Introduction

The rapid population growth and the intensifying human activities in the late twentieth century is harming the water ecology and the environment locally, regionally, and even globally. Problems such as the drying up of rivers, biodiversity loss, and water pollution seriously damage the structural function of watersheds [3, 39]. Since a healthy water ecosystem is indispensable to the sustainable utilization of water resources, ecological security, and the sustainable economic and social development for a watershed and a region, comprehensive methods are urgently needed to assess the status of ecosystems and monitor their change [43]. The fundamental monitoring methods that provide detailed information for correct water management can be classified into physical, chemical, and biological methods [34]. Biological monitoring methods can complement traditional physical and chemical monitoring methods [48], which tend to fall short in the following aspects: (1) they cannot comprehensively reflect the response of the ecosystem to the external pollution, (2) they cannot provide information on the effects of environmental changes on the biological communities in the water environment, and (3) they cannot truly reflect the situation in the vicinity of the monitoring points. This is especially relevant for flowing water bodies, for which accurate assessments can be difficult due to the rapidly changing hydrology.

As primary producers, phytoplankton is at the bottom of the food chain in aquatic ecosystems [56]. They have short life cycles and are sensitive to pollutants, and their composition is highly specific to the water body [69]. The nature and quantity of the phytoplankton communities change with the chemical composition of the water. Hence, phytoplankton is often used as an important indicator in monitoring water quality [30, 63]. Diatoms are photoautotrophic eukaryotic phytoplankton and are important components of natural water bodies. They are widely found in various water bodies such as rivers, streams, and lakes. They have high species diversity and short reproduction cycles, and can be easily collected and preserved [27, 49]. Diatoms are extremely sensitive to the changes in the water environment, such as water temperature, pH value, conductivity, nutrient concentration, and heavy metal levels [6]. For example, Cyclotella meneghiniana can indicate the nutritional status of water [15] and is also sensitive to water pH [66]. Achnanthes nodusa [46] and Synedra ulna [20] can indicate the level of Cu and Zn in the water body, respectively. Synedra acus, Diatoma hiemale, and Gomphonema parvulum can also reflect the pollution status of the water body [51]. Since the 1990s, researchers from the European Union, the United States, Australia, South Africa, China, and Brazil have carried out extensive works on diatoms [9, 17, 22, 26, 31, 41, 42, 59]. They used statistical methods to associated diatoms with environmental factors and established a series of functions and mathematical models to identify diatom communities, which greatly improved the accuracy in using diatoms to monitor the changes in water quality. These methods are now widely used to assess river health.

In 2000, the Water Framework Directive of the European Union recommended using diatoms as a routine biological indicator for water environment monitoring, after which many European countries updated their regulations accordingly [7]. Diatom indices are currently widely used in water environment monitoring. Most diatom indices are established based on the diatom species found in the water body, and they categorize the water bodies into different levels based on the ecological habits and pollution resistance of the detected diatoms. The first diatom index was established subsequently standardized the calculation of the diatom index (Descy et al., 1979). Many researchers have customized the diatom index based on the environmental conditions of their research area and established a number of diatom indices to evaluate the water quality of diverse water bodies, including specific pollution sensitivity index (SPI) [8], diatom model affinity (DMA) [38], saprobic index (SI) [47], relative abundance of diatoms (RAD) [19], percent of sensitive diatoms (PSD) [4], percent motile diatoms (PMD) [5], diatom assemblage index to organic pollution (DAIpo) [58], biological diatom index (BDI) [11], trophic diatom index (TDI) [25], pollution tolerance index (PTI) [28], pampean diatom index (IDP) [18], diatom quotient (DU) [24], generic diatom index (GDI) [59], etc.

Lakes are considered open if they are connected to oceans or seas via rivers and water can flow freely between the lake and the river. Over the past century, due to natural changes, lake reclamation, engineering projects, etc., more than 100 river-connected lakes in the middle and lower reaches of the Yangtze River have silted up, diminished, or experienced artificial segmentation. In this region, only Dongting Lake and Poyang Lake are currently both larger than 500 km2 in size and still maintain natural connection to the Yangtze River [68]. Dongting Lake has become the largest river-connected lake in the middle reaches of the Yangtze River in China. It is the most important flood plain wetland in China as well as a lake wetland with global significance with respect to protection [37]. Dongting Lake has always been dominated by diatoms, but the proportion of chlorophytes has gradually increased in recent years [64, 69], and the lake has reached a tip** point between being mesotrophic and mildly eutrophic [65]. Nevertheless, in this study, Dongting Lake was still mesotrophic overall (Fig. 2), although it was mildly eutrophic in the autumn and winter of 2017. The eastern lake area was constantly mildly eutrophic in recent years, likely due to the relatively high contribution from the region near S13. The periphery of S13 is a typical backwater area of eastern Dongting Lake, where the water is shallow and flows slowly, and the nutrient concentration (nitrogen and phosphorus) is always suitable for phytoplankton growth (particularly chlorophytes and cyanobacteria) [64]. Since 2013, cyanobacteria had become the dominant phytoplankton population a few times in the region around S13. The rising phytoplankton abundance also increased the chlorophyll a level, which further promoted the nutrient levels in the water.

Correlation between diatom indices and water quality factors

Dongting Lake is the largest river-connected lake in the middle and lower reaches of the Yangtze River in China. Dongting Lake is fed by seven rivers, and its outflow returns into the Yangtze River from Chenglingji section. The average water residence time is 14 to 18 days at Dongting Lake, and the water volume of the lake changes rapidly. The lake experiences fast water mixing, and the thermal stratification is not obvious [69]. As a result, most areas in the lake, especially the channel areas, are like rivers in their hydrological characteristics. In addition, the entire lake area is dominated by diatoms. Therefore, the diatom indices that are widely used in the biological evaluation of the water quality of rivers were examined in this study, to determine how different diatom indices responded to the water quality indicators at Dongting Lake.

The  ΣTLI accounts for the permanganate index and major pollutants such as nitrogen and phosphorus, and it is higher when the water body is more polluted. The indices RAD, GDI, PTI, and IDP all responded inversely to pollution stress and trophic status, and RAD and PTI responded extremely significantly to  ΣTLI. In contrast, DU and TDI responded positively to ΣTLI. The results are consistent with the correlation analysis in Fig. 5. It is worth commenting that PMD had a positive response to the degree of sedimentation in the water body. Areas with stronger sediment mixing have lower SD, which inhibits phytoplankton growth [10, 45] and reduces the Chla concentration (a proxy of the phytoplankton abundance), ultimately decreasing the ΣTLI of the water body. This is consistent with the very significantly negative correlation between PMD and ΣTLI found in Fig. 5.

The correlation analysis showed that three indices PTI, RAD, and PMD, which were mainly correlated with WT, Cond, DO, CODMn, BOD5, CODCr, and TP, could fully cover the relevant information in DU, GDI, TDI, and IDP. It is generally believed that higher temperature facilitates the reproduction and growth of chlorophytes and cyanobacteria [2, 40] but not diatoms, which prefer cooler water [1, 50]. Both Navicula and Melosira were dominant in Dongting Lake, but their abundances were correlated positively with PMD and negatively with GDI, respectively. As a result, WT correlated positively with GDI (p < 0.01) but negatively with PMD (p < 0.05). The observed significantly positive correlation between RAD and DO could be attributed to the fact that phytoplankton absorbs carbon dioxide and releases oxygen during photosynthesis [57] in northern China).

As stated previously, physical and chemical indicators have their limitations in reflecting the water quality, and biological indicators can reflect the quality of the water environment more comprehensively and accurately. In this study, the three indices of PTI, RAD, and PMD were selected using the 2017–2019 data and verified using the 2020–2022 data. Figure 7 shows that all three indices were significantly negatively correlated with the Nemerow index. Nemerow index is recommended by GB/T 14848-93 in China, which is simple to calculate and has always been the main method for water quality assessment in Dongting Lake [21, 37], the relative abundance of dominant diatoms are closely related to water quality in Daihai Lake [60], RAD has been selected as an important index for biological assessment of water quality in Dongting Lake during the construction of the aquatic biological integrity index [56], and TN and TP are main driving factors of seasonal succession of diatom community [13]. Of course, the suitability of RAD for water quality assessment in different regions will certainly vary, and we need to assess the accuracy of RAD based on the type and characteristics of the water body.

As a whole, RAD can play an important role in water quality assessment of different lakes in China, which is worthy of attentions by local authorities and need suitability assessment and promotion according to the regional characteristics.

Conclusions

In this study, we investigated water quality and phytoplankton in March, June, September and December from year 2017 to 2022, and selected seven diatom indices, including RAD, PMD, GDI, DU, PTI, TDI and IDP, to screen the adaptability of water quality assessment comparing with NI. The results from 2017 to 2019 showed that the diatom density in Dongting Lake ranged from 0.7 × 104 to 85.5 × 104 ind./L, with a certain decreasing trend. The spatial and temporal changes of major water quality factors were obvious, and the ∑TLI ranged from 45.99 to 50.72, with an average value of 47.85, indicating that the overall condition of Dongting Lake was medium nutrition. Correlation analysis showed that PTI, RAD and PMD could represent the information of DU, GDI, TDI and IDP, and were significantly positively correlated with DO (p < 0.01), while significantly negatively correlated with Cond, CODMn, BOD5, CODCr and ∑TLI (p < 0.001). The index verification results from year 2020 to 2022 showed that PTI, RAD and PMD were all significantly positively correlated with NI (p < 0.001). Taking into account the data integrity of the index calculation (PMD would appear the phenomenon of zero value) and the difficulty degree (PTI index needs to calculate the tolerance value of diatoms to external stress), RAD was finally selected as the biological indicator for evaluating the water quality of Dongting Lake. The water quality of Dongting Lake has been evaluated according to the lake standard for a long time in China, however, the water quality is difficult to meet the standard just because the concentration of TP was always exceeded the Class III water quality standard set by the state (0.05 mg/L).

The results of this study provided an alternative method for the water quality assessment of Dongting Lake, and also put forward a new path for the water quality assessment of the river-connected lakes. In the future, we will screen the diatom indices suitable for water quality assessment in a wider scope, and study the evaluation grade of the diatom indices combined with the existing water quality factor classification and evaluation criteria, in order to provide scientific basis for local administrative departments to carry out accurate and standardized management of river-connected lakes.