Spatial and temporal pattern of benthic macroinvertebrate assemblages in two large Chinese freshwater lakes subjected to different degrees of eutrophication

Abstract Freshwater lakes are subjected to increasing eutrophication and algal blooms worldwide, which have dramatically impacted the aquatic biodiversity. This situation is especially true in the Yangtze River Floodplain, China. However, few researches have been focused on differences in the spatial and temporal patterns of benthic macroinvertebrates assemblages between the two adjacent lakes in this floodplain. In this study, we conducted seasonal investigations in 2021 to compare the spatial and temporal patterns of macroinvertebrates in Lakes Taihu and Lake Chaohu. Non-metric multidimensional scaling (NMDS) analyses showed Lake Taihu was found to have differences in assemblages within the lake, but Lake Chaohu was found to have uniform assemblage across the entire lake. Macrophyte-dominated lake regions (eastern Lake Taihu) were characterised by the highest species richness and diversity indices than those in other lake regions. The pollution sensitive species (e.g. EPT: Ephemeroptera, Plecoptera and Trichoptera) were only distributed in the macrophytic areas of Lake Taihu. Corbicula fluminea, polychaetes Nephtys oligobranchia and Capitella sp. were dominant in the middle central lake regions in Lake Taihu. Lake Chaohu and the northern Lake Taihu, with turbid algal-dominated state, were characterised by dominance of Oligochaetes and Chironomids. These differences were attributed to varying habitats and environmental parameters among different lake regions. Canonical correspondence analysis (CCA) demonstrated that macrophytes, chemical oxygen demand (CODMn), temperature, ammonium (NH3-N) and electric conductivity (EC) were detected as important factors influencing such macroinvertebrate assemblage patterns between these two lakes. This study contributes to the scientific knowledge of within lake habitats, community assemblages, nutrients and pollution within these lakes.


Introduction
Due to increasing human disturbances, lake eutrophication has been considered to be one of the most serious environmental problems in the worldwide.Eutrophication can cause lakes to transform from a clear, macrophyte-dominated state to a turbid, algal-dominated state, and aquatic fauna interact closely with different state (Pan et al. 2012).Previous studies reveal that eutrophication and algal blooms affect the balance of freshwater ecosystem (Żbikowski and Kobak 2007;Ma et al. 2008;Liu et al. 2011;Peng et al. 2020).As an important component of lake ecosystem, macroinvertebrates play an important role in the process of nutrients circulation and energy flow in lake ecosystem (Xing et al. 2021).Due to long life span, slow migration ability and their sensitivity, macroinvertebrates can indicate different eutrophic stable state.The effects of environmental stress on water quality could be reflected by the changes of macroinvertebrate community structure and dominant species composition (Bilyard 1987;O'Connor, Walls, and Hughes 2000).Therefore, studies on macroinvertebrate community characteristics can help us to better understand which state a lake is in, thus provide important scientific basis data for water ecological protection and restoration of freshwater shallow lakes.
Lake Taihu and Lake Chaohu, the third and fifth largest freshwater shallow lakes in China, are located in the middle and lower reaches of the Yangtze River in China.The two lakes serve multiple functions such as, water supply, navigation, fishery, flood control, irrigation and tourism (Shang and Shang 2007;Wu et al. 2020;Zhao and Wang 2021).Since the 1980s, the discharge of domestic sewage and industrial wastewater flowing into the two lakes have resulted in the increase of nutrient concentration, the acceleration of lake eutrophication, and the frequent occurrence of cyanobacterial blooms, which have seriously threatened the ecological functions of the water bodies (Yu et al. 2011;Li et al. 2022;Wu, Chen, et al. 2022).
In recent decades, Lake Taihu and Lake Chaohu have experienced different degrees of eutrophication and cyanobacteria blooms (Shang and Shang 2007;Janssen et al. 2017;Wan et al. 2017).The two lakes have many similarities (location, climate, etc.), but Lake Taihu is found to have three typical stable states (macrophyte-dominated state, algae-dominated state and middle transitional state) within the lake but Lake Chaohu has only uniform stable state (algal-dominated state) across the entire lake.This study aimed to: (1) compare the spatial and temporal patterns of benthic macroinvertebrate assemblages in two large Chinese freshwater lakes subjected to different degrees of eutrophication; (2) explore the environmental factors driving the distribution patterns of macroinvertebrate community in the two lakes.

Study area and sampling sites
Lake Taihu is the third largest freshwater lake in China, located between 30°55 '40 "-31°32′ 58" N, 119°52 '32 "-120°36′ 10" E, with an area of 2, 338 km 2 and an average depth of 1.9 m.There are 22 major rivers entering Lake Taihu, including Wangyu River, Caoqiao River and Liangxi River (Figure 1).These main rivers flowing into Lake Taihu accounted for about 85% of the total lake resources, and the pollution loads into the lake are as high as about 75% (Qin et al. 2007).
The environmental characteristics of Lake Taihu shows high spatial heterogeneity.East Lake Taihu (Figure 1, T4 and T5) is located in the southeast regions, which is a typical macrophytic lake regions in the middle and lower reaches of the Yangtze River.The aquatic vegetation have high coverage, diversity and standing crops, among which submerged vegetation and floating vegetation are the main ecological types of aquatic vegetation in east Lake Taihu (Dong et al. 2014).In recent decades, the northern Lake Taihu (T1, T2 and T3) suffered severe eutrophication and frequent outbreaks of cyanobacterial blooms (Chen et al. 2003).Since 2000, the area of summer algal bloom outbreak accounts for more than half of the total area of Lake Taihu, and the frequency of cyanobacteria bloom is continuously increasing, which led to drinking water crisis and seriously hindering the production and life of the surrounding areas (Lucie 2007); The central lake (T6, T7, T8 and T9) is a transitional lake region with moderate eutrophication and little aquatic vegetation (Ji et al. 2015).
Lake Chaohu is located at 30°25 '28 "-31°43′ 28" N, 117°16 '54 "-117°51′ 46" E, with a surface area of 780 km 2 and a water depth of about 3 m.The main rivers entering Lake Chaohu are Nanfei River, Shiwuli River, Hangbu River, Fengle River, Pai River, Zhegao River, Zhaohe River and Baishishan River, and the lake water finally flows into the Yangtze River via Yuxi River (Figure 1).Before the 1950s, Lake Chaohu was a large natural lake with good water quality and high biodiversity in Lake Chaohu.After the 1980s, with the continuous economic and social development of the catchment area and the rapid increase of population, the discharge of a large amount of industrial wastewater and domestic sewage directly led to the increasing eutrophication and cyanobacterial bloom in Lake Chaohu (Zhang et al. 2015;Wu, Ji, et al. 2022).Aquatic vegetation is rarely found in Lake Chaohu where eutrophication is increasing (Pu et al. 2022;Wang et al. 2022).According to the distribution of aquatic vegetation and the degree of eutrophication, Lake Chaohu has only uniform stable state (algal-dominated state) across the entire lake (group 1).However, Lake Taihu can be divided into three types: algal-dominated lake regions (group 2: T1, T2 and T3), macrophyte-dominated lake regions (group 3: T4 and T5) and transition lake regions (group 4: T6, T7, T8 and T9).

Sampling and laboratory analysis
Three replicate sediment samples were quarterly collected with a modified Peterson grab with an area of 1/16 m 2 at each sampling site in 2021.Sediments were sieved through a 0.45 mm and the remaining detritus on the sieve were returned to the laboratory.Specimens were manually sorted in a white tray and were preserved with 4% formalin solution.Macroinvertebrate were sorted and identified to the lowest possible taxonomic levels in the laboratory, according to the relevant references (Brinkhurst 1986;Morse, Yang, and Tian 1994).
During the field sampling, water temperature (T) and pH were measured in the field.In the laboratory, Ammonium (NH 3 -N) was determined by phenol-hypochlorite method by ultraviolet spectrophotometer (Solorzano 1969).Total phosphorus (TP) concentration was determined following digestion according to Golterman, Clymo, and Ohmstad (1978).Total nitrogen (TN) concentration was measured by peroxodisulfate oxidation (Ebina, 1983).Dissolved oxygen (DO), electric conductivity (EC), chlorophyll a (Chla) and chemical oxygen demand (COD Mn ) were measured based on the standard methods (APHA 1989).

Data processing
Non-metric multidimensional scaling (NMDS) of macroinvertebrate communities was performed based on the Bray-Curtis similarity with abundance data.Analysis of similarity (ANOSIM) was used to investigate differences in the macroinvertebrate communities between groups.The statistic global R represented the degree of separation between groups.Preliminary detrended correspondence analysis (DCA) on the macroinvertebrate data revealed that the longest gradient length was 5.52 (> 3 standard deviations) in Lake Taihu and Chaohu.Therefore, canonical correspondence analysis (CCA) was used to explore relationships between the macrozoobenthos and environmental factors.In order to reduce multicollinearity among environmental factors, all variance inflation factors (VIFs) were kept below 10.The environmental factors influencing the composition of the macroinvertebrate were selected by 999 Monte Carlo permutation tests at the significant level (p < 0.01) (Wang et al. 2020).The environmental factors were ln (x + 1) transformed, except pH.In addition, the relative importance of physico-chemical factors and nutrients were assessed using the variation partitioning analysis (VPA).These analyses were performed using the software package CANOCO 4.5 (Ter Braak and Smilauer 2002) and the 'vegan' package in R software (version 3.4.3)(Oksanen et al. 2018).
One-way ANOVA and multiple comparison by Fisher's least significant differences (LSD) were used to determined significant differences (p < 0.05).The assumptions of normality were tested by the Kolmogorov-Smirnov test (software: SPSS 13.0).The repeated measures ANOVAs were performed by using SPSS 13.0 to analyse the effects of regions (eutrophic, transition and macrophyte) in four-seasons on shannon diversity, species richness and evenness index.

Environmental parameters
One-way ANOVA indicated that most environmental parameters (except T, DO and pH) were significantly different among 4 lake regions (p < 0.05, Table 1).Various aquatic plants were distributed in group 3, which were characterized by the lowest eutrophic levels (p < 0.05).In contrast, group 1 and group 2 had the greatest nitrogen and phosphorus nutrients and cyanobacteria blooms levels (p < 0.05).Sediments in group 4 were much sandier than in the other groups, however, the sediment was mainly silt in group 1 and group 2 (Table 1).

Macroinvertebrates communities
A total of 28 macroinvertebrate taxa were identified in Lake Taihu, including 8 Oligochaeta, 5 Chironomidae, 8 Mollusca, 2 Polychaeta and 5 other species.In Lake Chaohu, 21 taxa were identified, including 8 Oligochaeta, 7 Chironomidae, 3 Mollusca, 1 Polychaeta and 2 other species.Analysis of similarity (ANOSIM) and non-metric multidimensional scaling (NMDS) showed significantly different distribution patterns of macroinvertebrate assemblages among different groups (Figure 2).The spatial difference of macroinvertebrate community structure in Lake Chaohu was not significant.Oligochaeta and Chironomidae were exclusively dominant in the community structure of macroinvertebrates in Lake Chaohu, where the dominant species were Microchironomus sp., Limnodrilus hoffmeisteri, Aulodrilus pluriseta, Tubifex tubifex, and Tanypus chinensis.In Lake Chaohu, the Polychaetes Nephtys oligobranchira commonly occurred in western lake regions, but Molluscs were only collected in site C8.However, the community structure of macroinvertebrates varied significantly among different lake regions within Lake Taihu.Similar to Lake Chaohu, Oligochaeta, L. hoffmeisteri and T. tubifex were exclusively dominant in group 2. But the difference was that Corbicula fluminea and Bellamya aeruginosa commonly occurred in different sampling sites in group 2. The C. fluminea, Capitella sp. and N. oligobranchia were the dominant species in group 4. Diverse molluscs were abundant in group 3, and as indicator species of clean water, EPT species were only collected in this group, which was characterized by the highest shannon diversity, species richness and evenness (Figure 3).Repeated measures ANOVA showed that Shannon diversity, species richness and evenness were all significantly different among four groups, with higher values in groups 1 and 3 than groups 2 and 4 (Table 2, Figure 3).Whereas season was not a significant factor affecting the biological metrics, as the regional differences kept stable throughout the four seasons.

Correlations between macroinvertebrate community and environmental factors
The variation partition analysis (VPA) indicated that environmental factors explained 29.6% of macroinvertebrate community variability in Lake Taihu and Lake Chaohu.The physico-chemical factors explained 16.8% on the macroinvertebrate communities among the environmental factors.The nutrients explained a total of 8.6% of variation in the macroinvertebrate community.The physcio-chemical factors and nutrients explained 4.2% of the data variation in the macroinvertebrate community (Figure 4).The CCA analysis indicated that macrophytes, EC, NH 3 -N, COD Mn and temperature were detected as significant factors influencing these macroinvertebrate assemblage patterns between the two lakes.The eigenvalues of the first axis were 11.2% and 7.6%.The five significant factors explained 23.7% of the cumulative variance on the macroinvertebrate assemblages.The first axis showed the greatest correlations with COD Mn.The second axis showed the highest correlations with macrophytes (Figure 5a).
The CCA plot showed the relationships between macroinvertebrate species and environmental parameters (Figure 5b).EPT indicator insects (e.g.Helicosychidae and Perla sp.) and Gastropods (e.g. S. glabra and P. striatulus) were distributed in group 3 (group 3).However, the pollution-tolerant species (e.g. C. plumosus, A. pluriseta and L. hoffmeisteri) preferred in group 1 and group 2, which were characterised by highest COD Mn and nitrogen and phosphorus nutrients (group 1 and 2).Capitella sp., C. fluminea, and Gammarus sp. were mostly specific in group 4. Limnodrilus hoffmeisteri and Branchiura sowerbyi were clustered near the origin of the CCA plot, which indicated that these species were widely distributed in these two lakes (Figure 5b).

Discussion
This study found there were significant spatial differences in the community structure of macroinvertebrate within Lake Taihu, but no significant difference within Lake Chaohu.Macrophyte-dominated lake regions showed the highest speciess richness and diversity, and were characterized by dominance of gastropods.EPT sensitive species only occurred in these regions.On the contrary, algal-dominated lake regions had the lowest diversity and were dominated by pollution tolerate species.The CCA results indicated that macrophytes and eutrophic levels were most important factors influencing the community structure of macroinvertebrate.There were diverse aquatic plants with a coverage rate of more than 90%, and the main ecological types of aquatic vegetation in east Lake Taihu were submerged vegetation and floating leaf vegetation (Dong et al. 2014).As early as the 1970s, there were more than 50 aquatic vascular plant species in Lake Chaohu, but almost no macrophytes were found in Lake Chaohu in this study.The disappearance of macrophytes was one of the important reasons for the sharp decline in the number of macroinvertebrate in Lake Chaohu.
Many studies have demonstrated that the mechanism of influence of aquatic vegetation on macroinvertebrate was mainly reflected in the following aspects: (1) The higher spatial heterogeneity of aquatic plant areas could provide macroinvertebrate with higher ecological niches, and habitat diversity directly determines the diversity of macroinvertebrate (Galuppo et al. 2007, Pan et al. 2012;Shostell and Williams 2007); (2) Aquatic plants could offer sufficient food sources for macroinvertebrate and biofilms for scrapers to feed on (Gong, Xie, and Wang 2000;Tessier et al. 2004;Tews et al. 2004); (3) Submerged plants could provide gastropods with more favorable places for living and reproduction (Cai, Gong, and Qin 2012); (4) Aquatic plants could reduce the pollutants concentration by absorbing nutrient, and reduce the re-suspension of sediments to improve water transparency, thereby providing better habitats for macroinvertebrate (Scheffer 1998;Zhu et al. 2022); (5) Aquatic plants could also provide refuges against predation by fishes and other macroinvertebrate, greatly reducing the possibility of predation (Diehl 1988;Hargeby et al. 1994).
Due to the long-term discharge of industrial and agricultural pollutants, these two lakes suffered from severe eutrophication and algae blooms.Despite the higher eutrophic levels in the western area than in the eastern area, the community structure of macroinvertebrate tended to be simplified when eutrophication reached a certain level, resulting in no significant spatial difference in the community structure of macroinvertebrate within Lake Chaohu.Chironomids and Oligochaetes were dominant in group 1 and group 2. In previous study, the C. fluminea and gastropds were dominant in Lake Chaohu in the 1980s (Hu and Yao 1981).In this study, these Mollusca species almost disappeared in Lake Chaohu, but peaked at the sites T2 and T3 in Lake Taihu.The dynamic changes of the Mollusc species at the two sites deserved our attention in the future.The organic matters in the sediments of the eutrophic lake area were high in content and easily decomposable, and dissolved oxygen was quickly consumed during the decomposition of organic matters, resulting in an extremely low redox potential of the sediments in the algal lake area, which limited the survival of sensitive macroinvertebrate (Rasmussen and Kalff 1987;Cao et al. 2012;Obolewski et al. 2016).On the contrary, L. hoffmeisteri, T. chinensis and Microchironomus sp. were resistant species, with an extremely strong ability to tolerate pollution (Li et al. 2020).With the aggravation of eutrophication in the algal lake area, the pollution-resistant species had gradually replaced sensitive species, leading to a sharp decline in macrozoobenthos diversity (Devai and Moldovan 1983;Gong and Xie 2001;Timm, Mols, and Timm 2006).
In general, Polychaetes were found in the sea or in the Yangtze River estuary.In this study, Polychaete N. oligobranchia was found to be common and even dominant species in group 1 and group 4. Both lakes were connected to Yangtze River, which might contribute to species exchange with those in the East Sea or at the Yangtze River estuary.Previous studies showed that N. oligobranchia was also found in other large Yangtze-connected shallow lakes (e.g.Lake Poyang and Lake Hongze) (Zhang et al. 2012;Cai et al. 2014).
In addition, sediment types have an important impact on the composition and distribution of macroinvertebrate species (Li et al. 2019).The bivalve C. fluminea mainly fed by shredding and filter feeding.Sediments were relatively sandier and more oxidized, which are preferred by C. fluminea (Karatayev et al. 2003).Meanwhile, the disturbance from wind and waves caused the re-suspension of sediments, which provided suitable habitat and more food sources for C. fluminea (Cai et al. 2014).In contrast, Oligochaetes and Chironomids prefer to live muddy sediments in group 1 and group 4, which can reduce the living of mollusc by limiting their feeding behaviors (Schmidlin and Baur 2007).

Conclusions
The present study examined the spatio-temporal patterns of macroinvertebrate assemblages in two large freshwater Chinese lakes which subjected to different levels of eutrophication.The two large shallow freshwater lakes in the Yangtze River Basin have many similarities (location, climate, etc.), but Lake Taihu was found to have differences in assemblages within the lake and Lake Chaohu was found to have uniform assemblage across the entire lake.Macrophyte-dominated lake regions in Lake Taihu were characterised by the highest level of diversity and species richness than in the other lake regions.The EPT species only occurred in the macrophyte-dominated areas.Lake Chaohu and the northern Lake Taihu regions, with turbid algal-dominated state, were characterised by dominance of Oligochaetes and Chironomidae.These differences were attributed to varying habitats and environmental parameters among different lake regions.The results indicated that macrophytes, sediments characteristics and eutrophic levels were important factors influencing the community structure of macrozoobenthos between the two lakes.This study contributes to the scientific knowledge of within lake habitats, community assemblages, nutrients and pollution within these lakes.
currently working with School of Life Science, Huaibei Normal University, Huaibei City, China.Her research interests include protozoa.

Figure 1 .
Figure 1. the map of lake taihu and lake chaohu.

Figure 2 .
Figure 2. non-metric multidimensional scaling showed the difference of macroinvertebrate communities based on Bray-curtis similarity with absolute abundance.

Figure 3 .
Figure 3. Mean value and standard error (se) of relative abundance for invertebrate density, shannon-wiener diversity, species richness, and evenness at 4 region groups among 4 seasons.one-way anoVa and fisher's least significant difference (lsD) are used to determine the significant differences (p < 0.05).Different letters (i.e.a, b or c) indicate significant differences at p < 0.05.

Figure 4 .
Figure 4. Variance partitioning analysis showed the effect of environmental variables on macroinvertebrate communities.

Table 1 .
comparisons of environmental parameters among different lake regions.
Data are expressed as means ± se.one-way anoVa and fisher's least significant difference (lsD) are used to determine the significant differences (p < 0.05).Different letters (i.e.a, b or c) indicate significant differences at p < 0.05.Macrophytes,-: absent; +: present.

Table 2 .
results of repeated measures anoVas, showing the effects of regions (eutrophic, transition and macrophyte) in four-seasons on shannon diversity, species richness and evenness index.