Optical characteristics and factors that influence representative Lakes in the Taihu Lake basin, China

Abstract From 2018 to 2019, a survey of a typical lakes system in the Taihu Lake basin, China, was carried out. To provide reference values for the restoration and management of submerged vegetation in the lakes, the optical attenuation coefficient (Kd ) and euphotic depth (Zeu ) of the water body were calculated, and the distribution characteristics and factors that influenced the underwater optical field lands with different nutrition levels in the basin were analyzed. The main factor affecting the distribution of photosynthetically active radiation (PAR) in lakes in the Taihu Lake basin is suspended solids (SS), dominated by inorganic suspended matter (ISS). Chlorophyll a (Chl-a), dissolved organic carbon (DOC), and chromophoric dissolved organic matter (CDOM) all affect, although weakly, the optical characteristics of lake wetlands. In CDOM, humic-like components have a more significant impact on Kd . The Kd of lakes in the Taihu Lake basin is closely related to permanganate index (CODMn) and especially total phosphorus (TP); The growth of aquatic vegetation is increased in Shanghu, Dianshan-Yuandang, Yangcheng, and Wuli Lakes that have higher Zeu and Zeu /Depth values.


Introduction
Underwater photosynthetically active radiation (PAR) is an important indicator of the turbidity of water bodies and ecological conditions and is also an important factor affecting the photosynthesis and growth of aquatic plants (Karlsson et al. 2009;Brandão et al. 2016;Wang et al. 2020). PAR intuitively reflects the distribution characteristics of the underwater optical field. However, after the light enters the water, it is often affected by various factors such as suspended particulate matter, dissolved organic matter, and planktonic microorganisms, resulting in scattering, absorption, and gradual attenuation. When the underwater PAR attenuation reaches a level at which it cannot satisfy the normal requirements of aquatic plants, it causes the submerged vegetation to retreat and can induce phytoplankton outbreaks, often related to negative anthropogenic impacts on the wetland environment (Reinart and Pedusaar 2008;Kuwahara et al. 2015;Bai et al. 2016). At present, the optical attenuation coefficient (K d ) is commonly used to measure the attenuation characteristics of underwater PAR, and the euphotic depth (Z eu ) is used to represent the depth limit at which aquatic submerged plants can grow. As a metric, Z eu is a better indicator of the light penetration ability of water bodies more than Secchi depth (SD), and indicates the depth beyond which submerged plants cannot grow normally (Zhou et al. 2018;Li et al. 2019).
The quantitative indicators that affect the attenuation and distribution of underwater PAR include suspended solids (SS), turbidity (Turb), chlorophyll a (Chl-a), dissolved organic carbon (DOC), and chromophoric dissolved organic matter (CDOM). Different factors influence PAR in water bodies in different environments (Zhou et al. 2016;Yu et al. 2020;. For example, a study on the Chesapeake Bay area in the United States from Testa et al. (2019) found that SS and water salinity could significantly affect the distribution of the underwater light field. Bai et al. (2016) showed that the main factor affecting the distribution of the light field under Erhai Lake was turbidity. The research of  found that SS and CDOM were the main factors affecting the optical attenuation in Qiandao Lake, and that they variably affected the spectra of different bands. Through investigations of Fuxian Lake, Zhou et al. (2018) found that CDOM had a particularly important impact on lake optical characteristics during the rainy season, followed by SS. In this case, the land input processes directly restricted the distribution of the underwater light field. Walsby's research in the Baltic Sea (Walsby 1997) and Li's research in Chenghai Lake (Li et al. 2019) both found that planktonic biomass is an essential factor affecting the optical characteristics of water bodies. The review of the available literature shows that a better understanding of the distribution characteristics and the factors that influence the underwater light field in a specific area will play an important role in the local management and restoration of aquatic vegetation. An untargeted restoration process done without understanding the lake's optical environment could easily lead to failure and waste (He et al. 2014;Roddewig et al. 2020;Strom et al. 2020).
The Taihu Lake basin is located in a dense network of waterways in the core area of the Changjiang (Yangtze) River Delta in China. This area is also a highly economically developed area in China. The environment of the region has been impacted greatly by human activities. The discharge of production and domestic wastewater has caused serious eutrophication in the water bodies. Studies have shown that in recent decades, the coverage of aquatic vegetation in most lakes in the Taihu Lake basin (including Taihu Lake) has decreased significantly, and the aquatic ecosystem has gradually deteriorated (Liu et al. 2020). Therefore, targeted ecological restoration in the Taihu Lake basin could play an important role in improving people's quality of life and maintaining sustainable economic development. This research aims to: (1) understand the optical characteristics of water bodies of typical lake wetlands in the Taihu Lake basin; (2) explore the distribution of the underwater light field and factors that influence it; (3) quantify the relationship between optical attenuation characteristics and nutrients; and, (4) determine how improving the light received affects aquatic vegetation coverage in order to provide reference metrics for the restoration and management of submerged plants of the lakes in the basin.

Overview of the study area
In this study, Changdang, Gehu, Wuli, Shanghu, Yangcheng, Chenghu, Dianshan, and Yuandang Lakes were selected as the research objects ( Figure 1). Gehu Lake covers an area of 164.6 km 2 with an average depth of 1.2 m, making it the second largest lake in the Taihu Lake basin. Chenghu Lake is located in the lower reaches of Taihu Lake, with a water area of 45.5 km 2 and an average depth of 1.8 m. Shanghu Lake has an area of 8.2 km 2 and an average depth of 2.8 m. Wuli Lake is a small lake located deep inland in Meiliang Bay off of Taihu Lake, with an area of only 8.6 km 2 and an average depth of 2.4 m. Changdang and Yangcheng Lakes have areas of 89.4 km 2 and 119.8 km 2 , respectively, with an average depth of 1.1 m and 1.8 m, respectively. Aquaculture activities are being carried out in both lakes. Dianshan and Yuandang Lakes are located at the junction of Suzhou and Shanghai. As the two lakes are closely connected, they were analyzed as a whole in this study as "Dianshan-Yuandang Lake". The overall water area is about 72.7 km 2 and the average depth is 2.1 m.

Layout of sampling points and sample collection
Samples were collected from the Taihu Lake basin wetland system in April (spring), July (summer), and November (autumn) of 2018, and January of 2019 (winter). The sampling locations are shown in Figure 1. Mixed water samples were collected 50 cm below the surface of the water, stored in a polyethylene bottle and protected from light, then quickly returned to the laboratory for analysis.

Sample measurement methods
Water depth (Depth), SD, and PAR were measured in situ during sampling. Among them, Depth was measured using a portable depth sounder (SM-5A, Speedtech, USA), and SD was measured using a 30 cm black and white Secchi disk. PAR was measured using the XR-620CTD digital luxmeter (RBR Ltd., Canada). When measuring, the PAR intensity at the surface was measured first, followed by each deeper layer at a gradient of 0.1 m. Three data points were recorded for each layer, and the average value was used to represent that layer.
Other indicators were analyzed and measured in the laboratory. Total nitrogen (TN), total phosphorus (TP), permanganate index (COD Mn ), ammonia nitrogen (NH 3 -N), and Chl-a were determined using the potassium persulfate digestion method, molybdenum antimony spectrophotometric method, acid method, Nessler's reagent-Spectrophotometric method, and 90% acetone method, respectively. SS was determined using the filtration drying method. Organic suspended solids (OSS) and inorganic suspended solids (ISS) were determined using the loss on ignition method. The measurement methods were all referenced to values found in the literature (Editorial Board of Water and Wastewater Monitoring and Analysis Methods, and State Environmental Protection Administration of China 2002; Chen et al. 2014). The UV2700 spectrophotometer (Shimadzu, Kyoto, Japan) was used for absorbance measurement.
Part of the water sample was first passed through a 0.45 pore size cellulose acetate membrane to determine DOC, and then passed through a 0.22 lm pore size GF/C filter to determine CDOM. The first 20 mL of the filtrate after two filtrations was discarded in case the residual carbon on the filter membrane interfered with the measurement. The filtrate was stored in a polyethylene bottle that was washed using acid, and the sample was measured on the same day. DOC was measured using a total organic carbon analyzer (TOC-V CPN, Shimadzu). The spectrophotometer (UV2700, Shimadzu) was used to scan the absorbance of the filtrate between 200-800 nm at an interval of 1 nm for the CDOM absorption spectrum, and the absorption coefficient of CDOM was calculated on this basis, using methods found in the literature ). The CDOM content was approximately represented by the absorption coefficient at 350 nm (a(350)) (C ardenas et al. 2017). In addition, due to the complex composition and structure of CDOM, this study used the popular parallel factor analysis method to conduct the fluorophore component interpretation for CDOM, using model construction methods as defined in the literature  in which the separated components represented different CDOM components with different characteristics.

Calculation of optical attenuation coefficient and euphotic depth
Underwater PAR attenuation satisfies : (1) where, K d denotes the optical attenuation coefficient, z denotes the depth from the water surface to the measurement position, and E(z) and E(0) denote the PAR values at z m and 0 m depth, respectively. A set of no less than three data points at each sample were nonlinearly fitted with an exponential model to obtain the K d value, and was considered significant when the R 2 ! 0.95. Z eu was calculated using the following equation (Kirk 2011):

Data processing and analysis
ArcGIS 10.2 (ESRI Inc., Redlands, CA, USA) was used to plot the layout of sampling points. Excel 2016, SPSS 21.0 (IBM Corp., Armonk, NY, USA), and Origin 2019b (OriginLab Corp., Northampton, MA, USA) were used to calculate the mean and standard deviation, K d value fitting, ANOVA significance analysis, correlation analysis, and linear regression analysis. In the significance report, p < 0.01, p < 0.05, and p > 0.05 were considered to be extremely significant, significant, and insignificant, respectively.

Results
3.1. Nutrient characteristics of typical Lakes in the taihu lake basin Table 1 shows the mean value and standard deviation of COD Mn , TN, TP, NH 3 -N, and Chl-a in different seasons in the representative lakes of the Taihu Lake basin. The trophic level index (TLI) can integrate multiple indicators to comprehensively measure the nutritional status of water bodies. This method has been applied with mature classification standards. This study also uses TLI to preliminarily understand the nutritional characteristics of typical lakes in Taihu Lake basin. The calculation of TLI was based on the research of . According to calculations, the TLI of Changdang and Gehu Lakes exceeded 60, indicating that they were moderately eutrophic. Chenghu Lake had a TLI between 55 and 60 on average. The TLIs of Wuli, Yangcheng, and Dianshan-Yuandang Lakes were relatively lower, around 50-55, but were still classified as slightly eutrophic. The TLI of Shanghu Lake was lower than 50, which indicated a mesotrophic lake.

Optical characteristics of typical Lakes in the taihu lake basin
The optical characteristics of typical lakes in Taihu Lake basin in each season were shown in Table 2. The K d level was the highest in Changdang Lake, with an average of 5.22 ± 0.43 m À1 and a range of 3.21-8.71 m À1 , followed by Gehu Lake and Chenghu Lake, which could reach 6.28 ± 1.51 m À1 and 5.97 ± 1.43 m À1 , respectively, in summer. Wuli Lake, Yangcheng Lake, Dianshan-Yuandang Lake and Shanghu Lake had comparable K d value levels, with Yangcheng Lake having a slightly higher average of 4.99 ± 1.20 m À1 in summer. And were the lowest in Dianshan-Yuandang Lake, with an average of 2.33 ± 0.39 m À1 and a range of 0.88-5.23 m À1 . Seasonally, the K d value was lowest in the spring, then increased in autumn and winter and was the highest in the summer, with a maximum average of 7.91 ± 0.76 m À1 , which appeared in the summer at Changdang Lake. The K d of each lake in winter was relatively lower, with only Gehu Lake having a value higher than that in other lakes (p < 0.01). The lowest value occurred in the winter at Dianshan-Yuandang Lake, with a minimum of 1.21 ± 0.31 m À1 .
Overall, the SD of Yangcheng Lake was significantly higher than that of the other lakes (p < 0.01), with an average of 70.33 ± 2.64 cm and the maximum of 127 cm, followed by Data are expressed as the mean ± standard deviation of water quality indicators Dianshan-Yuandang Lake with the mean of 62.58 ± 5.84 cm. The SD of Wuli Lake and Shanghu Lake were at the same level, they remained at 65.05 ± 5.61 cm and 64.00 ± 5.52 cm respectively during the winter months when the SD was higher. The overall SD of Changdang Lake was the lowest, at 25.97 ± 1.52 cm, with a minimum of only 14 cm. Temporally, SD showed a pattern of highest to lowest values in summer < autumn < spring < winter in most lakes. The winter average of Dianshan-Yuandang Lake was the highest, at 93.86 ± 12.02 cm, and the summer average of Changdang Lake was the lowest, at 16.75 ± 2.05 cm.

SS characteristics of typical Lakes in the Taihu Lake basin
The concentration levels of SS, OSS and ISS in each season of each lake are shown in Table 2.
In general, the SS concentration in Changdang and Gehu Lakes was relatively higher in all seasons. In winter and spring, it was significantly higher than that of the other lakes (p < 0.01). The SS level of them in spring and autumn was similar. In summer, the average SS of Changdang Lake was the highest, at 68.77 ± 14.17 mg/L. The SS concentration in Chenghu Lake was similar to that in Gehu Lake and Changdang Lake in summer, but much lower in winter and spring compared with them, and only 14.26 ± 2.62 mg/L in spring. The average SS concentration of Yangcheng Lake in spring was the lowest, with 8.86 ± 6.33 mg/L and a range of 4.25-22.28 mg/L. Wuli Lake, Dianshan-Yuandang Lake and Shanghu Lake all showed values from the greatest to the least in summer > autumn > spring and winter.
The OSS concentration of each lake did not fluctuate greatly either in time or space. On average, that of Changdang Lake in summer was the highest, at 15.51 ± 1.69 mg/L with a range of 13.30-18.10 mg/L. The OSS level of Gehu Lake was slightly lower than that of Changdang Lake, and could reach 14.51 ± 2.63 mg/L in summer. The OSS levels of Wuli Lake, Yangcheng Lake, Dianshan-Yuandang Lake and Shanghu Lake were similar to each other, and the OSS concentration of Shanghu Lake was the lowest on average in winter at 2.87 ± 0.45 mg/L, with a range of 2.35-3.50 mg/L. The OSS concentration of Chenghu Lake was between the above two categories, with an average of 9.38 ± 2.51 mg/L in summer. In contrast, the ISS concentration fluctuated sharply, with the highest average being 54.86 ± 23.75 mg/L in Chenghu Lake in summer. The ISS concentration of Changdang Lake was lower than Chenghu Lake, with an average of 53.27 ± 12.48 mg/L in summer. Although the average value of Gehu Lake in summer was lower than that of Changdang Lake and Chenghu lake (50.12 ± 16.73 mg/L), the average level of ISS in the whole year was significantly higher than that of other lakes (p < 0.01). Wuli Lake in spring had the lowest average value of 4.74 ± 0.99 mg/L, ranging from 3.38-6.15 mg/L. Seasonally, with the exception of Yangcheng Lake, the ISS concentration in summer was significantly higher than that of the other seasons (p < 0.01).

Organic matter characteristics of typical Lakes in the Taihu Lake basin
The DOC of typical lakes in the Taihu Lake basin was stable, and seasonal fluctuations were not obvious (Figure 2a). The range of fluctuation was between 1.67 ± 0.04 -4.49 ± 0.25 mg/L on average. The two endpoints appeared in Shanghu Lake in spring and Changdang Lake in winter, respectively. The highest DOC was only 5.60 mg/L, which appeared in Gehu Lake in spring. Seasonally, there is great independence in the variation law of DOC in different lake water bodies without a uniform law. In contrast, the absorption coefficient a(350) had obvious seasonal characteristics (Figure 2b). The value of a(350) of each lake in summer was significantly higher than that in other seasons (p < 0.01), with the average value of Dianshan-Yuandang Lake the highest (7.68 ± 0.32 m À1 ). The lowest value occurred in Shanghu Lake in spring (1.22 m À1 ).
Through parallel factor analysis, the CDOM of each lakes was separated into three components, C1, C2, and C3. The C1 component represents a terrestrial humic-like component. The C2 component is close to the M peak in the traditional peak group, which is a type of humic-like component with a closer relationship to recent biological activities. The C3 component represents a type of autogenic protein-like component with high biological sensitivity (Coble 1996;Liu et al. 2019). The distribution of their corresponding fluorescence peaks, F 1 , F 2 , and F 3 , in each season of each lake is shown in Figure 3. The average F 1 value of Changdang Lake in summer was the highest, at 0.194 ± 0.026 RU. Except for autumn, the F 1 values of Changdang and Gehu Lakes in other seasons were significantly higher than those of Wuli, Yangcheng, Chenghu, and Shanghu Lakes (p < 0.05). The average F 2 of Dianshan-Yuandang Lake in summer was the highest, at 0.23 ± 0.013 RU, and that of Shanghu Lake in spring was the lowest, at 0.048 ± 0.003 RU. There was not much difference between F 1 and F 2 in terms of quantity, but the spatial distribution of F 2 values were more chaotic and the law was less obvious. The F 3 of Yangcheng, Chenghu, Shanghu, and Dianshan-Yuandang Lakes in all seasons were significantly higher than that of other lakes (p < 0.01). The highest value was in Yangcheng Lake, which could reach 1.397 ± 0.168 RU in summer on average. Seasonally across the whole basin, F 1 and F 2 both showed the decreasing pattern of summer > autumn > winter > spring. The two components of the whole basin could reach 0.173 ± 0.030 RU and 0.196 ± 0.035 RU in summer on average. The fluctuation of F 3 among different seasons was not obvious, with only F 3 in winter insignificantly lower than that in other seasons (p > 0.05).

Factors influencing the optical characteristics of typical Lakes in the Taihu Lake basin
A large number of studies have shown a significant negative correlation between K d and SD (Pierson et al. 2008;Kirk 2011;Zhang et al. 2012;Ma et al. 2016;Yu et al. 2019). This conclusion has also been verified in lakes of the Taihu Lake basin, except for the significant weaker correlation between the two parameters of Yangcheng Lake (r ¼-0.349, p < 0.05), other lakes all showed an extremely high significant negative correlation (p < 0.01). On the whole, SS was undoubtedly the main factor influencing the optical attenuation characteristics of lakes in the Taihu Lake basin during all seasons (Table 3). Except for Yangcheng Lake, all lakes showed that the effect of ISS was higher than that of OSS. Previous studies have demonstrated that resuspension of lake sediments and rising lake water temperatures are the main causes of elevated ISS levels in lakes, while this process also reduces phytoplankton and benthic biomass (Jin et al. 2022). For some shallow lakes in the Taihu Lake basin, the resuspension of sediments caused by wind and wave disturbance is the main source of SS in the water, especially ISS (Reinart and Pedusaar 2008;Qu et al. 2014). Especially in autumn and winter, when most submerged plants die, the resuspension of sediments becomes easier, and the influence on optical attenuation characteristics is also greater (He et al. 2014). Compared to the surrounding lakes, Taihu Lake in the centre of the basin has a more open water area, where the resuspension of sediment and the deterioration of the water column's light environment due to wind speed are more pronounced, and there is a close relationship between the release of nutrients from the lake and the resuspension of sediment (Tang et al. 2020).
Compared with SS, the impact of Chl-a on K d of all lakes was not notable, only having significant impacts in Shanghu (r ¼ 0.928, p < 0.01) and Wuli Lakes (r ¼ 0.768, p < 0.01). However, when analyzed during a single season, the optical attenuation characteristics of some lakes can still be significantly affected by Chl-a in autumn, such as Gehu (r ¼ 0.584, p < 0.01), Chenghu (r ¼ 0.982, p < 0.05), and Yangcheng Lakes (r ¼ 0.667, p < 0.01). This shows that the impact of Chl-a on the lake light field is greatly affected by objective conditions such as season and air temperature (Lv et al. 2012). Similar to the surrounding lakes in the basin, the relationship between the whole Taihu Lake and Chl-a was less pronounced than in SS, with waters more affected by Chl-a tending to have frequent cyanobacterial blooming (Zhang and Chen 2006).
The organic components DOC and CDOM in the lake also impacted the optical attenuation characteristics of the lakes, and the effect of CDOM was stronger than that of DOC. However, due to different lake conditions, the effects were also different. Among them, Shanghu (a(350): r ¼ 0.903, p < 0.01), Wuli (a(350): r ¼ 0.813, p < 0.01), Yangcheng (a(350): r ¼ 0.865, p < 0.01) 0.01; DOC: r ¼ 0.533, p < 0.01), Chenghu (a(350): r ¼ 0.796, p < 0.01), and Dianshan-Yuandang Lakes (a(350): r ¼ 0.576, p < 0.01); DOC: r ¼ 0.530, p < 0.01) were affected significantly. At the same time, lakes strongly affected by CDOM had a close positive correlation with fluorescence peaks F 1 and F 2 ( Table 4). The source and composition characteristics of these components could be obtained by comparison with previous studies. The C1 component is a type of terrestrial humic-like substance, and its fluorophore has a relatively higher degree of substitution and polycondensation. Compared with other components, this component may come from an older stratum (Yamashita et al. 2011). The C2 component is also a type of humic-like compound, which is closely related to recent activities of organisms, including moderate-intensity anthropogenic pollution (Yamashita et al. 2010;Liu et al. 2019). The C3 component is a type of protein-like substance with a higher degree of degradation. It is usually produced by biological activities and therefore has a higher biological sensitivity. However, it can also be transformed from some non-protein-like substances such as lignin phenols. This component is often related to higher intensities of imported pollutants Xiao et al. 2018). These results show that the main contribution to the optical attenuation characteristics of these lakes is some imported exogenous humic-like substance, including anthropogenic pollutants. Unlike these typical surrounding lakes in the basin, the CDOM fluorescence fraction of Taihu Lake is dominated by protein-like component, demonstrating that the authigenic sources within the lake are the main source of CDOM in Taihu Lake, and that the main reason for this is the frequent cyanobacterial blooming. Also, due to the longer lake renewal time, the rate of CDOM depletion in Taihu Lake is higher than that of the surrounding lakes ).
In addition, there is a significant positive correlation between the K d values of Shanghu and Yangcheng Lakes and F 3 (p < 0.05), which could be related to their aquaculture and algae growth. The artificial salvage and self-decay of a large number of Vallisneria natans (Lour.) Hara growing in Shanghu Lake as well as Elodea canadensis Michx, Ceratophyllum demersum L., and other aquatic plants growing in Yangcheng Lake promote the release of protein-like components, thereby enhancing the contribution of F 3 to K d . At the same time, as mentioned above, OSS was strong than ISS in the contribution of SS to K d in Yangcheng Lake, which could also be related to the metabolic activities of aquatic plants and animals such as crabs.

The relationship between the optical characteristics and the eutrophication level of water bodies of typical Lakes in the Taihu Lake basin
Studies have shown that since the new century, the TN and TP concentrations of Taihu Lake and lakes in this basin have gradually increased. The original large volumes of submerged vegetation in the lake have gradually disappeared, and the existing vegetation is dominated by emergent plants and floating-leaved plants with higher pollution and nutrient tolerance Penning et al. 2008;Bickel and Schooler 2015). The increase of nutrient levels and the demise of submerged vegetation bred large numbers of phytoplankton, which directly limited the distribution of the underwater light field (Pogozhev and Gerasimova 2011). Therefore, there is an inseparable relationship between the nutrient level of the lake and the optical attenuation. In this study, the K d values of the lakes in the Taihu Lake basin have a significant linear relationship with TN, TP, and COD Mn , and this relationship is more prominent in winter (Figure 4). This and a large number of previous studies have shown that Chl-a, CDOM, and particulate matters produced by phytoplankton themselves and their metabolism will directly affect the optical attenuation characteristics of water bodies (He et al. 2014;Bai et al. 2016). In winter, these components will be minimized in both quantity and influence, and the relationship between some elements such as nitrogen and phosphorus and K d are more easily discovered. In addition, among several indicators, the relationship between TP and K d was the closest. This conclusion has also been confirmed in previous studies of Taihu Lake, where phosphorus in the lake body is more likely to limit the distribution of the light environment in Taihu Lake compared to other nutrients (Zou et al. 2020). In addition to the K d value, studies have also shown that there is a close relationship between the transfer and transformation of phosphorous elements and the CDOM in the Taihu Lake basin Zhang et al. 2018). This indicates that phosphorous is not only an important factor affecting the distribution of the underwater light field of the lakes in the Taihu Lake basin, but also an important component restricting the biogeochemical processes of the basin.
4.3. The distribution of lake euphotic depth in the Taihu Lake basin and its response to aquatic vegetation The growth of aquatic vegetation is dependent upon good underwater lighting conditions. Z eu represents the lower limit of the depth at which submerged plants can grow normally. Therefore, understanding the existing vegetation distribution in the basin and the distribution characteristics of Z eu is of great significance to the restoration of aquatic vegetation in the basin (Lacoul and Freedman 2006;Zhou et al. 2018). This and other studies use the ratio of euphotic depth to water depth (Z eu /Depth) to describe the optical environment suitable for the growth of aquatic vegetation (Zhang et al. 2007). As mentioned above, the aquatic vegetation of most lakes in the Taihu Lake basin has shown a declining trend in recent decades , while the outlook for Shanghu, Yangcheng, Wuli, and Dianshan-Yuandang Lakes were relatively optimistic. This is consistent with the spatial distribution of the basin Z eu ( Figure 5). Shanghu Lake had the lowest nutrient levels and the best aquatic vegetation coverage. The vegetation was dominated by submerged plants such as Vallisneria natans (Lour.) Hara. The aquatic vegetation of Wuli Lake has undergone recovery after previously being damaged. A series of recent ecological management measures have allowed aquatic vegetation to be restored to a certain extent (Yan et al. 2004;. Its Z eu also showed a higher level, which was even better than that of Yangcheng Lake with higher vegetation coverage, implying the close relationship between the growth of submerged vegetation and the distribution of the underwater light field. The reason that the Z eu of Yangcheng Lake was lower than that of Wuli Lake might be that the large amount of metabolites discharged by aquatic products raised in Yangcheng Lake during the growth process existed in the form of suspended matter in the lake, which interfered with the propagation of light underwater. On the contrary, the aquatic vegetation coverage of Changdang Lake has been deteriorating (Wu et al. 2015). Although there were still submerged plants growing in the Shanghuang area south of the lake, this had little impact on the whole lake. The Z eu /Depth level of Changdang Lake was also at a lower level in the basin.

Conclusions
Through the analysis of Changdang, Gehu, Shanghu, Wuli, Yangcheng, Chenghu, and Dianshan-Yuandang Lakes in the Taihu Lake basin, the optical characteristics and influencing factors of representative lakes in the Taihu Lake basin were obtained. The main conclusions include: 1. The main factor affecting the distribution of the underwater light field in representative lakes in the Taihu Lake basin was SS, and the effect of ISS in most lakes was higher than that of OSS, followed by Chl-a. 2. DOC and CDOM also had a significant impact in some lakes. In lakes with greater impact of CDOM, most also had a continuous input of pollutants, including humuslike substances. 3. The K d value of typical lakes in the Taihu Lake basin was closely related to COD Mn and especially TP. 4. In Shanghu, Dianshan-Yuandang, Yangcheng, and Wuli Lakes, when the growth of aquatic vegetation was relatively better, the Z eu and Z eu /Depth values were both higher.