Response of benthic diatom assemblages to environmental heterogeneity in an urban river: implications for environmental monitoring

River systems play a crucial role in supporting human activities, including urban water supply, flood mitigation, and disposal of wastewater (Findlay & Taylor 2006). The health and stability of river ecosystems are essential for the sustainable development of urban areas. However, rapid economic development and continuous growth of urbanization are profoundly impacting the ecological status of rivers, especially in plain river network regions (Ren et al. 2022).

Benthic diatoms play a fundamental ecological role, significantly supporting primary productivity and nutrient cycles in river ecosystems (Masouras et al. 2021). Their rapid reproduction makes them more sensitive to environmental changes than other biotic groups (Seckbach & Kociolek 2011). Due to their distinctive morphological and ecological characteristics, benthic diatoms are considered one of the best bioindicators of environmental changes. Moreover, changes in benthic diatom assemblages can impact entire river ecosystems (Stubbington et al. 2017). Consequently, there is increasing attention on exploring the interaction between benthic diatom assemblages and environmental factors.

Environmental filtering is considered a major mechanism for structuring benthic diatom assemblages (Wu et al. 2021). Environmental factors determine community composition directly by affecting survival and indirectly by shaping competition (Cadotte & Tucker 2017). The interaction between benthic diatom assemblages and environmental factors has been extensively studied in various aquatic ecosystems, including lakes (Rivera et al. 2018), rivers (Jakovljević et al. 2016), reservoirs (Yang et al. 2023), and estuaries (Nunes et al. 2022). In rivers, benthic diatoms are particularly valuable as biological indicators for water monitoring (Masouras et al. 2021). Numerous studies have focused on understanding how benthic diatom assemblages respond to environmental variables, such as nutrients (Schneider et al. 2013), acidity (Andrén & Jarlman 2008), and hydrological alterations (Falasco et al. 2021). Despite extensive research on benthic diatoms in different river types, it remains unclear how environmental heterogeneity shapes the distribution of benthic diatom assemblages in urban rivers with highly heterogeneous habitats.

In this study, the succession of benthic diatoms during the flood and dry seasons was evaluated along the CaoE River in the urban plain river network of East China. Environmental factors were simultaneously measured within the benthic diatom habitats. The relationship between benthic diatom assemblages and environmental heterogeneity was further analyzed using redundancy analysis (RDA) and the Mantel test. Overall, the results enhance the understanding of the indicative role of benthic diatoms in the environmental assessment of urban rivers.

Physical parameters, including water temperature, pH, conductivity, and dissolved oxygen, were measured in situ using a portable multi-parameter water analyzer (Leici DZB-712, China). Surface water samples were collected for the analysis of chemical variables. The concentration of total nitrogen was determined by employing alkaline potassium persulfate digestion-UV spectrophotometry (Shimadzu UV-2600, Japan) (Ren et al. 2015). The concentration of total phosphorus was measured using a colorimetric molybdenum blue assay (Cao et al. 2016). The potassium permanganate index (COD_Mn) was quantified using a titration method by acid digestion with potassium permanganate and oxalic acid (Liu et al. 2021). To measure the chlorophyll a content, 500 mL of water was filtered through Whatman GF/C glass fiber filters, followed by extraction using 90% acetone. Four wavelengths (630, 645, 663, 750 nm) of acetone solution were recorded by a spectrophotometer (Shimadzu UV-2600, Japan). Chlorophyll a content was calculated according to Zhou et al. (2016).

Benthic diatoms on cobbles were collected for identification and enumeration following the methods of Kelly et al. (1998) and Wang et al. (2022). In brief, five submerged cobbles were picked randomly along the bank (50 m) to minimize random errors in diatom collection. Biofilms on cobbles were carefully scrapped using a sterile toothbrush. The cobbles were further rinsed with distilled water. All samples were thoroughly mixed in a stainless steel basin and preserved using 4% formaldehyde. Diatom collection at all sample sites followed the same procedure. Organic matter was digested using 35% hydrochloric acid and 30% hydrogen peroxide at 95°C for 4 h. The digestion solution was centrifuged to remove the supernatant, and the sediment was rinsed several times with distilled water. Cleaned diatoms were stored in absolute ethyl alcohol (Yang et al. 2023). For the identification of benthic diatoms, a light microscope with 1,000× magnification (Olympus BX-53, Japan) was used. A minimum of 350 frustules per sample was identified as the lowest possible taxonomic level following Lange-Bertalot et al. (2017), Krammer (1986, 1988, 1991), and Krammer & Lange-Bertalot (1991). Taxonomic names were verified using AlgaeBase (Guiry & Guiry 2020). The top 10 genera were selected for statistical analyses.

Alpha diversity of benthic diatom communities was calculated at the species level using PC-ORD 5 software. The multi-response permutation procedure (MRPP), analysis of similarities (ANOSIM), and Mantel test were conducted using R 4.0.2 (R Core Team). MRPP is a class of multivariate permutation tests of group difference (Cai 2006). Observed delta is the weighted mean within-group distance from actual data; expected delta is the weighted mean distance from permutations; A value is the chance corrected within-group agreement; p values are the significance of delta (Bakker 2023a). ANOSIM is a non-parametric technique based on ranks (Bakker 2023b), which was conducted to evaluate the temporal differences in benthic diatom communities. A Mantel test was performed to investigate the relationships between benthic diatom communities, environmental variables, and productivity. The Euclidean distance method was used for the calculation of environmental variables and productivity distances, and the Bray–Curtis distance method was used for the calculation of dominant genera and biodiversity distances. RDA was performed to analyze the relationship between diatom communities and environmental variables by using Canoco 4.5 software.

In terms of spatial heterogeneity (Figure 3(b) and 3(c)), significant differences were observed in conductivity in both the flood and dry seasons. A decreasing gradient was observed from downstream (pool habitat) to upstream (riffle habitat). This gradient is typical in coastal freshwater ecosystems with semi-tidal movements. According to Aslan et al. (2018), the flux and reflux of tides can affect water levels of coastal freshwater ecosystems, leading to conductivity increases of up to 4.6 times during flood tides. However, the construction of a dam in the CaoE River has blocked tidal effects, resulting in reduced estuarine conductivity and stabilizing the spatial gradient of conductivity.

In addition, chlorophyll a content showed a decreasing gradient from downstream (pool habitat) to upstream (riffle habitat). Given the uniformity of nutrient levels along the CaoE River (Figure 3(b) and 3(c)), it was suggested that nutrients were not the primary factor driving the high chlorophyll a content. Instead, the limited exchange capacity of the downstream water body, which promotes phytoplankton growth (Wu et al. 2022), appears to play the predominant role in the formation of higher chlorophyll a contents.

To assess the temporal similarity of diatom composition, MRPP and ANOSIM analyses were performed. The results showed a significant dissimilarity between the flood and dry seasons (MRPP: A value = 0.154, P = 0.001; ANOSIM: R = 0.981, P = 0.001) (Table 1). Significant differences were observed in species richness and Pielou's index rather than Shannon Wiener's index and Simpson's index (Figure 4(c)–4(f)). In the flood season (Figure 4(a)), dominant genera (defined as those with a relative abundance ≥10%) were Nitzschia (28.7%), Navicula (21.0%), and Melosira (14.0%). Dominant species (defined as those with a relative abundance ≥5%) included Nitzschia palea (24.1%), Melosira varians (14.0%), Cyclotella meneghiniana (8.9%), and Navicula rostellata (5.6%). Among these, N. palea, M. varians, and C. meneghiniana belong to pollution-tolerant species. Nitzschia palea is commonly found in α-meso to polysaprobic water (Beauger et al. 2020), while M. varians and C. meneghiniana exhibit high resistance to organic pollution (Szczepocka & Szulc 2009; Fontana et al. 2014). These findings are consistent with the evaluated COD_Mn levels observed during the flood season (Figure 3(a)).

In the dry season, changes in the composition of benthic diatom assemblages reflected changes in the habitat environment (Figure 4(b)). Dominant genera were Melosira (16.5%), Nitzschia (15.5%), and Achnanthidium (13.5%). Dominant species were M. varians (16.5%), Achnanthidium minutissimum (10.1%), N. palea (9.7%), C. meneghiniana (8.1%), and Cymbella turgidula (5.6%). Among these, A. minutissimum is considered an indicator of anthropogenic pressure, particularly associated with high nutrient levels (Zelnik & Sušin 2020). This change may be attributed to the increased total nitrogen during the dry season.

In general, conductivity emerged as one of the most important factors in shaping benthic diatom assemblages, consistent with findings from other studies (Stenger-Kovács et al. 2023). It has been reported that conductivity not only affects species composition but also the trait and guild of diatom communities. Elevated conductivity leads to the selection of conductivity-tolerant taxa, such as N. palea and C. meneghiniana (Stenger-Kovács et al. 2020; Tapolczai et al. 2021). Simultaneously, species richness and diversity of benthic diatoms tend to decrease at elevated conductivity in freshwater ecosystems (Perrett et al. 2021). In terms of changes in trait and guild composition, high conductivity tends to push assemblages toward a higher proportion of trait extremes, such as small or large-size categories (Stenger-Kovács et al. 2023). In addition, an increase in conductivity also elevates the proportion of motile diatoms (Stenger-Kovács et al. 2020). On the other hand, total nitrogen significantly explained the variance of benthic diatom assemblages during the flood season. This phenomenon aligns with observations from a study of a tropical river in Brazil (Bere & Tundisi 2011). It was proposed that the succession of benthic diatom assemblages along the gradient of conductivity was driven in part by nutrients.

During the dry season, the combined effects of total nitrogen, total phosphorus, and chlorophyll a provided a good explanation for the total variance. Mantel's test revealed significant correlations between total nitrogen and chlorophyll a with dominant genera of benthic diatoms (p < 0.01), and biodiversity was significantly related to total phosphorus (p < 0.05). RDA analysis (Figure 8(b)) showed that total nitrogen, chlorophyll a, and total phosphorus explained 24.6, 18.7, and 8.1% of the total variance, respectively. Consistent with the flood season, total nitrogen remained a primary factor influencing diatom composition during the dry season. Evaluated nitrogen concentrations may stimulate the growth of benthic diatoms while potentially leading to the extinction of certain taxa (Boulêtreau et al. 2006). In this study, total nitrogen was positively correlated with Nitzschia but negatively correlated with Gomphonema and Cocconeis (Figure 8(d)).

Recent research has highlighted the impact of productivity parameters, such as total phosphorus and chlorophyll a, on benthic diatom assemblages (Yang et al. 2023). Virta et al. (2019) reported a strong relationship between productivity and diatom diversity. Yang et al. (2023) also found that productivity indicators such as total phosphorus and chlorophyll a had a significant positive effect on diatom dissimilarity. In this study, chlorophyll a was positively associated with genera such as Cyclotella, Cymbella, and Aulacoseira, whereas Nitzschia and Navicula showed positive correlations with total phosphorus. Total phosphorus, as a productivity indicator, also showed a significant relationship with the diversity of benthic diatoms (Figure 7(b)). Therefore, benthic diatom assemblages are highly related to environmental variables, making them a reliable and effective tool for monitoring and assessing environmental conditions.

Benthic diatom assemblages can be good indicators of environmental conditions. This study aimed to assess how benthic diatom assemblages respond to environmental variations along the CaoE River in the plain river network of East China. The results revealed temporal heterogeneity in water temperature, dissolved oxygen, total nitrogen, and COD_Mn, and spatial heterogeneity in conductivity and chlorophyll a. In addition, the structure of benthic diatom assemblages also exhibited significant spatio-temporal variations. The dominant diatoms indicated that the water body of the CaoE River was under the condition of eutrophic and organic pollution. Statistical analyses highlighted the importance of environmental variables (such as conductivity, total nitrogen, total phosphorus, and chlorophyll a) in shaping benthic diatom assemblages in the CaoE River. Hence, this study explored how environmental heterogeneity influences the distribution and diversity of benthic diatoms in an urban river within the plain river network of East China. Furthermore, it provides an applicable way for the development and improvement of river monitoring and assessment with highly heterogeneous habitats.

This work was supported by the Shaoxing Natural Science Foundation (Grant No. 2022A13008) and the Shanxi Province Science Foundation for Youths (Grant No. 202203021222006).

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.