On the Importance of Sanitary Sewer Overflow on the Total Discharge of Microplastics from Sewage Water

The paper provides an investigation and understanding of the significance of various wastewater flows on microplastics retainment and emission to the environment. WWTPs and sewer overflows as an important pathway of microplastics to the environment are assessed by considering the removal of microplastics in WWTPs with different treatment processes and several sewer overflow types and their contribution to microplastic loads to recipients. On the example of the Baltic Sea basin, presented results indicate a considerable discharge of microplastic from WWTPs despite the relatively good overall removal efficiency. Results show that the discharge of microplastics from sewer overflows can be in the same magnitude as from treated wastewater although the total flow is much lower than that of treated wastewater. Sewer overflow events frequently occur and are expected to increase due to climate change and urbanization, unless infrastructure is adapted. At the same time, sewer overflows are often neglected in conventional wastewater handling.


Introduction
The problem with microplastic pollution of water environments has been recognized as one factor for the declining water quality globally and has also been pointed out as one potential factor for the observed biodiversity decline worldwide [1] [2] [3]. Several regional and global actions to address the problem of microplastic have been proposed, e.g. by the United Nations Conference on Sustainable Development in Rio+20 in 2012, which particularly mentioned the threat of plastics for the marine environment. Even if the key to reducing microplastics inputs to the environment is at the source, a ban of plastics in general such as done for other pollutants is not a realistic way forward as plastics have become essential for our modern society mainly because of their advantageous properties such as its durability. Even improved management of plastic waste including recycling and proper disposal will not prevent plastics to end up in the aquatic environment to some extent.
Microplastics, commonly defined as <5 mm in diameter, have their origin mainly from the fragmentation of larger plastic particles often termed as secondary microplastics [1]. Intentionally produced microplastics such as abrasives in personal care products or for industrial purposes are termed primary microplastics.
The potential impacts of microplastics on ecosystems are difficult to quantify. This is partly because of challenges to attribute impacts of microplastics from field observations and laboratory-based experimental are therefore often used to investigate physical effects. Even more difficult to quantify are chemical effects of microplastics as they may decay and because they may be carriers of various pollutants. Lipophilic organic contaminants, for example, can be absorbed by plastics and then taken up by animals. A clear causal relationship is then difficult to quantify. However, even though current knowledge assessing the risk from microplastics is insufficient, several studies indicate effects on living organisms caused by microplastics (e.g., [1] [4]- [10]) while other studies suggest similar effects as for natural particles [11]. Furthermore, microplastics included additives and adsorbed persistent organic pollutants on the surface of microplastics may pose an issue [12] [20]. There are also a great number of studies that report an efficient removal of microplastics in WWTPs [18] [21]- [26]. The vast volume of wastewater treated, however, implies significant amounts of microplastics entering the environment even modest amounts of microplastics in the WWTPs effluent. There are few studies on the significance of untreated wastewater released into recipients during different conditions such as technical failures, capacity limitations at the WWTPs or the sewer system. [23] and [27] indicated that sewer overflows may play a significant contribution to the total load of microplastics to the environment. However, so far, no attempts have been made to look more into this aspect.
The aim of this article is to provide a holistic investigation and understanding of the significance of various wastewater flows on microplastics retainment and emission to the environment. WWTPs and sewer overflows as an important pathway of microplastics to the environment are assessed by considering the removal of microplastics in WWTPs with different treatment processes and sev-C. Baresel

Material and Methods
Wastewater as a source for microplastics to the Baltic Sea has been investigated by means of 1) quantification of microplastic discharge from WWTPs to the Baltic Sea including the compilation of wastewater treatment practices and technologies; and 2) estimation of sanitary sewer overflow (SSO) as a source for microplastics to the Baltic Sea. Only net loads of microplastics on the Baltic Sea from WWTPs and SSOs were considered and as such no retention is considered in the presented study.

WWTPs in the Baltic Sea Basin
The quantification of microplastic discharge from WWTPs and SSO to the Baltic Sea is based on a compilation of exiting information from publicly accessible databases including [28] [29] [30] [31]. Because [28] includes more data on WWTPs than the [29] database, the study uses [28] as main data source complemented by data from [29] mainly for Russia and Belarus, which are not included in [28]. As the different data sources are not using some reporting format and plant identification number, also other data such as e.g. information about WWTP location coordinates have been used to merge necessary information. in-or outflow of wastewater is an important parameter for the calculation of microplastic discharge, water flows were calculated for WWTPs without such data. The estimation of water flow was based on an average annual flow of 125 m 3 per personal equivalent, which is derived from Swedish average flows to WWTPs [32] and data on German WWTP inflow [33]. This flow includes generated sewage water from households and industries but also other water flowing into sewer systems, i.e. surface runoff and groundwater.
For some countries, e.g. Germany and Estonia, reported flow data is judged to be incomplete or faulty as reported data varies outside common and expected ranges. Both reported flows and calculations of flows based on reported connected persons for WWTPs without flow data give annual sewage flow to WWTPs as low as 20 m 3 /(pe, yr) on average, which may be considered impossible to achieve even in completely new separated sewer systems and water saving installations in households. For a limited area, such average flows may be realistic but for a complete sewer connected to a WWTP, such areas do to the authors best knowledge not exist in the considered region of interest. However, reported flows were used in this study but a more throughout assessment of these data is recommended as actual flows may be significant higher and thus also related emissions in general and as calculated and reported here. In the case of Poland, reported flow data is erroneous as they are unrealistically low. Flow data for these WWTPs have been calculated based on the reported load and average flow assumptions.

Removal Efficiencies of WWTPs
To assess the removal efficiency of each WWTP, reported data on used treatment technologies from the data sources described has been combined with load data and related studies on treatment efficiency for microplastics using different

Microplastic Concentrations
The quantification of microplastic discharge from WWTPs to the Baltic Sea is based on the reported and calculated flows to considered WWTPs and the specific treatment available at each specific facility. Two different sets of microplastic concentrations and weights have been used in the study as shown in Table 2.
First, derived data from the project groups own microplastic analyses during several years and reported in several studies (e.g. [23] [34] [36] [37]) has been used. These measurements exclude non-synthetic fibres and consider only microplastics > 20 μm. The other data set is based on recently reported data by [25] and [38] and included measurements of microplastics down to the size of 10 µm. quantifications. Especially reported concentrations and weights by [38] for Danish WWTPs diverge significantly with extremely higher particle counts and much smaller particle weights. The authors themselves recognized this deviation and discuss possible explanations including the smaller particle size included.
Their results indicate that 90% of the particle sizes in raw wastewater are about 100 μm (and 91 μm in treated wastewater). However, a clear reason for these significant deviations is not provided and therefore two separate data sets are used in this study to investigate the impact of reported data.
Both sets of data sources have been assess and minimum, median and maximum values defined. However, these values are not based on strict statistical methods but rather the best possible estimate of these as pure Statistical methods are difficult to apply. This is mainly explained by the lack of standardization  in microplastic analyses leading to high variation of methods used in sampling, preparation, analyses and reporting of microplastics. This is a known problem in the field of microplastics research and especially when comparing various studies [1]. However, the current study has still set minimum, median and maximum values based on available data to examine the discharge of microplastics and the range of such discharges.

Sewer Overflows
Sanitary sewer overflows (SSO) are a condition where untreated wastewater is discharged from a sanitary sewer into the environment, normally due to lack of process or hydraulic capacity. In combined sewer systems, commonly in urban weather related events such as heavy rain or snowmelt.
The last category of considered SSOs comprises overflows at WWTPs due to a hydraulic capacity limitation of the WWTP but where the discharged water undergoes at least a partial treatment. As this partial treatment often targets particulate phosphorous, a good removal effect is also achieved on microplastics. As studies on the removal efficiency of microplastics in partial overflow treatment are not available, the removal efficiency for microplastics of 80% was assumed in this study.
There is no data available in the databases on SSO events and most countries have no active monitoring of such events whatsoever, which implies that even the study's own survey was not able to gather any usable data on the overflow quantities in different countries. As Sweden has been one of the countries that has tried to focus on the SSO during recent years, some data on SSO magnitudes could be used and applied as a best estimate for other countries as well. It is important to point out that even the used Swedish data [40]

Quantification of Microplastic Discharge from WWTPs
The evaluation of WWTPs in the Baltic Sea basin indicated that all considered WWTPs have at least a secondary treatment installed and thus an efficient removal of microplastics in WWTPs by 85% is given on average. This is true ex- Considering the total annual amount treated wastewater of > 8900 Mm 3 discharged to the Baltic Sea from these WWTPs, and accounting for each treatment facilities' removal efficiency for microplastics, a total microplastic discharge to the Baltic Sea from WWTPs is calculated to 13,700 × 10 9 particle/yr and 48,000 × 10 9 particle/yr as median for data set 1 respective 2 (see Table 3, I and V). Calculated total particle weight is 90 t/yr and 1.7 t/yr, respectively. Considering min and max values, min particle counts indicate significant differences in the calculated results for the two data sources (Table 3).
Reported microplastic discharges from WWTPs according to Table 3 Table 3 may still give a good indication. However, they must be handled with care as microplastics consist of many different plastics and depending on actual sources the composition in the wastewater can vary significantly between WWTPs.
A discharge of about 90 ton/yr via WWTP effluent to recipients flowing into the Baltic Sea as indicated in Table 3 (I) can be compared with reported values by [23] for the total microplastic discharge from Swedish WWTPs of 4.7 -42 ton/yr. Calculated discharge of microplastic based on the second data source comprised a much wider range with significant maximum discharge while the median discharge both in particle number and weight seems rather underestimated.
Considering the mentioned uncertainties in reported flow data for e.g. Germany and other countries, actual discharge of microplastic may even be higher. This is because reported flow data is assessed to be significantly lower than expected and for similar regions.

Quantification of Microplastic Discharge from Sanitary Sewer Overflows (SSO)
The total amount of sewage discharged to recipients from sanitary sewer over- With the considered microplastic concentrations for the different SSO-types, total microplastic discharges are calculated and presented in Table 3 (row II-IV and VI-VIII). The calculation of microplastic discharge to recipients from sanitary sewer overflows (SSO) in the Baltic Sea basin indicate that more than half of the total discharge of microplastics originating from sewage may come from SSOs. While technical SSO and SSO after treatment at WWTPs together only contribute with less than 5%, weather event SSO alone may account for >50% of total discharges. This even though they only stand for 1.5% of the total sewage flow to WWTPs in the basin.
The results presented in

Conclusions
Presented results indicate a considerable discharge of microplastic to recipients of the Baltic Sea from WWTPs, this, despite relatively good overall removal efficiency in existing facilities. Even so, there is room for improvements in some countries; average removal efficiency for microplastics of 93% must be considered The study shows that the discharge of microplastics from sewer overflows can be in the same magnitude as from treated wastewater although the total flow is much lower than that of treated wastewater. Sewer overflow events frequently occur and are expected to increase due to climate change and urbanization, unless infrastructure is adapted. At the same time, sewer overflows are often neglected in conventional wastewater handling, as traditional pollutant concentrations such as nutrients and organics often are much lower due to dilution with stormwater. For microplastics, however, this does not have to be the case.
The present findings may, therefore, be used to increase focus on mitigation measures on preventing discharge of microplastics to the environment by either reducing the risk for SSOs or by implementing a basic treatment of sewer overflows in similar ways as already done in some WWTPs. For this, the monitoring of SSO events would be required, which also would facilitate for a refinement of the estimated microplastic discharge amounts suggested by this study.