The distribution and ecological risks of antibiotics in surface water in key cit

Dun-yu Lü , Chu Yu , Zi-jun Zhuo , , Shu-ran Meng , Song-o Liu

a Institute of Hydrogeology and Environmental Geology, Chinese Academy of Geological Sciences, Shijiazhuang 050061, Chinab Key Laboratory of Groundwater Sciences and Engineering, Ministry of Natural Resources, Zhengding 050803, China

c China University of Geosciences (Beijing), Beijing 100083, China

Keywords:Surface water Antibiotics COVID-19 epidemic Distribution characteristic Ecological risk Hydrogeological engineering Kaifeng City Yellow River China

ABSTRACT A large number of antibiotics have been discharged into rivers by human activities, posing a threat to aquatic ecosystems.The surface water of the Yellow River Basin also suffers antibiotic pollution, which hin ders the improvementin the aquatic ecological environment.This study investigated and analyzed the characteristics and assessed the ecological risks of antibiotic pollution in surface water bodies such as canals, rivers and fish ponds in Kaifeng, Henan Province, which is a key city along the lower reaches of the Yellow River.The test results are as follows.A total of 15 types of antibiotics were detected in the surface water.They had a total antibiotic concentration of 12.2-249.9 μg/L, of which tetracyclines (TCs)and quinolones accounted for the highest percentages.Six types of quinolones had detection rates of up to 100%, and doxycycline (DC) and oxytetracycline (OTC) had average concentrations of 29.52 μg/L1 and 13.71 μg/L, respectively.The major canals with water diverted from the Yellow River had total concentrations of quinolones and TCs of 22.0 μg/L and 14.9 μg/L, respectively, which were higher than those in previous studies.This phenomenon may be related to the decrease in the water flow of the Yellow River during the dry season and the increase in the antibiotic consumption of residents in the context of the Covid-19 outbreak.The upper reaches of the Huiji River in the Xiangfu District had higher antibiotic content than other districts in Kaifeng.Specifically, TCs accounted for 72.38%-91.84% of all antibiotics,and the DC and OTC concentrations were significantly higher than other antibiotics in the upper reaches.As indicated by the ecological risk assessment results, TCs had the highest ecological risks to green algae.Among them, DC had medium-high risks; TC, OTC, and chlortetracycline (CTC) had medium-high risks;trimethoprim (TMP) and lomefloxacin (LOM) had low risks; other TC antibiotics had no risk.Compared with green algae, most antibiotics showed higher ecological risks to daphnia and lower ecological risks to fish.DC and OTC dominate antibiotic pollutants in the surface water in Kaifeng City, and especially in Xiangfu District, where DC and OTC have medium-high risks.The TCs in the major Yellow River showed medium risks to both green algae and daphnia.It can be speculated that the antibiotic pollution in the Yellow River might pose a certain threat to the ecological security of water in Kaifeng City.

1.Introduction

Antibiotics are widely applied today for treating human and animal diseases and are massively used as growth promoters in agriculture, aquaculture, and livestock farming(Finley RL et al., 2013; Wallinga D et al., 2015).With the characteristics of persistent input and pseudo-persistence,antibiotics in the environment have been frequently detected in different water environments worldwide, such as rivers,lakes, groundwater, seawater, and drinking water.Exposure to these environments will induce the formation of resistance genes and drug-resistant bacteria, which pose a serious threat to human health throughout the world.Given this, theGlobal Environment Outlook6 issued by the United Nations Environment Programme alerts that antimicrobial resistance will be the leading cause of human death globally by 2050(UNEP, 2019).

China is the largest producer and consumer of antibiotics in the world.In 2013, the country consumed a total of 92700 tof antibiotics, of which about 46% were discharged into rivers through sewage (Zhang QQ et al., 2015).Moreover, antibiotic pollution has occurred in important surface waters in China,including the Pearl River, the Yangtze River, the Yellow River, the Haihe River (Liu X et al., 2019), the Songhua River(Ma JS et al., 2021), the Poyang Lake (Ding HJ et al., 2017),and the Taihu Lake (Xu J et al., 2014).The Yellow River has been the mother river of the Chinese nation since ancient times, and the Yellow River Basin occupies a very important position in China’s social and economic development and ecological security (Shi JS et al., 2014).In 2018, the total population and the regional GDP of this basin accounted for 30.3% and 26.5% of the population and GDP of the country,respectively.In 2019, the ecological protection and highquality development of the Yellow River Basin became major national strategies.TheOutline of the Yellow River Basin’s Ecological Protection and High-Quality Development Plan(2021) pointed out that the Yellow River Basin has long suffered severe environmental pollution and that strengthening the systematic management of the environmental pollution is one primary task in the future.

Antibiotic pollution in the Yellow River Basin cannot be ignored.According to research reports, dozens of antibiotics dominated by fluoroquinolones (FQs), macrocyclic lipids(MLs), and sulfonamides (SAs) have been detected in the surface water of the Yellow River Basin.The antibiotics detected in the Yellow River Basin have fewer types and lower contents than those in the Pearl River, the Yangtze River, the Haihe River, and the Liaohe River Basin.Their concentrations are 1-100 ng/L and their total concentration is 100-300 ng/L in general.From 2005 to 2016, a total of 21 types of antibiotics were detected in the Yellow River, and trimethoprim had the highest content of up to 62.5 μg/L among the detected antibiotics (Li S et al., 2018; Zhao FQ,2021; Zhao FQ, 2022).

Antibiotics in the water environment pose risks to aquatic organisms, and related research has received increasing attention.The Honghu Lake has high oxytetracycline,sulfadiazine, ciprofloxacin, and tetracycline contents, which pose medium-high risks to algae (Wang Z et al., 2017).The antibiotics in the Danjiangkou Reservoir have much lower ecological risks to different aquatic organisms in different seasons than those in major rivers in China and the Baiyangdian Lake (Li WH et al., 2012; Hu Y et al., 2018).The sulfamethoxazole has the highest ecological risks in Taihu Lake (Lin T et al., 2016).Antibiotics can also accumulate in aquatic organisms through food chains (Chen H et al., 2015), further leading to the accumulation of resistance genes in the water environment.Moreover, using the water contaminated by antibiotics as an irrigation water source may also lead to contamination of antibiotics and resistance genes in irrigated soils (Pan M and Chu LM, 2018).

Kaifeng City, which is located in the southwest of the large alluvial fan in the lower reaches of the Yellow River, is an important agricultural planting area in China and is also a core city in the Central Plains Economic Zone planned by the state (Cui Z et al., 2016; Fig.1).The Yellow River is one of the main water sources for the agricultural irrigation of the city, the designed irrigation area with water diverted from the Yellow River accounts for 90% of the total farmland area.Therefore, the ecological security of the Yellow River is inseparable from the city’s social and economic development.Moreover, the Yellow River floodplain area was once the city’s main aquaculture area.In addition, Kaifeng City has a developed water system, where canal systems and natural rivers coexist, which makes Kaifeng City known as the water city in northern China.In recent years, Kaifeng City has taken the prevention of water pollution and the improvement in the ecological environment of the Yellow River as one major task to achieve ecological protection and high-quality development in the region (The People’s Government of Henan Province,2020).However, the degree and the ecological risks of antibiotic pollution in the surface water are not well determined yet, which is not conducive to the effective implementation of comprehensive water pollution prevention.

Fig.1.Administrative divisions, main rivers and channels, and sampling point distribution of Kaifeng City, Henan Province, China.

This study mainly investigated the Huiji River, which is one major river in Kaifeng City with a total length of 191 km and a basin area of 4130 km2.It originates from the Heichi reservoir in the northwest of the urban area and flows out of the urban area from northwest to southeast, with several tributaries joining it along its pathway.The water sources of the river mainly include water diverted from the Yellow River, water receding from farmland, industrial sewage, and household wastewater.The Huiji River is the largest sewage river in Kaifeng City, and agricultural planting areas along the river coast have long been irrigated by sewage (Yang WH et al., 2016).The central urban reach of the Huiji River is planned as an area of recreational and landscape water, the section with a length of 8 km in the lower reaches is designated as a sewage disposal area, and the other reaches in the main urban area are planned as agricultural water area.This study analyzed the antibiotic pollution levels of different surface waters, such as the Yellow River, canals, and ponds,and assessed their ecological risks.The purpose of this study is to provide a scientific reference for the prevention of water pollution in Kaifeng City and other cities in the lower reaches of the Yellow River.

2.Materials and methods

2.1.Sample collection

Samples were collected from the surface water in the urban area of Kaifeng and were analyzed in November, 2020(Fig.1).Specifically, the samples were mainly taken from the major Yellow River, the canals for agricultural irrigation and water receding from farmland, natural rivers, and ponds, with sampling points approaching the national and provincial monitoring sections as far as possible.The basic information on the ten sampling points is shown in Table 1.A total of 12 sets of water samples were collected, including two sets of parallel samples.Fixed-depth samplers were employed for sampling.They were wrapped with tin foil to avoid the photolysis of antibiotics during the sampling, and the sampling depth was controlled at 0.5 m below the water surface (Ying JL et al., 2022).The collected samples were stored in 500 mL dark brown glass bottles with a PTFE liner and a screw cap, to each of which 75 mg of ascorbic acid(C6H8O6) and 0.125 g of ethylenediaminetetraacetic acid disodium (Na2EDTA) had been added.Subsequently, a 1∶1 hydrochloric acid solution was immediately added to the samples to adjust their pH to 2.Then, the samples were immediately placed into refrigerators containing ice cubes for storage at 0-4 °C.Afterward, the samples were sent to the laboratory for extraction within three days.

Table 1.Geographical location, water sample types, and other basic physical characteristics of sampling points.

2.2.Materials and reagents

Twenty-four types of antibiotic standards used in this study included the erythromycin (ETM), roxithromycin(RTM), lincomycin (LCM), azithromycin (AZM), and tylosintartrate (TYL) of MLs; the sulfadiazine (SDZ),sulfamonomethoxine (SMM), sulfamethoxazole (SMX),sulfacetamide (SCM), sulfadimethoxine (SDM),sulfamethazine (SMZ), trimethoprim (TMP), and sulfaquinoxaline (SQX) of SAs; the ofloxacin (OFL),norfloxacin (NOR), ciprofloxacin (CIP), enrofloxacin (ENR),lomefloxacin (LOM), fleroxacin (FLE), and difloxacin (DIF)of FQs, and the doxycycline (DC), tetracycline (TC),oxytetracycline (OTC), and chlortetracycline (CTC) of tetracyclines (TCs).Among these standards, LCM was purchased from Toronto Research Chemicals Inc., DIF and LOM were purchased from ANPEL Laboratory Technologies(Shanghai) Inc., OTC was purchased from China National Institute for Food and Drug Control, and other standards were purchased from Dr.Ehrenstorfer GmbH.

Methanol (CH3OH), acetone (CH3COCH3), formic acid(HCOOH), and acetonitrile (CH3CN) were all chromatographic reagents, and C6H8O6and Na2EDTA were all analytical reagents.Moreover, HLB solid-phase extraction(SPE) columns (60 mg, 3 mL) and MCX SPE columns (60 mg, 3 mL) were purchased from the American Water company.

2.3.Sample testing and analysis

SPE and ultra-performance liquid chromatographytandem mass spectrometry (UPLC-MS/MS) were employed for sample pretreatment and detection.The test instruments included Waters ACQUITY UPLC I-Class/Xevo TQ-S IVD ultra-high-performance liquid chromatography-triple quadrupole mass spectrometers (equipped with an electric spray ion source), ACQUITY BEH C18 Columns (100 mm ×2.1 mm, 1.7 μm), and 12 position N-EVAP nitrogen evaporators supplied by the Beijing Kanglin Science &Technology Co., Ltd.

40 mL of water samples were measured after being filtered using 0.45 μm filter membranes.Subsequently, the samples passed through HLB SPE columns (activated with 3 mL of CH3OH and 3 mL of ultrapure water in advance) at a flow rate of 1 drop/s.After the samples completely flew out,the HLB SPE columns were rinsed with 2 mL of ultra-pure water and were then depressured and drained.Subsequently,they were eluted with 6 mL of CH3OH eluant and were then depressured and drained again.The eluent was concentrated to be nearly dry at 40 °C using the N-EVAP nitrogen evaporators.The residues were dissolved with 1.00 mL of 80% CH3OH solution and were then filtered using 0.22 μm filter membranes for tests.

The same mobile phases were adopted for MLs, SAs,FQs, and TCs.Mobile phase-A was a 0.1% formic acid methanol/acetonitrile mixed solution, and mobile phase-B was a 0.1% formic acid solution.The gradient elution process was as follows.The initial volume of mobile phase A was set to 25%.It was increased to 60% at the 18th min and started to decrease at the 20th min after being stabilized for 2 mins.Then, it decreased to 25% at the 21th min and remained unchanged till the 25th min.

2.4.Quality control

The external standard method was adopted for the quantitative analysis of the samples.The standard curve had a concentration range of 0-200 μg/L, and good fitting effects(r2>0.99) were achieved.The blank samples (laboratory and sampling site), on-site spiked samples, and parallel samples were analyzed using instruments.The antibiotic concentrations in the blank samples did not exceed the detection limits.The sample test methods used in this study had detection limits of 0.001-0.007 μg/L and a recovery of 67.4%-103%.

2.5.Ecological risk assessment

As stated in theTechnical Guidelines for Environment and Health Risk Characterization of Chemical Substances(Trial)(the Ministry of Ecology and Environment of the People’s Republic of China, 2020) [hereinafter referred to as the Technical Guidelines (Trial)], the risk characterization ratio (RCR) of the environment can be used to determine whether a chemical substance has caused risks to the ecological environment, as well as the degree of risks.The calculation method is as follows:

wherePECwateris the predicted concentration of a chemical substance in a water environment, which can be replaced with the monitored concentration, ng/L;PNECwateris the predicted no-effect concentration for aquatic organisms, ng/L;exoToxwateris the value of key ecotoxicological effects for aquatic organisms, which is usually represented by the effect concentration EC50or EC10for the most sensitive organism,the median lethal concentration LC50, or the no observed effect concentration NOEC, ng/L;AFwateris an assessment factor, which is an empirical constant and is generally taken as 1000 and 100 for theexoToxwatervalues of acute and chronic toxicity tests, respectively (Van Leeuwen KVL, 1996;Ding HJ et al., 2017; Li S et al., 2018).

TheTechnical Guidelines (Trial)state thatRCRwater>1 indicates that a chemical substance has unreasonable risks to the environment, otherwise it indicates that a chemical substance has not been found to have unreasonable risks to the environment.More detailed grading criteria are presently adopted in relevant studies in China and abroad.Specifically,RCRwater<0.01, 0.011 denote that a chemical substance has no risk, low risks, medium risks, and high risks to the environment,respectively (Sánchez-Bayo F et al., 2002; Hernando MD et al.al., 2006).

When theRCRvalues of antibiotics vary in a large range,log10RCRwatercan be used to more intuitively and effectively reflect the differences in the ecological risks of different types of antibiotics.In this case, the grading criteria are that log10RCRwater<-2, -20 denote no risk, low risks, medium risks,and high risks, respectively.

The selection ofexoToxwaterdirectly influences the risk assessment results.At present, comprehensive antibiotic toxicity data are mainly obtained by querying the U.S.EPA’s Ecotoxicology Knowledgebase (ECOTOX) and prediction using the Ecological Structure Activity Relationships Program(ECOSAR).Owing to different experimental environments,quite different toxicity data of the organisms at the same trophic level can be obtained by querying the ECOTOX,sometimes yielding differences of 2-3 orders of magnitude for the same antibiotic (Li S et al., 2018).The ECOSAR model is an ecotoxicity analysis program developed based on quantitative structure-activity relationships (Ding HJ, 2018;Wu AM et al., 2017).For most antibiotics, the toxicity data predicted using the ECOSAR model are within the range of experimental data and thereby are highly practical.

Since the chronic toxicity of antibiotics in the environment has received more attention (Li S et al., 2019),this study mainly used the chronic toxicity data predicted using the ECOSAR model for risk assessment.For the toxicity data of antibiotics whose individual predicted values differed significantly from experimental values, other references were queried.

3.Results and discussion

3.1.Composition and contents of antibiotics in surface water

Among the concerned 24 types of antibiotics, 15 types were detected in the surface water samples from the study area, and the nine types of antibiotics that were not detected included ETM, AZM, TYL, SDZ, SMM, SCM, SDM, SMZ and SQX.The concentrations of antibiotics detected in the samples varied greatly from less than the detection limits to 177.3 μg/L (Table 2).Among these antibiotics, 12 types yielded detection rates of ≥50%.The type number of the detected antibiotics was in the order of FQs>TCs>SAs=MLs.Seven types of FQs were detected.They had detection rates of 100% except for FLE, whose detection rate was 10%.The detection rates of TCs were 20%-100%, while those of SAsand MLs were relatively low and were 50%-80% and 30%-80%, respectively.

Table 2.Detected rates and concentrations of antibiotics in surface water.

The detected antibiotics had a total concentration of 12.2-249.9 μg/L, TCs and FQs accounted for 7.52%-87.7%and 12.3% -91.84% of all the detected antibiotics,respectively.The total concentration of various antibiotics was in the order of TCs>FQs>MLs>SAs.The TCs had a total concentration of 1.5-229.5 μg/L.They were dominated by DC and OTC, whose maximum detected concentrations were 177.3 μg/L and 447.9 μg/L, respectively, and average concentrations of 29.52 μg/L and 13.71 μg/L, respectively.Other TCs had maximum and average concentrations of both less than 10 μg/L.The FQs had a total concentration of 10.7-22.0 μg/L and various detected FQs had average concentrations of 0.69-5.78 μg/L.The MLs and SAs had total concentrations of 0.2-6.9 μg/L and 0.1-1.9 μg/L,respectively, and various MLs and SAs had relatively low average concentrations.

Overall, the contents of various antibiotics detected in the surface water in this study are higher than those of previous studies.Under the criteria of a detection rate of ≥50% and a maximum detected concentration of ≥0.1 μg/L, the study area suffers serious pollution of 12 types of antibiotics, whose average concentration is in the order of DC > OTC > LOM >NOR > TC > LCM > OFL > CIP > ENR > SMX > DIF >TMP.

A total of ten types of antibiotics were detected in the Yellow River water (Y1), including six types of FQs, three types of TCs and one type of SAs.The detected antibiotics had a total concentration of up to 36.9 μg/L.They mainly included FQs and TCs, which accounted for 59.5% and 40.3%of all the detected antibiotics.In addition, SAs only accounted for 0.2% (Table 3).Compared with previous studies of the Henan section of the middle-lower reaches of the Yellow River (Xu WH et al., 2009; Xi NN, 2017), this study found that the antibiotics in the Yellow River water were still dominated by FQs whereas their types and contents significantly increased.Compared with 2014 and 2006, three and four types of FQs were newly detected in this study,respectively, and the FQ content increased six and 48 times,respectively.Three types TCs were detected in this study.They had high content, which was only slightly lower than that of FQs.MLs were not detected.For the SAs detected in this study, their type number and content were lower than those detected in the same period in 2014.Since SAs are widely used in aquaculture (Li YL, 2017), the decrease in their content may be related to the dose reduction action of drugs used in aquaculture in China since 2019 (Cheng ZK et al., 2022).

Table 3.Antibiotic concentrations in the Henan section of the middle-lower reaches of the Yellow River.

Studies have shown that the antibiotic content in surface water varies significantly with the season.It is mainly affected by water flow and is generally higher in the dry season than in the wet season (Xu WH et al., 2006; Wang WH, 2018; Liang XM et al., 2020).The sampling and investigation in this study were conducted in the dry season of the Yellow River when the water flow was only about one-third of that in the wet season (Li XN, 2006).The small water flow is one possible reason for the high antibiotic content in the Yellow River water.In addition, with the outbreak of the Covid-19 epidemic in early 2020 in China and the world,studies have found that the sales of antibiotics in some regions and countries have soared, and antibiotic abuse is on the rise(Akhtar Z et al., 2021; Giacomelli A et al., 2021; Sulis G et al., 2021).Especially at the beginning of the Covid-19 outbreak, antibiotics were widely used in the prevention and treatment of patients with mild, severe, and critical infections,although the rate of secondary bacterial infection was much lower than the antibiotic consumption rate (Cherry W et al.,2021).The FQs detected in the Yellow River are exactly the antibiotics widely used in the treatment of respiratory diseases(Wang Z et al., 2017).Although there is no available relevant research and report in China, it can be inferred from the longterm antibiotic consumption habits of the Chinese that the epidemic-induced increase in antibiotic consumption is likely the main reason for the significant increase in the antibiotic content in the water environment.

3.2.Distribution of antibiotics in surface water

Fig.2.Total content (μg/L) distribution of MLs, SAs, FQs and TCs of each sampling point.

There were close hydraulic connections among the ten sampling points, except for Y10, which was an independent sampling point of a pond.The surface water system in the study area naturally flows from northwest to southeast in general.Along the runoff direction, the antibiotic content increases significantly from the major Yellow River and the Huangbian River in the northwest to the upper reaches of the Huiji River in the southeast (Fig.2).The average antibiotic content of sampling points Y4-Y6 was 3.75 times that of sampling points Y1-Y3.Moreover, the total antibiotic content first increased and then decreased from the upper to the lower reaches of the Huiji River and decreased to the level of the upper reaches at sampling point Y7, which lies in the lower reaches of the river.The change in the antibiotic content in the surface water may be related to the antibiotic structure.The TCs of sampling points Y4-Y6 accounted for 72.38%-91.84% of the total TCs, and the TCs in sampling points Y1-Y3 and Y7 in the upper and lower reaches accounted for 32.91%-54.24% of the total TCs.Compared with FQs, TCs in the water environment are unstable and prone to hydrolysis and photolysis (Wang JH et al., 2020; Zhu XD et al., 2012).Observations show that the TC in actual water bodies has a photolysis rate of 0.072-0.118 /min and an average half-life of only 8.2 min on sunny days in winter (Hu XX et al., 2012).Therefore, the degradation of TCs may be the main reason for the decrease in the total antibiotic amount in the water of the lower reaches.The variation coefficients of TCs and FQs in the surface water were 1.61 and 0.27,respectively, indicating that there are man-made sources of pollution that increase the TC content in the surface water.In addition, sampling points Y1-Y3 lay in the major Yellow River and the urban landscape canals.These canals are equipped with linings on both sides, thus reducing the river pollution caused by the downward infiltration of surface contaminants.This is also a possible reason for the low antibiotic content of sampling points Y1-Y3.In addition, the strict control of contaminant discharge in the central urban area may also limit the antibiotic content in the canals.

Sampling points Y4-Y6 lay in natural rivers and their DC and OTC contents were much higher than the content of other antibiotics at these points (Fig.3).Both DC and OTC are widely used in humans or animals (Zhang PW, 2018).Relevant analytical results showed that there was a significant correlation between the DC and OTC content of Y4-Y6, witha correlation coefficient of 0.998 (P<0.01).Therefore, it is speculated that the DC and OTC had similar pollution sources.Sampling points Y4-Y6 lied in Xiangfu District,Kaifeng City, which is a key district involving the livestock and poultry breeding industry in Henan Province.In recent years, with a continuous increase in the scale of the livestock and poultry breeding industry, the abuse of chemical fertilizers and pesticides and the random discharge of livestock manure, rural household waste, and sewage have been increasingly prominent (Yang YT, 2017).Therefore, the livestock and poultry breeding industry is very likely the main cause of the antibiotic pollution in rivers through the direct discharge of the manure and sewage or the application of untreated livestock and poultry manure to farmland as organic fertilizers.Unlike the artificial canals, the natural rivers will not reduce river pollution since the antibiotic pollution sources on the ground surface are prone to infiltrate into the ground along with rainwater leaching and eventually entering rivers.This is another possible reason for the higher antibiotic content in the lower reaches of rivers than that in the upper reaches.

Fig.3.Thermogram of antibiotic concentrations (μg/L) detected at each sampling point.

Sampling points Y9 and Y8 lay in the canal system flowing through farmland, with the canal water mainly originating from the Yellow River and receding from farmland.The antibiotic content was roughly stable from Y1 to Y9 and decreased from Y9 to Y8 in the water flow direction.It is speculated that planting has a slight effect on the antibiotic pollution of the surface water in Kaifeng City.Sampling point Y10 lay in a fish pond and had the lowest total antibiotic concentration among all sampling points.Except for FQs and DC, no other antibiotics were detected,indicating that aquaculture did not exacerbate the antibiotic pollution of the surface water.

3.3.Ecological risk assessment of antibiotics in the surface water

The toxicity data of the assessed antibiotics were obtained through prediction using the ECOSAR model and querying related studies, as shown in Table 4.This study did not assess the ecological risks of CTC in the surface water to daphnia and fish due to the lack of valid experimental toxicity data.

Table 4.Toxicity data and assessment factors of detected antibiotics.

Fig.4 shows the ecological risks of antibiotics detected in the surface water to the organisms with different trophic levels.For green algae, the detected antibiotics yielded the log10RCR values of -3.3-1.34 (Fig.4a).Overall, TCs had the highest ecological risks to green algae.Among them, the DC had an average log10RCR of 0.57, which was the highest among the assessed antibiotics, indicating that DC has the highest ecological risks.In particular, the DC of sampling points Y4-Y6 and Y9 all had an average log10RCR of greater than 0, indicating high risks, while that of other sampling points had an average log10RCR of -1-0, indicating medium risks.The TC, OTC, and CTC had average log10RCR of-0.34, -0.09, and -0.37, respectively, suggesting mediumrisks to green algae.The TMP and LOM yielded average log10RCR of -1.43 and -1.71, respectively, indicating low risks to green algae.Other assessed antibiotics all had an average log10RCR of less than -2, indicating no risk to green algae.

Fig.4.Ecological risks of antibiotics detected in surface water to different trophic species.(a), (b), (c) are the log10RCR values of antibiotics of sampling points for green algae, daphnid and fish.

For daphnia, the detected antibiotics yielded log10RCR of-2.86-1.47 (Fig.4b).The antibiotics except for OTC had higher log10RCR for daphnia than they had for green algae,indicating that most antibiotics have higher ecological risks to daphnia than to green algae.Similar to the risks to green algae, TCs also have the highest ecological risks to daphnia.Among them, DC had an average log10RCR of 0.69,indicating thus it has the highest ecological risks to daphnia.Spatially, the DC of sampling points Y4-Y6 and Y9 had high risks to daphnia, while that at other points had medium risks to daphnia.TC and OTC had log10RCR of -1-0, suggesting medium risks.Moreover, SMX and TMP had average log10RCR of -0.07 and -0.44, respectively, indicating that SAs have medium risks to daphnia, which was higher than the risks they had to green algae.The SMX at individual points,such as Y5 and Y7, showed high risks of daphnia.In addition,the RTM and LCM of MLs showed low risks.For FQs, the LOM and ENR showed medium risks and no risk,respectively, and other antibiotics showed low risks to daphnia.

For fish, the detected antibiotics had a log10RCR of-3.91-0.37 (Fig.4c).The antibiotics except for LCM and SMX had lower log10RCR for fish than they had for green algae, indicating that most antibiotics pose lower ecological risks to fish than those they had to green algae.DC, LCM, and SMX all yielded an average log10RCR of -1-0, suggesting medium risks.Among them, DC had the highest average log10RCR and the highest ecological risk to fish.The DC of sampling point Y4 had an average log10RCR of greater than 0,suggesting high risks to fish.By contrast, the DC of sampling points Y5, Y6, Y8, and Y10 showed medium risks and the DC of other sampling points showed low risks.The TC, OTC,and TMP showed an average log10RCR of -2--1, suggesting low risks.Other antibiotics showed no risk.

It is noteworthy that the TCs of sampling point Y1 had medium ecological risks to both green algae and daphnia.As mentioned above, this phenomenon may be related to the increase in antibiotic consumption in the context of the outbreak of COVID-19, as well as the sampling and investigation period of this study.The Yellow River is the main source of water supply in Kaifeng City.From 1997 to 2013, water diverted from the Yellow River accounted for 99.23% of the surface water supply and 37.5% of the total water supply.Therefore, the water supply diverted from the Yellow River is very important to the industrial and agricultural production of Kaifeng City, as well as the life of its residents (Ma Z et al., 2016).The medium risks of the antibiotics in the Yellow River water indicate that the antibiotic pollution of the Yellow River water poses a threat to the water ecological security of Kaifeng City.

Nevertheless, antibiotics in the water environment are generally unstable, especially those in the surface waters,which are prone to degrade by light.Therefore, to further understand the contamination levels and ecological risks of the surface waters in Kaifeng City, it is necessary to conduct more comprehensive surveys and monitoring.In addition,there will be a gap between the toxicological data predicted using a model and the actual data in local areas during an ecological risk assessment.Therefore, it is also critical to accurately assess the ecological risks of antibiotics to strengthen the toxicity tests of native species under the condition of exposure to single and combined antibiotics and to build the ecotoxicity databases of China.

4.Conclusions

The key findings of this study are as follows.

(i) Antibiotics have been frequently detected in the surface water bodies in Kaifeng City, such as canals, rivers, and fish ponds.Moreover, the detected antibiotics have high contents,indicating the wide presence of antibiotic pollution in the surface water environment.Under the criteria of a detection rate of ≥ 50% and a maximum detected concentration of ≥ 0.1 μg/L, 12 types of antibiotics have been found to cause serious pollution.Among them, six types of FQs have detection ratesof up to 100%.The DC and OTC have the highest average concentrations of 29.52 μg/L and 13.71 μg/L, respectively,and thereby are the major sources of antibiotic pollution.

(ii) The major Yellow River has total concentrations of FQs and TCs of 22.0 μg/L and 14.9 μg/L, respectively, which are higher than those of previous studies.This phenomenon is related to the decrease in the water flow of the Yellow River during the dry season and the increase in the antibiotic consumption of residents in the context of the COVID-19 outbreak.

(iii) In terms of spatial distribution, the antibiotic concentrations in the upper reaches of the Huiji River in Xiangfu District are higher than those in other areas of Kaifeng City.TCs account for 72.38%-91.84% of all antibiotics in river water.Among them, DC and OTC have significantly higher concentrations than other antibiotics.These results indicate that the Xiangfu District suffers serious antibiotic pollution, which mainly originates from the livestock and poultry breeding industry of Kaifeng City.

(iv) The ecological risk assessment results show that TCs have the highest ecological risks to green algae.Among them,DC has medium-high risks, and TC, OTC, and CTC show medium risks.In addition, TMP and LOM have low risks, and other antibiotics show no risk to green algae.Compared with green algae, most antibiotics have higher ecological risks to daphnia and lower ecological risks to fish.The TCs in major Yellow River have medium ecological risks to both green algae and daphnia.It is speculated that the antibiotic pollution in the Yellow River may pose a threat to the water ecological security of Kaifeng City.

CRediT authorship contribution statement

Dun-Yu Lü, Chu Yu, Zi-Jun Zhuo, Shu-Ran Meng, and Song-Bo Liu conceived of the presented idea.Zi-Jun Zhuo and Shu-Ran Meng contributed to sample preparation.Song-Bo Liu carried out the experiments.Dun-Yu Lü Chu Yu, Zi-Jun Zhuo, and Shu-Ran Meng contributed to the interpretation of the results.Chu Yu took the lead in writing the manuscript.All authors provided critical feedback and helped shape the research, analysis, and manuscript.

Declaration of competing interest

The authors declare no conflicts of interest.

Acknowledgment

This research was jointly supported by the project of the China Geological Survey (DD20211309), the National Natural Science Foundation of China (41602273), and the High-Level Talent Funding Program of Hebei province(A202101004).