The effect of wind speed decline on macroinvertebrates in Lake Taihu, China

The effect of wind speed decline on macroinvertebrates in Lake Taihu, China

Science of the Total Environment 662 (2019) 481–489 Contents lists available at ScienceDirect Science of the Total Environment journal homepage: www...

2MB Sizes 0 Downloads 7 Views

Science of the Total Environment 662 (2019) 481–489

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

The effect of wind speed decline on macroinvertebrates in Lake Taihu, China Kai Peng a,b, Yongjiu Cai a, Boqiang Qin a,b, Zhijun Gong a,b,⁎ a Taihu Laboratory for Lake Ecosystem Research, State Key Laboratory of Lake Science and Environment, Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences, Nanjing 210008, China b University of Chinese Academy of Sciences, Beijing 100049, China

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Wind speed decreased significantly in the past eight years in Lake Taihu. • Wind speed decline changed the structure of the macroinvertebrate community. • Different species respond differently to wind speed decline. • Wind speed affects macroinvertebrates mainly through three different aspects.

a r t i c l e

i n f o

Article history: Received 17 September 2018 Received in revised form 16 January 2019 Accepted 17 January 2019 Available online 23 January 2019 Editor: Daniel Wunderlin Keywords: Wind-wave disturbance Macroinvertebrate Wind speed decline

a b s t r a c t A decline in wind speed will have a series of effects on a lake ecosystem. Attention should be paid to macroinvertebrates, a critical ecosystem component, but few previous studies have considered these organisms. To study the influence of wind speed on macroinvertebrates, we selected Meiliang Bay and Zhushan Bay in Lake Taihu, which have high spatial homogeneity. The response of the benthic community and dominant species to a decrease in wind speed was studied using eight years of field survey data through NMDS analysis and regression analysis. The results showed that the decrease in wind speed significantly changed the structure of the macroinvertebrate community. Wind speed may affect macroinvertebrates directly or indirectly, mainly through wind-wave disturbance or changes in dissolved oxygen and food resources. The responses of different species and different growth stages to wind speed varied. According to the regression results, among the 8 most dominant species, the abundance of chironomids and malacostracans was positively correlated with wind speed, and that of bivalves and some oligochaetes was negatively correlated with wind speed. However, while the abundance of oligochaetes was negatively correlated with wind speed during the larval period, it was positively correlated with wind speed during the adult period. With future declines in wind speed, corresponding changes in the dominant species in Lake Taihu will have a series of effects on lake ecosystem, and more attention should be paid to these processes in future studies. © 2019 Elsevier B.V. All rights reserved.

⁎ Corresponding author at: Taihu Laboratory for Lake Ecosystem Research, State Key Laboratory of Lake Science and Environment, Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences, Nanjing 210008, China. E-mail address: [email protected] (Z. Gong).

https://doi.org/10.1016/j.scitotenv.2019.01.267 0048-9697/© 2019 Elsevier B.V. All rights reserved.

482

K. Peng et al. / Science of the Total Environment 662 (2019) 481–489

1. Introduction

2. Methods and materials

Wind-induced disturbance plays a vital role in biogeochemical cycling (Søndergaard et al., 1992) and aquatic community structure (Cai et al., 2012; Schallenberg and Burns, 2010; Wang et al., 2016) and has a complex influence on eutrophication, especially in large, shallow lakes. Wind-induced disturbances cause sediment resuspension (Li et al., 2007; Qin et al., 2004), sediment nutrient release (Søndergaard et al., 1992), and other changes in physicochemical conditions, which then have a direct or indirect impact on biological processes (de Faria et al., 2017; Huang et al., 2016; Shao et al., 2013; Wu and Hua, 2014). A decline in surface wind speed has been observed in many regions of the world, especially in the Northern Hemisphere (Guo et al., 2015; Mcvicar et al., 2012a; Vautard et al., 2010). Through a global review of 148 studies, Mcvicar et al. (2012b) revealed that the average wind speed in near-surface terrestrial environments has declined by 0.014 m s−1 a−1 during the past 30 years. In China, which is located in the Northern Hemisphere, the average rate of decrease in mean wind speed is −0.018 m s−1 a−1 (Guo et al., 2015). Global-scale warming and human activity may be responsible for the wind speed decline (Jiang et al., 2010; Xu et al., 2006), and the trend of wind speed decline will thus continue in the future. Many studies have focused on elucidating the effect of changes in wind speed on lakes. A regional decrease in wind velocity will cause lakes to become calmer (Wu et al., 2016). The upwelling and entrainment of deep-water nutrients into surface waters become less likely in large and deep lakes as a result of wind speed decline, and at the same time, the combined effect of increasing temperatures and decreasing wind speeds on aquatic ecosystem functions and services may be larger than the effect of local anthropogenic activity or overfishing (Cohen et al., 2016). Researchers are gradually beginning to pay attention to the impact of wind speed on freshwater ecosystems (Mcvicar et al., 2012b; Wu et al., 2016). In lake ecosystems, macroinvertebrates are a critical ecosystem component because they process and transfer organic material, promote nutrient cycling (Vannote et al., 1980; Vaughn and Hakenkamp, 2010) and influence the level of eutrophication. Numerous studies have investigated the effect of wind-generated surface waves on macroinvertebrates (Bertin et al., 2015; Brodersen, 1995). However, most of these studies considered this topic based on the community scale or trait-based reconstruction on a relatively short time scale. Bertin et al. (2015) determined through reconstruction that both environmental heterogeneity and wind flows are predominant in shaping the spatial structure of macroinvertebrate communities in a highaltitude wetland, and in Lake Taihu, Cai et al. (2017) found that windinduced disturbance and spatial variables were more important than anthropogenic factors in shaping these communities. Fewer studies have investigated the changes in response to wind decline at the species level. First, there are few lakes for which long-term survey data are available, especially shallow lakes, which are strongly influenced by wind-induced disturbance throughout the world. Additionally, due to differences in the characteristics among macroinvertebrate species, spatial heterogeneity in physical and chemical environmental factors (especially water depth and dissolved oxygen) is an important determinant of the distribution and community structure of macroinvertebrates, complicating efforts to quantitatively distinguish the importance of different factors. Based on past experience and previous studies, we predicted that wind-induced disturbance would directly or indirectly (through changing environmental factors) affect macroinvertebrates. Specifically, we predicted that the effects of wind-induced disturbance on benthic invertebrates would generally (1) change the community structure of the invertebrates and (2) affect species density, and considering the direct and indirect effects of wind-induced disturbance on macroinvertebrates, those with different feeding habits, life history characteristics, body sizes, etc., would be affected differently.

2.1. Study area and sampling sites Lake Taihu (Fig. 1), the third largest freshwater lake in China, is a large, shallow, eutrophic lake (mean depth = 1.9 m) located in the southern Changjiang (Yangtze) River Delta, which is strongly affected by wind-induced disturbance (Qin et al., 2007). The lake has an area of 2338.1 km2, a catchment area of 36,500 km2, a volume of 4.4 billion m3, a maximum length of 68.5 km (north-south direction) and a maximum width of 56 km (east to west) (Duan et al., 2009; Li et al., 2013; Qin, 2008). The wind directions around the lake are dominated by prevailing southeasterly winds in summer and northwesterly winds in winter, and the mean residence time is approximately 309 days (Qin, 2008). To understand the response of macroinvertebrates to the decline in wind speed, we mainly focused on the relationship between macroinvertebrates and wind-induced disturbance in the northern bays of Lake Taihu. The northern bays (Meiliang Bay and Zhushan Bay) are covered by soft organic sediment and have the most lacustrine sediment in the area (Qin et al., 2007). Thus, these bays provided a good degree of sediment spatial homogeneity for our research. 2.2. Sampling and sample analysis Benthic samples were collected quarterly at 8 stations (Fig. 1) from 2009 to 2016 in February, May, August, and November, representing winter, spring, summer, and autumn, respectively. The two bays are both covered by soft sediment and have high sediment spatial homogeneity. Samples were collected with three 0.025 m2 modified Peterson grabs and were sieved in situ through 250-μm-aperture mesh. The residual material retained in the sieve was transported to the laboratory on the same day. In the laboratory, the samples were sorted on a white tray, and the specimens were preserved in 7% buffered formalin solution. The specimens were identified to the species or genus level and counted. During invertebrate sampling, vertically integrated water samples and surface sediment samples were collected at the 8 locations, placed in an acid-cleaned plastic container, kept cool and sheltered, and then transported to the laboratory. We chose total nitrogen (TN) and total phosphorus (TP) to represent the anthropogenically induced stress to the benthic community, and chemical analyses of water and sediment samples included TN and TP, which were measured after conversion to soluble inorganic forms by wet persulphate oxidation (Ebina et al., 1983; Zhu et al., 2013). Daily wind data from 1 January 1970 to 31 December 2016 from the WuXi weather station (No. 58354, 31°36′N, 120°21′E), located near Meiliang Bay and Zhushan Bay, were obtained from the China Meteorological Data Sharing Service System (http://cdc.cma.gov.cn). Water level data were obtained from the Taihu Laboratory for Lake Ecosystem Research (TLLER), located on the shore of Meiliang Bay in northern Lake Taihu. 2.3. The relationship between wind speed decline and macroinvertebrate We use eight years of field survey data from a study area in northern Lake Taihu with relatively stable water quality and high homogeneity of the sediment. We used the mean macroinvertebrate abundances from the whole study area to eliminate the effect of different disturbance intensities among different sampling sites. The intensity of wind-induced disturbance was expressed by two variables: wind speed and water level. The effects of the wind-induced disturbance intensity on the macroinvertebrate community structure and dominant species were studied. We fitted equations for wind speed decline in 1970–2016 and 2000–2016 (the wind speed decline began to accelerate in 2000) and then used an equation combining wind speed and dominant

K. Peng et al. / Science of the Total Environment 662 (2019) 481–489

483

Fig. 1. The study area and the sampling sites.

macroinvertebrate species to predict the changes in macroinvertebrates and wind speed after 20 years. 2.4. Data analysis Non-metric multidimensional scaling (NMDS) analysis was used to show the changes in community structure during the eight years. To understand the relationship between the community structure of macroinvertebrates and wind speed, the wind speed data were mapped to the site scores from the NMDS analysis. The NMDS analysis used macroinvertebrate density data, and based on Bray-Curtis distances, produced three-dimensional solutions meeting the criterion of a final stress of b0.2 (Clarke, 1993). The NMDS analysis was performed using the function metaMDS in the R package vegan (Oksanen et al., 2016). Wind and water level data were mapped to the NMDS analysis site scores using the R package ggplot2 (Wickham, 2009). Permutational multivariate analysis of variance (PERMANOVA) (Anderson, 2001) was used to compare the differences between the different wind speed conditions. The intensity of disturbance is influenced by both wind speed and water depth. Because the impact of wind-induced disturbance on macroinvertebrates is persistent, we used quarterly averages of wind speed and water level data (corresponding to the times of sampling) to analyse the relationship between macroinvertebrates and wind speed. To explore the relationships among the individual species, wind speed and water level, stepwise multiple linear regressions with forward selection were used. We used the abundances of the eight most common species. Because the data were counts (which do not follow a normal distribution and are subject to large sampling errors), we adopted the method of sacrificing degrees of freedom to reduce the error by taking the mean of all 8 sites. For the mean-value data, we removed the zero values and used the empirical exponent of 0.25 to

transform the data to better approximate a normal distribution (Downing, 1979). These changes were made for two reasons. 1) The changes reduce the between-group error. The environmental factors differed slightly among different sampling sites, especially the distance from the shore, and the intensity of the wind-induced disturbance also differed among sites. The use of average values can solve these problems. 2) These changes reduced the within-group error, which also reduced the sampling error according to the central limit theorem. Thus, we believe that the mean is close to the real value. The analysis of variance (ANOVA) and regression analyses were carried out in R. For the ANOVA, if the data met the assumptions of normality and homogeneity of variances, then one-way ANOVA was used; otherwise, the nonparametric Kruskal-Wallis test was used. For more detailed methods, see Kabacoff (2011). Differences are reported as significant if p b 0.05. 3. Results 3.1. Structure of the macroinvertebrate community In total, 42 taxa were recorded from the 8 sites during the eight years, including five polychaetes, seven oligochaetes, 12 aquatic insects (chironomids), seven bivalves, five gastropods and six other species. Microchironomus tabarui, Corbicula fluminea, Paranthura sp., Grandidierella taihuensis, Tanypus chinensis, Limnodrilus hoffmeisteri, Rhyacodrilus sinicus, and Branchiura sowerbyi were the 8 most dominant taxa (their taxonomic group information is shown in Table 2). Owing to the high relative homogeneity of the sediment, the taxonomic composition varied little. The average macroinvertebrate abundance varied greatly among years between 931 ind./m2 and 3610 ind./m2, and the average macroinvertebrate abundance varied spatially from 1338 ind./m2 to 3031 ind./m2, showing distinct spatiotemporal heterogeneity. In general, because the sites were in an algae-dominated area,

484

K. Peng et al. / Science of the Total Environment 662 (2019) 481–489

Fig. 2. Change in TN and TP in the study area during 2009–2016.

pollution-tolerant oligochaetes were the most widely distributed species, and L. hoffmeisteri had the highest density (14,640 ind./m2) in the winter of 2010. 3.2. Variation in environmental factors Except for some outlier points, the yearly variation in the TP and TN at the eight sites did not obviously change during the ten years (Fig. 2). The one-way ANOVA yielded p values of 0.13 and 0.058 for water TN and TP and 0.128 and 0.555 for sediment TN and TP, respectively, and we accordingly cannot reject the null hypothesis. These results indicate that the nutrients did not change significantly in the eight years. This lack of change may indicate that changes in the benthic community composition are not caused primarily by anthropogenic effects.

Influenced by seasonal precipitation in the Lake Taihu basin, the water level had a strong seasonal trend (Fig. 3). The water level was stable before 2015, but it was significantly higher in 2016 than in other years. Over the past eight years, the wind speed has changed dramatically (Fig. 3). In 2009–2014, wind speed exhibited an obvious decline, while in 2015–2016, the wind speed tended to be stable. 3.3. Relationship between the benthic community and wind speed A three-dimensional MDS ordination analysis was used to display the benthic community composition. A two-dimensional MDS ordination (MDS1, MDS2) for the benthic macrofauna in samples during the eight years and the yearly average site scores is shown (Fig. 4). The yearly average scores suggest that the benthic community could

Fig. 3. The seasonal wind speed and water level changes during the 8 years.

K. Peng et al. / Science of the Total Environment 662 (2019) 481–489

485

Fig. 4. Two-dimensional MDS ordination and the relationship of site score with wind speed and water level, for which the values indicate the annual average site score.

be divided into three stage based on years: the first two years composed stage 1 (2009–2010), the next four years composed stage 2 (2011–2014), and the last two years composed stage 3 (2015–2016). A PERMANOVA was performed on these benthic MDS ordination score data, and the results (Table 1) revealed an overall significant difference among the three stages (R = 0.316, p b 0.001). At the same time, the wind speed, water level and the MDS ordination score were correlated (Fig. 4). Most of the stage 1 points were red, indicating that the wind speeds were higher than 2.75 m/s. Similar to the wind speed results, stage 2 points were lower than stage 1 points and higher than stage 3 points. Specifically, stage 3 points had lower wind speeds, and most of the speeds were lower than 2.25 m/s. Before 2015, the change in water level was relatively stable and thus had little influence on the MDS ordination scores. However, the water level varied greatly from 2015 to 2016, especially in summer, and thus had a strong influence on the MDS ordination scores. 3.4. The relationship between dominant species and wind-induced disturbance The linear regression results showed that 7 of the 8 most dominant taxa were significantly associated with wind speed and/or water level (Table 2, Fig. 5). The F-statistics showed that 5 of these taxa had a p value of b0.01. In the linear formula including the 7 dominant species, two of the species (M. tabarui and T. chinensis) were negatively

Table 1 The results of a PERMANOVA based on the similarity (Bray-Curtis distance) of standardized values. Significant differences between sample groups were determined using the R statistic, and the differences between groups were evaluated based on the significance level p. Global test: R = 0.316, p b 0.001 Sample group Stage 1 vs stage 2 Stage 1 vs stage 3 Stage 2 vs stage 3

Test statistic, R

Significance level, p

0.30 0.20 0.265

b0.001 b0.001 b0.001

correlated with wind speed and positively correlated with water level, which means that they were negatively correlated with wind-induced disturbance. Two of the species (C. fluminea and R. sinicus) were positively correlated with wind speed and negatively correlated with water level, which means that they were negatively correlated with wind-induced disturbance. One species (Paranthura sp.) was only negatively correlated with wind speed, and we can therefore consider it to have been negatively correlated with wind-induced disturbance. The last two dominant species were anomalous: they were negatively correlated (G. taihuensis) or positively correlated (L. hoffmeisteri) with wind speed and water level. 3.5. The changes in the dominant species in the future The water level has been stable during the past 16 years (Fig. 6). The variance analysis results showed that there was a significant difference in the annual variation of the water level between 2016 and some other years (2013, 2014, 2015, and 2011). Therefore, we hypothesize that there will be no significant change in the water level in Lake Taihu in the future. However, in recent years, the wind speed has fallen, especially in the last twenty years (Fig. 7). Therefore, according to the trends in wind change over the last twenty years and nearly fifty years, we fit two models (Fig. 6). If the wind speed in the future continues to decline according to current trends, we hypothesize that the relationship between the predation of fish and the increase in macroinvertebrates, especially chironomids and oligochaetes, is linear. Based on the model combining wind speed decline and dominant benthic species, in twenty years, there will be a series of changes in the number of populations of these species (Table 2). Based on the coefficient observed in the linear regression, species in Chironomidae experienced the largest proportional increases, and these species may increase by 40–70% in the future. R. sinicus had the largest proportional reduction, and in twenty years, it may be reduced by 23.5–40.0%. Although L. hoffmeisteri and C. fluminea exhibited a small proportional increase, they had a large starting number in Lake Taihu (both with an average density of N1000 ind./m2); thus, the absolute increases in these species are also larger, and they should therefore receive a considerable amount of attention.

486

K. Peng et al. / Science of the Total Environment 662 (2019) 481–489

Table 2 Linear equations of relationships between dominant species vs the wind speed and water level and the changes over the next twenty years. x: wind speed; y: water level. Fit1 and Fit2 represent using formula 1 and formula 2 to estimate species changes after 20 years, respectively. Phylum

Class

Species

Linear equation

r2

p

Fit1%

Fit2%

Arthropoda Mollusca Arthropoda Arthropoda Arthropoda Annelida Annelida Annelida

Insecta Bivalvia Malacostraca Malacostraca Insecta Oligochaeta Oligochaeta Oligochaeta

M. tabarui C. fluminea Paranthura sp. G. taihuensis T. chinensis L. hoffmeisteri R. sinicus B. sowerbyi

z = −1.71x + 2.29y − 1.01 z = 0.60x − 1.37y + 6.32 z = −0.57x + 3.72 z = −0.29x − 3.40y + 15.33 z = −1.98x + 1.64y + 2.21 z = 26.88x + 19.47y − 7.87xy − 61.37 z = 2.07x − 1.98y + 3.94 Not significant

0.497 0.19 0.12 0.23 0.44 0.13 0.48

0.001 0.02 0.052 0.011 0 0.10 0.015

45.5 −5.9 11.0 3.4 43.8 −7.8 −23.5

68.6 −9.7 17.6 5.5 66.3 −12.8 −40.0

4. Discussion 4.1. The relationship between macroinvertebrates and wind-induced disturbance Macroinvertebrates are affected by many factors, and although our study area was highly homogeneous, the nutrient changes were not

significant, and these effects could not be completely eliminated. Additionally, other factors were not considered, and all of these factors jointly determine the interannual fluctuations in macroinvertebrates. At the same time, seasonal fluctuations in macroinvertebrates are also important. However, reproduction and death are also affected by environmental factors, and the seasonal fluctuations are therefore both the cause and the result. It is difficult to eliminate the influence of seasonal

Fig. 5. The relationships between the dominant species and the wind speed and water level. Red lines represent the residuals; x, mean wind speed; y, water level; and z, number of individuals (empirical exponent of 0.25 transformed) (a. M. tabarui, b. C. fluminea, c. G. taihuensis, d. T. chinensis, e. L. hoffmeisteri, f. R. sinicus, and g. Paranthura sp.). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

K. Peng et al. / Science of the Total Environment 662 (2019) 481–489

487

Fig. 6. Changes in the average annual water level during 2001–2016 and the results of the analysis of variance for Lake Taihu. * indicates p b 0.05, ** indicates p b 0.01, *** indicates p b 0.001.

fluctuations. This lack of elimination may explain the modest r2 values of our linear regression models. The statistical results based on p values also showed that wind disturbances were significant factors affecting interannual fluctuations in macroinvertebrates. Wind-induced bottom disturbance is affected by the water level and wind speed. Wind-induced disturbances may affect macroinvertebrates in three ways. 1) Disturbances may increase the dissolved oxygen (DO) at the sediment-water interface, which will be beneficial for macroinvertebrates, especially species with a higher oxygen demand, for which DO may become the main factor limiting survival (Connolly et al., 2004). 2) Disturbances may directly influence macroinvertebrates. These direct effects may differ among different growth periods and different species, and disturbances may lower the survival rate of larvae and species with small body sizes. For Annelida and Arthropoda, high disturbance may cause more species loss. 3) Disturbance may alter the chances of obtaining food, and as such, filter-feeders may be among the first taxa to suffer from disturbance when the wind speed declines. The water level, in addition to the influence of disturbance intensity, also directly changes other environmental factors and then affects macroinvertebrates. On one hand, as the water depth increases, DO is

Fig. 7. Changes in the annual average wind speed over the last fifty years at WuXi station (No. 58354). Formula 1 (blue line) represents the trend for all years, and formula 2 (red line) represents the trend beginning in 2000. Confidence intervals are presented in grey. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

gradually reduced. On other hand, water level affects light and other conditions at the bottom of the water body, which may affect the growth of benthic algae and change both the abundance and composition of macroinvertebrate food. Owing to the different factors influencing macroinvertebrates both positively and negatively, it is difficult from a theoretical perspective to deduce the factors that ultimately influence all population changes, but we could potentially find some trends in data over long time series if the wind speed changed obviously. For the 8 most dominant taxa in our study, two Chironomidae species (M. tabarui and T. chinensis) were negatively associated with disturbance. In the adult stage, because the adults have reduced biting mouthparts and do not feed, they can survive only several weeks at most (Oliver, 2003), and because high wind speeds can decrease the mating rate, during the short period of survival, many of them may not have a chance to mate. On the other hand, when the protective gelatinous matrix surrounding the eggs expands in water, wind can also influence the hatching rate. In the larvae stage, both of these species are small gatherers, and thus wind speed decline will less directly affect them, while more clams in the lake sediment environment will provide beneficial feeding conditions for the larvae. One of the dominant taxa responding positively to wind, B. sowerbyi, is in Oligochaeta. To escape predation, most adult oligochaetes and their cocoons in certain areas are found in sediment layers, despite hypoxic environmental conditions (Newrkla and Mutayoba, 1987; Newrkla and Wijegoonawardana, 1987). Therefore, the direct influence of disturbance may be weak in Oligochaeta, but since disturbance can increase DO in the surface sediment, it is reasonable to expect such taxa to have a positive relationship with disturbance. C. fluminea also responded positively to wind speed, but the coefficient of the equation was slightly different from that of the other dominant species. The absolute value of the ratio of the independent variable coefficients of the most dominant species was close to 1 (1 ± 0.38), which means that wind speed-induced disturbance and the water level had similar effects on biological processes. The absolute value of this ratio for C. fluminea was 0.3, which means that the water level had a stronger impact than did wind speed. In fact, when the larvae are released from the mother, they settle and burrow into the substratum (Cataldo and Boltovskoy, 1998). Due to protection by the shell, the impact of disturbance on C. fluminea is limited, regardless of whether the given individual is a larva or an adult. Hence, disturbance may influence C. fluminea through food and oxygen. C. fluminea has a high demand for oxygen (Sousa et al., 2010). For filter-feeders, disturbance can increase the

488

K. Peng et al. / Science of the Total Environment 662 (2019) 481–489

Fig. 8. The relationship between wind speed and the dominant species (L. hoffmeisteri and B. sowerbyi) at different growth stages. Confidence intervals are presented in grey.

chance of obtaining food, and previous studies have shown that the suspended silt concentration has no effect on the growth of such organisms (Foe and Knight, 1985). Thus, the wind speed and water level each have a positive relationship with C. fluminea. Because C. fluminea is a filter-feeder, uses phytoplankton as a food source and has a high demand for oxygen, the direct influence of the water level on this species may be strong. Based on these characteristics, combined with the linear model, we may infer that the direct impact of the water level may be higher than that of wind-induced disturbance, although both have an impact. The other two dominant species of oligochaetes were L. hoffmeisteri and B. sowerbyi. It seems unlikely that L. hoffmeisteri would have a positive relationship with both the water level and the mean wind speed but, instead, would show a significant negative interaction between the water level and mean wind speed. Because deeper water is associated with less oxygen and weaker disturbance, generally, it is reasonable for the two independent variables to have opposite effects. At the same time, there was no statistically significant correlation between the two independent variables in the other oligochaete. In fact, the opposite results seem reasonable when we consider the different growth stages of the species. We divided the two growth stages of oligochaetes into young and adult stages. According to previous studies (Carroll and Dorris, 1972; Yan and Liang, 2002) and the practice of picking samples in the laboratory, more larvae of these taxa occur than adults in August and November, while in February and May, adult individuals are dominant. During the adult period, the abundance of both L. hoffmeisteri and B. sowerbyi was positively associated with wind speed, while in the larval stage, the abundance of each was negatively associated with wind speed (Fig. 8). In Tubificidae, however, during the reproductive season, owing to their sexual propagation (Kennedy, 1965, 1966), wind-induced disturbance reduced their reproductive success ratio. However, wind speed can directly affect the survival rate of larvae through disturbance, whereas in the adult stage, the effect of disturbance becomes limited, and the increase in bottom DO caused by disturbance promotes growth. The abundances of both malacostracans (Paranthura sp. and G. taihuensis) were negatively correlated with wind speed. The malacostracans produce a current of water that moves and reduces the thickness of the boundary layer that impedes the diffusion of oxygen and other gases over the gills (Sutcliffe, 1984). The abdominal muscles in these two species are well developed, and the species have strong athletic abilities and diverse feeding habits (Chadderton et al., 2003; Väinölä et al., 2008). Therefore, the impact of

food resources on the species is very limited, and the two species may be influenced by wind speed through direct disturbance and oxygen changes. The effects of disturbance on the larvae may be greater than those on adults because the larvae do not burrow into the surface sediments, but at the same time, the larval body has a high demand for oxygen (Sutcliffe, 1984). In general, disturbance has a direct negative effect on the larvae but a positive indirect effect on them by increasing DO, while during the adult period, the effect of disturbance and DO will become very limited. Therefore, the coefficient of the equation represented only the comprehensive effect of the disturbance caused by wind speed. 4.2. Influence of wind-induced disturbance on the benthic community structure The results of the NMDS analysis showed that the structure of the benthic community changed with the decrease in wind speed. Different species have different living habits, reproductive development times, body sizes, DO demands, and food compositions. For rare species, owning to the low frequency of occurrence, it is difficult to predict the effects of wind speed decline through field investigation, and it is therefore difficult to accurately predict the evolution of the community structure in a specific direction. In recent years, given the lack of a fundamental change in the eutrophication level, the main factors influencing the structure of the macroinvertebrate community in Lake Taihu were at the trophic level. However, our study also showed that a change in wind speed changed the structure of the macroinvertebrate community within a certain range. 5. Conclusion Increasing attention has been paid to the impact of significant wind speed decline on lake ecosystems. Here, through eight years of field investigation, we analysed the effects of wind speed decline on macroinvertebrates in Meiliang Bay and Zhushan Bay of Lake Taihu. The results of NMDS analysis based on benthic communities show that the decline of wind speed changes the community structure of benthic organisms. The results of multiple regression analysis show that there is a significant correlation between wind speed and typical dominant species. Wind speed affects macroinvertebrates mainly through windwave disturbances or changes in dissolved oxygen and food resources.

K. Peng et al. / Science of the Total Environment 662 (2019) 481–489

Acknowledgements The study was jointly supported by the National Natural Science Foundation of China [grant number 41671110, 41621002, 41501215, 41661134036] the Key Research Program of Frontier Sciences, CAS [grant number QYZDB-SSW-DQC016] and the “One-Three-Five” Strategic Planning of NIGLAS [grant number NIGLAS2017GH05]. References Anderson, M.J., 2001. A new method for non-parametric multivariate analysis of variance. Austral Ecol. 26 (1), 32–46. Bertin, A., Alvarez, E., Gouin, N., Gianoli, E., Montecinos, S., Lek, S., Gascoin, S., Lhermitte, S., 2015. Effects of wind-driven spatial structure and environmental heterogeneity on high-altitude wetland macroinvertebrate assemblages with contrasting dispersal modes. Freshw. Biol. 60 (2), 297–310. Brodersen, K.P., 1995. The effect of wind exposure and filamentous algae on the distribution of surf zone macroinvertebrates in Lake Esrom, Denmark. Hydrobiologia 297 (2), 131–148. Cai, Y., Gong, Z., Qin, B., 2012. Benthic macroinvertebrate community structure in Lake Taihu, China: effects of trophic status, wind-induced disturbance and habitat complexity. J. Great Lakes Res. 38 (1), 39–48. Cai, Y., Xu, H., Vilmi, A., Tolonen, K.T., Tang, X., Qin, B., Gong, Z., Heino, J., 2017. Relative roles of spatial processes, natural factors and anthropogenic stressors in structuring a lake macroinvertebrate metacommunity. Sci. Total Environ. 601–602, 1702–1711. Carroll, J.H., Dorris, T.C., 1972. The life history of Branchiura sowerbyi. Am. Midl. Nat. 87 (2), 413–422. Cataldo, D., Boltovskoy, D., 1998. Population dynamics of Corbicula fluminea (Bivalvia) in the Paraná River Delta (Argentina). Hydrobiologia 380 (1–3), 153–163. Chadderton, W.L., Ryan, P.A., Winterbourn, M.J., 2003. Distribution, ecology, and conservation status of freshwater Idoteidae (Isopoda) in southern New Zealand. J. R. Soc. N. Z. 33 (2), 529–548. Clarke, K.R., 1993. Non-parametric multivariate analyses of changes in community structure. Austral Ecol. 18 (1), 117–143. Cohen, A.S., Gergurich, E.L., Kraemer, B.M., McGlue, M.M., McIntyre, P.B., Russell, J.M., Simmons, J.D., Swarzenski, P.W., 2016. Climate warming reduces fish production and benthic habitat in Lake Tanganyika, one of the most biodiverse freshwater ecosystems. Proc. Natl. Acad. Sci. U. S. A. 113 (34), 9563–9568. Connolly, N.M., Crossland, M.R., Pearson, R.G., 2004. Effect of low dissolved oxygen on survival, emergence, and drift of tropical stream macroinvertebrates. J. N. Am. Benthol. Soc. 23 (2), 251–270. de Faria, D.M., Cardoso, L.D., Marques, D.D., 2017. Epiphyton dynamics during an induced succession in a large shallow lake: wind disturbance and zooplankton grazing act as main structuring forces. Hydrobiologia 788 (1), 267–280. Downing, J.A., 1979. Aggregation, transformation, and the design of benthos sampling programs. Journal De L'office Des Recherches Sur Les Pêcheries Du Canada 36 (12), 1454–1463. Duan, H.T., Ma, R.H., Xu, X.F., Kong, F.X., Zhang, S.X., Kong, W.J., Hao, J.Y., Shang, L.L., 2009. Two-decade reconstruction of algal blooms in China's Lake Taihu. Environ. Sci. Technol. 43 (10), 3522–3528. Ebina, J., Tsutsui, T., Shirai, T., 1983. Simultaneous determination of total nitrogen and total phosphorus in water using peroxodisulfate oxidation. Water Res. 17 (12), 1721–1726. Foe, C., Knight, A., 1985. The effect of phytoplankton and suspended sediment on the growth of Corbicula fluminea (Bivalvia). Hydrobiologia 127 (2), 105–115. Guo, H., Xu, M., Hu, Q., 2015. Changes in near-surface wind speed in China: 1969–2005. Int. J. Climatol. 31 (3), 349–358. Huang, J., Xi, B.D., Xu, Q.J., Wang, X.X., Li, W.P., He, L.S., Liu, H.L., 2016. Experiment study of the effects of hydrodynamic disturbance on the interaction between the cyanobacterial growth and the nutrients. J. Hydrodyn. 28 (3), 411–422. Jiang, Y., Luo, Y., Zhao, Z., Tao, S., 2010. Changes in Wind Speed Over China During 1956–2004. p. 421. Kabacoff, R., 2011. R in Action. Manning Publications Co. Kennedy, C.R., 1965. The distribution and habitat of Limnodrilus claparède (Oligochaeta: Tubificidae). Oikos 16 (1/2), 26–38. Kennedy, C.R., 1966. The life history of Limnodrilus udekemianus Clap. (Oligochaeta: Tubificidae). Oikos 17 (1), 10–18.

489

Li, T., Wang, D.S., Zhang, B., Liu, H.J., Tang, H.X., 2007. Morphological characterization of suspended particles under wind-induced disturbance in Taihu Lake, China. Environ. Monit. Assess. 127 (1–3), 79–86. Li, Y., Tang, C., Wang, C., Tian, W., Pan, B., Hua, L., Lau, J., Yu, Z., Acharya, K., 2013. Assessing and modeling impacts of different inter-basin water transfer routes on Lake Taihu and the Yangtze River, China. Ecol. Eng. 60 (11), 399–413. Mcvicar, T.R., Roderick, M.L., Donohue, R.J., Niel, T.G.V., 2012a. Less bluster ahead? Ecohydrological implications of global trends of terrestrial near-surface wind speeds. Ecohydrology 5 (4), 381–388. Mcvicar, T.R., Roderick, M.L., Donohue, R.J., Li, L.T., Niel, T.G.V., Thomas, A., Grieser, J., Jhajharia, D., Himri, Y., Mahowald, N.M., 2012b. Global review and synthesis of trends in observed terrestrial near-surface wind speeds: implications for evaporation. J. Hydrol. 416 (3), 182–205. Newrkla, P., Mutayoba, S., 1987. Why and where do oligochaetes hide their cocoons? Hydrobiologia 155 (1), 171–178. Newrkla, P., Wijegoonawardana, N., 1987. Vertical distribution and abundance of benthic invertebrates in profundal sediments of Mondsee, with special reference to oligochaetes. Hydrobiologia 155 (1), 227–234. Oksanen, J., Kindt, R., Legendre, P., Hara, B.O., Henry, M., Stevens, H., 2016. vegan Community Ecology Package Version 2.4-1 September 2016. Oliver, D.R., 2003. Life history of the Chironomidae. Annu. Rev. Entomol. 16 (1), 211–230. Qin, B., 2008. Lake Taihu, China. Springer Netherlands. Qin, B., Hu, W., Gao, G., Luo, L., Zhang, J., 2004. Dynamics of sediment resuspension and the conceptual schema of nutrient release in the large shallow Lake Taihu, China. Chin. Sci. Bull. 49 (1), 54–64. Qin, B., Xu, P., Wu, Q., Luo, L., Zhang, Y., 2007. Environmental issues of Lake Taihu, China. Hydrobiologia 581 (1), 3–14. Schallenberg, M., Burns, C.W., 2010. Effects of sediment resuspension on phytoplankton production: teasing apart the influences of light, nutrients and algal entrainment. Freshw. Biol. 49 (2), 143–159. Shao, K.Q., Gao, G., Tang, X.M., Wang, Y.P., Zhang, L., Qin, B.Q., 2013. Low resilience of the particle-attached bacterial community in response to frequent wind-wave disturbance in freshwater mesocosms. Microbes Environ. 28 (4), 450–456. Søndergaard, M., Kristensen, P., Jeppesen, E., 1992. Phosphorus release from resuspended sediment in the shallow and wind-exposed Lake Arresø, Denmark. Hydrobiologia 228 (1), 91–99. Sousa, R., Rufino, M., Gaspar, M., Antunes, C., Guilhermino, L., 2010. Abiotic impacts on spatial and temporal distribution of Corbicula fluminea (Müller, 1774) in the River Minho estuary, Portugal. Aquat. Conserv. Mar. Freshwat. Ecosyst. 18 (1), 98–110. Sutcliffe, D.W., 1984. Quantitative aspects of oxygen-uptake by Gammarus (Crustacea, Amphipoda) - a critical-review. Freshw. Biol. 14 (5), 443–489. Väinölä, R., Witt, J.D.S., Grabowski, M., Bradbury, J.H., Jazdzewski, K., Sket, B., 2008. Global diversity of amphipods (Amphipoda; Crustacea) in freshwater. Hydrobiologia 595 (1), 241–255. Vannote, R.L., Minshall, G.W., Cummins, K.W., Sedell, J.R., Cushing, C.E., 1980. The river continuum concept. Can. J. Fish. Aquat. Sci. 37 (2), 130–137. Vaughn, C.C., Hakenkamp, C.C., 2010. The functional role of burrowing bivalves in freshwater ecosystems. Freshw. Biol. 46 (11), 1431–1446. Vautard, R., Cattiaux, J., Yiou, P., Thépaut, J.N., Ciais, P., 2010. Widespread Land Surface Wind Decline in the Northern Hemisphere. American Geophysical Union. Wang, H., Zhang, Z., Liang, D., Du, H., Pang, Y., Hu, K., Wang, J., 2016. Separation of wind's influence on harmful cyanobacterial blooms. Water Res. 98, 280–292. Wickham, H., 2009. ggplot2: Elegant Graphics for Data Analysis. Springer Publishing Company, Incorporated. Wu, D., Hua, Z.L., 2014. The effect of vegetation on sediment resuspension and phosphorus release under hydrodynamic disturbance in shallow lakes. Ecol. Eng. 69, 55–62. Wu, T., Huttula, T., Qin, B., Zhu, G., Ropponen, J., Yan, W., 2016. In-situ erosion of cohesive sediment in a large shallow lake experiencing long-term decline in wind speed. J. Hydrol. 539, 254–264. Xu, M., Chang, C.P., Fu, C., Qi, Y., Robock, A., Robinson, D., Zhang, H.M., 2006. Steady decline of east Asian monsoon winds, 1969–2000: evidence from direct ground measurements of wind speed. J. Geophys. Res. 111 (D24). Yan, Y.J., Liang, Y.L., 2002. Abundance and production of Limnodrilus hoffmeisteri (Oligochaeta: tubificidae) in algae-dominated Lake Houhu (Wuhan, China). Chin. J. Oceanol. Limnol. 20 (1), 81–85. Zhu, M.Y., Zhu, G.W., Zhao, L.L., Yao, X., Zhang, Y.L., Gao, G., Qin, B.Q., 2013. Influence of algal bloom degradation on nutrient release at the sediment-water interface in Lake Taihu, China. Environ. Sci. Pollut. Res. 20 (3), 1803–1811.