key: cord-0864336-npn0gh94
authors: Wang, Chuan; Li, Qianzheng; Ge, Fangjie; Hu, Ze; He, Peng; Chen, Disong; Xu, Dong; Wang, Pei; Zhang, Yi; Zhang, Liping; Wu, Zhenbin; Zhou, Qiaohong
title: Responses of aquatic organisms downstream from WWTPs to disinfectants and their by-products during the COVID-19 pandemic, Wuhan
date: 2021-11-18
journal: Sci Total Environ
DOI: 10.1016/j.scitotenv.2021.151711
sha: f1f991500ea6f4530142445600ad70970f9a19db
doc_id: 864336
cord_uid: npn0gh94

The outbreak of COVID-19 has led to the large-scale usage of chlorinated disinfectants in cities. Disinfectants and disinfection by-products (DBPs) enter rivers through urban drainage and surface runoff. We investigated the variations in residual chlorine, DBPs, and different aquatic organisms in the Hanjiang, Fuhe, and Qinglinghe Rivers in Wuhan during the COVID-19 pandemic. The sampling sites were from the wastewater treatment plant outlets to the downstream drinking water treatment plant intakes. Total residual chlorine and DBPs (dichloromethane and trichloromethane) detected in the river water ranged from 0 to 0.84 mg/L and 0 to 0.034 mg/L, respectively. The residual chlorine and DBPs showed a gradual reduction pattern related to water flow, and the concentration at intakes did not exceed the Chinese drinking water source quality standards. Phytoplankton and zooplankton densities were not significantly correlated with residual chlorine and DBPs. The fluctuations in phytoplankton resource use efficiency (RUE) and zooplankton RUE in the Fuhe River, with the highest residual chlorine, and the Qinglinghe River with the highest DBPs, were higher than those in the Hanjiang River. For benthic macroinvertebrates, the number of functional feeding groups in the Hanjiang River was higher than that in the Fuhe and Qinglinghe Rivers. The water and sediment bacterial communities in the Hanjiang River differed significantly from those in the Fuhe and Qingling Rivers. The denitrification function involved in N metabolism was stronger in the Fuhe and Qinglinghe Rivers. Structural equation modelling revealed that residual chlorine and DBPs impacted the diversity of benthos through direct and indirect effects on plankton. Although large-scale chlorine-containing disinfectants use occurred during the investigation, it did not harm the density of the detected aquatic organisms in water sources. With the regular use of chlorinated disinfectants for indoor and outdoor environments in response to the SARS-CoV-2 globally, it is still necessary to study the long-term and accumulated responses of water ecosystems exposed to chlorine-containing disinfectants.

The outbreak of coronavirus disease 2019 was declared by the World Health Organization as a public health emergency of international concern and was recognised as a pandemic (https://www.who.int/emergencies/diseases/novel-coronavirus-2019). Consequently, there has been an intensification in the need to manage urban drainage and the health of receiving water bodies (Kataki et al,. 2021; Zhang et al., 2020a) . The primary and secondary treatment of wastewater in wastewater treatment plants (WWTPs) can remove viruses to some extent, and disinfection is key to prevent infectious pathogens from entering the environment . Chlorine-containing disinfectants are commonly used for disinfection in WWTPs and drinking water treatment plants (DWTPs) because of their efficiency and cost-effectiveness (Collivignarelli et al., 2018; Srivastav et al., 2020) . Since February 2020, the China Ministry of Ecology and Environment has issued guidelines for hospital wastewater disinfection. It was suggested to use liquid chlorine, chlorine dioxide, or bleach for disinfection at a concentration of 50 mg/L (as available chlorine). The contact time for disinfection should be no less than 1.5 h, with residual chlorine limits should be no less than 6.5 mg/L (as free chlorine)

would balance the E. coli limit values and acute toxicity of Daphnia magna, Vibrio fischeri, and Pseudokirchneriella subcapitata (Collivignarelli et al., 2017) . Scholars in China reported that residual chlorine (>0.5 mg/L) could maintain biological stability, as assessed by assimilable organic carbon and biodegradable organic carbon in the reclaimed water (Ren and Chen, 2021) . Artificial neural network modelling and automated control have also been developed to improve disinfection quality and reduce chlorine disinfectant consumption (Khawaga et al., 2019; Kadoya et al., 2020) .

Increasing the disinfectant concentration will result in lower levels of viral RNA, but the level of disinfection by-product (DBP) residues is higher, which will lead to an ecological risk from the effluent water (Zhang et al., 2020b) . Therefore, researchers have been looking for a balance between the disinfection effect and the toxicity of residual chlorine.

The effluent's biological and genetic toxicity is often reported to be higher after chlorine disinfection (Zhong et al., 2019; Le Roux et al., 2017) . The disinfection of sewage is vital in controlling the spread of many diseases that are caused by microorganisms but it can also produce harmful levels of DBPs, which are closely related to human health disorders, such as anaemia, reproductive system defects, continuously introduced into the aquatic environment (Ao et al., 2021) . These compounds can serve as DBP precursors, and their potential toxicity could be enhanced after chlorination (Zhang et al., 2019) . The global use of chlorine disinfectants also increases microbial chlorine resistance, potentiating the exposure to difficult-to-treat resistant pathogens (Ekundayo et al., 2021) .

Through water circulation and disinfection of tap water and food processing, residual chlorine and DBPs can cause human health problems through drinking water and food (Maheshwari et al., 2020; Simpson and Mitch, 2021) . Upstream wastewater enters the drinking water source of downstream cities along the receiving river (Richardson and Plewa, 2020) . Soluble microbial products discharged into surface water can affect the formation of DBPs in DWTPs (Zhang et al., 2020c) .

The impact of disinfectants and DBPs on aquatic organisms and water ecosystems via water transmission from WWTPs to the water sources of DWTPs has been has been of concern to researchers since large-scale disinfection in China, especially in Wuhan City, during the spread of COVID-19 (Zhang et al., 2020a) .

However, we currently know very little about this. Therefore, we surveyed the Hanjiang, Fuhe, and Qinglinghe Rivers in Wuhan City during the COVID-19

pandemic. The sample sites of each river were from WWTP where the sewage water Lugol's iodine solution, and returned to the laboratory for qualitative and quantitative species analyses. Benthic macroinvertebrates were collected using a Peterson mud harvester, washed in situ with a 40-mesh screen, selected with tweezers, and stored in a 4% formalin solution for microscopic examination and weighing. Approximately 500 mL of each water sample was filtered using qualitative filter paper followed by Millipore 0.22-μm hydrophilic nylon membranes. The membrane discs and fresh sediment were separately placed in sterilised centrifuge tubes and stored at -80°C until DNA extraction. -N) were determined as previously described (Bai et al., 2020) . The sediment TN and TP were determined using the persulfate digestion method. The sediment organic matter (SOM) was determined as loss on ignition.

The chlorine concentration (including total residual chlorine and free chlorine) was measured using a Q-CL501B residual chlorine and total chlorine analyser (Shenzhen Sinsche Technology CO. Ltd, Shenzhen City, China). The measurement was based on the DPD (N, N-diethyl-1,4-phenylenediamine) colorimetric method.

DBPs including trichloromethane (TCM), dichloromethane (DCM), and tribromethane (TBM) in water and sediment samples were detected by purge and trap/gas chromatography-mass spectrometry (Shimadzu GCMS-QP2020, Shimadzu generate the raw reads. The reads were spliced according to the overlapping relationship, and the sequence quality was qualified and filtered, followed by OTU cluster analysis and species classification analysis. total residual chlorine, trichloromethane and dichloromethane, respectively.

Significant differences between each paired sample are shown by asterisks (*, P <0.05; **, P <0.01).

A total of 135 species belonging to 83 genera from 8 phyla were identified. The J o u r n a l P r e -p r o o f relative abundance and biomass percentage of phytoplankton are shown in Fig. 4a and 4b. Cyanophyta was dominant in abundance, whereas Dinoflagellate were dominant in biomass. The total number of phytoplankton species in the Hanjiang, Fuhe, and Qinglinghe Rivers were 63, 89, and 101, respectively (Fig. 4c) . The phytoplankton biomass in the Hanjiang, Fuhe, and Qinglinghe Rivers ranged from 0.01-0.04 mg/L, 0.45-3.15 mg/L, and 0.04-66.87 mg/L, respectively (Fig. 4d) .

The algae cell density in the Hanjiang, Fuhe, and Qinglinghe Rivers ranged from 1.62 × 10 4 -2.08 × 10 5 cells/L, 1.29 × 10 6 -6.84 × 10 6 cells/L, and 9.54 × 10 4 -4.17 × 10 7 cells/L, respectively (Fig. S3a) The Qingling River had the highest RUEpp, followed by the Fuhe and Hanjiang Rivers (Fig. 5a) . Bacillariophyta mainly contributed to the RUE of the Hanjiang River, and the dominant contribution did not change along with water flow. The RUE of the 

A total of 96 species of zooplankton were identified, including 30 protozoa, 44 rotifers, 12 cladocera, and ten copepods. The relative abundance and biomass percentages of zooplankton are shown in Fig. 6a and 6b , respectively. Protozoa were dominant in abundance, while rotifers were dominant in biomass. The total zooplankton species in the Hanjiang, Fuhe, and Qinglinghe Rivers were 39, 81, and 67, respectively (Fig. 6c) . The zooplankton biomass in the rivers of Hanjiang, Fuhe, and Qinglinghe ranged from 0.02-0.34 mg/L, 0.63-3.51 mg/L, and 0.04-5.49 mg/L, respectively (Fig. 6d) .

The zooplankton density in the Hanjiang, Fuhe, and Qinglinghe Rivers ranged from 2.6-92.8 ind./L, 690.5-2976.04 ind./L, and 6,018.2-14,206.5 ind./L, respectively (Fig. S4a) . The Qinglinghe River had the highest zooplankton biomass and density, followed by the Fuhe and Hanjiang Rivers. The dominant taxon in the Hanjiang River was copepoda, except for HJ3, which is about 3 km downstream of the Shiyang WWTP (Fig. S4b) . The dominant taxon in the Fuhe River changed from protozoa to rotifera from the Sanjintan WWTP to Dijiao DWTP samples (Fig. S4c) .

The percentage of protozoa density decreased from 85.44% to 13.03%, and that of rotifera increased from 13.24% to 86.89%. The dominant species changed from dominant species changed from Arcella sp. to Tintinnopsis wangi.

The Hanjiang River had the highest RUEzp, followed by the Fuhe and Qingling rivers (Fig. 7a) . Copepoda mainly contributed to the RUEzp of the Hanjiang River, and the dominant contribution did not change with the water flow. The RUEzp of the Fuhe and Qingling Rivers is dominant from rotifera to cladocera and from rotifera to copepoda from the WWTP to 3 km downstream, respectively ( Fig. 7b-d) The number 1-4 represent the sampling sites from WWTP to DWTP in sequence. a)

The RUE of each taxon at different sampling sites; b-d) the RUE contribution of each taxon for the Hanjiang, Fuhe, and Qinglinghe Rivers, respectively.

The benthic macroinvertebrates identified belonged to the phyla Annelida, Mollusca, and Arthropoda. They accounted for 25%, 35%, and 40% of the total number, respectively (Fig. 8a) . The Hanjiang, Qinglinghe, and Fuhe Rivers were dominated by annelids, arthropods, and molluscs, respectively.

According to the functional feeding groups in the three rivers, the identified zoobenthos were classified into scrapers, gatherer-collectors, filter-collectors, and predators (Fig. 8b) . The Hanjiang, Qinglinghe, and Fuhe Rivers were dominated by filter-collectors, gatherer-collectors, and scrapers, respectively. The Hanjiang River had the most abundant functional feeding groups, and all four groups were detected.

The density and biomass of zoobenthos in the Hanjiang River were 128 ind./m 2 and 19.09 g/m 2 , respectively. Sensitive species were found only in the Hanjiang River.

The density and biomass of zoobenthos in the Qinglinghe River were 80 ind./m 2 and 0.2 g/m 2 , and only pollution-tolerant species were found there. The density and biomass of zoobenthos in the Fuhe River were 128 ind./m 2 and 343.122 g/m 2 , respectively. Moreover, medium-tolerant species were found there.

The three rivers were all in polluted states when evaluated using the Shannon-Wiener diversity index. The Fuhe River was the most severely polluted when evaluated using the Pielou evenness index. The benthic pollution index showed

Journal Pre-proof that the rivers were in a moderately polluted state (Table S1) . 

A total of 2,062,281 high-quality 16S rDNA sequence reads were obtained from sediment and water samples using the Illumina HiSeq platform. After equalising the high-quality reads, the equalised reads were assigned to 6,778 operational taxonomic units (OTUs) at 97% similarity. All the OTUs were clustered into 18 phyla or unclassified (relative abundance > 0.1%), with 45.29% classified as Proteobacteria ( Fig. S5) , followed by Actinobacteria (16.32%), Bacteroidetes (8.87%), Chloroflexi (4.75%), and Verrucomicrobia (3.85%). Acinetobacter (5.38%) and Pseudomonas River was unique among the three rivers, both in the sediment and water samples.

Verrucomicrobia was more abundant, and Chloroflexi was less abundant in the Hanjiang River sediment than those in the Fuhe and Qinglinghe sediment (Fig. S5 a) .

Proteobacteria were more abundant, and Bacteroidetes and Cyanobacteria were less abundant in the Hanjiang water than those in the Fuhe and Qinglinghe water (Fig. S5   b) .

The OTUs of sediment and water samples were clustered by the Bray-Curtis algorithm using a weighted calculation method, considering the presence or absence of species and the abundance of species. Hanjiang was a separate branch in both sediment and water samples, while Fuhe and Qinglinghe were interlaced (Fig. 9a) .

NMDS at both the genus and order levels showed that Hanjiang was a distinct group and the difference between river sediment was larger than that between the water samples ( Fig. 9b-c) . ANOSIM revealed a significant difference in the community structure between Hanjiang and Fuhe and between Hanjiang and Qinglinghe at the OTU level (Fig. 9d-e) . When analysed at the genus, family, order, class, and phylum levels ( Table S2) .

J o u r n a l P r e -p r o o f More biomarkers were found in the sediment samples than in the water samples, as revealed by Linear Discriminant Analysis (LDA) (Fig. S7.a-b) . Based on the LAD scores, the list of top biomarker taxa in the HJ sediment was class Alphaproteobacteria and phylum Verrucomicrobia (Fig. S7a) Aquabacterium in the family Burkholderiales_incertae_sedis (Fig. S7b) . The relative abundances of genus Pseudomonas in Hanjiang were 8.30 and 36.11 times greater than Fuhe and Qinglinghe, respectively. The relative abundances of genus was stronger in Hanjiang water, while the synthesis of sulphur-containing amino acids was stronger in Fuhe and Qinglinghe water (Fig. 10b) . Assimilatory nitrate reduction in the sediment showed a tendency contrary to that in water (Fig. 10c) . significantly related to the largest number of taxa in the water samples ( Fig. 11a-b) .

Sediment OM% and water TP were the environmental factors with the highest degree of explanation for species information tested by RDA (Fig. S8) . DBP formation depends on the quantity and quality of DOM, the sources of which include terrestrial foliar litter, atmospheric dry and wet deposition, algae, microplastics, etc. (Chen et al., 2020; He et al., 2020; Lee et al., 2020) .

Trihalomethanes ( (Zhang et al., 2020a) . Adverse effects on the early development of Cyprinus carpio and acute toxicity following chlorine exposure in black tiger shrimp and mussels have been reported (Verma et al., 2007; Lin and Husnah, 2002; Rajagopal et al., 2003) . The abnormality and damage were possibly caused by altered protein metabolism, impairment of the gluconeogenic pathway, and subsequently the glycolytic pathway, and alteration of membrane transport and neurotransmission (Verma et al., 2007) . A strong inhibitory effect on Chlorella sp. growth was observed at 0.20 mg/L of residual chlorine (Jin et al., 2014) . In this study, although the residual chlorine concentration in the water samples was higher than 0.20 mg/L, the inhibition of total phytoplankton cell density, RUEpp, and diversity were not apparent. The 24-h LC 50 values of the total residual chlorine for glochidia were between 70 and 220 ug/L, 2.5 to 37 times higher than that for cladocerans (Valenti et al., 2010) . However, positive correlations were found between free chlorine and Rotifera RUE (P = 0.001), Protozoa density (P = 0.003), Rotifera density (P = 0.028), and Cladocera density (P = 0.047). Rotifera provided food for benthic macroinvertebrates, which might have caused the indirect impact of residual chlorine on benthic macroinvertebrates. Ecosystems have a buffer and recovery capability against external pressure; therefore, acute toxicity and short-term inhibitory effects in natural water bodies do not appear when a single contaminant concentration reaches the critical value of the biological response under experimental conditions. However, long-term and accumulated effects are still worthy of attention.

Apart from the impact on aquatic organisms, a positive association between an increased risk of bladder cancer, colorectal cancer, poor pregnancy outcomes, with long-term exposure to chlorinated water has been shown (Jr et al., 2007; Bove et al., 2002) . The discharge of chlorine-containing tailwater can lead to the loss of nitrifying microorganisms, destroying the ammonia nitrogen-nitrite-nitrate cycle system in river water, affect the natural water nitrogen cycle process, and cause harm to the aquatic environment and aquatic organisms . Through the metabolic transformation of diversified microorganisms and the transmission of the food chain, negative effects will be buffered through the self-stable regulatory mechanism of the ecosystem. Functional diversity is better than structural diversity in maintaining the stability of aquatic ecosystems. Structural diversity was not significantly affected, but functional diversity showed signs of decline. This requires further attention and further assessment to determine whether this decline is recoverable.

Potential risks caused by chlorine disinfectants and DBPs have increased with the increasing use of pharmaceutical and personal care products. Irbesartan, the medicine for treatment of high blood pressure, transformed into five completely new DBPs, the highest toxicity of which reached 12 times than that of irbesartan (Romanucci et al., distribution systems (Kennedy et al., 2021) . Therefore, their impact on aquatic ecosystems has received continued attention. For aquatic organisms, the direct impact of residual chlorine and DBPs on benthic organisms is far greater than their indirect transmission through aquatic organisms. Other indirect impacts are not included in the model, such as important consumer fish in the aquatic food chain.

Apart from evaluating ecological integrity in the receiving water, the protection of the drinking water sources and aquifers also needs to be considered. The protection and restoration of drinking water sources are key methods for DBP source control.

Despite the residual chlorine was found higher than the standard for drinking water quality in China at some sites downstream the WWTPs, the ecosystems and drinking water sources were not significantly negatively impacted, which is inseparable from the setting of the ecological buffer zone around the drinking water sources. Therefore, continuous protection is an effective method for drinking water safety and an important guarantee for using an ecological buffer zone to strengthen the impact resistance of the water ecosystem and maintain a self-organized balance.

The variations in residual chlorine and DBPs and different types of aquatic organisms were studied from the three WWTPs to the downstream drinking water sources of DWTPs during the COVID-19 epidemic when large-scale chlorine-containing disinfectant was used. The residual chlorine and DBPs gradually J o u r n a l P r e -p r o o f decreased along the water flow, which met the quality standards of the drinking water sources. The community structure of phytoplankton, zooplankton, benthic macroinvertebrates, and bacteria in the water and sediment changed variously. The

Hanjiang River has the largest volume among the three rivers. Therefore, due to the dilution effect, the concentrations of residual chlorine and DBPs were lowest in the Hanjiang water, and the fluctuation of various aquatic organisms was the lowest in the Hanjiang River. The diversity of benthic macroinvertebrates was influenced by the direct effects of residual chlorine and DBPs and the indirect effects of bacterioplankton and zooplankton. Although there was no significant inhibition of the density of phytoplankton and zooplankton, the long-term and accumulated effects on aquatic organisms are still worthy of further investigation.

Peracetic acid-based advanced oxidation processes for decontamination and disinfection of water: A review

Spatial and seasonal variation of water parameters, sediment properties, and submerged macrophytes after ecological restoration in a long-term (6 year) study in Hangzhou west lake in China

Drinking Water Contaminants and Adverse Pregnancy Outcomes: A Review

Guidelines for Canadian drinking water quality summary table, Water and Air Quality Bureau， Healthy Environments and Consumer Safety Branch， Health Canada

Role of wastewater treatment in COVID-19 control

Study on secondary risk of water environment under enhanced disinfection of wastewater treatment plant during epidemic prevention and control

Disinfection in Wastewater Treatment Plants: Evaluation of Effectiveness and Acute Toxicity Effects

Overview of the Main Disinfection Processes for Wastewater and Drinking Water Treatment Plants

Seasonal and spatial evolution of trihalomethanes in a drinking water distribution system according to the treatment process

Global bibliometric meta-analytic assessment of research trends on microbial chlorine resistance in drinking water/water treatment systems

Cyanobacteria dominance influences resource use efficiency and community turnover in phytoplankton and zooplankton communities

Investigation of trihalomethanes formation potential in Karoon River water

Characterization of dissolved organic matter derived from atmospheric dry deposition and its DBP formation

Effects of residual chlorine discharged in water on the growth of phytoplankton

Case control study of the geographic variability of exposure to disinfectant byproducts and risk for rectal cancer

Regularized regression analysis for the prediction of virus inactivation efficiency by chloramine disinfection

Concerns and strategies for wastewater treatment during COVID-19 pandemic to stop plausible transmission. Resources Conservation and Recycling

Regulated disinfection byproduct formation over long residence times

Model identification and control of chlorine residual for disinfection of wastewater

Chloramination of wastewater effluent: Toxicity and formation of disinfection byproducts

Characteristics of microplastic polymer-derived dissolved organic matter and its potential as a disinfection byproduct precursor

Drinking Water Disinfection Byproducts (DBPs) and Human Health Effects: Multidisciplinary Challenges and Opportunities

Toxicity of chlorine to different sizes of black tiger shrimp (Penaeus monodon) in low-salinity shrimp pond water

Disinfection by-products formation and acute toxicity variation of hospital wastewater under different disinfection processes. Separation and Purification Technology

Optimization of disinfectant dosage for simultaneous control of lead and disinfection-byproducts in water distribution networks

Study on disinfection by-products and treatment technology of tail water in urban sewage treatment plants. Applied Chemical lower Phong river dissolved organic matters and formations of unknown chlorine dioxide and chlorine disinfection by-products by Orbitrap mass spectrometry

Tolerance of five species of tropical marine mussels to continuous chlorination

Effect of residual chlorine on the interaction between bacterial growth and assimilable organic carbon and biodegradable organic carbon in reclaimed water

To regulate or not to regulate? What to do with more toxic disinfection by-products?

Disinfection by-products and ecotoxic risk associated with hypochlorite treatment of irbesartan

Chlorine and ozone disinfection and disinfection byproducts in postharvest food processing facilities: A review

Disinfection by-products in drinking water: Occurrence, toxicity and abatement

Influence Analysis and Suggestion of Disinfectants on the Wastewater Treatment Plant and Water Environment during Epidemic Period

USEPA (2006) National primary drinking water regulations: stage 2 disinfectants and disinfection by-products rule

Chlorine toxicity to early life stages of freshwater mussels (Bivalvia: Unionidae)

Persistent sub-lethal chlorine exposure elicits the temperature induced stress responses in Cyprinus carpio early fingerlings

Disappearance rates of chloramines in river water

Impact of DBPs on the fate of zebrafish; Behavioral and lipid profile changes

Potential spreading risks and disinfection challenges of medical wastewater by the presence of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) viral RNA in septic tanks of Fangcang Hospital

Disinfection threatens aquatic ecosystems

DNA pyrimidine bases in water: Insights into relative reactivity, byproducts formation and combined toxicity during chlorination

Chlorination of enoxacin (ENO) in the drinking water distribution system: Degradation, byproducts, and toxicity

Disinfection byproducts and their toxicity in wastewater effluents treated by the mixing oxidant of ClO2/Cl2

The manuscript was written through contributions of all authors. All authors have given 

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.J o u r n a l P r e -p r o o f