ABSTRACT
Antibiotics have been recognized as emerging pollutants due to their ecological and human health risks. This paper aims to enhance the ecological risk assessment (ERA) framework for antibiotics, to illustrate the distribution of these risks across different locations and seasons, and to identify the antibiotics that pose high ecological risk. This paper focuses on 52 antibiotics in seven major basins of China. Relying on the optimized approach of ERA and antibiotic monitoring data published from 2017 to 2021, the results of ERA are presented in multilevel. Across the study area, there are marked variations in the spatial distribution of antibiotics' ecological risks. The Huaihe River Basin, the Haihe River Basin, and the Liaohe River Basin are the top three in the ranking of present ecological risks. The research results also reveal significant differences in temporal variation, underscoring the need for increased attention during certain seasons. Ten antibiotics with high contribution rates to ecological risk are identified, which is an important reference to formulate an antibiotic control list. The multilevel results provided both risk values and their ubiquities across a broad study region, which is a powerful support for developing ecological risk management of antibiotics.
HIGHLIGHTS
Optimized the calculation method of multilevel ecological risks.
Provided ecological risk spatiotemporal distribution of antibiotics for the seven major river basins of China (2017–2021).
Provided the control priority for the seven major river basins.
Provided the top concerned antibiotics with high contribution to ecological risks of the seven major river basins.
INTRODUCTION
Antibiotics, commonly used for treating bacterial infections, are harmful to ecosystems due to their toxicities and lead to antibiotic resistance of bacteria. Consequently, antibiotics are recognized as emerging pollutants internationally (Zhang & Li 2020). After consumption, 70–90% of antibiotics are excreted in their original form or as active metabolites. Current wastewater treatment technologies often fail to effectively remove them, leading to the continuous release of substantial amounts of antibiotics into the aquatic environment. This contributes to widespread and long-term ecological risks (Liu et al. 2018; Li et al. 2020; Zhang et al. 2021). It is often reported that China as a major producer and user of antibiotics has formed a habit of antibiotic abuse. Antibiotic ecological risk has become a focal point in environmental research and management in China (Lyu et al. 2020). The ‘Action Plan for the Control of New Pollutants’ issued by the Chinese State Council Office on 4 May 2022 clearly states that antibiotics are categorized as emerging pollutants and pose ecological and health risks for China. This action plan requires the initiation of a comprehensive investigation of chemical substances, environmental monitoring, and environmental risk assessment, followed by the formulation of supplementary lists, control strategies for emerging pollutants, as well as the development of local standards. Therefore, the ecological risk assessment (ERA) covering sufficient results of antibiotic monitoring can present the ecological risk of antibiotics at a national scale, which is a necessary step for environmental management.
In principle, the ecological risk of a substance is determined by its environmental monitoring concentration (MEC) and toxicity to organisms. Normally, predicted no-effect concentration (PNEC) is used to indicate the toxicity. Logically, lower PNEC represents the lower MEC that can cause ecological hazards, which means higher toxicity could lead to more serious ecological risk. Based on this principle, monitoring of antibiotics in surface waters and ERA has been widely conducted, yielding abundant research outcomes (Li et al. 2020; Shaheen et al. 2022). However, these efforts have evident limitations. Firstly, most studies primarily utilize the risk quotient (RQ), which is a simple, deterministic quotient expressed as MEC divided by PNEC (Xu et al. 2021; Mheidli et al. 2022). When MEC and PNEC values for antibiotics are numerous in a study region, the single method of RQ cannot comprehensively characterize ecological risk, causing results to be overestimated or underestimated (Yang et al. 2023). Moreover, it is inevitable that there may be outliers within either MEC or PNEC that may lead to bias or misinterpretation of risks (Donnachie et al. 2016). Thus, a probabilistic ERA of antibiotics is emerging. Based on RQ, a semi-probabilistic method is used to determine the risk values by multiplying RQ with the frequency of concentrations exceeding PNEC (Liu et al. 2020a, 2020b). The advantage of this method is that it provides risk assessment results reflecting the risk overflow rate that represents the ubiquity of ecological risk levels across a broad study region as opposed to point estimates provided by the traditional deterministic ERA, thereby reducing the influence of outliers (Chen et al. 2023). When considering PNEC as another determinant of RQ, the related probabilistic methods should also be included, such as probabilistic ERA and species sensitivity distribution (Hu et al. 2018; Le Page et al. 2019). However, these methods require plentiful ecological toxicity data, while most antibiotics, as emerging pollutants, do not have sufficient PNEC values at present (Yang et al. 2018; Liu et al. 2020a, 2020b; Rathi et al. 2021). The semi-probabilistic method focusing on MEC therefore is suitable for conducting the present ERA of antibiotics. Secondly, there is a wide variety of antibiotics, over a hundred commonly used in clinical practice and dozens with environmental monitoring data. On the other hand, existing researches have focused solely on monitoring and risk assessment of a limited number of antibiotics (Xi et al. 2019; Yin 2021). Thirdly, the majority of existing articles concentrate on localized antibiotic monitoring without extensive comparison and risk screening, making it difficult to provide comprehensive data support for the formulation of priority lists of antibiotics. Lastly, some studies fail to consider the seasonality of the ecological risk of antibiotics, which hinders them from reflecting the temporal distribution patterns nationally or in the study region. As a result, there is a lack of data supporting the ecological risk management of antibiotics, especially for policy making.
Considering the above-mentioned issues, the first objective of the paper was to optimize the assessment method of ecological risk by combining RQ with the overflow probability (F) to form the priority index (PI). The second objective was to analyze monitoring data of antibiotics (from 2017 to 2021) in seven major river basins of China where the human populations are concentrated and then to calculate multilevel ecological risk values. The third objective was to reveal whether antibiotic ecological risks vary with season and region, and to list high-risk antibiotics across seven major river basins of China, intending to provide data support for the prioritized monitoring of antibiotics and their ecological risk management.
METHODOLOGY AND DATA SOURCES
Multilevel ecological risk calculation
The RQ method, while indicating risk values for specific water bodies, is inadequate for capturing the ubiquity of ecological risk levels across a broad study region. Thus, the second level involves F to express the proportion of MEC in the river basin that exceeds PNEC, aiming to comprehensively reflect the possibility of ecological risk throughout the river basin (NORMAN Association 2013). Corresponding to the three classifications of RQ in the first level, F is divided into low-risk overflow rate (FL), medium-risk overflow rate (FM), and high-risk overflow rate (FH).
Based on the results of the above two levels' calculations, the third level employs PI to express the risk value as the product of RQ and F. According to three classifications of both RQ and F, PI is expressed as PIL, PIM, and PIH. The summation of the three classifications of PI presents the overall ecological risk, which is used to compare ecological risks within the study region of large-scale.
Study area and data sources
This study focuses on the surface waters of seven major river basins in China, including Heilongjiang, Liaohe, Haihe, Yellow River, Huaihe, Yangtze, and Pearl River Basins (Zhang et al. 2018). Seven major river basins, covering 47.3% of China's land area, contribute 82.6% of China's gross domestic product (GDP) and contain 82.7% of China's population (Song 2023). The monitoring data are sourced from environmental monitoring publications, master's and doctoral theses published on the China National Knowledge Infrastructure (https://www.cnki.net) from 2017 to 2021, as well as papers published on the Web of Science. The monitoring data are drawn from 75 publications, covering 70 surface water bodies and 107 types of antibiotics. The statistics on monitoring data are presented in Table 1. The monitoring data are categorized into dry and wet periods based on the sampling date. Kinds of literature with data sources and monitoring sites of each basin are listed in Supplementary Table S1. Boundary data for seven major river basins are obtained from the National Earth System Science Data Center (http://www.geodata.cn/).
Statistics on antibiotic monitoring data of surface water in seven major river basins from 2017 to 2021
Basins . | Surface waters involved . | Antibiotics involved . | Sampling sites (publication involved) . | Population density, people/square kilometer (Song 2023) . |
---|---|---|---|---|
Heilongjiang | 2 | 21 | 42 (4) | 82.8 |
Liaohe | 3 | 22 | 74 (3) | 131 |
Haihe | 6 | 39 | 207 (7) | 470 |
Yellow River | 6 | 30 | 101 (7) | 152.1 |
Huaihe | 2 | 37 | 72 (2) | 617.8 |
Yangtze River | 34 | 91 | 595 (36) | 256.3 |
Pearl River | 17 | 71 | 186 (16) | 338.5 |
SUM | 70 | 107 | 1,277 (75) | – |
Basins . | Surface waters involved . | Antibiotics involved . | Sampling sites (publication involved) . | Population density, people/square kilometer (Song 2023) . |
---|---|---|---|---|
Heilongjiang | 2 | 21 | 42 (4) | 82.8 |
Liaohe | 3 | 22 | 74 (3) | 131 |
Haihe | 6 | 39 | 207 (7) | 470 |
Yellow River | 6 | 30 | 101 (7) | 152.1 |
Huaihe | 2 | 37 | 72 (2) | 617.8 |
Yangtze River | 34 | 91 | 595 (36) | 256.3 |
Pearl River | 17 | 71 | 186 (16) | 338.5 |
SUM | 70 | 107 | 1,277 (75) | – |
Toxicity data
Toxicity data of antibiotics for three trophic levels (fish, daphnia, and algae) were obtained from the US EPA ECOTOX toxicity database (https://cfpub.epa.gov/ecotox/). When selecting parameters of toxicity data, five indicators were set as toxicity endpoints (LC50, EC50, IC50, LOEC, and NOEC), and freshwater was chosen as the exposure medium. After querying and calculations, 52 antibiotics with toxicological data were listed in Supplementary Table S2.
Ranking of multilevel ERA for seven major river basins of China
Basins . | RQSUM . | F . | PI . | RQSUM (Yin 2021) . | Population density . | ||
---|---|---|---|---|---|---|---|
FL . | FM . | FH . | |||||
Huaihe River | 1 | 7 | 3 | 3 | 1 | 7 | 1 |
Liaohe River | 2 | 5 | 1 | 1 | 3 | 2 | 6 |
Haihe River | 3 | 2 | 4 | 2 | 2 | 1 | 2 |
Yellow River | 4 | 3 | 2 | 6 | 6 | 5 | 5 |
Yangtze River | 5 | 4 | 5 | 5 | 4 | 4 | 4 |
Heilongjiang River | 6 | 1 | 6 | 7 | 5 | 6 | 7 |
Pearl River | 7 | 6 | 7 | 4 | 7 | 3 | 3 |
Basins . | RQSUM . | F . | PI . | RQSUM (Yin 2021) . | Population density . | ||
---|---|---|---|---|---|---|---|
FL . | FM . | FH . | |||||
Huaihe River | 1 | 7 | 3 | 3 | 1 | 7 | 1 |
Liaohe River | 2 | 5 | 1 | 1 | 3 | 2 | 6 |
Haihe River | 3 | 2 | 4 | 2 | 2 | 1 | 2 |
Yellow River | 4 | 3 | 2 | 6 | 6 | 5 | 5 |
Yangtze River | 5 | 4 | 5 | 5 | 4 | 4 | 4 |
Heilongjiang River | 6 | 1 | 6 | 7 | 5 | 6 | 7 |
Pearl River | 7 | 6 | 7 | 4 | 7 | 3 | 3 |
Note: The ranking decreases in the order of 1–7.
RESULTS AND DISCUSSION
Spatial distribution of ecological risk of antibiotic
Using monitoring data from a total of 1,277 sampling sites across seven major river basins from 2017 to 2021, the average RQ, F, and PI values were calculated. The ranking of multilevel ERA results is shown in Table 2. In terms of ranking order, it can be observed that there are relatively minor differences between RQ and PI. The Huaihe River Basin, the Liaohe River Basin, and the Haihe River Basin are the top three concerning both RQ and PI rankings, while the Pearl River Basin is the smallest and ranked last. Compared to the previous research results shown in the second column on the right of Table 2 (Yin 2021), the ranking positions for most basins are similar, except for the Huaihe River Basin and the Pearl River Basin. The exception could be due to the different sources of exposure and toxicity data. The exposure data used by Yin 2021 were from 2010 to 2020 and related to only 13 antibiotics, and toxicity data were collected from the literature rather than the toxicity database used by this article.
Results of multilevel ecological risk assessment of seven major basins in China, and monitoring sites distribution.
Results of multilevel ecological risk assessment of seven major basins in China, and monitoring sites distribution.
The spatial distribution of ecological risks depends on many factors that could be related to social and environmental conditions, such as population density, precipitation, temperature, water flow, photolysis, biodegradation, etc. (Chen et al. 2018; Qiao et al. 2018; Huang et al. 2019). As half of antibiotics are used in animal farming, the ratio of animal husbandry in the economic structure of a study area is also a significant factor for the distribution (Zhou et al. 2020; Patyra et al. 2023). The population density of each basin shown in Table 2 presents a certain relationship with the spatial distribution, except for the Liaohe River Basin and the Pearl River Basin. Other factors in social and environmental conditions could play stronger roles for the exception.
There are 1,277 sampling sites across seven major river basins from 2017 to 2021. Figure 2 also shows the distribution of monitoring sites and locations of megacities. All 22 megacities with a population of more than 5 million people are distributed in seven major river basins. The distribution reveals that the monitoring efforts are currently concentrated around megacities, thus results of this study largely represent the ecological risks of the megacities. Moreover, most of the monitoring sites are located in downstream areas of the main rivers, particularly in the Yellow River Basin and the Pearl River Basin. Therefore, due to the uneven distribution of monitoring sites, there could be significant variations of ecological risk in other regions of the basins, suggesting more surveillance of antibiotics over a broader geographical region. This finding and suggestion have been confirmed by other articles (Li et al. 2018; Liu et al. 2018). Qiao et al. (2018) highly recommended the establishment of a systematic surveillance network that includes regular, continuous measurement of antibiotics (Qiao et al. 2018). Therefore, this article strongly urges the necessity of enhancing both monitoring and related ERA to verify and manage the significant risks identified herein.
Temporal distribution of ecological risk of antibiotic
Temporal distribution of multilevel ecological risk assessment in seven major river basins.
Temporal distribution of multilevel ecological risk assessment in seven major river basins.
Identification of antibiotics of particular concern
Thermodynamic chart of the contribution rate of antibiotics to PI in seven major river basins (Unit: %). Note: the antibiotics marked by ‘*’ have been listed in the watch list of substances by the European Union.
Thermodynamic chart of the contribution rate of antibiotics to PI in seven major river basins (Unit: %). Note: the antibiotics marked by ‘*’ have been listed in the watch list of substances by the European Union.
The black-bordered boxes in Figure 4 highlight 10 antibiotics present in both wet and dry periods, underscoring the need for these antibiotics to receive special attention at a national level. Regarding the increase of ranking order from bottom to top, the higher ranking of antibiotics indicates the higher priority of concern at a national scale. There have been nine antibiotics listed in the watch list of substances of ‘Union-wide monitoring in the field of water policy pursuant to Directive 2008’ since 2015 (European Commission 2022), and seven of them are included in the figure, which are marked by the star in Figure 4. Most of these 10 antibiotics were pointed out by many review papers due to the low removal efficiency in the practical treatment process (Langbehn et al. 2021; Ilurdoz et al. 2022; Morin-Crini et al. 2022). Wastewater treatment is a key point of the fate of antibiotics in the aquatic environment, and the low removal efficiency could explain the high contribution of these antibiotics to river basins (Michael et al. 2013; Wang et al. 2019).
As Table 1 shows, only a small number of publications are available about monitoring data of the Huaihe River Basin, the Liaohe River Basin, and the Heilongjiang River Basin, thus further monitoring and validation are needed for the abnormally high contribution rates of tetracycline and roxithromycin. Regarding antibiotics with high contributions, it is necessary to seek alternative drugs with similar therapeutic effects but lower ecological risks.
CONCLUSIONS AND RECOMMENDATIONS
The optimized approach with the multilevel risk assessment presents not only the risk values but also the ubiquity of ecological risk levels across a broad study region. Based on the results of ERA by the optimized approach, it is recommended to preferentially strengthen the monitoring of antibiotics and ERA in the Huaihe River Basin, the Haihe River Basin, and the Liaohe River Basin, and to enhance control measures for antibiotic use. Considering seven major river basins cover most parts of China's land area, GDP, and population, the presented results and conclusions mostly stand for the ecological risk of antibiotics at the national scale. Monitoring density of antibiotics is highly uneven across the basins, and future efforts of monitoring of antibiotics should be extended to the upper and middle reaches of the main rivers, as well as other tributary regions.
Since ecological risk values of antibiotics exhibit significant differences between the dry and wet periods, it is advisable to focus on controlling antibiotic usage and improving wastewater treatment for the period with high ecological risk.
Antibiotics of particular concern are identified, which is helpful in formulating a national control list of antibiotics. On the other hand, there are certain differences between the river basins regarding certain types of antibiotics and their contributions to ecological risk, suggesting the formulation of antibiotic risk management on a regional level based on its antibiotic consumption and environmental conditions.
ACKNOWLEDGEMENTS
This work was funded by Natural Science Foundation of Inner Mongolia Autonomous Region (2020MS05018, 2021MS05017, 2020BS04008).
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
CONFLICT OF INTEREST
The authors declare there is no conflict.