team 2018). dissimilarity between the three substrate types were more or less the same (Gammaridae, Corixidae, Oli- gochaeta and Caenidae) but with different contribu- tions between the different pairs of substrate types. Indicator species analysis revealed two indicator taxa for sites with stony substrates (Oligochaeta and Table 1 Summary table Substrate type: Stony, n = 24 Substrate type: Macrophytes, Substrate type: Sandy, n = 35 of the SIMPER results for (Similarity: 46.88%) n = 38, (Similarity: 39.72%) (Similarity: 41.06%) benthic taxa contribution to similarity within each Taxa % %Cum Taxa % %Cum Taxa % %Cum substrate type (PRIMER 7 Software) Chironomidae 24.15 24.15 Chironomidae 29.43 29.43 Chironomidae 39.85 39.85 Oligochaeta 19.56 43.71 Corixidae 12.29 41.72 Corixidae 16.64 56.49 Gammaridae 17.26 60.98 Oligochaeta 11.90 53.63 Oligochaeta 10.29 66.78 Corixidae 13.78 74.76 Gammaridae 11.18 64.81 Gammaridae 10.06 76.84 Caenidae 8.36 73.17 Vol.: (0123456789) 1 3 Aquat Ecol Table 2 SIMPER results showing the contribution of taxa to meaning that this taxa combination occurred in less average dissimilarity between different pairs of substrate types than half of sites belonging to this group. Chironomi- (PRIMER 7 Software) dae and Coenagrionidae showed also high sensitivity Contribution (%) in sites covered with macrophytes. On the other hand, sites with sandy substrate had only one representative Stony and Macrophytes Average dissimilarity = 58.08 taxon (Atyidae) and no taxa combination; half of sites Gammaridae 9.12 belonging to this group included this particular com- Corixidae 6.84 bination (sensitivity = 0.49). Oligochaeta 6.43 The nonparametric Kruskal–Wallis test indi- Caenidae 5.77 cated no significant differences between sampling Stony and Sandy Average dissimilarity = 58.55 periods (p = 0.136; Mavromati et  al. 2021). The sta- Gammaridae 10.04 tistical analysis showed that the substrate type had Corixidae 8.15 a statistically significant effect on the Number of Oligochaeta 7.52 Taxa (H = 8.233, p = 0.016), the relative abundance Caenidae 6.43 of Oligochaeta (H = 9.646, p = 0.008) and the rela- Macrophytes and Sandy Average dissimilarity = 60.20 tive abundance of Odonata (H = 5.997, p = 0.050) Gammaridae 8.45 (Table 5). The pairwise comparison for taxa richness Corixidae 7.34 showed that the statistically different substrate types Caenidae 6.10 were stony substrate-macrophytes and macrophytes- Oligochaeta 6.06 sandy substrate (p < 0.05). The increase in taxa rich- ness was evident with greater substrate complexity, as macrophytes supported the greatest number of taxa Table 3 Indicator taxa for each substrate type while sandy substrates the fewest (Fig.  2). The rela- tive abundance of Odonata was higher in substrates Indicator taxa associated with macrophytes; whereas, the lowest Stony substrates Macrophytes percentages were recorded in stony substrates. The same results were shown in the pairwise comparisons Oligochaeta** Coenagrionidae* as statistically significant differences were observed Erpobdellidae** only between stony substrate and macrophytes The significance of the indicator value is shown: *p ≤ 0.05, (p < 0.05). On the other hand, the relative abundance **p ≤ 0.01 of Oligochaeta was higher in stony substrates, fol- lowed by macrophytes and finally by sandy sub- Erpobdellidae) and one indicator taxon (Coenagrio- strates. The pairwise comparison showed statistically nidae) for sites dominated by macrophytes (Table  3). significant differences between stones-macrophytes The best combinations of indicator taxa for each sub- and stones-sand. The remaining indicators of benthic strate group are shown in Table 4; the species combi- macroinvertebrate assemblages, which were tested in nation of sites with stony substrate (Gammaridae and our study showed no statistically significant  differ - Oligochaeta) showed high sensitivity; the majority ences between the three substrate types. of sites covered with stones included this particular According to PERMANOVA test, macroinver- combination. Their predictive power was rather low, tebrate assemblages were significantly different Table 4 Selected taxa Substrate type Selected taxa combination Predictive Sensitivity Square root combinations of indicator power of indicator taxa for each substrate value type. Prediction power, sensitivity and indicator Stony substrate Gammaridae + Oligochaeta 0.41 0.71 0.54 values (square rooted) of Macrophytes Chironomidae + Coenagrionidae 0.50 0.68 0.58 each taxa combination Sandy substrate Atyidae 0.52 0.49 0.50 Vol:. (1234567890) 1 3 Aquat Ecol Table 5 Nonparametric Kruskal–Wallis  test between several The analysis of multivariate homogeneity of group indicators of benthic macroinvertebrate assemblages and dif- dispersions (variances) revealed that the within-group ferent substrate types spread of macroinvertebrate communities did not dif- Metric Kruskal–Wallis H P value fer among substrate types (F = 2.3302, p = 0.095). 2, 94 This result clarifies the nature of multivariate effects Taxa 8.233 0.016 of macroinvertebrate assemblages and suggests that EPT 2.736 0.255 differences in benthic fauna were mainly due to dif- EPT (%) 0.524 0.769 ferences in centroid locations. ECO (%) 1.391 0.499 ETO (%) 0.704 0.703 EPO (%) 1.219 0.544 Discussion EPTCBO (%) 1.182 0.554 EPC (%) 0.612 0.736 The results presented in this paper allowed us to Oligochaeta (%) 9.646 0.008 assess the role of the littoral substrates in benthic Chironomidae (%) 2.154 0.341 macroinvertebrate fauna. We found that the studied Gastropoda (%) 5.247 0.073 substrate types supported to a certain degree distinct Bivalvia (%) 2.594 0.273 macroinvertebrate assemblages resulting from differ - Mollusca (%) 2.7933 0.255 ent levels of habitat complexity. However, the pres- Odonata (%) 5.997 0.050 ence of ubiquitous taxa across all sites was evident Simpson 0.329 0.848 in our results and was highlighted in two different Shannon 0.169 0.919 analyses, the SIMPER analysis and the nonparamet- Significant correlations (p < 0.05) are shown in bold ric Kruskal–Wallis test (Gething et al. 2020), provid- ing mixed results concerning community composition among substrate types (F2, 94 = 2.1876, p = 0.003). and diversity. No distinct clusters were observed in the ordination According to SIMPER results, Chironomidae space, indicating only slightly distinct benthic com- showed the highest contribution to the similarity of munities (Fig.  3). The plot of the NMDS analysis all three groups of substrates; its percentage to the revealed three polygons that defined the maximum overall community structure is rather high, espe- area of each group’s site scores in the two-dimen- cially in sandy substrates. It is well documented that sional ordination space, which seem to overlap at Chironomidae include several species with broad a certain extent. Macroinvertebrates occurring in environmental preferences including different sub- stony substrates seem to be a subset of those occur- strate types (Verdonschot 2006; Lencioni et al. 2018; ring in the other two substrate types. The hulls are Dorić et  al. 2020). Čerba et  al. (2022) concluded in fit to the raw data as they appear in the plot, which their study that substrate types significantly affected creates angular polygons. Despite the low degrees Chironomidae community composition and abun- of freedom, pseudo-F value of PERMANOVA anal- dance. A fauna list of Chironomidae larvae in main- ysis was large enough to reject the null hypothesis land Greece highlights exactly this: most species were of no differences in the centroid locations and/or indicative of distinct climatic, geological and hydro- the dispersion of groups between the three substrate chemical features (Płóciennik and Karaouzas 2013). types. The larger the pseudo-F value, the greater the On the other hand, Rossaro et al. (2014) refer to Chi- difference is supposed to be and is different from ronomidae taxa as opportunistic, being found in dif- the p value. Differences in the centroids locations of ferent habitats rather than restricted in a single habitat the three substrates are also evident in Fig. 4 where only. They use the term “preference” to describe their also the level of dispersion within each substrate ideal habitat type instead of “exclusivism” (Rossaro is shown. These three centroid locations represent et  al. 2014). The fauna of sandy substrate of lakes is the three different substrate types and according to mainly composed of Chironomidae taxa accompanied the boxplot, it is clear that the centroid locations of with a few other taxa, which explains their high con- sites covered with macrophytes and those of stony tribution to the similarity results of SIMPER analysis. substrates differ substantially. Ntislidou et  al. (2021) studied the macroinvertebrate Vol.: (0123456789) 1 3 Aquat Ecol Fig. 2 Boxplots of indica- tors of macroinvertebrate assemblages (Number of Taxa, Relative Abundance of Oligochaeta and Relative Abundance of Odonata) across different substrate types fauna of the profundal and sublittoral zones of three Caenidae discriminated the group of sites covered Greek eutrophic lakes and found that the assemblages with macrophytes from the other two groups and was of the muddy substrates of these zones were domi- found to be the taxon associated with the group dis- nated by Chironomidae and Oligochaeta taxa. similarities. This result agrees neither with the study of Pilotto et al. (2015) which associates Caenidae spe- cies with stony substrates and the presence of zebra mussels which they use as food resources, nor with the results of McGoff and Irvine (2009) who found a negative correlation between Caenis luctuosa, the macrophyte Percentage Volume Inhabited (PVI) and the extent of macrophytes lakewards. The percentage of dissimilarity was higher than 55% among all three groups of substrate types; none- theless the taxa that contributed mostly to the aver- age dissimilarity were the same with different per - centages of contribution at each group. The IndVal analysis (De Caceres and Legendre 2009) showed more clear results and revealed indicator species for stony substrates and for substrates covered with mac- rophytes. No indicator species were found for sandy substrates, probably because of their unique macroin- Fig. 3 Non-metric multidimensional scaling (NMDS) plot of vertebrate assemblage in comparison with the other benthic macroinvertebrate community composition between groups of sites and the relatively low number of taxa different substrate types in the studied lakes Vol:. (1234567890) 1 3 Aquat Ecol Our results showed that there was an increase in taxa richness in coarser substrates which agrees with the initial hypothesis of Graça et al. (2015). They ana- lysed the macroinvertebrate community in streams and characterized fine substrates poorer in terms of abundance of macroinvertebrates and taxa richness. Surface complexity seems to be positively correlated with diversity and abundance of benthic fauna espe- cially in studies concerning streams (Taniguchi and Tokeshi 2004; Barnes et al. 2013). On the other hand, our results showed that the relative abundance of Oli- gochaeta was greater in stony substrates in compari- son with the other two types, which is a surprising result according to the literature (Rieradevall et  al. 1999; Graca et al. 2015; Buendia et al. 2011). In par- ticular, Graça et al. (2015) argue that finer sediments are more appropriate for taxa like Chironomidae and Fig. 4 Boxplots of the level of dispersion within the three sub- Oligochaeta, which prefer spending time within sub- strate types which was calculated as the mean distance of each strate particles. The benthic fauna of sandy substrates sample point to the centroid of the respective substrate type. in our dataset was mainly composed of Chironomi- The centre line in the box displays the median and the mar- gins of the box specify the 25th and 75th percentile. Whisk- dae, resulting in low abundances of their remaining ers extend to the smallest (lower whisker) or the largest (upper taxa, in comparison with the other two groups of whisker) value within the range of 1.5 × interquartile range sites. Sychra et  al. (2010) suggested in their study that some species of Oligochaeta, such as Naididae (De Caceres et  al. 2012). To overcome this obstacle with phytal preferences are quite abundant in reed- the extended version of indicator species analysis beds near the shore. Oligochaeta and Chironomidae was used to give us indicator species also for sandy are considered to be generally tolerant organisms and substrates. The predictive values were quite simi- according to Mavromati et  al. (2021) they are the lar in all three groups of sites. Sites with stony sub- dominant taxa in sites with high proportion of artifi- strate highlighted Gammaridae and Oligochaeta as cial shoreline in poor and bad ecological quality sta- a combination of taxa with high sensitivity, which tus in Mediterranean lakes. is an estimation of the frequency of the families at Odonata larvae followed the opposite pattern and these sites. Studies indicate that species belonging to exhibited greater abundances in sites covered with Gammaridae show a clear preference in stony sub- macrophytes, followed by sandy and stony sub- strates, especially when they are colonized by Dre- strates. Waters and Giovanni (2002) associate Lesti- issena sp. (Stewart et al. 1998; Hesselschwerdt et al. dae (Odonata) with vascular macrophytes present in 2008). High sensitivity of the taxa combination of depositional habitats. The same pattern was evident Chironomidae and Coenagrionidae was also found in in Graça et  al. (2015) where it is argued that Odo- the groups of sites covered with macrophytes. Pilotto nata and Trichoptera inhabit coarser, well sorted et al. (2015) suggested that species of Coenagrionidae substrates. The links between aquatic vegetation and such as Ischnura elegans prefer sites with submerged macroinvertebrate abundance and diversity are evi- macrophytes and natural littoral habitats. dent in the majority of studies but the results are still Differences in community composition and diver - confounding with the type of sediment that aquatic sity were evident across different substrate types; macrophytes need for growth (Waters and Giovanni Gething et  al. (2020) characterize substrate type as 2002; Zenker and Baier 2009). It is crucial to check surrogates for increasing habitat complexity. In our in littoral zones which one is actually affecting mac- results, sites covered with macrophytes supported roinvertebrate assemblages; macrophytes or the sedi- greater number of taxa followed by sites with stony ment that they rely on? Another study focused only substrates and finally the ones covered with sand. in Chironomidae community composition, concluded Vol.: (0123456789) 1 3 Aquat Ecol that sites covered with macrophytes provided greater in order to be able to draw conclusions without the availability of food resources, microhabitats to confounding effects. inhabit and shelter from predators (Čerba et al. 2022). It is crucial to acknowledge that some macroinver- McGoff and Irvine (2009) associated positively lit- tebrate taxa are depending on certain substrate types; toral macroinvertebrate abundance with both mac- this can be useful when developing and implement- rophyte PVI (Percentage Volume Inhabited)  and ing site-adapted conservation and restoration meas- the extent of macrophytes lakewards in their study ures (Brauns et  al. 2007). In this context, priority is between Lake Habitat Quality Assessment and mac- advised to be placed to the conservation and/or res- roinvertebrate community structure. toration of the natural littoral habitats that host rich Overall, the substrate type seems to be a primary macroinvertebrate assemblages. The need to conserve factor affecting macroinvertebrate assemblages (along and were appropriate restore aquatic vegetation is with other natural factors such as lake size and depth) obvious as our results suggested that sites dominated (Timm and Möls 2012). We recorded this spatially with macrophytes exhibit the greatest number of taxa variability of benthic fauna but our results were rather and the most diverse group of Odonata. Controlling weak. Statistically speaking, the distinct macroinver- the extent and frequency of vegetation cutting and tebrate assemblages were a result of differences in dredging could lead to habitat heterogeneity resulting location of their centroids and not homogeneity of on the dispersal of the macroinvertebrate community dispersions but still we could consider other factors (Gething et al. 2020). affecting our dataset such as the dissimilarity meas- Acknowledgements The present study was conducted in ures (Bray–Curtis, Jaccard etc.) which could have the frame of the Greek National Water Monitoring Network, altered the interpretation of the results (Anderson according to the JMD 140384/2011, implemented by The Gou- et al. 2006). As Barnes et al. (2013) pointed out, spe- landris Natural History Museum, Greek Biotope/Wetland Cen- cies richness is mainly affected by complexity and not tre (EKBY). The network is supervised by the General Direc- torate for Waters of the Ministry of Environment and Energy. heterogeneity but still they advocate to take into con- The data used in this research come from Act MIS 5001204 sideration not only habitat structure but the processes financed by the European Union Cohesion Fund (Partner - involved in these relationships. The effect of sub- ship Agreement 2014–2020), and from Acts MIS 371010, strate types alone on species richness in lakes is less 371138, 371140, 371145 financed by the European Regional Development Fund (National Strategic Reference Framework studied and the fact that we did not find any  strong 2007–2013). K. Argiriou, M. Bozatzidou, G. Nakas, and V. relationships could be attributed to the lack of  more Navrozidou contributed to sorting and taxa identification. H. detailed categories (macrophyte cover, substrate size). Hadjicharalambous contributed to data analysis. EKBY’s per- Macrophytes seem to play an important role in regu- sonnel conducted samplings and contributed to data analysis. lating the balance of all trophic relationships occur- Data availability The dataset generated during and analysed ring in the littoral zone of lakes. Tolonen et al. (2003) during the current study are available from the corresponding suggested that in an oligo-mesotrophic lake with a author on reasonable request. well-established macrophyte zone, the composition Declarations and size of zoobenthos differed significantly along the gradient of vegetation density horizontally from Conflict of interest The authors have no competing interests shore to open water. On the other hand, they sug- to declare that are relevant to the content of this article. gested that shallow eutrophic lakes might be affected by other natural factors such as water level fluctuation Open Access This article is licensed under a Creative Com- mons Attribution 4.0 International License, which permits and nutrient loading. Sediment deposition can signifi- use, sharing, adaptation, distribution and reproduction in any cantly alter the composition of the bottom substrate, medium or format, as long as you give appropriate credit to the enriching it with organic matter. Jurca et  al. (2021) original author(s) and the source, provide a link to the Crea- pointed out that there was an absence of species with tive Commons licence, and indicate if changes were made. The images or other third party material in this article are included specific mesohabitat preferences in eutrophic lakes, in the article’s Creative Commons licence, unless indicated as there is a clear linkage of macrophyte presence and otherwise in a credit line to the material. If material is not species richness with eutrophication. All these factors included in the article’s Creative Commons licence and your should be taken into consideration when sampling for intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly benthic macroinvertebrates in littoral zones of lakes Vol:. (1234567890) 1 3 Aquat Ecol De Cáceres M, Legendre P (2009) Associations between spe- from the copyright holder. To view a copy of this licence, visit cies and groups of sites: indices and statistical inference. http:// creat iveco mmons. org/ licen ses/ by/4. 0/. Ecology 90(12):3566–3574 De Cáceres M, Legendre P, Wiser SK, Brotons L (2012) Using species combinations in indicator value analyses. Methods Ecol Evol 3(6):973–982. https:// doi. org/ 10. 1111/j. 2041- References 210X. 2012. 00246.x Dorić V, Koh M, Mihaljević Z (2020) First record of Nubensia Alvarez Cobelas M, Rojo C, Angeler DG (2005) Mediterra- nubens (Edwards, 1929) (Diptera: Chironomidae) from nean limnology: current status, gaps and the future. J Lim- Croatia. Natura Croatica: Periodicum Musei Historiae nol 64(1):13–29. https:// doi. org/ 10. 4081/ jlimn ol. 2005. 13 Naturalis Croatici 29(1):139–142 Anderson MJ, Ellingsen KE, McArdle BH (2006) Multivari- Dou Q, Du X, Cong Y, Wang L, Zhao C, Song D, Liu H, Huo ate dispersion as a measure of beta diversity. Ecol Lett T (2022) Influence of environmental variables on mac- 9(6):683–693. https:// doi. org/ 10. 1111/j. 1461- 0248. 2006. roinvertebrate community structure in Lianhuan Lake. 00926.x Ecol Evol 12:e8553. https:// doi. org/ 10. 1002/ ece3. 8553 Anderson MJ, Gorley RN, Clarke KR (2008) PERMANOVA+ Dudgeon D, Arthington AH, Gessner MO, Kawabata ZI, for PRIMER: guide to software and statistical methods. Knowler DJ, Lévêque C, Naiman RJ, Prieur-Richard AH, PRIMER-E, Plymouth Soto D, Stiassny MLJ, Sullivan CA (2006) Freshwater Arbizu PM (2020) Pairwiseadonis: Pairwise multilevel com- biodiversity: importance, threats, status and conservation parison using adonis. R package version 0.4 challenges. Biol Rev 81:163–182. https:// doi. org/ 10. 1017/ Arthington AH (2021) Grand challenges to support the fresh- S1464 79310 50069 50 water biodiversity emergency recovery plan. Front Envi- Dufrene M, Legendre P (1997) Species assemblages and ron Sci 9:664313. https:// doi. org/ 10. 3389/ fenvs. 2021. indicator species: the need for a flexible asymmetrical approach. Ecol Monogr 67:345–366 Barnes JB, Vaughan IP, Ormerod SJ (2013) Reappraising the European Commission (2018) Commission decision (EU) effects of habitat structure on river macroinvertebrates. 2018/229 of 12 February 2018 establishing, pursuant to Freshw Biol 58:2154–2167. https:// doi. org/ 10. 1111/ fwb. directive 2000/60/EC of the European Parliament and of the Council, the values of the member state monitoring Bartels A, Berninger UG, Hohenberger F, Wickham S, Peter- system classifications as a result of the intercalibration mann JS (2021) Species data of littoral macroinvertebrate exercise and repealing commission decision 2013/480/ communities of alpine lakes along an elevational gradi- EU. Off J Eur Communities L47:1–91 ent (Hohe Tauern National Park, Austria). PLoS ONE Free G, Solimini AG, Rossaro B, Marziali L, Giacchini R, 16(11):e0255619. https:// doi. org/ 10. 1371/ journ al. pone. Paracchini B, Ghiani M, Vaccaro S, Gawlik BM, Fresner 02556 19 R, Santner G, Schonhuber M, Cardoso AC (2009) Model- Brauns M, Garcia XF, Walz N, Pusch MT (2007) Effects of ling Lake macroinvertebrate species in the shallow sublit- human shoreline development on littoral macroinver- toral: relative roles of habitat, lake morphology, aquatic tebrates in lowland lakes. J Appl Ecol 44:1138–1144. chemistry and sediment composition. Hydrobiologia https:// doi. org/ 10. 1111/j. 1365- 2664. 2007. 01376.x 633:123–136. https:// doi. org/ 10. 1007/ s10750- 009- 9869-7 Buendia C, Gibbins CN, Vericat D, Lopez-Tarazon J, Batalla Gething KJ, Ripley MC, Mathers KL, Chadd RP, Wood PJ RJ (2011) Influence of naturally high fine sediment loads (2020) The influence of substrate type on macroinverte- on aquatic insect larvae in a montane river. Scottish Geogr brate assemblages within agricultural drainage ditches. J 127(4):315–334. https:// doi. org/ 10. 1080/ 14702 541. Hydrobiologia 847:4273–4428. https:// doi. org/ 10. 1007/ 2012. 670006 s10750- 020- 04416-6 De Caceres M, Jansen F, De Caceres, MM (2016). Package Graça M, Ferreira W, Firmiano K, França J, Callisto M (2015) ‘indicspecies’. Indicators. 8(1) Macroinvertebrate identity, not diversity, differed across Čerba D, Koh M, Vlaičević B, TurkovićČakalić I, Milošević patches differing in substrate particle size and leaf litter D, StojkovićPiperac M (2022) Diversity of periphytic packs in low order, tropical Atlantic forest streams. Lim- chironomidae on different substrate types in a floodplain netica 29:29–40. https:// doi. org/ 10. 23818/ limn. 34. 03 aquatic ecosystem. Diversity 14(4):264. https:// doi. org/ 10. Hesselschwerdt J, Necker J, Wantzen K (2008) Gammarids in 3390/ d1404 0264 lake constance: habitat segregation between the invasive Clarke KR, Gorley RN, Somerfield PJ, Warwick RM (2014) Dikerogammarusvillosus and the indigenous Gammarus- Change in marine communities: an approach to statis- roeselii. Fundam Appl Limnol 173:177–186. https:// doi. tical analysis and interpretation, 3rd edn. PRIMER-E, org/ 10. 1127/ 1863- 9135/ 2008/ 0173- 0177 Plymouth IPBES (2019) Global assessment report on biodiversity and Clarke KR, Gorley RN (2015) PRIMER v7: User Manual/ ecosystem services of the intergovernmental science- Tutorial. PRIMER-E, Plymouth policy platform on biodiversity and ecosystem services. Costa SS, Melo AS (2008) Beta diversity in stream macroin- IPBES secretariat, Bonn, Germany, p 1148. https:// doi. vertebrate assemblages: among-site and among-microhab- org/ 10. 5281/ zenodo. 38316 itat components. Hydrobiologia 598:131–138 Jurca T, Donohue L, Mcgoff E, Tunali S, Irvine K (2021) Unravelling the effect of multiple stressors on ecologi- cal structure of littoral lake macroinvertebrates. Int Rev Vol.: (0123456789) 1 3 Aquat Ecol methods, intercalibration and practical recommendations. Hydrobiol 106:202–212. https:// doi. org/ 10. 1002/ iroh. Sci Total Environ 543:123–134 20200 2048 Poikane S, Zohary T, Cantonati M (2020a) Assessing the eco- Kagalou I, Ntislidou C, Latinopoulos D, Kemitzoglou D, logical effects of hydromorphological pressures on Euro- Tsiaoussi V, Bobori DC (2021) Setting the phosphorus pean lakes. Inland Waters 10(2):241–255. https:// doi. org/ boundaries for Greek natural shallow and deep lakes for 10. 1080/ 20442 041. 2019. 16548 00 water framework directive compliance. Water 13(5):739. Poikane S, Herrero FS, Kelly MG, Borja A, Birk S, van de https:// doi. org/ 10. 3390/ w1305 0739 Bund W (2020b) European aquatic ecological assessment Latinopoulos D, Ntislidou C, Kagalou I (2016) Multipurpose methods: a critical review of their sensitivity to key pres- plans for the sustainability of the Greek Lakes: emphasis sures. Sci Total Environ 740:140075. https:// doi. org/ 10. on multiple stressors. Environ Process 3:589–602. https:// 1016/j. scito tenv. 2020. 140075 doi. org/ 10. 1007/ s40710- 016- 0152-4 R Core Team (2018) R: a language and environment for sta- Legendre P, Legendre L (2012) Numerical ecology. Elsevier tistical computing. r foundation for statistical computing, Lencioni V, Cranston PS, Makarchenko E (2018) Recent Vienna. Retrieved 22 Apr 2019, from https:// www. Rproj advances in the study of chironomidae: an overview. J ect. org Limnol 77:1–6. https:// doi. org/ 10. 4081/ jlimn ol. 2018. 1865 Reid AJ, Carlson AK, Creed IF, Eliason EJ, Gell PA, Johnson Mavromati E, Kemitzoglou D, Tsiaoussi V (2021) A new PTJ, Kidd KA, MacCormack TJ, Olden JD, Ormerod SJ, WFD—compliant littoral macroinvertebrate index for Smol JP, Taylor WW, Tockner K, Vermaire JC, Dudgeon monitoring and assessment of Mediterranean lakes D, Cooke SJ (2019) Emerging threats and persistent con- (HeLLBI). Environ Monit Assess 193:745. https:// doi. servation challenges for freshwater biodiversity. Biol Rev org/ 10. 1007/ s10661- 021- 09493-1 94:849–873. https:// doi. org/ 10. 1111/ brv. 12480 Mavromati E, Kagalou I, Kemitzoglou D, Apostolakis A, Rieradevall N, Bonada N, Prat N (1999) Substrate and depth Seferlis M, Tsiaoussi V (2018) Relationships among preferences of macroinvertebrates along a transect in a land use patterns, hydromorphological features and Pyrenean high mountain lake (Lake Redó, NE Spain). physicochemical parameters of surface waters: WFD Limnetica 17:127–134 lake monitoring in Greece. Environ Process 5:139–151 Rossaro B, Marziali L, Montagna M, Magoga G, Zaupa S, McGoff E, Irvine K (2009) A test of the association between Boggero A (2014) Factors controlling morphotaxa dis- lake habitat quality assessment and macroinvertebrate tributions of Diptera chironomidae in freshwaters. Water community structure. Aquatic Conserv: Mar Freshw 14(7):1014. https:// doi. org/ 10. 3390/ w1407 1014 Ecosyst 19:520–533. https:// doi. org/ 10. 1002/ aqc. 1024 Salmon Q, Colas F, Westrelin S, Dublon J, Baudoin JM Ntislidou C, Lazaridou M, Tsiaoussi V, Bobori DC (2018) (2022) Floating littoral zone (FLOLIZ): a solution to A new multimetric macroinvertebrate index for the eco- sustain macroinvertebrate communities in regulated logical assessment of Mediterranean lakes. Ecol Ind lakes? Ecol Eng 176:106509. https:// doi. org/ 10. 1016/j. 93:1020–1033. https:// doi. org/ 10. 1016/j. ecoli nd. 2018. ecole ng. 2021. 106509 05. 071 Solimini AG, Free G, Donohue I, Irvine K, Pusch M, Ros- Ntislidou C, Rossaro B, Lazaridou M, Bobori D (2021) What saro M, Sandin L, Cardoso AC (2006) Using benthic drives benthic macroinvertebrate dispersal in different macroinvertebrates to assess ecological status of lakes lake substrata? The case of three Mediterranean lakes. Aquat Ecol 55:1033–1050. https:// doi. org/ 10. 1007/ current knowledge and way forward to support WFD s10452- 021- 09880-3 implementation. Institute for Environment and Sustain- Oksanen J, Blanchet G, Kind R, Legendre P, Minchin P, ability, Luxembourg, p 49 O’Hara, RB Simpson G, Solymos P, Stevenes MHH, Stewart TW, Miner JG, Lowe RL (1998) Macroinvertebrate Wagner H (2012) Vegan: community ecology package. R communities on hard substrates in Western Lake Erie: package version 2.0-2. structuring effects of Dreissena. J Great Lakes Res Pereira TS, Pio JFG, Calor AR, Copatti CE (2017) Can the 24(4):868–879. https:// doi. org/ 10. 1016/ S0380- 1330(98) substrate influence the distribution and composition of 70868-8 benthic macroinvertebrates in streams in Northeastern Sychra I, Adámek Z, Petřivalská K (2010) Distribution and Brazil? Limnologica 63:27–30. https:// doi. org/ 10. 1016/j. diversity of littoral macroinvertebrates within extensive limno. 2016. 12. 003 reed beds of a lowland pond. Ann Limnol - Int J Lim Pilotto F, Bazzanti M, DiVito V, Frosali D, Livretti F, Mast- 46(4):281–289. https:// doi. org/ 10. 1051/ limn/ 20100 26 rantuono L, Pusch MT, Sena F, Solimini AG (2015) Rela- Taniguchi H, Tokeshi M (2004) Effects of habitat complex- tive impacts of morphological alteration to shorelines and ity on benthic assemblages in a variable environment. eutrophication on littoral macroinvertebrates in Mediter- Freshw Biol 49:1164–1178. https:// doi. org/ 10. 1111/j. ranean lakes. Freshw Sci 34:410–4221365- 2427. 2004. 01257.x Płóciennik M, Karaouzas I (2014) The Chironomidae (Diptera) Timm H, Mols T (2012) Littoral macroinvertebrates in fauna of Greece: ecological distributions and patterns, Estonian lowland lakes: the effects of habitat, season, taxalist and new records. Ann Limnol Int J Lim 50(1):19– eutrophication and land use on some metrics of biologi- 34. https:// doi. org/ 10. 1051/ limn/ 20130 66 cal quality. Fundam Appl Limnol 180:145–156. https:// Poikane S, Johnson RK, Sandin L, Schartau AK, Solimini AG, doi. org/ 10. 1111/j. 1365- 2427. 2004. 01257.x Urbanič G, Arbačiauskas K, Aroviita J, Gabriels W, Miler Tolonen K, Hämäläinen H, Holopainen IJ, Mikkonen K, O, Pusch MT, Timm H, Bohmer J (2016) Benthic mac- Karjalainen J (2003) Body size and substrate associa- roinvertebrates in lake ecological assessment: a review of tion of littoral insects in relation to vegetation structure. Vol:. (1234567890) 1 3 Aquat Ecol Hydrobiologia 499:179–190. https:// doi. org/ 10. 1023/A: Wentworth CK (1922) A scale of grade and class terms for 10263 25432 000 clastic sediments. J Geol 30:377–392. https:// doi. org/ 10. Urbanič G, Petkovska V, Pavlin M (2012) The relationship 1086/ 622910 between littoral benthic invertebrates and lakeshore Wickam H (2017) Tidyverse: easily install and load the modification. Fundam Appl Limnol 180:157–173. ‘Tidyverse’. R package version 1.2.1. https:// cran.r- proje https:// doi. org/ 10. 1127/ 1863- 9135/ 2012/ 0207ct. org/ web/ packa ges/ tidyv erse/ index. html Verdonschot PFM (2006) Beyond masses and blooms: Zenker A, Baier B (2009) Relevance of abiotic criteria used the indicative value of oligochaetes. Hydro- in German lake typology for macroinvertebrate fauna. biologia 564:127–142. https:// doi. org/ 10. 1007/ Hydrobiologia 636:379–392. https:// doi. org/ 10. 1007/ s10750- 005- 1714-zs10750- 009- 9967-6 Vinson MR, Hawkins CP (1998) Biodiversity of stream insects: variation at local, basin and regional scales. Publisher’s Note Springer Nature remains neutral with regard Annu Rev Entomol 43:271–293 to jurisdictional claims in published maps and institutional Waters NMC, Giovanni CRSG (2002) Distribution and affiliations. diversity of benthic macroinvertebrates associated with aquatic macrophytes. J Freshw Ecol 17(2):223–232. https:// doi. org/ 10. 1080/ 02705 060. 2002. 96638 90 Vol.: (0123456789) 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Aquatic Ecology Springer Journals

Does littoral substrate affect macroinvertebrate assemblages in Mediterranean lakes?

Download PDF
13 pages

Loading next page...
 
/lp/springer-journals/does-littoral-substrate-affect-macroinvertebrate-assemblages-in-zRw4eRnhFo

References (66)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2023
ISSN
1386-2588
eISSN
1573-5125
DOI
10.1007/s10452-023-10037-7
Publisher site
See Article on Publisher Site

Abstract

Aquat Ecol https://doi.org/10.1007/s10452-023-10037-7 Does littoral substrate affect macroinvertebrate assemblages in Mediterranean lakes? Efpraxia Mavromati  · Dimitra Kemitzoglou · Vasiliki Tsiaoussi Received: 3 February 2023 / Accepted: 25 May 2023 © The Author(s) 2023 Abstract The objective of this study was to inves- Keywords Lakes · Macroinvertebrates · Substrate tigate the effects of substrate type in macroinverte- type · Species richness · Community composition brate assemblages in Mediterranean lakes. Samplings have taken place in the littoral zone of 21 lakes in Greece, between 2015 and 2018. We compared ben- Introduction thic macroinvertebrate assemblages among three substrate types of their littoral zones; sandy, covered Freshwater biodiversity is worldwide declining at with macrophytes and stony substrate. Benthic mac- unprecedented rates (IPBES 2019). Factors under- roinvertebrate assemblages at sites with extended lying this biodiversity loss have been thoroughly macrophyte cover differed only slightly in composi- studied over the years and can be categorized in six tion and abundance from the ones found in stony and groups: hydrological alterations, habitat degradation sandy substrates. Coenagrionidae were indicative of and loss, pollution, overexploitation, invasive spe- sites covered with macrophytes and Oligochaeta and cies and climate change (Dudgeon et  al. 2006; Reid Erpobdellidae were representative of stony substrates. et  al. 2019; Arthington 2021). Most of these factors The type of substrate proved to be a statistically sig- are linked to human activities, which resulted in many nificant factor influencing the number of benthic restoration activities being carried out, focusing on macroinvertebrate taxa, the relative abundance of mitigating these anthropogenic impacts (Gething Oligochaeta and the relative abundance of Odonata. et al. 2020). However, in many cases there are knowl- In the context of designing site-adapted management edge gaps that hamper restoration efforts. For exam- measures, priority could be given to the conservation ple, although it is well known that lakes are subject and restoration of aquatic vegetation in lake litto- to morphological and hydrological modifications, the ral zones, which host rich macroinvertebrate assem- effects of these changes have  been less understood blages with abundant taxa of Odonata. (Poikane et al. 2020a). In Europe, we have increased our knowledge of freshwater ecosystems structure and function over the years, owing mainly to the implementation of Communicated by Télesphore Sime-Ngando. the Water Framework Directive (WFD). In this con- E. Mavromati (*) · D. Kemitzoglou · V. Tsiaoussi  text, benthic macroinvertebrates have been  among The Goulandris Natural History Museum, Greek Biotope/ the most widely used biological quality elements for Wetland Centre, 14th km Thessaloniki - Mihaniona, ecological assessments purposes especially in rivers 57001 Thermi, Greece and lakes, as reported in the 2018 EC Intercalibration e-mail: emavromati@ekby.gr Vol.: (0123456789) 1 3 Aquat Ecol Decision (European Commission 2018). Recently, their confounding effects (Solimini et  al. 2006). As they have become more popular in bio-assessment a rule of thumb, waterbodies in good ecological sta- methods as they have been proved sensitive not only tus could support a variety of species-indicators for to eutrophication and acidification but also to mor - undisturbed waters with high abundance values. phological changes and general degradation (Urbanič Except from their response to human stressors, et  al. 2012; Poikane et  al. 2016, 2020b; Mavromati freshwater macroinvertebrate taxa vary in response et al. 2021). Ntislidou et al. (2021) characterize them to habitat heterogeneity and substrate type. Gener- as “ecosystem engineers” playing an important role to ally, high habitat heterogeneity and coarser substrates aquatic ecosystem services as they participate in vari- present great invertebrate diversity and abundance ous biogeochemical processes. In the Mediterranean (Zenker and Baier 2009). According to Graça et  al. region, one such WFD-compliant assessment system (2015), coarse substrates are more complex than fine based on littoral benthic macroinvertebrates of Greek ones, as they retain coarser particulate organic matter lakes has recently been developed (HeLLBI; Mavro- and abundant with microalgae with higher biomass, mati et al. 2021). HeLLBI was designed and applied which can be used as food resources. Most studies to respond to anthropogenic impacts, and in particular accept the hypothesis that stony substrates present eutrophication and morphological changes (artificial higher richness and abundance of benthic taxa and in shoreline). particular abundance of collector–gatherers and filter- The distribution of macroinvertebrate communi- collectors (Pereira et al. 2017). ties in freshwater ecosystems is affected by both natu- Additionally, the presence of macrophytes in the ral and human factors (Dou et al. 2022). The natural littoral zone results in higher habitat heterogene- factors include all these environmental parameters ity. Dense macrophyte stands could alter inverte- that characterize a single site or water body, such as brate communities in numerous ways. Not only do temperature, water depth, dissolved oxygen, pΗ, con- they serve as food resources directly and indirectly ductivity, the spatial heterogeneity of habitats (Free through the growth of periphyton, but they can also et  al. 2009; Dou et al. 2022). Human activities affect access and deplete sediment nutrient loads (Waters macroinvertebrate communities either by nutrient and Giovanni 2002; Tolonen et al. 2003; Zenker and enrichment caused by land use activities or due to Baier 2009; Salmon et al. 2022). shoreline modification (Ntislidou et al. 2018; Bartels The overall objective of the study is to investigate et  al. 2021; Mavromati et  al. 2021). Differences in the effects of substrate types in benthic macroinver - biological communities across various environmental tebrate assemblages in Mediterranean lakes differing parameters have been extensively studied in streams in trophic status and other environmental parameters. and rivers but in lakes, especially in the Mediterra- In particular, we want to investigate the differences nean region, these studies are limited (Vinson and in benthic macroinvertebrate assemblages among Hawkins 1998). the sandy substrates, substrates covered with macro- Biological diversity can be studied several ways; phytes and stony substrates in Mediterranean natural it can be characterized by the diversity of species lakes, identify the indicator taxa characterizing each within sites and quantified by measuring the number substrate type and explore the role of macrophytes in of taxa or by diversity indices (Costa and Melo 2008). shaping these macroinvertebrate assemblages. Another way of measuring diversity is by describing the dissimilarities of biological assemblages between different environments (Costa and Melo 2008). Materials and methods As bio-indicators, freshwater macroinvertebrate taxa respond differently to stressors and they have Study area and sampling procedure the ability to incorporate the effects of the stressors they are exposed to, in combination and over time. As Benthic macroinvertebrates data were collected from they show spatial variation along different lake zones 21 lakes (Fig.  1), including two transboundary lakes, and depths, habitats and even lakes, finding out which Megali and Mikri Prespa, all belonging to the Greek natural factors affect their  assemblages is essential National Water Monitoring Network (Mavromati in order to separate them from human pressures and et al. 2021). Vol:. (1234567890) 1 3 Aquat Ecol Fig. 1 Twenty-one studied lakes of the Greek National Moni- votida, 13: Paralimni, 14: Petron, 15: Stymfalia, 16: Tricho- toring Network. 1: Amvrakia, 2: Cheimaditida, 3: Dystos, 4: nida, 17: Vegoritida, 18: Volvi, 19: Voulkaria, 20: Yliki, 21: Ismarida, 5: Kastoria, 6: Koroneia, 7: Kourna, 8: Lysimacheia, Zazari 9: Megali Prespa, 10: Mikri Prespa, 11: Ozeros, 12: Pam- The lakes included in this study belong to three In total, 97 littoral sampling sites have been sur- different natural lake types, according to the mixing veyed in the spring season during the 2015–2018 regime and depth gradient (Kagalou et  al. 2021). sampling campaign. Most lakes were sampled As most Mediterranean lakes, they face multiple once. Six lakes were sampled for two years and pressures including nutrient loading from point and one was sampled for three years. A nonparametric non-point sources, water abstraction and morpho- Kruskal–Wallis test was applied in this dataset and logical changes (Latinopoulos et  al. 2016). Moreo- showed that there were no differences among the ver, they seem to be affected by regional landscape sampling years (Mavromati et  al. 2021). The num- characteristics, in comparison with most cold tem- ber and location of the sampling sites for each lake perate and tropical lakes, by groundwater hydrology were selected according to lake size, habitat maps, and by the Mediterranean climate (Alvarez Cobelas and land use data of the lakes and their catchment et al. 2005; Mavromati et al. 2018). areas. Vol.: (0123456789) 1 3 Aquat Ecol Samples were collected using a semi-quantitative Abundance of Ephemeroptera, Plecoptera, Trichop- approach, which consists of a three-minute kick/ tera, Coleoptera, Bivalvia and Odonata (%EPTCBO), sweep with a standard hand net (500 μm mesh size), Simpson’s Diversity Index (Simpson) and Shannon at the littoral zone of each lake (up to 1.2 m depth of Diversity Index (Shannon). water). This particular approach was selected to cover potential effects of lakeshore modifications and it was conducted during the development of the littoral mac- Statistical analysis roinvertebrate assessment method (Mavromati et  al. 2021). At each sampling site, the cover with aquatic The  nonparametric Kruskal–Wallis  test was applied macrophytes (e.g. Phragmites australis, Potamogeton to examine statistically significant differences of mac- sp.) was recorded as a percentage. Furthermore, vis- roinvertebrate community indicators between differ - ual assessments of substrate composition were made ent substrate types (stony substrate, macrophytes and based on the predominant substratum size using the sandy substrate), followed by Wilcoxon signed rank size categories given in the Wentworth scale (Went- pairwise tests for the statistically significant indica- worth 1922). The substrate composition was further tors. Nonparametric tests were chosen, as the data grouped into two categories: sandy substrate (< 2 mm did not meet the assumptions of parametric tests. diameter) and stony substrate (> 2  mm diameter). Boxplots were used to show distributions of numeric As a result, each site sampling site was categorised values of statistically significant biological commu- into one of the three substrate types: stony substrate nity indicators. Boxplots were prepared in R using the (> 2  mm diameter), macrophyte and sandy substrate ggplot2 function (R Core Team 2018). (< 2 mm diameter). Non-metric Multidimensional Scaling (NMDS) Sieving was carried out on site; sorting, identifica- is an ordination technique, which enables complex tion and counting were carried out in the laboratory, multivariate data to be visualised in two dimensions. and the samples were preserved in vials containing This technique was employed to check for differences 70% ethanol. The littoral invertebrate fauna was iden- in benthic macroinvertebrate assemblages with sam- tified to family level, except oligochaetes, which were ples being a priori grouped by substrate type. NMDS identified as a subclass. Finally, the total number of centroids were calculated as the centre points of all individuals of each taxon was recorded and their rela- replicates for each sampling method in multidimen- tive abundance was calculated. sional space, as Gething et  al. (2020) suggested, using Bray–Curtis similarity coefficients on fourth Indicators of benthic macroinvertebrate assemblages root-transformed data. According to Anderson et  al. (2008), data transformation has been recommended Several indicators of benthic macroinvertebrate as a way to reduce the contribution of highly abun- assemblages (16 in total) were calculated from the dant species in relation to less abundant ones in the whole dataset: Number of Taxa (Taxa), Number of calculation of Bray–Curtis measure; when the trans- EPT taxa (EPT), Relative Abundance of Mollusca formation is severe (e.g. fourth root), rare species will (%Mollusca), Relative Abundance of Chironomi- have higher contribution to the analysis (Clarke et al. dae (%Chironomidae), Relative Abundance of Oli- 2014). This test statistic calculates a pseudo-F value, gochaeta (%Oligochaeta), Relative Abundance of similar to the F value in ANOVA. Larger pseudo-F Odonata (%Odonata), Relative Abundance of Bival- values indicate more pronounced group separation; via (%Bivalvia), Relative Abundance of Gastropoda however, its significance is usually of more interest (%Gastropoda), Relative Abundance of Ephemer- than its magnitude. The ordination diagram was also optera, Plecoptera and Trichoptera (%EPT), Rela- enhanced with the convex polygons around each cen- tive Abundance of Ephemeroptera, Coleoptera and troid (Costa and Melo 2008). Differences in the ben- Odonata (%ECO), Relative Abundance of Ephemer- thic fauna between the three different substrate types optera, Trichoptera and Odonata (%ETO), Relative were statistically tested via a permutational multi- Abundance of Ephemeroptera, Plecoptera and Odo- variate analysis of variance (PERMANOVA) using nata (%EPO), Relative Abundance of Ephemerop- the adonis2 function in vegan (Oksanen et  al. 2012; tera, Plecoptera and Coleoptera (%EPC), Relative Arbizu 2020; Gething et al. 2020). Vol:. (1234567890) 1 3 Aquat Ecol In order to analyse the multivariate homogeneity Results of group dispersions (variances) of macroinverte- brates assemblages and characterise their variability, The dataset used in this study included 77 taxa the betadisper function in vegan package was calcu- (together with Oligochaete subclass). The taxonomic lated based on Bray–Curtis distances (Oksanen et al. groups with higher numbers of taxa were Diptera 2012). Ordination plots were prepared and statistical (13), Gastropoda (10), Coleoptera (10) and Odonata analyses were performed in R version 3.5.3 (R Core (7). The highest number of taxa were observed in Team 2018). Lakes Lysimacheia and Vegoritida (23 and 21 taxa, The Similarity Percentages Analysis (SIMPER) respectively) and the lowest in Lake Koroneia (one was performed in Primer v7 software, to examine taxon). The lake with the highest number of EPT which taxa contributed most to the average similar- taxa was Vegoritida (6 taxa); on the other hand, Lake ity of sampling sites within each substrate type and Koroneia had no taxa belonging to orders Ephemer- the average dissimilarity between different types optera, Plecoptera and Trichoptera. of substrate (Clarke and Gorley 2015). Το identify According to SIMPER, the benthic fauna of our dominant taxa within each macroinvertebrate assem- littoral sites was dominated by five taxa: Chirono- blage for each substrate type, the indicator value midae, Oligochaeta, Corixidae, Gammaridae and index was applied using the indval function within Caenidae. The results of the analysis describing the the ladsv package and the indicators function within macroinvertebrates assemblages of the 97 sampling the indicspecies package (De Caceres et  al. 2016). sites are shown in Table 1. Chironomidae contributed The former is performed by a permutation test to most to the average similarity of all sites; especially assess the statistical significance of the association in sites with sandy substrate they almost represented between species and site groups, yielding a percent- 40% of all taxa. Benthic macroinvertebrate commu- age indicator value for each species (De Caceres and nities which occurred at sites with extended macro- Legendre 2009; Legendre and Legendre 2012). An phyte cover differed only slightly from the ones found indicator value of 0.25 was accepted as being ecologi- in stony and sandy substrates, in composition and cally relevant (Dufrene and Legendre 1997), and all abundance. Caenidae were mostly found in substrates significant indicators with a fidelity value below 0.25 dominated by macrophytes. The average dissimilar- were removed to exclude rare taxa (De Caceres et al. ity between stony substrates and sites covered with 2012). All analyses were performed with the use of macrophytes was 58.08%, between stony and sandy ladsv (Dufrene and Legendre 1997), indicspecies (De substrates was 58.55% and between sandy substrates Caceres and Legendre 2009), and tidyverse (Wickam and sites covered with macrophytes was 60.20% 2017) R packages in R environment version 3.5.3 (R (Table  2). The taxa that contributed to the average Core Team 2018). dissimilarity between the three substrate types were more or less the same (Gammaridae, Corixidae, Oli- gochaeta and Caenidae) but with different contribu- tions between the different pairs of substrate types. Indicator species analysis revealed two indicator taxa for sites with stony substrates (Oligochaeta and Table 1 Summary table Substrate type: Stony, n = 24 Substrate type: Macrophytes, Substrate type: Sandy, n = 35 of the SIMPER results for (Similarity: 46.88%) n = 38, (Similarity: 39.72%) (Similarity: 41.06%) benthic taxa contribution to similarity within each Taxa % %Cum Taxa % %Cum Taxa % %Cum substrate type (PRIMER 7 Software) Chironomidae 24.15 24.15 Chironomidae 29.43 29.43 Chironomidae 39.85 39.85 Oligochaeta 19.56 43.71 Corixidae 12.29 41.72 Corixidae 16.64 56.49 Gammaridae 17.26 60.98 Oligochaeta 11.90 53.63 Oligochaeta 10.29 66.78 Corixidae 13.78 74.76 Gammaridae 11.18 64.81 Gammaridae 10.06 76.84 Caenidae 8.36 73.17 Vol.: (0123456789) 1 3 Aquat Ecol Table 2 SIMPER results showing the contribution of taxa to meaning that this taxa combination occurred in less average dissimilarity between different pairs of substrate types than half of sites belonging to this group. Chironomi- (PRIMER 7 Software) dae and Coenagrionidae showed also high sensitivity Contribution (%) in sites covered with macrophytes. On the other hand, sites with sandy substrate had only one representative Stony and Macrophytes Average dissimilarity = 58.08 taxon (Atyidae) and no taxa combination; half of sites Gammaridae 9.12 belonging to this group included this particular com- Corixidae 6.84 bination (sensitivity = 0.49). Oligochaeta 6.43 The nonparametric Kruskal–Wallis test indi- Caenidae 5.77 cated no significant differences between sampling Stony and Sandy Average dissimilarity = 58.55 periods (p = 0.136; Mavromati et  al. 2021). The sta- Gammaridae 10.04 tistical analysis showed that the substrate type had Corixidae 8.15 a statistically significant effect on the Number of Oligochaeta 7.52 Taxa (H = 8.233, p = 0.016), the relative abundance Caenidae 6.43 of Oligochaeta (H = 9.646, p = 0.008) and the rela- Macrophytes and Sandy Average dissimilarity = 60.20 tive abundance of Odonata (H = 5.997, p = 0.050) Gammaridae 8.45 (Table 5). The pairwise comparison for taxa richness Corixidae 7.34 showed that the statistically different substrate types Caenidae 6.10 were stony substrate-macrophytes and macrophytes- Oligochaeta 6.06 sandy substrate (p < 0.05). The increase in taxa rich- ness was evident with greater substrate complexity, as macrophytes supported the greatest number of taxa Table 3 Indicator taxa for each substrate type while sandy substrates the fewest (Fig.  2). The rela- tive abundance of Odonata was higher in substrates Indicator taxa associated with macrophytes; whereas, the lowest Stony substrates Macrophytes percentages were recorded in stony substrates. The same results were shown in the pairwise comparisons Oligochaeta** Coenagrionidae* as statistically significant differences were observed Erpobdellidae** only between stony substrate and macrophytes The significance of the indicator value is shown: *p ≤ 0.05, (p < 0.05). On the other hand, the relative abundance **p ≤ 0.01 of Oligochaeta was higher in stony substrates, fol- lowed by macrophytes and finally by sandy sub- Erpobdellidae) and one indicator taxon (Coenagrio- strates. The pairwise comparison showed statistically nidae) for sites dominated by macrophytes (Table  3). significant differences between stones-macrophytes The best combinations of indicator taxa for each sub- and stones-sand. The remaining indicators of benthic strate group are shown in Table 4; the species combi- macroinvertebrate assemblages, which were tested in nation of sites with stony substrate (Gammaridae and our study showed no statistically significant  differ - Oligochaeta) showed high sensitivity; the majority ences between the three substrate types. of sites covered with stones included this particular According to PERMANOVA test, macroinver- combination. Their predictive power was rather low, tebrate assemblages were significantly different Table 4 Selected taxa Substrate type Selected taxa combination Predictive Sensitivity Square root combinations of indicator power of indicator taxa for each substrate value type. Prediction power, sensitivity and indicator Stony substrate Gammaridae + Oligochaeta 0.41 0.71 0.54 values (square rooted) of Macrophytes Chironomidae + Coenagrionidae 0.50 0.68 0.58 each taxa combination Sandy substrate Atyidae 0.52 0.49 0.50 Vol:. (1234567890) 1 3 Aquat Ecol Table 5 Nonparametric Kruskal–Wallis  test between several The analysis of multivariate homogeneity of group indicators of benthic macroinvertebrate assemblages and dif- dispersions (variances) revealed that the within-group ferent substrate types spread of macroinvertebrate communities did not dif- Metric Kruskal–Wallis H P value fer among substrate types (F = 2.3302, p = 0.095). 2, 94 This result clarifies the nature of multivariate effects Taxa 8.233 0.016 of macroinvertebrate assemblages and suggests that EPT 2.736 0.255 differences in benthic fauna were mainly due to dif- EPT (%) 0.524 0.769 ferences in centroid locations. ECO (%) 1.391 0.499 ETO (%) 0.704 0.703 EPO (%) 1.219 0.544 Discussion EPTCBO (%) 1.182 0.554 EPC (%) 0.612 0.736 The results presented in this paper allowed us to Oligochaeta (%) 9.646 0.008 assess the role of the littoral substrates in benthic Chironomidae (%) 2.154 0.341 macroinvertebrate fauna. We found that the studied Gastropoda (%) 5.247 0.073 substrate types supported to a certain degree distinct Bivalvia (%) 2.594 0.273 macroinvertebrate assemblages resulting from differ - Mollusca (%) 2.7933 0.255 ent levels of habitat complexity. However, the pres- Odonata (%) 5.997 0.050 ence of ubiquitous taxa across all sites was evident Simpson 0.329 0.848 in our results and was highlighted in two different Shannon 0.169 0.919 analyses, the SIMPER analysis and the nonparamet- Significant correlations (p < 0.05) are shown in bold ric Kruskal–Wallis test (Gething et al. 2020), provid- ing mixed results concerning community composition among substrate types (F2, 94 = 2.1876, p = 0.003). and diversity. No distinct clusters were observed in the ordination According to SIMPER results, Chironomidae space, indicating only slightly distinct benthic com- showed the highest contribution to the similarity of munities (Fig.  3). The plot of the NMDS analysis all three groups of substrates; its percentage to the revealed three polygons that defined the maximum overall community structure is rather high, espe- area of each group’s site scores in the two-dimen- cially in sandy substrates. It is well documented that sional ordination space, which seem to overlap at Chironomidae include several species with broad a certain extent. Macroinvertebrates occurring in environmental preferences including different sub- stony substrates seem to be a subset of those occur- strate types (Verdonschot 2006; Lencioni et al. 2018; ring in the other two substrate types. The hulls are Dorić et  al. 2020). Čerba et  al. (2022) concluded in fit to the raw data as they appear in the plot, which their study that substrate types significantly affected creates angular polygons. Despite the low degrees Chironomidae community composition and abun- of freedom, pseudo-F value of PERMANOVA anal- dance. A fauna list of Chironomidae larvae in main- ysis was large enough to reject the null hypothesis land Greece highlights exactly this: most species were of no differences in the centroid locations and/or indicative of distinct climatic, geological and hydro- the dispersion of groups between the three substrate chemical features (Płóciennik and Karaouzas 2013). types. The larger the pseudo-F value, the greater the On the other hand, Rossaro et al. (2014) refer to Chi- difference is supposed to be and is different from ronomidae taxa as opportunistic, being found in dif- the p value. Differences in the centroids locations of ferent habitats rather than restricted in a single habitat the three substrates are also evident in Fig. 4 where only. They use the term “preference” to describe their also the level of dispersion within each substrate ideal habitat type instead of “exclusivism” (Rossaro is shown. These three centroid locations represent et  al. 2014). The fauna of sandy substrate of lakes is the three different substrate types and according to mainly composed of Chironomidae taxa accompanied the boxplot, it is clear that the centroid locations of with a few other taxa, which explains their high con- sites covered with macrophytes and those of stony tribution to the similarity results of SIMPER analysis. substrates differ substantially. Ntislidou et  al. (2021) studied the macroinvertebrate Vol.: (0123456789) 1 3 Aquat Ecol Fig. 2 Boxplots of indica- tors of macroinvertebrate assemblages (Number of Taxa, Relative Abundance of Oligochaeta and Relative Abundance of Odonata) across different substrate types fauna of the profundal and sublittoral zones of three Caenidae discriminated the group of sites covered Greek eutrophic lakes and found that the assemblages with macrophytes from the other two groups and was of the muddy substrates of these zones were domi- found to be the taxon associated with the group dis- nated by Chironomidae and Oligochaeta taxa. similarities. This result agrees neither with the study of Pilotto et al. (2015) which associates Caenidae spe- cies with stony substrates and the presence of zebra mussels which they use as food resources, nor with the results of McGoff and Irvine (2009) who found a negative correlation between Caenis luctuosa, the macrophyte Percentage Volume Inhabited (PVI) and the extent of macrophytes lakewards. The percentage of dissimilarity was higher than 55% among all three groups of substrate types; none- theless the taxa that contributed mostly to the aver- age dissimilarity were the same with different per - centages of contribution at each group. The IndVal analysis (De Caceres and Legendre 2009) showed more clear results and revealed indicator species for stony substrates and for substrates covered with mac- rophytes. No indicator species were found for sandy substrates, probably because of their unique macroin- Fig. 3 Non-metric multidimensional scaling (NMDS) plot of vertebrate assemblage in comparison with the other benthic macroinvertebrate community composition between groups of sites and the relatively low number of taxa different substrate types in the studied lakes Vol:. (1234567890) 1 3 Aquat Ecol Our results showed that there was an increase in taxa richness in coarser substrates which agrees with the initial hypothesis of Graça et al. (2015). They ana- lysed the macroinvertebrate community in streams and characterized fine substrates poorer in terms of abundance of macroinvertebrates and taxa richness. Surface complexity seems to be positively correlated with diversity and abundance of benthic fauna espe- cially in studies concerning streams (Taniguchi and Tokeshi 2004; Barnes et al. 2013). On the other hand, our results showed that the relative abundance of Oli- gochaeta was greater in stony substrates in compari- son with the other two types, which is a surprising result according to the literature (Rieradevall et  al. 1999; Graca et al. 2015; Buendia et al. 2011). In par- ticular, Graça et al. (2015) argue that finer sediments are more appropriate for taxa like Chironomidae and Fig. 4 Boxplots of the level of dispersion within the three sub- Oligochaeta, which prefer spending time within sub- strate types which was calculated as the mean distance of each strate particles. The benthic fauna of sandy substrates sample point to the centroid of the respective substrate type. in our dataset was mainly composed of Chironomi- The centre line in the box displays the median and the mar- gins of the box specify the 25th and 75th percentile. Whisk- dae, resulting in low abundances of their remaining ers extend to the smallest (lower whisker) or the largest (upper taxa, in comparison with the other two groups of whisker) value within the range of 1.5 × interquartile range sites. Sychra et  al. (2010) suggested in their study that some species of Oligochaeta, such as Naididae (De Caceres et  al. 2012). To overcome this obstacle with phytal preferences are quite abundant in reed- the extended version of indicator species analysis beds near the shore. Oligochaeta and Chironomidae was used to give us indicator species also for sandy are considered to be generally tolerant organisms and substrates. The predictive values were quite simi- according to Mavromati et  al. (2021) they are the lar in all three groups of sites. Sites with stony sub- dominant taxa in sites with high proportion of artifi- strate highlighted Gammaridae and Oligochaeta as cial shoreline in poor and bad ecological quality sta- a combination of taxa with high sensitivity, which tus in Mediterranean lakes. is an estimation of the frequency of the families at Odonata larvae followed the opposite pattern and these sites. Studies indicate that species belonging to exhibited greater abundances in sites covered with Gammaridae show a clear preference in stony sub- macrophytes, followed by sandy and stony sub- strates, especially when they are colonized by Dre- strates. Waters and Giovanni (2002) associate Lesti- issena sp. (Stewart et al. 1998; Hesselschwerdt et al. dae (Odonata) with vascular macrophytes present in 2008). High sensitivity of the taxa combination of depositional habitats. The same pattern was evident Chironomidae and Coenagrionidae was also found in in Graça et  al. (2015) where it is argued that Odo- the groups of sites covered with macrophytes. Pilotto nata and Trichoptera inhabit coarser, well sorted et al. (2015) suggested that species of Coenagrionidae substrates. The links between aquatic vegetation and such as Ischnura elegans prefer sites with submerged macroinvertebrate abundance and diversity are evi- macrophytes and natural littoral habitats. dent in the majority of studies but the results are still Differences in community composition and diver - confounding with the type of sediment that aquatic sity were evident across different substrate types; macrophytes need for growth (Waters and Giovanni Gething et  al. (2020) characterize substrate type as 2002; Zenker and Baier 2009). It is crucial to check surrogates for increasing habitat complexity. In our in littoral zones which one is actually affecting mac- results, sites covered with macrophytes supported roinvertebrate assemblages; macrophytes or the sedi- greater number of taxa followed by sites with stony ment that they rely on? Another study focused only substrates and finally the ones covered with sand. in Chironomidae community composition, concluded Vol.: (0123456789) 1 3 Aquat Ecol that sites covered with macrophytes provided greater in order to be able to draw conclusions without the availability of food resources, microhabitats to confounding effects. inhabit and shelter from predators (Čerba et al. 2022). It is crucial to acknowledge that some macroinver- McGoff and Irvine (2009) associated positively lit- tebrate taxa are depending on certain substrate types; toral macroinvertebrate abundance with both mac- this can be useful when developing and implement- rophyte PVI (Percentage Volume Inhabited)  and ing site-adapted conservation and restoration meas- the extent of macrophytes lakewards in their study ures (Brauns et  al. 2007). In this context, priority is between Lake Habitat Quality Assessment and mac- advised to be placed to the conservation and/or res- roinvertebrate community structure. toration of the natural littoral habitats that host rich Overall, the substrate type seems to be a primary macroinvertebrate assemblages. The need to conserve factor affecting macroinvertebrate assemblages (along and were appropriate restore aquatic vegetation is with other natural factors such as lake size and depth) obvious as our results suggested that sites dominated (Timm and Möls 2012). We recorded this spatially with macrophytes exhibit the greatest number of taxa variability of benthic fauna but our results were rather and the most diverse group of Odonata. Controlling weak. Statistically speaking, the distinct macroinver- the extent and frequency of vegetation cutting and tebrate assemblages were a result of differences in dredging could lead to habitat heterogeneity resulting location of their centroids and not homogeneity of on the dispersal of the macroinvertebrate community dispersions but still we could consider other factors (Gething et al. 2020). affecting our dataset such as the dissimilarity meas- Acknowledgements The present study was conducted in ures (Bray–Curtis, Jaccard etc.) which could have the frame of the Greek National Water Monitoring Network, altered the interpretation of the results (Anderson according to the JMD 140384/2011, implemented by The Gou- et al. 2006). As Barnes et al. (2013) pointed out, spe- landris Natural History Museum, Greek Biotope/Wetland Cen- cies richness is mainly affected by complexity and not tre (EKBY). The network is supervised by the General Direc- torate for Waters of the Ministry of Environment and Energy. heterogeneity but still they advocate to take into con- The data used in this research come from Act MIS 5001204 sideration not only habitat structure but the processes financed by the European Union Cohesion Fund (Partner - involved in these relationships. The effect of sub- ship Agreement 2014–2020), and from Acts MIS 371010, strate types alone on species richness in lakes is less 371138, 371140, 371145 financed by the European Regional Development Fund (National Strategic Reference Framework studied and the fact that we did not find any  strong 2007–2013). K. Argiriou, M. Bozatzidou, G. Nakas, and V. relationships could be attributed to the lack of  more Navrozidou contributed to sorting and taxa identification. H. detailed categories (macrophyte cover, substrate size). Hadjicharalambous contributed to data analysis. EKBY’s per- Macrophytes seem to play an important role in regu- sonnel conducted samplings and contributed to data analysis. lating the balance of all trophic relationships occur- Data availability The dataset generated during and analysed ring in the littoral zone of lakes. Tolonen et al. (2003) during the current study are available from the corresponding suggested that in an oligo-mesotrophic lake with a author on reasonable request. well-established macrophyte zone, the composition Declarations and size of zoobenthos differed significantly along the gradient of vegetation density horizontally from Conflict of interest The authors have no competing interests shore to open water. On the other hand, they sug- to declare that are relevant to the content of this article. gested that shallow eutrophic lakes might be affected by other natural factors such as water level fluctuation Open Access This article is licensed under a Creative Com- mons Attribution 4.0 International License, which permits and nutrient loading. Sediment deposition can signifi- use, sharing, adaptation, distribution and reproduction in any cantly alter the composition of the bottom substrate, medium or format, as long as you give appropriate credit to the enriching it with organic matter. Jurca et  al. (2021) original author(s) and the source, provide a link to the Crea- pointed out that there was an absence of species with tive Commons licence, and indicate if changes were made. The images or other third party material in this article are included specific mesohabitat preferences in eutrophic lakes, in the article’s Creative Commons licence, unless indicated as there is a clear linkage of macrophyte presence and otherwise in a credit line to the material. If material is not species richness with eutrophication. All these factors included in the article’s Creative Commons licence and your should be taken into consideration when sampling for intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly benthic macroinvertebrates in littoral zones of lakes Vol:. (1234567890) 1 3 Aquat Ecol De Cáceres M, Legendre P (2009) Associations between spe- from the copyright holder. To view a copy of this licence, visit cies and groups of sites: indices and statistical inference. http:// creat iveco mmons. org/ licen ses/ by/4. 0/. Ecology 90(12):3566–3574 De Cáceres M, Legendre P, Wiser SK, Brotons L (2012) Using species combinations in indicator value analyses. Methods Ecol Evol 3(6):973–982. https:// doi. org/ 10. 1111/j. 2041- References 210X. 2012. 00246.x Dorić V, Koh M, Mihaljević Z (2020) First record of Nubensia Alvarez Cobelas M, Rojo C, Angeler DG (2005) Mediterra- nubens (Edwards, 1929) (Diptera: Chironomidae) from nean limnology: current status, gaps and the future. J Lim- Croatia. Natura Croatica: Periodicum Musei Historiae nol 64(1):13–29. https:// doi. org/ 10. 4081/ jlimn ol. 2005. 13 Naturalis Croatici 29(1):139–142 Anderson MJ, Ellingsen KE, McArdle BH (2006) Multivari- Dou Q, Du X, Cong Y, Wang L, Zhao C, Song D, Liu H, Huo ate dispersion as a measure of beta diversity. Ecol Lett T (2022) Influence of environmental variables on mac- 9(6):683–693. https:// doi. org/ 10. 1111/j. 1461- 0248. 2006. roinvertebrate community structure in Lianhuan Lake. 00926.x Ecol Evol 12:e8553. https:// doi. org/ 10. 1002/ ece3. 8553 Anderson MJ, Gorley RN, Clarke KR (2008) PERMANOVA+ Dudgeon D, Arthington AH, Gessner MO, Kawabata ZI, for PRIMER: guide to software and statistical methods. Knowler DJ, Lévêque C, Naiman RJ, Prieur-Richard AH, PRIMER-E, Plymouth Soto D, Stiassny MLJ, Sullivan CA (2006) Freshwater Arbizu PM (2020) Pairwiseadonis: Pairwise multilevel com- biodiversity: importance, threats, status and conservation parison using adonis. R package version 0.4 challenges. Biol Rev 81:163–182. https:// doi. org/ 10. 1017/ Arthington AH (2021) Grand challenges to support the fresh- S1464 79310 50069 50 water biodiversity emergency recovery plan. Front Envi- Dufrene M, Legendre P (1997) Species assemblages and ron Sci 9:664313. https:// doi. org/ 10. 3389/ fenvs. 2021. indicator species: the need for a flexible asymmetrical approach. Ecol Monogr 67:345–366 Barnes JB, Vaughan IP, Ormerod SJ (2013) Reappraising the European Commission (2018) Commission decision (EU) effects of habitat structure on river macroinvertebrates. 2018/229 of 12 February 2018 establishing, pursuant to Freshw Biol 58:2154–2167. https:// doi. org/ 10. 1111/ fwb. directive 2000/60/EC of the European Parliament and of the Council, the values of the member state monitoring Bartels A, Berninger UG, Hohenberger F, Wickham S, Peter- system classifications as a result of the intercalibration mann JS (2021) Species data of littoral macroinvertebrate exercise and repealing commission decision 2013/480/ communities of alpine lakes along an elevational gradi- EU. Off J Eur Communities L47:1–91 ent (Hohe Tauern National Park, Austria). PLoS ONE Free G, Solimini AG, Rossaro B, Marziali L, Giacchini R, 16(11):e0255619. https:// doi. org/ 10. 1371/ journ al. pone. Paracchini B, Ghiani M, Vaccaro S, Gawlik BM, Fresner 02556 19 R, Santner G, Schonhuber M, Cardoso AC (2009) Model- Brauns M, Garcia XF, Walz N, Pusch MT (2007) Effects of ling Lake macroinvertebrate species in the shallow sublit- human shoreline development on littoral macroinver- toral: relative roles of habitat, lake morphology, aquatic tebrates in lowland lakes. J Appl Ecol 44:1138–1144. chemistry and sediment composition. Hydrobiologia https:// doi. org/ 10. 1111/j. 1365- 2664. 2007. 01376.x 633:123–136. https:// doi. org/ 10. 1007/ s10750- 009- 9869-7 Buendia C, Gibbins CN, Vericat D, Lopez-Tarazon J, Batalla Gething KJ, Ripley MC, Mathers KL, Chadd RP, Wood PJ RJ (2011) Influence of naturally high fine sediment loads (2020) The influence of substrate type on macroinverte- on aquatic insect larvae in a montane river. Scottish Geogr brate assemblages within agricultural drainage ditches. J 127(4):315–334. https:// doi. org/ 10. 1080/ 14702 541. Hydrobiologia 847:4273–4428. https:// doi. org/ 10. 1007/ 2012. 670006 s10750- 020- 04416-6 De Caceres M, Jansen F, De Caceres, MM (2016). Package Graça M, Ferreira W, Firmiano K, França J, Callisto M (2015) ‘indicspecies’. Indicators. 8(1) Macroinvertebrate identity, not diversity, differed across Čerba D, Koh M, Vlaičević B, TurkovićČakalić I, Milošević patches differing in substrate particle size and leaf litter D, StojkovićPiperac M (2022) Diversity of periphytic packs in low order, tropical Atlantic forest streams. Lim- chironomidae on different substrate types in a floodplain netica 29:29–40. https:// doi. org/ 10. 23818/ limn. 34. 03 aquatic ecosystem. Diversity 14(4):264. https:// doi. org/ 10. Hesselschwerdt J, Necker J, Wantzen K (2008) Gammarids in 3390/ d1404 0264 lake constance: habitat segregation between the invasive Clarke KR, Gorley RN, Somerfield PJ, Warwick RM (2014) Dikerogammarusvillosus and the indigenous Gammarus- Change in marine communities: an approach to statis- roeselii. Fundam Appl Limnol 173:177–186. https:// doi. tical analysis and interpretation, 3rd edn. PRIMER-E, org/ 10. 1127/ 1863- 9135/ 2008/ 0173- 0177 Plymouth IPBES (2019) Global assessment report on biodiversity and Clarke KR, Gorley RN (2015) PRIMER v7: User Manual/ ecosystem services of the intergovernmental science- Tutorial. PRIMER-E, Plymouth policy platform on biodiversity and ecosystem services. Costa SS, Melo AS (2008) Beta diversity in stream macroin- IPBES secretariat, Bonn, Germany, p 1148. https:// doi. vertebrate assemblages: among-site and among-microhab- org/ 10. 5281/ zenodo. 38316 itat components. Hydrobiologia 598:131–138 Jurca T, Donohue L, Mcgoff E, Tunali S, Irvine K (2021) Unravelling the effect of multiple stressors on ecologi- cal structure of littoral lake macroinvertebrates. Int Rev Vol.: (0123456789) 1 3 Aquat Ecol methods, intercalibration and practical recommendations. Hydrobiol 106:202–212. https:// doi. org/ 10. 1002/ iroh. Sci Total Environ 543:123–134 20200 2048 Poikane S, Zohary T, Cantonati M (2020a) Assessing the eco- Kagalou I, Ntislidou C, Latinopoulos D, Kemitzoglou D, logical effects of hydromorphological pressures on Euro- Tsiaoussi V, Bobori DC (2021) Setting the phosphorus pean lakes. Inland Waters 10(2):241–255. https:// doi. org/ boundaries for Greek natural shallow and deep lakes for 10. 1080/ 20442 041. 2019. 16548 00 water framework directive compliance. Water 13(5):739. Poikane S, Herrero FS, Kelly MG, Borja A, Birk S, van de https:// doi. org/ 10. 3390/ w1305 0739 Bund W (2020b) European aquatic ecological assessment Latinopoulos D, Ntislidou C, Kagalou I (2016) Multipurpose methods: a critical review of their sensitivity to key pres- plans for the sustainability of the Greek Lakes: emphasis sures. Sci Total Environ 740:140075. https:// doi. org/ 10. on multiple stressors. Environ Process 3:589–602. https:// 1016/j. scito tenv. 2020. 140075 doi. org/ 10. 1007/ s40710- 016- 0152-4 R Core Team (2018) R: a language and environment for sta- Legendre P, Legendre L (2012) Numerical ecology. Elsevier tistical computing. r foundation for statistical computing, Lencioni V, Cranston PS, Makarchenko E (2018) Recent Vienna. Retrieved 22 Apr 2019, from https:// www. Rproj advances in the study of chironomidae: an overview. J ect. org Limnol 77:1–6. https:// doi. org/ 10. 4081/ jlimn ol. 2018. 1865 Reid AJ, Carlson AK, Creed IF, Eliason EJ, Gell PA, Johnson Mavromati E, Kemitzoglou D, Tsiaoussi V (2021) A new PTJ, Kidd KA, MacCormack TJ, Olden JD, Ormerod SJ, WFD—compliant littoral macroinvertebrate index for Smol JP, Taylor WW, Tockner K, Vermaire JC, Dudgeon monitoring and assessment of Mediterranean lakes D, Cooke SJ (2019) Emerging threats and persistent con- (HeLLBI). Environ Monit Assess 193:745. https:// doi. servation challenges for freshwater biodiversity. Biol Rev org/ 10. 1007/ s10661- 021- 09493-1 94:849–873. https:// doi. org/ 10. 1111/ brv. 12480 Mavromati E, Kagalou I, Kemitzoglou D, Apostolakis A, Rieradevall N, Bonada N, Prat N (1999) Substrate and depth Seferlis M, Tsiaoussi V (2018) Relationships among preferences of macroinvertebrates along a transect in a land use patterns, hydromorphological features and Pyrenean high mountain lake (Lake Redó, NE Spain). physicochemical parameters of surface waters: WFD Limnetica 17:127–134 lake monitoring in Greece. Environ Process 5:139–151 Rossaro B, Marziali L, Montagna M, Magoga G, Zaupa S, McGoff E, Irvine K (2009) A test of the association between Boggero A (2014) Factors controlling morphotaxa dis- lake habitat quality assessment and macroinvertebrate tributions of Diptera chironomidae in freshwaters. Water community structure. Aquatic Conserv: Mar Freshw 14(7):1014. https:// doi. org/ 10. 3390/ w1407 1014 Ecosyst 19:520–533. https:// doi. org/ 10. 1002/ aqc. 1024 Salmon Q, Colas F, Westrelin S, Dublon J, Baudoin JM Ntislidou C, Lazaridou M, Tsiaoussi V, Bobori DC (2018) (2022) Floating littoral zone (FLOLIZ): a solution to A new multimetric macroinvertebrate index for the eco- sustain macroinvertebrate communities in regulated logical assessment of Mediterranean lakes. Ecol Ind lakes? Ecol Eng 176:106509. https:// doi. org/ 10. 1016/j. 93:1020–1033. https:// doi. org/ 10. 1016/j. ecoli nd. 2018. ecole ng. 2021. 106509 05. 071 Solimini AG, Free G, Donohue I, Irvine K, Pusch M, Ros- Ntislidou C, Rossaro B, Lazaridou M, Bobori D (2021) What saro M, Sandin L, Cardoso AC (2006) Using benthic drives benthic macroinvertebrate dispersal in different macroinvertebrates to assess ecological status of lakes lake substrata? The case of three Mediterranean lakes. Aquat Ecol 55:1033–1050. https:// doi. org/ 10. 1007/ current knowledge and way forward to support WFD s10452- 021- 09880-3 implementation. Institute for Environment and Sustain- Oksanen J, Blanchet G, Kind R, Legendre P, Minchin P, ability, Luxembourg, p 49 O’Hara, RB Simpson G, Solymos P, Stevenes MHH, Stewart TW, Miner JG, Lowe RL (1998) Macroinvertebrate Wagner H (2012) Vegan: community ecology package. R communities on hard substrates in Western Lake Erie: package version 2.0-2. structuring effects of Dreissena. J Great Lakes Res Pereira TS, Pio JFG, Calor AR, Copatti CE (2017) Can the 24(4):868–879. https:// doi. org/ 10. 1016/ S0380- 1330(98) substrate influence the distribution and composition of 70868-8 benthic macroinvertebrates in streams in Northeastern Sychra I, Adámek Z, Petřivalská K (2010) Distribution and Brazil? Limnologica 63:27–30. https:// doi. org/ 10. 1016/j. diversity of littoral macroinvertebrates within extensive limno. 2016. 12. 003 reed beds of a lowland pond. Ann Limnol - Int J Lim Pilotto F, Bazzanti M, DiVito V, Frosali D, Livretti F, Mast- 46(4):281–289. https:// doi. org/ 10. 1051/ limn/ 20100 26 rantuono L, Pusch MT, Sena F, Solimini AG (2015) Rela- Taniguchi H, Tokeshi M (2004) Effects of habitat complex- tive impacts of morphological alteration to shorelines and ity on benthic assemblages in a variable environment. eutrophication on littoral macroinvertebrates in Mediter- Freshw Biol 49:1164–1178. https:// doi. org/ 10. 1111/j. ranean lakes. Freshw Sci 34:410–4221365- 2427. 2004. 01257.x Płóciennik M, Karaouzas I (2014) The Chironomidae (Diptera) Timm H, Mols T (2012) Littoral macroinvertebrates in fauna of Greece: ecological distributions and patterns, Estonian lowland lakes: the effects of habitat, season, taxalist and new records. Ann Limnol Int J Lim 50(1):19– eutrophication and land use on some metrics of biologi- 34. https:// doi. org/ 10. 1051/ limn/ 20130 66 cal quality. Fundam Appl Limnol 180:145–156. https:// Poikane S, Johnson RK, Sandin L, Schartau AK, Solimini AG, doi. org/ 10. 1111/j. 1365- 2427. 2004. 01257.x Urbanič G, Arbačiauskas K, Aroviita J, Gabriels W, Miler Tolonen K, Hämäläinen H, Holopainen IJ, Mikkonen K, O, Pusch MT, Timm H, Bohmer J (2016) Benthic mac- Karjalainen J (2003) Body size and substrate associa- roinvertebrates in lake ecological assessment: a review of tion of littoral insects in relation to vegetation structure. Vol:. (1234567890) 1 3 Aquat Ecol Hydrobiologia 499:179–190. https:// doi. org/ 10. 1023/A: Wentworth CK (1922) A scale of grade and class terms for 10263 25432 000 clastic sediments. J Geol 30:377–392. https:// doi. org/ 10. Urbanič G, Petkovska V, Pavlin M (2012) The relationship 1086/ 622910 between littoral benthic invertebrates and lakeshore Wickam H (2017) Tidyverse: easily install and load the modification. Fundam Appl Limnol 180:157–173. ‘Tidyverse’. R package version 1.2.1. https:// cran.r- proje https:// doi. org/ 10. 1127/ 1863- 9135/ 2012/ 0207ct. org/ web/ packa ges/ tidyv erse/ index. html Verdonschot PFM (2006) Beyond masses and blooms: Zenker A, Baier B (2009) Relevance of abiotic criteria used the indicative value of oligochaetes. Hydro- in German lake typology for macroinvertebrate fauna. biologia 564:127–142. https:// doi. org/ 10. 1007/ Hydrobiologia 636:379–392. https:// doi. org/ 10. 1007/ s10750- 005- 1714-zs10750- 009- 9967-6 Vinson MR, Hawkins CP (1998) Biodiversity of stream insects: variation at local, basin and regional scales. Publisher’s Note Springer Nature remains neutral with regard Annu Rev Entomol 43:271–293 to jurisdictional claims in published maps and institutional Waters NMC, Giovanni CRSG (2002) Distribution and affiliations. diversity of benthic macroinvertebrates associated with aquatic macrophytes. J Freshw Ecol 17(2):223–232. https:// doi. org/ 10. 1080/ 02705 060. 2002. 96638 90 Vol.: (0123456789) 1 3

Journal

Aquatic Ecology – Springer Journals

Published: Sep 1, 2023

Keywords: Lakes; Macroinvertebrates; Substrate type; Species richness; Community composition

There are no references for this article.