Changsha city is located in the northeastern region of Hunan Province, downstream of the Xiangjiang River and on the western edge of the Changliu Basin (111°53′ E and 27°51’ N). The length from east to west is approximately 230 km, and the width from north to south is approximately 88 km. Most of the rivers in Changsha city belong to the Xiangjiang River system. Moreover, 15 tributaries flow into the Xiangjiang River, mainly the Liuyang River, Laodao River, Jinjiang River, and Weishui River. There are 302 tributaries with lengths greater than 5 km, including 289 in the Xiangjiang River basin. The hydrological characteristics of Changsha include a complete water system and a dense river network. There is a large amount of water and abundant water energy resources. The average annual surface runoff within the city is 8.265 billion m³, with runoff depths ranging from 550 to 850 mm. The annual average runoff of the Xiangjiang River flowing through Changsha city is 69.25 billion m³, and the river is navigable throughout the year.

Eight typical wetland parks in Changsha city were selected as the research objects (Figure 1): Yudai Lake (YD), Liuyang River (LYH), Meixi Lake (MX), Jinshan Lake (TS), Xiangjiang Jiangtan Park (XH), Xianjia Lake (XJ), Yuehu Park (YH), and Yanghu Wetland Park (YSD). Two surface water samples and two sediment samples were collected from each park, and the sediment samples were collected from below the water sample. The samples were collected in March, 2023. All tools were cleaned with ultrapure water at least 3 times before sampling. A 2 L iron sampling bucket was placed at a depth of 20–30 cm underwater for sampling. Fifteen liters of water was collected at each sampling point and filtered through a 0.045 mm stainless steel sieve. Then, all of the solids retained on the sieve were rinsed with deionized water into a sampling bottle for preservation. This study included microplastics with sizes within the range 0.045–5 mm. The sampling points for sediment microplastic collection were consistent with the collection times and locations of the microplastics in the surface water. The collections of surface sediments from the lakes were carried out with a Peterson grab at the site location (41). Five hundred grams of sediment was packed into a self-sealing bag, labeled, and brought back to the laboratory for refrigeration.

Each water sample was added to 20 mL of 30% H₂O₂ solution in the laboratory and digested at room temperature for 24 h until visible organisms were removed from the samples. A total of 800 mL of a saturated NaCl solution (1.2 g cm⁻³) was added to the reaction mixture, which was thoroughly stirred and allowed to stand for 24 h before extracting the supernatant. Glass ultrafine fiber filter paper (0.45 μm) was used for vacuum filtration of the supernatant. Finally, the filter paper was stored in a culture dish and immediately covered with aluminum foil for future observation. Each sediment sample was placed separately in a glass culture dish and covered with aluminum foil to prevent indoor pollution. Then, the glass culture dish was placed in a drying oven and allowed to stand for 24 h at 60°C. After the sediment sample was completely dried, it was mixed evenly and initially screened through a 5 mm stainless steel sieve. Large impurities such as stones and branches were removed. Subsequently, 100 g of the dried sample was removed and placed in a 1 L beaker. A saturated sodium chloride solution was added to the beaker, which was subsequently stirred with a glass rod for 2 min and allowed to stand for 24 h. Finally, the supernatant was collected. This experimental process was repeated three times. All of the supernatants were collected and passed through a 0.45 μm nylon membrane filter (with a negative pressure of no more than 40 kPa). All of the substances on the filter membrane were transferred to a 200 mL beaker, and an excess of 30% H₂O₂ solution was added (37). The solution was then shaken at 65°C and 80 rpm for 24 h for digestion. The dissolved solution was again filtered through a nylon membrane with a pore size of 0.45 μm and a diameter of 47 mm. Finally, the membrane was placed in a clean glass Petri dish for inspection. The PM_2.5 in the laboratory atmospheric was 8 μg m³. It is noteworthy that the extraction experiment of microplastics was conducted in a super clean workbench, where the air was filtered. In addition, in order to eliminate experimental errors caused by containers and air, we also set up a CK control, which only added digestion solution without adding water or sediment samples. The results of microplastic content in all samples were obtained after removing the plastic content in the CK control.

The chemical compositions of the microplastics in the sample were identified with a Raman spectrometer and a laser infrared imaging spectrometer (Agilent 8,700 LDIR) (32). Briefly, the obtained filter membrane was immersed in an ethanol solution for ultrasonic treatment, so that the substances on the filter membrane were dispersed in the ethanol solution. The filter membrane in the ethanol solution was removed and cleaned multiple times with ethanol. Then, the ethanol solution was concentrated and added to a high reflection glass. After the ethanol was completely evaporated, LDIR testing was performed. The particle analysis method and microplastics spectrum library were used for the analyses. The automatic testing method was also used (matching degree > 0.65, particle size range 20–500 μm).

To avoid contamination during extraction of the microplastics, glass containers were used during the experiments. All open containers were covered with aluminum foil to prevent microplastics from contaminating the experimental samples in the environment. The high-purity water used for cleaning the instruments and preparing the solutions was obtained via filtration through a 0.45 μm glass fiber membrane. Pure water filtered through a 0.45 μm glass fiber filter membrane was used for the blank experiment. The whole experimental process was consistent with the experimental steps for the above water samples. A total of 10 microplastic particles were found from the 3 blank samples, and the blank value was deducted from the following results.

Environmental factors was collected that could affect the abundance of microplastics within a 3-kilometer radius of each park, mainly including the number of restaurants, express delivery, and supermarkets, population size, vehicle flow, the amount of household waste generated, industrial and domestic wastewater emissions and park visitors. The main methods of collection were door-to-door statistics, statistical yearbooks, and social development statistical bulletins.

To evaluate the ecological risk of microplastic pollution in typical wetland parks, the potential ecological risk index (PERI) was calculated based on the abundance of microplastics and polymer types. The formula was: { E k = T K × Ci C O T k = P k × S k PERI = ∑ k = 1 M E k where C_i is the abundance of microplastics; C₀ is the standard reference value, and Everaert, Van Cauwenberghe, De Rijcke, Koelmans, Mees, Vandegehuchte and Janssen (42) calculated the safe abundances of microplastics in surface water (6,650 n L⁻¹) and sediment (540 n kg⁻¹); T_k is the chemical toxicity coefficient of polymer k; P_k is the percentage of polymer k; S_k is the hazard fraction of polymer k (43), and the specific values are shown in Table 1; and E_k and PERI are the potential ecological risk factors and potential ecological risk indices, respectively. The threshold values for the potential ecological risk classification standards were I (PER<10), II (10 ≤ PER<100), III (100 ≤ PER<1,000), and IV (1,000 ≤ PER<10,000).

The abundance of microplastics in the water and sediment of the parks was expressed in n L⁻¹ (the number of microplastics contained in each liter of water sample) or n kg⁻¹ (the number of microplastics contained in each kilogram of sediment sample). The data were analyzed with R version 4.2.2.¹ Heatmaps were generated to demonstrate the spatial distributions of the microplastics across the sampling sites. To better understand the differences among the various types of microplastics and city parks, the abundance data were standardized with Z score transformation, and bidirectional clustering was conducted on both the microplastic types and sampling sites based on the Bray–Curtis distance. Spearman correlation analyses were used to evaluate the relationship between the microplastic abundance in surface water and that in sediment.

The microplastic abundances in the surface waters and sediments of eight typical parks in Changsha are shown in Figure 2. The abundances of microplastics in the surface waters of all parks ranged from 150 to 525 n/L, and the abundances of microplastics in the sediments ranged from 120 to 585 n/kg. Among the microplastics in the surface water, MX had the highest abundance (525 n L⁻¹). This may be because MX is located in the center of Changsha, surrounded by residential areas around the lake, and with more artificial labor. The abundances of microplastics in XH, YD, and YHD were also relatively high, at approximately 400 n L⁻¹. This may be due to the high pedestrian flow in the parks. For example, YD is located on the campus of Central South University, and YHD is the largest ecological wetlands park with a high number of tourists. The abundances of microplastics in TS and YH were relatively low, approximately 150 n L⁻¹. This may be related to their locations in the suburbs and the low numbers of tourists. Among the abundant microplastics in the sediments, MX and YHD also had the highest levels of microplastics (585 n kg⁻¹), which was consistent with the levels of microplastics in the surface waters, confirming that MX had the most severe microplastic pollution. Surprisingly, the abundances of microplastics in the sediments of TS, XH, XJ, YD, and YH were relatively low, at approximately 120 n kg ⁻¹, which was inconsistent with the microplastic levels in the surface water. These results indicated that the abundance of microplastics in surface water does not represent the abundance of microplastics in the sediment. It was also possible that some microplastics were distributed in water, while others were distributed in sediment.

We arranged the samples according to their geographical information, such as MX and XJ (located in the middle, close to each other, with the same direct water source); YD, XH, and YSD (located in the south, all close to the Xiangjiang River); LYH and YH (located in the northeast, close to each other); TS (located in the northwest). The distribution patterns of various microplastics species among the different sampling points are shown in Figure 3. A total of 23 different polymers were identified by separating the microplastics from the surface waters of the eight parks (Figure 3A). Among them, a total of 10 polymers were identified in the surface water of LYH, and CPE showed the highest abundance at 80 n L⁻¹. A total of 15 polymers were identified in the surface water of MX, and PET and FLU showed the highest abundances at 85 and 95 n L⁻¹, respectively. A total of 10 polymers were identified in the surface water of TS, and PET, JF, and PP showed the highest abundances at 35, 30, and 35 n L⁻¹, respectively. A total of 16 polymers were identified in the surface water of XH, and the highest abundance was that of PET at 305 n L⁻¹. A total of 12 polymers were identified in the surface water of XJ, and the highest abundance was that of PET occurring at 395 n L⁻¹. A total of 10 polymers were identified in the surface water of YD, and the highest abundances were those of PET and FLU at 30 and 40 n L⁻¹, respectively. A total of 11 polymers were identified in the surface water of YH, and the highest abundances were those of ACR and PU at 45 and 35 n L⁻¹, respectively. A total of 16 polymers were identified in the surface water of YHD, and the highest abundance was that of FLU at 165 n L⁻¹. Based on the Bray–Curtis distance, a hierarchical clustering analysis of the microplastic compositions in the water was carried out (Figure 3B). The results showed that PET, unlike the other microplastics, exhibited a larger proportion in the waters from all sites except YH. The other microplastics were divided into three categories. The first category included PS, PBAT, PE, and SBR, which were the smallest groups, especially in MX, YHD, YD, and TS. The second category included FLU, CP, CPE, ACR, PU, JF, PP, PVC, and PR, which exhibited substantial proportions in MX, YHD, YD, and TS and showed a complementary trend with LYH and YH (increasing and decreasing), while in XH and XJ, they were all among the middle group. The third category included SBS, PTFE, PMMA, BR, PC, PD, PLA, EVA, and EAA, which exhibited lower proportions in XH and XJ, while in MX, YHD, YD, JS, and YH, they were generally present at low levels but had greater levels in individual areas (Figure 3B).

A total of 19 different polymers were identified by separating the microplastics from the sediments of the 8 parks (Figure 3C). Among them, a total of 15 polymers were identified in the sediment of LYH, and PVC showed the highest abundance at 95 n L⁻¹. A total of 11 polymers were identified in the sediment of MX, and the abundance of ACR was the highest at 430 n L⁻¹. A total of eight polymers were identified in the sediment of TS, and PET and FLU showed the highest abundances at 40 and 30 n L⁻¹, respectively. A total of 10 polymers were identified in the sediment of XH, and PU showed the highest abundance at 45 n L⁻¹. A total of nine polymers were identified in the XJ sediment, and PVC and EVA showed the highest abundances at 35 and 25 n L⁻¹, respectively. A total of 10 polymers were identified in the YD sediment, and CP, PU and ACR showed the highest abundances at 50, 25, and 24 n L⁻¹, respectively. A total of nine polymers were identified in the YH sediment, and the highest abundance was that of ACR at 155 n L⁻¹. A total of 13 polymers were identified in the YHD sediment, and the highest abundance was that of CP at 365 n L⁻¹. Based on the Bray–Curtis distance, a hierarchical clustering analysis of the microplastic composition structure in the water was carried out (Figure 3D). In the sediments, ACR and CP had the highest abundances, and their distributions differed from those of the other microplastics. These were the main microplastic components in MX, YH, YHD, and YD. The other microplastics were divided into three major categories, and BR, SBS, ABS, PTFE, and FR were the least common, especially in YD, XH, TS, LYH, and XJ. The second category included PVC, CPE, EVA, FLU, PP, PET, and PU. In contrast to those in the first category, these were generally the most common components in YD, XH, JS, LYH, and XJ, but they were relatively infrequent in MX, YH, and YHD. The third category included PRs, MBSs, EAAs, PEs, and PSs, with distribution characteristics similar to those of the first category (Figure 3D).

Acrylate copolymers (ACRs) generally refer to acrylic acid hydroxypropyl acrylate copolymers. Acrylate is a general term for esters of acrylic acid and its homologs, which are monomers used in manufacturing adhesives, synthetic resins, special rubbers, and plastics. Acrylate copolymers are used in the coatings, adhesives, leather, chemical fibers, papermaking, and printing industries. Raman spectroscopy revealed that ACR exhibited peaks at 1,445 and 1,716 cm⁻¹ for a symmetric stretching vibration of N=C=O and an asymmetric stretching vibration for carbon atoms in the benzene ring (Figure 4). Polypropylene (PP) is formed by addition polymerization of propylene. PPs are widely used in the production of fiber products such as clothing and blankets, medical equipment, automobiles, bicycles, parts, transportation pipelines, chemical containers and food and drug packaging. The largest areas of polypropylene use in China include products such as woven bags, packaging bags, and tying ropes, which account for approximately 30% of the total consumption. Raman spectroscopy revealed that PP exhibited peaks at 1,458 and 1,363 cm⁻¹, corresponding to a CH₂ bending vibration and a CH₂ rocking vibration, respectively (Figure 4). Polystyrene (PS) is synthesized by free radical addition of styrene monomers. PS is mainly used for insulation, shock resistance, packaging materials, and floating products. The universal type (R) is suitable for packaging materials, while the flame-retardant type (F) is suitable for building and insulation materials. PS can also be used to make foamed plastic products and films. Raman spectroscopy revealed that PS exhibited peaks at 1,000, 1,445, 1,530, and 1,602 cm⁻¹, corresponding to the symmetric stretching vibrations of C-C bonds in the benzene ring, the symmetric stretching vibration of N=C=O, the stretching motion of C-C, and the stretching vibrations of C and O atoms in carboxyl groups, respectively (Figure 4). Fluorine rubber (FLU) provides heat resistance, oxidation resistance, oil resistance, corrosion resistance, and atmospheric aging resistance. It is used for mechanical seals, pumps, reactors, mixers, compressor casings, valves, various instruments, and other equipment, such as valve seats, valve stem fillers, diaphragms, and gaskets. It has been widely used in the aerospace, aviation, automotive, petroleum, and household appliance fields. Raman spectroscopy revealed that FLU exhibited peaks at 1,151 and 1,168 cm⁻¹ for a C-C stretching vibration, a CH bending vibration, a CH₃ oscillation, and a C-C rocking vibration, respectively (Figure 4). Polyethylene (PE) is an important fiber material and a common component of fragmented microplastics. PEs are often used in cling films, vest-style plastic bags, plastic food bags, milk bottles, buckets, water bottles, etc. PEs are mainly divided into three categories: linear low-density polyethylene (LLDPE), low-density polyethylene (LDPE), and high-density polyethylene (HDPE). Raman spectroscopy revealed that PE exhibited peaks at 1,439 and 1,461 cm⁻¹, indicating cross-and swaying vibrations of CH₂ groups (Figure 4). The ethylene-vinyl acetate copolymer is used as a hot melt adhesive (EVA) and is composed of the basic resin, tackifiers, viscosity regulators, and antioxidants. EVAs with vinyl acetate contents less than 5% are used produce films, wires and cables, LDPE modifiers, adhesives, etc. EVA products with vinyl acetate contents ranging from 5 to 10% are elastic films, etc. EVAs with vinyl acetate contents ranging from 20 to 28% are mainly used for hot melt adhesives and coating products. Raman spectroscopy revealed that EVA exhibited a peak at 1,220 cm⁻¹ corresponding to a torsional vibration of CH₂, a rocking vibration of CH, and a stretching vibration of C-C. Polyurethane (PU) can be made into foam plastics, elastomers, fiber plastics, fibers, leather shoe resins, coatings, adhesives and sealants, among which foam plastics account for the largest proportion. In addition, polyurethane, an important waterproof coating, is widely used in building roofs, exterior walls, roof slabs, basements, kitchens and bathrooms, roads and bridges, and other parts. Raman spectroscopy revealed that PU exhibited peaks at 1,283, 1,445, and 1,732 cm⁻¹, corresponding to C-C stretching vibrations of benzene rings and carboxyl groups, a symmetric stretching vibration for N=C=O, a bending vibration of CH₂, and a stretching vibration of C=O, respectively. Polyethylene terephthalate (PET) is an important synthetic fiber that can be spun and reprocessed into polyester. Polyesters are characterized by high strengths, good elasticities, strong heat resistances, and strong thermoplastic properties and have broad application prospects in the clothing, construction, and automobile industries. Raman spectroscopy revealed that PET exhibited peaks at 1,168, 1,435, and 1716 cm⁻¹, corresponding to the stretching and rocking vibrations of C-C, the bending vibrations of CH₂, and the asymmetric stretching vibrations of carbon–carbon bonds in the benzene ring, respectively. Polyvinyl butyral (PVB) is a condensate of polyvinyl alcohol and butyraldehyde and is mainly used in manufacturing laminated glass, coatings, and adhesives. In addition, PVB is widely used for bonding wood, ceramics, metals, plastics, leather, laminated materials, etc. Raman spectroscopy revealed that PET exhibited peaks at 1,283, 1,439, 1,445, and 1,461 cm⁻¹ for the C-C stretching vibrations of the benzene rings and carboxyl groups, the symmetric stretching vibration of N=C=O, and the cross and sway vibrations of CH₂.

The physical and chemical properties of the surface waters and sediments in the eight parks were measured, and the results are shown in Figure 5. The total N content in the MX surface water was the highest at 1.28 mg L⁻¹, while the total N contents in the TS and XJ waters were the lowest at 0.65 and 0.41 mg L⁻¹, respectively. This was mainly because there were more residential areas and a greater population around MX, while there were no residential areas around TS and XJ, resulting in a more beautiful ecological environment. The NH₄⁺ concentrations in the surface waters of MX and YD were 1.28 and 1.24 mg L⁻¹, respectively, which were the highest among those in all the parks. The NH₄⁺ contents in the TS and XJ water samples were the lowest, at 0.48 and 0.32 mg L⁻¹, respectively. The total P content in the surface water of XH was 5.15 mg L⁻¹, which was the highest among all parks. The total P contents in the surface waters of LYH, TS, and XJ were relatively low, with values of 2.15, 1.84, and 1.98 mg L⁻¹, respectively. The pH of the park’s water samples range from 7.22 to 7.71, and the pH of the sediment samples range from 7.37 to 7.82, both of which were alkaline. Among them, the pH of MX water samples and sediment samples were 7.71 and 7.82, respectively, while the pH of YD water samples and sediment samples were 7.63 and 7.64, respectively. This may be due to the large number of residents around the park, which produced more alkaline water in daily life, leading to an increase in pH in the park’s water. The COD contents in the surface waters of MX and YD were 21.3 and 21.7, respectively, which were the highest among all parks. The COD contents in the TS and XH water samples were the lowest, at 9.3 and 8.2, respectively. The BOD contents in the surface waters of LYH and MX were 3.8 and 4.1, respectively, which were the highest among all the parks. The COD contents in the JS and XJ water samples were the lowest, at 2.8 and 2.6, respectively. Among the sediments of the eight parks, MX and YH had the highest total P contents at 334 and 366 mg kg⁻¹, respectively. The total P contents in the TS and XH sediments were the lowest, at 97 and 110 mg kg⁻¹, respectively. Based on measurements of the heavy metal contents in the sediments, the YH and YHD sediments had the highest Pb contents, at 50 and 59 mg kg⁻¹, respectively. The Pb content in the TS sediment was the lowest, at 1.8 mg kg⁻¹. The Zn contents in the LYH and MX sediments were the highest, at 198 and 182 mg kg⁻¹, respectively. The Zn contents in the TS and XH sediments were the lowest, at 43 and 47 mg kg⁻¹, respectively. The highest Cu content in the MX sediment was 124 mg kg⁻¹. The Cu contents in the XH and YHD sediments were the lowest, at 38 and 37 mg kg⁻¹, respectively.

Human activity is the only source of microplastics and directly determines their distributions in the environment. A regression analysis revealed a significant positive correlation between human activities and the abundance of microplastics in park water (Figure 6). Among them, the correlations of population, industrial discharge and domestic wastewater discharge with the abundance of microplastics in park water were the strongest, followed by the correlations of the number of stores (supermarkets, express delivery stores, hospitals, and restaurants) and household waste production with the abundance of microplastics in park water. The correlations of car flow and tourists and with the abundance of microplastics in park water were the weakest. These results indicated that peoples’ daily family lives was the main factor affecting the distributions of microplastics in park water. Daily life generated various types of household waste and wastewater, which were discharged into rivers or bottom waters, thereby affecting the abundance of microplastics in park water. There were certain correlations between the abundance of microplastics in park sediments and the population, industrial and domestic wastewater discharge, and number of stores (supermarkets, express delivery stores, hospitals, and restaurants), while the correlations of the abundance of microplastics in park sediments with the amount of domestic waste generated, car traffic, and tourist traffic were relatively weak. This indicated that peoples’ daily family lives affected the abundances of microplastics in park sediments. However, the abundance of microorganisms in the sediment could have an impact on the abundance of microplastics. Therefore, it was necessary to continue exploring the environmental factors that affected the abundances of microplastics in sediments.

The potential ecological risk index (PERI) was used to evaluate the ecological risks caused by microplastic pollutants in the waters and sediments of the eight parks in Changsha (Figure 7). The PERIs of the microplastics in the various samples varied greatly, ranging from 53 to 8,549. Among the water samples, those from LYH, MX, and YD had PERI values in the IV risk zone; XH, XJ, and YHD had PERI values in the III risk zone; and TS and YH had PERI values in the II risk zone. In the sediment samples, MX and YD had PERIs in the IV risk zone; LYH, XH, XJ, and YHD had PERIs in the III risk zone; and TS and YH had PERIs in the II risk zone. These results indicated that the ecological risks from microplastics in the surface waters of the parks were basically consistent with those of the microplastics in the sediments. In addition, all the samples from the parks were above the Level II risk zone, indicating that there were ecological risks to the environments of these parks. The governance of microplastics and the management of the parks should be strengthened in the future.