Next Article in Journal
Effects of Humus and Solidification Agents on the Solidification/Stabilization Process of Organic-Rich River Sludge: Characteristics of the Stabilized Sludge
Previous Article in Journal
A Helping Hand: Fungi, as Well as Bacteria, Support Ecophysiological Descriptors to Depict the Posidonia oceanica Conservation Status
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Water
Volume 17
Issue 8
10.3390/w17081152
water-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Surface Water Monitoring with Sedimentation Boxes: Assessing the Sampling Performance and Its Effect on Microplastic Concentration

by
Cristina Julieta Saravia
1,*,
Mathias Ricking
1,
Peter Grathwohl
2,
Claus Gerhard Bannick
1 and
Nathan Obermaier
1
1
German Environment Agency, Corrensplatz 1, 14195 Berlin, Germany
2
Department of Geosciences, University of Tübingen, Schnarrenbergstr. 94–96, 72076 Tübingen, Germany
*
Author to whom correspondence should be addressed.
Water 2025, 17(8), 1152; https://doi.org/10.3390/w17081152
Submission received: 3 March 2025 / Revised: 7 April 2025 / Accepted: 9 April 2025 / Published: 12 April 2025
(This article belongs to the Section Water Quality and Contamination)
Browse Figures
Review Reports Versions Notes

Abstract

:
Currently, there are still no harmonized and thus reproducible methods for microplastics (MP) sampling. Infrequent spot sampling with, e.g., nets, pumps, or containers, does not reflect the large spatial and temporal variety of MP abundance, and there is little experience with time-integrated, passive sampling methods. However, passive samplers have been applied thoroughly to recover suspended particulate matter (SPM) from water bodies. The physical and chemical characteristics of MP are in range with those of other materials belonging to SPM, and we state that MP are an integral component of SPM. In general, passive samplers like the sedimentation box decrease the flow velocity, enhancing the sedimentation of SPM within the device. The retention rates of particles in sedimentation boxes depend on various factors such as the flow velocity, the SPM size and density, but precise information remains scarce. Therefore, we performed laboratory tests to assess the retention rates of the polymers polystyrene and polyethylene and analyzed the dependency of sedimentation on the flow velocity and particle sizes. The quantification of MP in samples collected by sedimentation boxes underestimates the concentration of smaller-sized particles due to their lower retention rate, and MP concentrations should be reported accordingly. Subsequently, we carried out a series of field experiments with sedimentation boxes and showed that MP can be retained from different water bodies with diverse characteristics. Due to their robust sampling mechanism, sedimentation boxes are promising devices for time-integrated, long-term sampling of MP.

Graphical Abstract

1. Introduction

The ubiquitous occurrence of microplastics (MP) in surface waters around the globe, even in remote areas, is widely studied [1,2,3]. Concern about their potential adverse effects on human and environmental health reinforces the need for reliable concentration data [4,5,6]. Different spot sampling techniques (e.g., nets, pumps, containers) have been used to identify MP. While many snapshot investigations prove that MP occur in water bodies, there is little long-term data and basically no information on spatial and temporal variation [7]. Overall, sampling and quantifying different polymers in water bodies is challenging due to low concentrations, the microscopic size, and the need for a reliable distinction from “natural” suspended particulate matter (SPM).
Several MP-focused meta-studies highlight the large variety of sampling, sample preparation, and detection methods (e.g., Lu et al. [8] or Hidalgo-Ruz et al. [9]). These different methods significantly influence the obtained results, leading to studies not being comparable with each other (see, e.g., Primpke et al. [10] for the vast result differences between pyrolysis gas chromatography/mass spectrometry and hyperspectral Fourier-transform infrared spectroscopy (FTIR) imaging spectroscopy).
The commonly used sampling methods for MP from water bodies cover a limited water volume and usually a rather short time span (minutes to hours). Frequent spot sampling would be required to determine a time-averaged MP concentration, which is needed to calculate MP fluxes in rivers. Alternatively, time-integrated sampling methods such as passive samplers cover larger time spans (from days to weeks). Within this study, the term “passive sampler” refers to a sampling device through which water flows and SPM is separated from the water phase without using manual or automated labor in the device itself.
Passive samplers for particulate matter are devices that generally work with the principle of sedimentation due to a decreased flow velocity within the sampler induced by an expansion of the inner diameter. They are well-known for the sampling of SPM [11] and may also be suitable to sample MP. Nevertheless, current literature reveals that passive samplers are insufficiently validated to provide quantitative results on both SPM and MP in surface waters.
The most commonly used passive sampler is the TIMS (time-integrated mass-flux sediment) sampler, which is directly applied in flowing water and has been developed by Phillips et al. [12]. The sampler has been used to sample SPM from rivers and creeks. However, only a few studies have assessed the retention rates under controlled laboratory conditions [12,13,14,15]. A comparison of the data reveals that retention rates decrease with increasing flow velocity and decreasing particle size. Hence, they indicate a systematic response of the devices’ retention performance to altering parameters. Other parameters that are suspected to have an impact on retention rates have not been considered yet, such as the density or shape of the used material or the influence of organic material or flocs.
Furthermore, there have been attempts to validate the TIMS sampler in the field [12,14,15,16,17,18,19,20,21,22]. For instance, Perks et al. [19] compared TIMS samples to SPM concentrations estimated with a previously calibrated turbidity sensor. Although there is a statistically significant correlation between both methods, they determine a level of underestimation of 66–96% for the concentration of suspended solids by the TIMS sampler. However, given the typically high variability of the flow velocity in rivers, retention rates can be expected to change during the sampling period, and the total sampled volume cannot be determined reliably. Without having information on the flow, it is impossible to determine the concentration or flux of SPM.
Besides the TIMS sampler, there are other passive samplers which are directly placed in flowing water, such as the BiSam sampler [23,24,25] or are typically operated in monitoring stations, such as the sedimentation tank [23,24].
In general, passive samplers like the sedimentation box can also be operated with a pump to sample SPM from water bodies (compare ISO 5667-23, 2011 [26]). This allows for a constant flow velocity inside the sampling device and the monitoring of the sampled water volume. Sedimentation boxes have been designed by the German Environmental Specimen Bank and are applied in the routine monitoring of SPM from rivers [11].
Harhash et al. [27] investigate the retention performance of sedimentation boxes operated with a pump. They performed laboratory experiments with different mineral, organic, and MP particles of varying sizes. Masson et al. [28] compared sedimentation box samples to continuous flow centrifuge samples. Both studies found significantly coarser particles in the sedimentation box, underscoring the higher retention rate of larger particles compared to smaller particles.
The suitability of passive samplers to collect SPM is documented in many studies and we hypothesize that these experiences are transferrable to MP. SPM does not include or exclude specific materials; it rather describes the physical state of particulate matter. Hence, MP and their different properties add to the range of physical properties of SPM. SPM includes a wide range of materials and their respective properties (e.g., see Koelmans et al. [29] for different categories of natural particles), and we state that the differences between MP and other materials in SPM are in line with differences between materials described as SPM.
The density of organic matter such as wood, leaves, or algal debris ranges from 0.9 to 1.3 g/cm3 [30]. This range is similar to the density range of polymers in general and to the densities of our focus polymers: polypropylene (PP, 0.91–0.94 g/cm3, [31]), polyethylene (PE, 0.91–0.97 g/cm3, [32]), PS (1.05–1.06 g/cm3, [33]), and tire and road wear particles (TRWP, around 1.8 g/cm3, [34]). In contrast to particulate organic matter, inorganic particles are mostly minerals and have densities between 2.5 g/cm3 and 3.0 g/cm3. However, biofouling may alter the density of particles. Several studies have demonstrated that biofilm growth leads to both the sinking of formerly buoyant MP particles and the resuspension of formerly sinking MP particles [35,36,37,38]. Furthermore, the adhesion of minerals on MP particles or the formation of aggregates can increase their density and facilitate sedimentation. It has been found that this process may be enhanced through biofouling [39] or the presence of surfactants [40].
In the UK, Walling et al. [41] found over 95% of the SPM in most of the investigated rivers to have a particle size below 63 µm. This is in good accordance with other studies, e.g., Rügner et al. [42], finding median particle sizes between 5 and 34 µm for SPM in rivers in South-West Germany, or Le et al. [43], who determined median SPM sizes between 10 and 20 µm in the lower Mekong, mostly as flocs. By definition, MP particles are in range with these particle sizes, and investigations on MP particle size in water bodies show similar size ranges. Crew et al. [44] found that 95% of all sampled microplastic particles had a size below 400 μm; in the MiWa project, even 96% were below 20 μm [45].
The flow characteristics (e.g., velocity, turbulence) of the surrounding water define both the size and density characteristics of SPM and thus, the transport mechanisms of MP particles [46]. MP is present in various shapes. Fibers, foils, and particles most likely behave differently in similar flow conditions. This adds to the difficulty of assessing polymer-specific transport patterns. All that considered, it seems easier to directly measure MP in SPM than to predict if certain MP will end up in SPM.
The aim of our study is to investigate the potential of sedimentation boxes as time-integrative sampling devices for MP in surface waters. To approximate actual MP concentrations in surface waters and to reduce the uncertainty in quantitative MP results, it is pivotal to understand the retention of MP in the sedimentation boxes. We operate them with a pump to reduce the limitations of currently used passive samplers.
We conducted laboratory experiments under controlled conditions to gain additional insights into MP retention rates. Then we carried out several sampling campaigns at rivers. The first step was to apply sedimentation boxes directly in the river, which provided qualitative MP results. We then operated the boxes with a pump in regular river monitoring stations. Taken together with the laboratory experiments, we aim to quantify MP concentrations in the sampled rivers.

2. Materials and Methods

Between 2019 and 2021, we applied sedimentation boxes in various experiments to sample MP from different rivers in Central and Eastern Europe. In this study, we present the results from three different monitoring campaigns as examples of the application of sedimentation boxes. The first sampling campaign was part of the Joint Danube Survey 4 (JDS4) in 2019. The survey was carried out to assess water quality parameters along the River Danube and its main tributaries. Sedimentation boxes were used to assess MP abundance and its regional variation. Further information can be found in Kittner et al. [47]. While they used the MP concentration data to compare the pollution of the rivers, we evaluated the data to derive information on the sampling performance of the sedimentation boxes, taking the properties of the sampled rivers into account.
In 2020, we carried out further monitoring campaigns at the River Elbe and Teltowkanal. The aim of these campaigns was to assess the sampling performance of the boxes and develop a robust sampling strategy in terms of handling and workload. In addition, we determined the retention rates of MP particles under controlled conditions in laboratory experiments.

2.1. Sedimentation and Operating Principle of Passive Samplers

To elucidate the sensitivity of different parameters affecting the retention of SPM in sedimentation boxes, we give a brief overview of the physical basics of sedimentation and the operating principle of passive samplers.
The behavior of a suspended particle in a stationary fluid can be described by Stokes’ law [48]. Due to gravity, different forces act on the particle and its terminal sinking or rising velocity is determined by the relation between gravitational forces (FG) and the sum of buoyant (FB) and frictional (FF) forces (see Figure 1a). Among other parameters, these forces are not only dependent on the size (volume Vp and diameter dp) but also on the density ρ of the particle (only particles with a density larger than water are supposed to sink). While frictional forces are proportional to the particle diameter, gravitational forces are proportional to the particle diameter in third power. Hence, the terminal sinking velocity of a particle is proportional to the particle diameter in second power (see Figure 1b), and larger particles settle down or rise faster than smaller particles. Moreover, in turbulent fluids, the smaller the particle, the stronger its movement is influenced by the surrounding flow.
In passive samplers with a horizontal design, particles are transported horizontally with the same velocity as the surrounding water. Simultaneously, they move vertically with their terminal sinking (or rising) velocity vs. Hence, depending on their specific sinking velocity, particles are retained if they sediment before reaching the outlet of the sampler, i.e., if their retention time is long enough (see Figure 1c). In vertical passive samplers, particles are retained if their terminal sinking velocity vs is larger than the vertical flow velocity vd (see Figure 1b).
In a sedimentation box, the flow direction of the water is continuously changed by vertical baffles. Figure 2 shows a simulation of the flow inside the box. In the first chamber, the jet hits the first baffle, and the flow becomes fully turbulent. Due to the expansion of the box in comparison to the inlet, the flow velocity decreases and the flow becomes more laminar in the subsequent chambers. In the second and third chambers, the flow alters between a horizontal and vertical direction. Areas with a particularly low flow velocity are close to the walls and in the corners of the box. If a particle reaches these areas, it can be trapped in the sampler without being fully sedimented. This is illustrated in Figure 3, where the particles accumulate in the corners of the device.
In consequence, the retention of particles in the sedimentation box depends on the size and density of the particles as well as the flow velocity inside the box. The sinking velocity, and thus, the expected retention of particles increases with increasing diameter and density. Furthermore, the retention time of a particle increases with decreasing flow velocity inside the box. If it is lower than in the sampled river, formerly suspended particulate matter can be retained by sedimentation (provided their density is higher than the density of water).
We used Stokes’ law to estimate a suitable flow within the sedimentation box. With an expansion of the inner diameter by the factor 34, our chosen flow rate of 5.2 L/min produces a flow velocity of 0.004 m/s inside the box and should lead to a complete retention of PS particles > 454 µm (considering a density of 1.04 g/cm3). A flow rate of 2.6 L/min produces a flow velocity of 0.002 m/s inside the box and should lead to the complete retention of PS particles > 321 µm. Smaller particles should be retained partially with the respective flow rates.
The aforementioned calculations strongly simplify the fate of SPM in the sedimentation box and neglect factors such as different particle shapes, adhesion to walls, position within the flow, or particle concentration. Furthermore, when using passive samplers directly in the river, the flow velocity and flow pattern inside the device might not be constant due to a varying flow in the river.

2.2. Laboratory Experiment Methods

We conducted laboratory experiments to determine the retention rates of sedimentation boxes for PS and PE particles. For the experiments with PS, we produced artificial suspensions of reverse osmosis water and PS particles with a median particle size of 250 µm and a density of 1.04 g/cm3. To produce a stock suspension, we heated 2 L of reverse osmosis water, added 100 g Tween80®, left it to boil for approx. 5 minutes, and stirred it for 24 h at ambient temperature. Tween80© is a surfactant which is based on Polysorbate 80 and facilitates the suspension of the hydrophobic PS particles. Subsequently, we added 90 g PS particles. For each experiment, we added the 2 L stock suspension to 2 m3 reverse osmosis water, resulting in a concentration of 45 mg/L PS. NaCl was added until it reached a concentration of 0.1 g/L to increase electrical conductivity. During the experiments, the suspension was automatically stirred (agitator: Kiesel, Heilbronn, Germany, 1400–1680 rpm) in two stainless-steel barrels, which were connected to each other at the bottom.
The experiments with PE were carried out similarly; however, we produced a stock suspension with surface water. The water was taken at the surface of a local lake in Berlin-Marienfelde, Germany. We added 45 g of UHMW-PE (Ultra High Molecular Weight Polyethylene) with a median particle size of 147 µm and a density of 0.94 g/cm3 to 2 L surface water. At first, the PE particles remained on the water surface but were suspended after one day of constant stirring at ambient temperature. We added the stock suspension to 1 m3 surface water, resulting in a concentration of 45 mg/L PE. The suspension was automatically stirred in a stainless-steel barrel.
We used three parallel sedimentation boxes for the PS experiments. Due to the reproducible results of the parallel boxes, we conducted the PE experiment with a single sedimentation box to reduce the workload. For the PS experiments, a submersible stainless-steel pump (Optima, Ebara pumps) was connected to a stainless-steel distribution piece, which divides the flow into three approximately equal parts. For the PE experiment, the submersible pump was connected directly. One-inch needle valves made of brass were used to regulate the flow at the inlet of the sedimentation boxes. Behind the sedimentation boxes, electromagnetic flowmeters (Picomag, Endress & Hauser, Reinach, Switzerland) measured flow and total volume. Sieves with a mesh size of 50 µm (Bückmann, Mönchengladbach, Germany) were used at the outlet of the system. These were placed approx. 50 cm above ground to guarantee completely filled sedimentation boxes.
A total of 600 L of the MP suspension was pumped through each of the sedimentation boxes at a flow of 5.2 L/min for run 1 and 2.6 L/min for run 2 of the PS experiments and with a flow rate of 5.2 L/min for the PE experiment. Before and after each run, the sedimentation boxes were flushed with 100 L MP-free reverse osmosis water for PS and surface water for PE. After the completion of each experiment, the retained material on the outlet sieves was transferred into glass bottles. The content of each sedimentation box was directed over 5 µm sieves using a vacuum pump. The remaining suspension in the barrels was directed over 50 µm sieves. All samples were frozen at −20 °C, freeze-dried at −60 °C and 1 mbar (Delta 1–24 LSCplus, Christ, Osterode am Harz, Germany), weighed, and the PSD was determined (Partica LA 960, HORIBA, Kyoto, Japan).

2.3. Field Experiment Methods

For the monitoring campaigns, the boxes were either placed directly in the stream, using the natural flow of the river (method “in situ”) or operated with a pump (method “pump”). We used slightly different models at the different sampling sites. All samples obtained within the Joint Danube Survey 4 (JDS4) in 2019 were taken using the in situ method. Due to practical reasons (e.g., sampling sites without access to electricity), the JDS4 team placed the boxes directly into the stream. The used sedimentation boxes had a volume of 67 L, six 1″ inlets, and four 1″ outlets. They were placed approx. 0.5 m below the water surface and operated for 14 days. The flow velocity inside the box decreases with a factor of 37. Further information can be found in Kittner et al. [47].
To evaluate whether the JDS4 sampling method is transferable to the Teltowkanal, we investigated the flow velocity in the river. Therefore, we used an acoustic Doppler velocimeter and measured the flow velocity at five spots in front of an installed sedimentation box for a total of 65 min. The flow velocity showed significant differences with a minimum minute average of −0.22 m/s, a maximum minute average of 0.09 m/s, and a mean minute average flow velocity of −0.01 m/s during the sampling period.
Based on these results, the corresponding uncertainty about the flow through the device and the associated loss of information when using the sedimentation boxes directly in the stream, we used a pump-based setup in river monitoring stations for the sampling campaigns at the River Elbe and Teltowkanal in Germany. The sampling locations were the monitoring station Cumlosen, lying on the right side of the River Elbe at km 470, and Kleinmachnow, lying on the left side of Teltowkanal at km 6 (kilometers at Teltowkanal are counted opposite to the flow direction, from west to east). Three parallel boxes were operated at each sampling location to assess the reproducibility.
For the pump method, we used a setup, which is similar to the one used for the laboratory experiments. We used boxes with a single 1″ inlet, to facilitate the connection with the hose. At Teltowkanal, we used boxes with a volume of approx. 60 L and an expansion factor of 78; at the River Elbe, we used smaller boxes with a volume of approx. 20 L and an expansion factor of 34. We connected the boxes to the existing circular pipe system of the monitoring station. Using a ball valve, we set a constant flow velocity of 5.2 L/min (or 0.002 m/s in the large and 0.004 m/s in the small box) for the sampling duration of 48 h and controlled it with an electromagnetic flowmeter (Picomag, Endress & Hauser, Reinach, Switzerland). We chose this flow velocity and sampling duration based on our initial calculation and the subsequent results of preliminary experiments, where we could show that enough sedimentary material for all further analyses could be retained within the sedimentation box. Furthermore, lower flow velocities (e.g., 2.6 L/min) sometimes caused clogging in the ball valve, leading to an irregular flow.
After the completion of the sampling period, we separated the content of the boxes by fractionated filtration, using a sequence of stainless-steel sieves (mesh size 500 and 100 µm for the in situ method; 500, 100, 50, and 10 µm for the pump method). For the filtration over 10 µm, the application of a vacuum was needed, and due to the high solids content and filter cake formation, we used aliquots representing 10% of the total sample volume for this fraction. We analyzed the filtrate of this step to obtain the <10 µm fraction. Samples taken with the pump method were frozen at −20 °C and subsequently freeze-dried at −60 °C and 1 mbar (Delta 1–24 LSCplus freeze-dryer, Christ, Osterode am Harz, Germany). Samples from the in situ method were air-dried in an oven at 40 °C. We weighed all subsamples and homogenized them using a mortar and pestle after drying.
The total organic carbon (TOC) of the samples was determined individually for most fractions, following DIN EN 15936 and using the “varioTOC” (Elementar, Langenselbold, Germany). Between 2.6 and 43.8 mg of the subsamples were weighed for triplicate determination onto silver foil and treated with 10% hydrochloric acid to remove inorganic carbon. After air-drying at 40 °C, samples were combusted at 950 °C under oxygen, and the combustion products were transferred to a catalyst (copper oxide) to achieve complete oxidation to CO2. After a drying step and removal of other volatile compounds by adsorption, CO2 is quantified by an infrared detector (NDIR, nondispersive infrared sensor, at 38 °C). The NDIR was calibrated using a soil reference material (0.651% TOC, calibration range 0.0058–0.916 mg TOC) and potassium hydrogen phthalate (47.05% TOC, calibration range 0.48–3.79 mg TOC). The standard deviation of the method is typically less than 10%.
To quantify MP content, we carried out the polymer analysis with a thermogravimetric detection method, namely the TED-GC/MS (thermal extraction desorption-gas chromatography/mass spectrometry). Although the method is suitable for analyzing a larger number of polymer types, we could only unambiguously detect PE, PP, PS, and styrene-butadiene rubber (SBR), a marker for TRWP. The sample analysis with TED-GC/MS provides information on polymer types and masses, yet information on the number, size, or shape is lost. However, we can roughly approximate the size according to the mass results of the fractionated filtration. Due to the analysis process, the minimum mass reflecting the limit of detection is different for each polymer. Detailed information on polymer analysis with TED-GC/MS is provided by Kittner et al. [47].
To characterize the river during the sampling period, information on the annual discharge at or close to the sampling site was derived from different sources (e.g., Cumlosen: River Basin Community Elbe [49] or Lanžhot: Czech Hydrometeorological Institute [50]). The actual discharge during the sampling period might have been very different from the annual average due to seasonal or event-induced variation; however, it was not assessed during the sampling.

3. Results

3.1. Laboratory Experiments

To determine the PS and PE retention rates of the sedimentation boxes, we calculated a particle-size specific mass balance for each run and each sedimentation box. Therefore, the retained content of each sedimentation box was divided by the input material minus the remaining material in the barrels for each size class.
For the PS experiments, the retained masses were similar for the three sedimentation boxes, with a mean coefficient of variation of 3.7% (see Figure S1 in the Supplementary Material). The same applies to the particle size distribution, which was similar in the three boxes and in the sieves at the outlet (see Figures S2 and S3 in the Supplementary Material). In comparison to the material retained with the outlet sieves, the particle size distribution of the sedimentation box samples is shifted toward larger particles for all runs. This coarsening effect also reflects in the retention rates, which are displayed per particle size class in Figure 4. For the PS experiments, the mean value of the three parallel boxes is shown. For all runs, the retention rates increase steadily with an increasing particle size until approaching 100% for PS particles with a size of 517 µm and a flow rate of 2.6 L/min. Throughout all particle sizes, the retention rates for PS are higher for the lower flow rate of 2.6 L/min. At the same flow rate of 5.2 L/min, the retention rates for PE and PS are very similar to each other.

3.2. Field Experiments

Sedimentation boxes are relatively easy to operate both in situ and in monitoring stations. When operated in situ, they need to be fixed securely; the JDS4 team mostly installed them beneath bridges with steel ropes. The installation in a monitoring station can be done in under one hour by two people. Once the boxes are installed, they are usually maintenance-free and do not require any additional attention, such as a backwash etc.
The laboratory experiments with sedimentation boxes operated in parallel indicate reproducible sampling, and we repeated the experiment with multiple sedimentation boxes in the field to verify our findings. The mean coefficient of variation of total retained SPM mass between the three parallel boxes was 4.2% at River Elbe (Cumlosen) and 5.5% at Teltowkanal (Kleinmachnow). Furthermore, the masses of the different size fractions were similar between the parallel boxes and between two subsequent runs, although showing greater variability than the total masses (see Figures S4 and S5 in the Supplementary Material). Here, the overall mean coefficients of variation were 12.1% for Elbe and 18.6% for Teltowkanal.
Sedimentation boxes were easy to handle by two people and provided sufficient dry mass for all subsequent analyses (average total mass 33.8 g in Cumlosen and 34.6 g in Kleinmachnow).
Table 1 shows large differences in flow characteristics in the sampled rivers, ranging from small, lowland channels such as the Teltowkanal with a low average discharge to large streams such as the River Danube in Bulgaria. Although we did not directly analyze SPM in the rivers, due to the differences in average discharge (Qavg) from 8 m3/s for the Teltowkanal [51] to 6001 m3/s for the River Danube at Ruse [52]), it can be expected that the concentration and characteristics of SPM in the rivers were highly diverse. This is underlined by the results of the fractionated filtration and TOC determination of the retained material. TOC is in a range between 0.4 and 14.6%, the contents are higher for larger size fractions on average. The mass share of the fraction < 100 µm lies between 32.3 and 91.8% (with an average of approx. 70%).
Despite the large differences between the sampled rivers and the diverse SPM characteristics, we detected MP in all samples in every fraction. We found PE in all samples and fractions, whereas PP, PS, and SBR were not detected everywhere. The fractionated filtration of the sedimentation box content approximates the size distribution of the detected MP. Hence, MP of all size classes and polymer types have been retained with sedimentation boxes, even some smaller than 10 µm.
The polymer concentrations in the retained solids range between 0.003 and 7.9 µg/mg for PP, PS, and SBR. Samples obtained with the in situ method have a lower average concentration (0.27 µg/mg) for these polymers than samples obtained with the pump method (0.94 µg/mg). Furthermore, for samples obtained with the in situ method, total MP concentrations were higher in the 100–500 µm fraction than in the <100 µm fraction (5.3 vs. 0.7 µg/mg). For samples obtained with the pump method, the tendency of higher total MP concentrations in larger size fractions is less pronounced (5.4 µg/mg for fractions < 100 µm and 11.1 µg/mg for fractions > 100 µm). In comparison to PP, PS, and SBR, the determined PE concentrations are considerably higher (4.71 µg/mg on average). When determining MP concentrations, the variability between the parallel boxes is considerably larger than it is for total masses. Figure S6 in the Supplementary Material displays the coefficients of variation for the different polymer types and fractions. These are considerably larger for the large-size fractions and range between 4 and 117%, with a mean of 36%.
When operating the sedimentation box with a pump, MP concentrations in rivers can be estimated based on the retention rates derived in the laboratory experiments. Figure 5 shows exemplary results for the estimated PS concentrations in the River Elbe and the Teltowkanal. Due to the known flow rate and total volume, the retained PS mass per sampled volume can be calculated (Figure 5, “retained by sedimentation box”). Including the specific retention rates, we calculated PS concentrations in the rivers per size fraction (Figure 5, “river”). We estimated the retention rates as 80% for the >500 µm fraction, 50% for the 100–500 µm fraction, and 20% for the 50–100 µm fraction. Particles < 50 µm were not part of the laboratory experiments, and their retention rates remain unknown.

4. Discussion

Sedimentation boxes have a simple and robust design, are easy to handle and install, and require no maintenance. Hence, sampling SPM and MP from surface waters with the device is cost-effective and may enable researchers to sample at a greater number of spots in comparison to other devices.
In this study, we aimed to evaluate the sampling performance of sedimentation boxes. We were able to identify different aspects which have an impact on the sampling performance and drive the uncertainty when determining MP concentrations in rivers. First, the retention rates of sedimentation boxes have a random or intrinsic error, which defines the reproducibility of the results. We quantified this error by comparing the results of three boxes operated in parallel. Furthermore, we expected the particle size and the flow velocity in the device to be decisive impact factors on retention rates. By conducting laboratory experiments under controlled conditions, we were able to quantify these factors and their variability. There are additional factors, such as the interaction with natural SPM, which likely increase the uncertainty of MP retention rates when sedimentation boxes are used to sample rivers. However, their influence could not be quantified with our setup. Moreover, the applied analysis method has a method-specific intrinsic error, which adds to the uncertainty of the derived MP concentrations. This error was not further investigated.
The coefficients of variation of the retained total mass are around 4%, and for the mass of different size fractions, they are well below 20%, regarding the samples provided by the three parallel sedimentation boxes. This did not change when switching from laboratory to field experiments. The MP concentrations obtained from rivers are subject to larger variation. Within a size fraction and between the parallel sedimentation boxes, the overall mean coefficient of variation is 36%. The uncertainty is larger for large-size fractions with a lower total mass and lower MP concentrations. Here, single MP particles may be decisive for the total concentration, which increases the variability between parallel boxes. However, the variation between MP concentrations cannot fully be attributed to the intrinsic error of sampling with the sedimentation box, as it cannot be distinguished from variability introduced within the process of sample preparation and MP detection with TED-GC/MS. To the best of our knowledge, there are no studies available which quantify the intrinsic error of MP analysis with TED-GC/MS.
Only a few studies report the intrinsic error of commonly applied MP sampling methods. When operating parallel manta nets, different variability between the replicate samples has been found, e.g., a mean coefficient of variation of 35% [53] or 89% [54] between MP concentrations. The replicate samples in Campanale et al. [55] differed with a coefficient of variation of 35% for the total number of MP particles and 52% after fractionation. However, due to different MP detection methods, the comparison of the results of the sedimentation boxes is limited.
Hence, when maintaining constant conditions (i.e., flow rate, particle composition, and concentration), the intrinsic error of sampling with sedimentation boxes is similar to or lower than the one derived for manta nets.
Our laboratory experiments have shown that the retention rates of particles with the same characteristics (e.g., density, shape) increase with increasing particle size. This is in line with the principle of Stokes’ law, which states that the sinking velocity of particles increases with their size. The retention rates in our experiments closely follow a square root function of the particle diameter, as expected from Stokes’ law. However, complete retention of particles < 454 µm (for a flow of 5.2 L/s) and <321 µm (for a flow of 2.6 L/s), as calculated using the equations in Figure 1, was not observed. For both flow rates, particles of that and larger sizes were only retained partially. This may be due to the particle shape, which was not spherical or areas with non-laminar flow in the box (see Figure 2). Size-specific retention rates confirm the findings of many studies, which observe a shift in particle size towards larger particles for the material retained with passive samplers in comparison to the input or surrounding material [12,13,15,16,17,18,19,21,23].
Our laboratory experiments have shown that retention rates increase with a decreasing flow rate, as expected. By halving the flow velocity, retention rates increase by an average factor of 1.3, i.e., almost 2 . Lower flow rates lead to a lower flow velocity and longer retention times inside the sedimentation box. Therefore, smaller particles with lower sinking velocities can be retained to a higher degree in comparison to higher flow rates. This effect has also been observed in laboratory experiments operating the TIMS sampler with different flow velocities [12,13]. When using the pump method, the flow velocity inside the box needs to be smaller than in the sampled river to allow the retention of formerly suspended particles. Although flow velocities in the Teltowkanal were very low (0.01 m/s on average), the flow velocity inside the box was lower (0.002 m/s); hence, sampling due to enhanced sedimentation was still feasible. A constant flow rate is crucial for quantitative results. Our velocity measurements have shown that an operation of sedimentation boxes in situ, i.e., with the natural flow of a river is prone to error due to highly variable flow rates inside the box. This effect is especially pronounced in rivers with unsteady flow regimes, such as the Teltowkanal, but certainly extends to other rivers.
In advance of the experiments, we expected the retention rates of PS particles to be higher, considering the retention rates determined by Harhash et al. [27], who conducted similar laboratory experiments, yet with a higher flow rate. For PS particles with a diameter of 228 µm and a flow rate of 8.5 L/min, they determined a retention rate of approx. 67%, while our experiments showed retention rates of 71% (2.6 L/min) and 51% (5.2 L/min) for PS particles with a diameter of 229 µm. A possible explanation could be different particle shapes or different MP concentrations in the suspension.
Remarkably, PE particles have been retained very similarly to PS particles in the laboratory experiments, although their density is lower than water (0.94 g/cm3). Due to their low density, it cannot be expected that they sediment in the boxes, as this is against Stokes’ law. With the sedimentation box being closed, PE particles are probably retained in the upper section of the sedimentation box. Positively buoyant particles can be trapped between the baffles of the sedimentation box or before the outlet, which is approx. 4 cm below the water level. As the density difference between the polymer and water is the same as for PS, the same mechanism applies in the opposite direction, leading to similar size-specific retention rates. Additionally, the low flow velocity close to the walls and especially in the corners of the box may favor the adhesion of the particles (see Figure 3).
Our laboratory experiments have shown that sampling with sedimentation boxes is sensitive to sampling conditions and alters the sample composition compared to the input material. However, the retention of particles using an artificial MP suspension is highly reproducible, and the retention rates are in an approx. square root function in relation to the particle diameter. Hence, in the given setup, particle size distribution and concentration can be recalculated from the obtained sample.
In the field experiments, polymers with lower densities than water (PE, PP) were retained by the sedimentation boxes. Furthermore, MP particles with a size < 10 µm were retained at the River Elbe and the Teltowkanal. In addition to possible sampling through positive buoyancy, these observations emphasize the relevance of additional factors, which are not considered when regarding an isolated MP particle. Possible explanations are the advection to the device’s walls, biofouling, or aggregation with larger and high-density minerals. In addition, Mancini et al. [56] found that the fast-settling (natural) particles scavenged MP particles and thereby increased their sedimentation velocity. This may favor the retention of small and low-density MP particles.
Although retention rates could only be averaged roughly for an entire size fraction, together with the controlled sampling conditions they allow the estimation of PS concentrations at River Elbe and Teltowkanal. Laermanns et al. [57] found PS concentrations in the water phase of River Elbe at 0.02–0.56 µg/L, which is in the same order of magnitude as our results. In the project “MiWa”, the Teltowkanal was sampled with a continuous flow centrifuge and a PS concentration of 2.9 µg/L was determined [58], which is in good accordance with our results too. However, due to different sampling and analysis methods, a direct comparison to other studies is limited. The MP concentration detected here and reported in the literature indicates that their fraction in SPM is typically below 1% (see also Table 1).
The determination of quantitative MP data is impossible using the in situ method of the sedimentation boxes due to the missing information on the precise flow velocity and volume that was flowing through the box. In rivers with very low or alternating flow velocities, such as the Teltowkanal, the operation with a pump ensures a constant flow through the box, which could otherwise come to a standstill. It is still unclear if polymers that have not been detected are not retained by the sedimentation box or if these polymers are not present in the sampled water bodies. Moreover, the retained polymer mass needs to exceed the limit of detection to be quantifiable. Although our monitoring campaigns, which operate sedimentation boxes in situ are not suitable to provide quantitative results, qualitative information on MP abundance in the rivers can be derived. Assuming similar physical properties and thus retention rates, the relatively constant MP concentrations in all size fractions indicate a similar particle size distribution for MP and natural SPM.
The PE concentrations determined with TED-GC/MS must be treated with caution. Several studies have shown that during combustion, natural organic material such as leaves or certain fats may form thermal decomposition products which are similar to markers used for PE quantification with TED-GC/MS or pyrolysis-GC/MS [47,59,60]. Lykkemark et al. [61] did not find similar matrix effects leading to an overestimation of PS and PP with pyrolysis-GC/MS. A comparison between the displayed PE concentrations and TOC values in Table 1 leads to a Pearson correlation coefficient of 0.78, indicating a good linear correlation. The correlation between PP, PS, and SBR concentrations with TOC is considerably weaker, with Pearson correlation coefficients ranging between 0.15 and 0.45. A good correlation between PE and TOC may be caused by a misinterpretation of organic compounds as PE. Thus, an overestimation of PE cannot be ruled out when applying these detection methods.
When sedimentation boxes are applied for the sampling of natural media such as rivers, we expect additional factors (other than flow velocity and particle size and density) to increase the uncertainty of retention rates. In the laboratory experiments, we determined the retention rates for artificial suspensions with pristine PS and PE particles of a limited size range (77–517 µm). Previously mentioned factors, such as interaction with natural SPM, weathering, biofouling, or different shapes, likely have an impact on the retention of MP in sedimentation boxes, but we did not include these factors. Moreover, the retention rates of particles < 77 µm remain unknown. These additional factors could not be assessed with the setup of our experiments. Hence, we cannot quantify the resulting increase in uncertainty for retention rates. Further investigations are needed to assess the impact of these additional factors on the retention of SPM and MP.

5. Conclusions

To date, there are no harmonized sampling methods for the monitoring of MP in water bodies because MP sampling is still challenging due to the low concentration, the high fluctuation, and the microscopic size. It has been shown that neither sampling techniques nor definitions distinguish between “natural” SPM and MP and that the properties of MP are in range with those of “natural” SPM. Hence, MP are already an integrated part of reported SPM.
Passive integrative sampling devices have been used to sample SPM in many different water bodies around the world. Yet, only a few studies have investigated their retention performance or identified factors that have an impact on retention. We investigated one type of passive sampler, sedimentation boxes, and operated them with a pump to achieve constant flow velocities, record the total sampled volume, and calculate retention rates.
Our laboratory experiments show that highly reproducible retention rates can be determined for the sedimentation boxes under controlled conditions. The retention rates increase with increasing particle size and a decreasing flow rate. Hence, both factors need to be considered in sampling strategies and when recalculating MP concentrations in the sampled water body.
We resume data from several experiments with sedimentation boxes in rivers across Europe. Despite the diverse river characteristics, we found MP in all samples and we conclude that sedimentation boxes (and most likely other passive samplers) are suitable to sample surface waters for MP qualitatively. Based on the determined retention rates, we could calculate PS concentrations in the sampled rivers, where we have operated the sedimentation boxes with a pump. The concentrations are in good accordance with the findings of previous studies.
Enclosed sedimentation boxes are suitable for retaining low-density polymers (<1 g/cm3) and small MP (<50 µm). Although particles may be underrepresented by sedimentation boxes due to the particle-size-specific retention rate, their actual concentration can be recalculated through their respective retention rates. Therefore, a standardized series of experiments with sedimentation boxes is needed to determine polymer- and size-specific retention rates under the effect of varying parameters such as flow velocity or SPM concentration.
In consequence, sampling MP with sedimentation boxes requires a low workload and provides highly reproducible results. In contrast to commonly used MP sampling methods such as manta nets, samples are time-integrated, and MP particles are not excluded due to their density or size. Although we were able to quantify some parameters that have an impact on the retention rates, the influence of others remains unknown (e.g., weathering, interaction with SPM, advection to the devices’ walls). Hence, we cannot fully determine the sensitivity of the method toward varying sampling conditions. Nevertheless, sedimentation boxes could be a valuable tool to obtain long-term information on MP in water bodies.

Supplementary Materials

The following supplementary material (Figures S1–S6) can be downloaded at: https://www.mdpi.com/article/10.3390/w17081152/s1.

Author Contributions

Conceptualization, C.J.S. and N.O.; methodology, C.J.S. and N.O.; investigation, C.J.S. and M.R.; writing—original draft preparation, C.J.S. and N.O.; writing—review and editing, M.R., P.G. and C.G.B.; visualization, C.J.S.; supervision, P.G. and C.G.B.; funding acquisition, C.G.B. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by the German Federal Ministry of Education and Research (BMBF) within the projects RUSEKU (02WPL1442A-K) and PLAID (01DS19004).

Data Availability Statement

The original contributions presented in this study are included in the article or supplementary material. Further inquiries can be directed to the corresponding author.

Acknowledgments

We would like to thank U. Braun and her team from unit III 2.5 at the German Environment Agency as well as K. Altmann and her team from unit 6.6 at the Federal Institute for Materials Research and Testing for analytical support. Furthermore, we would like to thank Florian Schwertfirm at KM Turbulenz for providing the results of the simulation of the sedimentation box within project RUSEKU.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
FTIRFourier-transform infrared spectroscopy
JDS4Joint Danube Survey 4
MPMicroplastics
NDIRNon-dispersive infrared sensor
PEPolyethylene
PPPolypropylene
PSPolystyrene
PSDParticle size distribution
QavgAverage discharge
SBRStyrene-butadiene rubber
SPMSuspended Particulate Matter
TED-GC/MSThermal extraction desorption-gas chromatography/mass spectrometry
TIMSTime-integrated mass-flux sediment
TOCTotal organic carbon
TRWPTire and road wear particles
UHMW-PEUltra high molecular weight polyethylene

References

  1. González-Pleiter, M.; Edo, C.; Velázquez, D.; Casero-Chamorro, M.C.; Leganés, F.; Quesada, A.; Fernández-Piñas, F.; Rosal, R. First detection of microplastics in the freshwater of an Antarctic Specially Protected Area. Mar. Pollut. Bull. 2020, 161, 111811. [Google Scholar] [CrossRef] [PubMed]
  2. Niu, X.; Wang, X.; Dong, H.; Ciren, N.; Zhang, H.; Chen, X.; Zhuoga, S.; Jia, X.; Xu, L.; Zhou, Y. Microplastics in remote region of the world: Insights from the glacier of Geladandong, China. Appl. Geochem. 2024, 168, 106026. [Google Scholar] [CrossRef]
  3. Godoy, V.; Calero, M.; González-Olalla, J.M.; Martín-Lara, M.A.; Olea, N.; Ruiz-Gutierrez, A.; Villar-Argaiz, M. The human connection: First evidence of microplastics in remote high mountain lakes of Sierra Nevada, Spain. Environ. Pollut. 2022, 311, 119922. [Google Scholar] [CrossRef] [PubMed]
  4. Zhao, B.; Rehati, P.; Yang, Z.; Cai, Z.; Guo, C.; Li, Y. The potential toxicity of microplastics on human health. Sci. Total Environ. 2024, 912, 168946. [Google Scholar] [CrossRef]
  5. Huang, W.; Song, B.; Liang, J.; Niu, Q.; Zeng, G.; Shen, M.; Deng, J.; Luo, Y.; Wen, X.; Zhang, Y. Microplastics and associated contaminants in the aquatic environment: A review on their ecotoxicological effects, trophic transfer, and potential impacts to human health. J. Hazard. Mater. 2021, 405, 124187. [Google Scholar] [CrossRef]
  6. Li, X.; Chen, Y.; Zhang, S.; Dong, Y.; Pang, Q.; Lynch, I.; Xie, C.; Guo, Z.; Zhang, P. From marine to freshwater environment: A review of the ecotoxicological effects of microplastics. Ecotoxicol. Environ. Saf. 2023, 251, 114564. [Google Scholar] [CrossRef]
  7. Talbot, R.; Chang, H. Microplastics in freshwater: A global review of factors affecting spatial and temporal variations. Environ. Pollut. 2022, 292, 118393. [Google Scholar] [CrossRef]
  8. Lu, H.-C.; Ziajahromi, S.; Neale, P.A.; Leusch, F.D.L. A systematic review of freshwater microplastics in water and sediments: Recommendations for harmonisation to enhance future study comparisons. Sci. Total Environ. 2021, 781, 146693. [Google Scholar] [CrossRef]
  9. Hidalgo-Ruz, V.; Gutow, L.; Thompson, R.C.; Thiel, M. Microplastics in the Marine Environment: A Review of the Methods Used for Identification and Quantification. Environ. Sci. Technol. 2012, 46, 3060–3075. [Google Scholar] [CrossRef]
  10. Primpke, S.; Fischer, M.; Lorenz, C.; Gerdts, G.; Scholz-Böttcher, B.M. Comparison of pyrolysis gas chromatography/mass spectrometry and hyperspectral FTIR imaging spectroscopy for the analysis of microplastics. Anal. Bioanal. Chem. 2020, 412, 8283–8298. [Google Scholar] [CrossRef]
  11. Schubert, B.; Heininger, P.; Keller, M.; Claus, E.; Ricking, M. Monitoring of contaminants in suspended particulate matter as an alternative to sediments. TrAC Trends Anal. Chem. 2012, 36, 58–70. [Google Scholar] [CrossRef]
  12. Phillips, J.M.; Russell, M.A.; Walling, D.E. Time-integrated sampling of fluvial suspended sediment: A simple methodology for small catchments. Hydrol. Process. 2000, 14, 2589–2602. [Google Scholar] [CrossRef]
  13. Doriean, N.J.C.; Teasdale, P.R.; Welsh, D.T.; Brooks, A.P.; Bennett, W.W. Evaluation of a simple, inexpensive, in situ sampler for measuring time-weighted average concentrations of suspended sediment in rivers and streams. Hydrol. Process. 2019, 33, 678–686. [Google Scholar] [CrossRef]
  14. Elliott, E.A.; Monbureau, E.; Walters, G.W.; Elliott, M.A.; McKee, B.A.; Rodriguez, A.B. A novel method for sampling the suspended sediment load in the tidal environment using bi-directional time-integrated mass-flux sediment (TIMS) samplers. Estuar. Coast. Shelf Sci. 2017, 199, 14–24. [Google Scholar] [CrossRef]
  15. Smith, T.B.; Owens, P.N. Flume- and field-based evaluation of a time-integrated suspended sediment sampler for the analysis of sediment properties. Earth Surf. Process. Landf. 2014, 39, 1197–1207. [Google Scholar] [CrossRef]
  16. Foets, J.; Wetzel, C.; Martínez-Carreras, N.; Teuling, A.; Iffly, J.F.; Pfister, L. Technical note: A time-integrated sediment trap to sample diatoms for hydrological tracing. Hydrol. Earth Syst. Sci. 2020, 24, 4709–4725. [Google Scholar] [CrossRef]
  17. Lučić, M.; Mikac, N.; Bačić, N.; Vdović, N. Appraisal of geochemical composition and hydrodynamic sorting of the river suspended material: Application of time-integrated suspended sediment sampler in a medium-sized river (the Sava River catchment). J. Hydrol. 2021, 597, 125768. [Google Scholar] [CrossRef]
  18. McDonald, D.M.; Lamoureux, S.F.; Warburton, J. Assessment of a time-integrated fluvial suspended sediment sampler in a high arctic setting. Geogr. Annaler. Ser. A Phys. Geogr. 2010, 92, 225–235. [Google Scholar] [CrossRef]
  19. Perks, M.; Warburton, J.; Bracken, L. Critical assessment and validation of a time-integrating fluvial suspended sediment sampler. Hydrological Processes 2014, 28, 4795–4807. [Google Scholar] [CrossRef]
  20. Schindler Wildhaber, Y.; Michel, C.; Burkhardt-Holm, P.; Bänninger, D.; Alewell, C. Measurement of spatial and temporal fine sediment dynamics in a small river. Hydrol. Earth Syst. Sci. 2012, 16, 1501–1515. [Google Scholar] [CrossRef]
  21. Clunies-Ross, P. Glacial Suspended Particulate Matter: Character, Composition and Adsorption Potential in Freshwater Environments. Ph.D. Thesis, University of Canterbury, Christchurch, New Zealand, 2017. [Google Scholar]
  22. Horton, S.L.; Stephenson, W.J.; Dickson, M.E. Trapping of fine-grained sediment on a supply-limited intertidal shore platform at Kaikōura, New Zealand. Geomorphology 2023, 442, 108926. [Google Scholar] [CrossRef]
  23. Heininger, P.; Schild, R.; de Beer, K.; Planas, C.; Roose, P.; Sortkjaer, O. Ermittlung der gewässerseitigen Einträge von Polyzyklischen Aromatischen Kohlenwasserstoffen (PAKs) in die Nordsee auf der Basis einer harmonisierten Methodik (internationales Pilotprojekt. Umweltbundesamt Texte 2002, 299, 286. [Google Scholar]
  24. Pohlert, T.; Hillebrand, G.; Breitung, V. Effects of sampling techniques on physical parameters and concentrations of selected persistent organic pollutants in suspended matter. J. Environ. Monit. JEM 2011, 13, 1579–1588. [Google Scholar] [CrossRef] [PubMed]
  25. Reinemann, L.; Schemmer, H. Neuartige Schwebstoffsammler zur Gewinnung von Schwebstoffen aus fließenden Gewässern. DGM 1994, 38, 22–25. [Google Scholar]
  26. ISO 5667-23:2011; Water quality—Sampling—Part 23: Guidance on Passive Sampling in Surface Waters. International Organization for Standardization: Geneva, Switzerland, 2011.
  27. Harhash, M.; Schroeder, H.; Zavarsky, A.; Kamp, J.; Linkhorst, A.; Lauschke, T.; Dierkes, G.; Ternes, T.A.; Duester, L. Efficiency of five samplers to trap suspended particulate matter and microplastic particles of different sizes. Chemosphere 2023, 338, 139479. [Google Scholar] [CrossRef]
  28. Masson, M.; Angot, H.; Le Bescond, C.; Launay, M.; Dabrin, A.; Miège, C.; Le Coz, J.; Coquery, M. Sampling of suspended particulate matter using particle traps in the Rhône River: Relevance and representativeness for the monitoring of contaminants. Sci. Total Environ. 2018, 637–638, 538–549. [Google Scholar] [CrossRef]
  29. Koelmans, A.A.; Redondo-Hasselerharm, P.E.; Nor, N.H.M.; de Ruijter, V.N.; Mintenig, S.M.; Kooi, M. Risk assessment of microplastic particles. Nat. Rev. Mater. 2022, 7, 138–152. [Google Scholar] [CrossRef]
  30. Harris, P.T. The fate of microplastic in marine sedimentary environments: A review and synthesis. Mar. Pollut. Bull. 2020, 158, 111398. [Google Scholar] [CrossRef]
  31. Maddah, H.A. Polypropylene as a Promising Plastic: A Review. Am. J. Polym. Sci. 2016, 6, 1–11. [Google Scholar]
  32. Jordan, J.L.; Casem, D.T.; Bradley, J.M.; Dwivedi, A.K.; Brown, E.N.; Jordan, C.W. Mechanical Properties of Low Density Polyethylene. J. Dyn. Behav. Mater. 2016, 2, 411–420. [Google Scholar] [CrossRef]
  33. Pugh, T.L.; Heller, W. Density of polystyrene and polyvinyltoluene latex particles. J. Colloid Sci. 1957, 12, 173–180. [Google Scholar] [CrossRef]
  34. Unice, K.M.; Weeber, M.P.; Abramson, M.M.; Reid, R.C.D.; van Gils, J.A.G.; Markus, A.A.; Vethaak, A.D.; Panko, J.M. Characterizing export of land-based microplastics to the estuary—Part I: Application of integrated geospatial microplastic transport models to assess tire and road wear particles in the Seine watershed. Sci. Total Environ. 2019, 646, 1639–1649. [Google Scholar] [CrossRef] [PubMed]
  35. Amaral-Zettler, L.A.; Zettler, E.R.; Mincer, T.J.; Klaassen, M.A.; Gallager, S.M. Biofouling impacts on polyethylene density and sinking in coastal waters: A macro/micro tipping point? Water Res. 2021, 201, 117289. [Google Scholar] [CrossRef]
  36. Lee, S.-Y.; An, J.; Kim, J.; Kwon, J.-H. Enhanced settling of microplastics after biofilm development: A laboratory column study mimicking wastewater clarifiers. Environ. Pollut. 2022, 311, 119909. [Google Scholar] [CrossRef]
  37. Semcesen, P.O.; Wells, M.G. Biofilm growth on buoyant microplastics leads to changes in settling rates: Implications for microplastic retention in the Great Lakes. Mar. Pollut. Bull. 2021, 170, 112573. [Google Scholar] [CrossRef]
  38. Wright, R.J.; Erni-Cassola, G.; Zadjelovic, V.; Latva, M.; Christie-Oleza, J.A. Marine Plastic Debris: A New Surface for Microbial Colonization. Environ. Sci. Technol. 2020, 54, 11657–11672. [Google Scholar] [CrossRef]
  39. Leiser, R.; Jongsma, R.; Bakenhus, I.; Möckel, R.; Philipp, B.; Neu, T.R.; Wendt-Potthoff, K. Interaction of cyanobacteria with calcium facilitates the sedimentation of microplastics in a eutrophic reservoir. Water Res. 2021, 189, 116582. [Google Scholar] [CrossRef]
  40. Sutherland, B.R.; Dhaliwal, M.S.; Thai, D.; Li, Y.; Gingras, M.; Konhauser, K. Suspended clay and surfactants enhance buoyant microplastic settling. Commun. Earth Environ. 2023, 4, 393. [Google Scholar] [CrossRef]
  41. Walling, D.E.; Owens, P.N.; Waterfall, B.D.; Leeks, G.J.L.; Wass, P.D. The particle size characteristics of fluvial suspended sediment in the Humber and Tweed catchments, UK. Sci. Total Environ. 2000, 251–252, 205–222. [Google Scholar] [CrossRef]
  42. Rügner, H.; Schwientek, M.; Egner, M.; Grathwohl, P. Monitoring of event-based mobilization of hydrophobic pollutants in rivers: Calibration of turbidity as a proxy for particle facilitated transport in field and laboratory. Sci. Total Environ. 2014, 490, 191–198. [Google Scholar] [CrossRef]
  43. Le, H.-A.; Gratiot, N.; Santini, W.; Ribolzi, O.; Tran, D.; Meriaux, X.; Deleersnijder, E.; Soares-Frazão, S. Suspended sediment properties in the Lower Mekong River, from fluvial to estuarine environments. Estuar. Coast. Shelf Sci. 2020, 233, 106522. [Google Scholar] [CrossRef]
  44. Crew, A.; Gregory-Eaves, I.; Ricciardi, A. Distribution, abundance, and diversity of microplastics in the upper St. Lawrence River. Environ. Pollut. 2020, 260, 113994. [Google Scholar] [CrossRef] [PubMed]
  45. Triebskorn, R.; Braunbeck, T.; Grummt, T.; Hanslik, L.; Huppertsberg, S.; Jekel, M.; Knepper, T.P.; Krais, S.; Müller, Y.K.; Pittroff, M.; et al. Relevance of nano- and microplastics for freshwater ecosystems: A critical review. TrAC Trends Anal. Chem. 2019, 110, 375–392. [Google Scholar] [CrossRef]
  46. Waldschläger, K.; Schüttrumpf, H. Erosion Behavior of Different Microplastic Particles in Comparison to Natural Sediments. Environ. Sci. Technol. 2019, 53, 13219–13227. [Google Scholar] [CrossRef]
  47. Kittner, M.; Kerndorff, A.; Ricking, M.; Bednarz, M.; Obermaier, N.; Lukas, M.; Asenova, M.; Bordos, G.; Eisentraut, P.; Hohenblum, P.; et al. Microplastics in the Danube River Basin: A First Comprehensive Screening with a Harmonized Analytical Approach. ACS EST Water 2022, 2, 1174–1181. [Google Scholar] [CrossRef]
  48. Stieß, M. Mechanische Verfahrenstechnik—Partikeltechnologie 1; Springer: Heidelberg, Germany, 2009. [Google Scholar]
  49. River Basin Community Elbe. Das Fachinformationssystem (FIS) der FGG Elbe. 2020. Available online: https://www.elbe-datenportal.de/FisFggElbe/content/start/BesucherUnbekannt.action (accessed on 10 November 2024).
  50. Czech Hydrometeorological Institute. Flood Forecasting Service. 2022. Available online: https://hydro.chmi.cz/index.php?lng=ENG (accessed on 10 November 2024).
  51. Wasserportal Berlin. Gewässerkundliche Messdaten—Messstelle Lichterfelde; Wasserportal Berlin: Berlin, Germany, 2022. [Google Scholar]
  52. Pekárová, P.; Pramuk, B.; Halmová, D.; Miklánek, P.; Prohaska, S.; Pekár, J. Identification of long-term high-flow regime changes in selected stations along the Danube River. J. Hydrol. Hydromech. 2016, 64, 393–403. [Google Scholar] [CrossRef]
  53. Frank, Y.A.; Vorobiev, D.S.; Kayler, O.A.; Vorobiev, E.D.; Kulinicheva, K.S.; Trifonov, A.A.; Soliman Hunter, T. Evidence for Microplastics Contamination of the Remote Tributary of the Yenisei River, Siberia—The Pilot Study Results. Water 2021, 13, 3248. [Google Scholar] [CrossRef]
  54. Yonkos, L.T.; Friedel, E.A.; Perez-Reyes, A.C.; Ghosal, S.; Arthur, C.D. Microplastics in Four Estuarine Rivers in the Chesapeake Bay, U.S.A. Environ. Sci. Technol. 2014, 48, 14195–14202. [Google Scholar] [CrossRef]
  55. Campanale, C.; Stock, F.; Massarelli, C.; Kochleus, C.; Bagnuolo, G.; Reifferscheid, G.; Uricchio, V.F. Microplastics and their possible sources: The example of Ofanto river in southeast Italy. Environ. Pollut. 2020, 258, 113284. [Google Scholar] [CrossRef]
  56. Mancini, M.; Serra, T.; Colomer, J.; Solari, L. Suspended sediments mediate microplastic sedimentation in unidirectional flows. Sci. Total Environ. 2023, 890, 164363. [Google Scholar] [CrossRef]
  57. Laermanns, H.; Reifferscheid, G.; Kruse, J.; Földi, C.; Dierkes, G.; Schaefer, D.; Scherer, C.; Bogner, C.; Stock, F. Microplastic in Water and Sediments at the Confluence of the Elbe and Mulde Rivers in Germany. Front. Environ. Sci. 2021, 9, 794895. [Google Scholar] [CrossRef]
  58. Jekel, M.; Anger, P.; Bannick, C.G.; Barthel, A.-K.; Braun, U.; Braunbeck, T.; Dittmar, S.; Eisentraut, P.; Elsner, M.; Gnirß, R.; et al. Mikroplastik im Wasserkreislauf—Probennahme, Probenaufbereitung, Analytik, Vorkommen und Bewertung (MiWa); Universitätsverlag der TU Berlin: Berlin, Germany, 2020. [Google Scholar] [CrossRef]
  59. Cassandra, R.; Yufei, P.; Elvis, D.O.; Jake, W.O.B.; Kevin, V.T. Extraction and Pyrolysis-GC-MS analysis of polyethylene in samples with medium to high lipid content. J. Environ. Expo. Assess. 2022, 1, 13. [Google Scholar] [CrossRef]
  60. Dierkes, G.; Lauschke, T.; Schweyen, P.; Ternes, T. Polyethylen überall! Wirklich? Poster. 2024. Available online: https://www.researchgate.net/publication/380532457_Polyethylen_uberall_Wirklich (accessed on 10 January 2025).
  61. Lykkemark, J.; Mattonai, M.; Vianello, A.; Gomiero, A.; Modugno, F.; Vollertsen, J. Py–GC–MS analysis for microplastics: Unlocking matrix challenges and sample recovery when analyzing wastewater for polypropylene and polystyrene. Water Res. 2024, 261, 122055. [Google Scholar] [CrossRef] [PubMed]
Water 17 01152 g001
Figure 1. (a): Behavior of a spherical particle in a stationary fluid according to Stokes. FB: buoyant force, FF: force of friction, FG: force of gravity, ρfl, ρp: densities of the fluid and the particle, Vp, dp: volume and diameter of the spherical particle, g: gravitational acceleration, η: dynamic viscosity, vs: terminal sinking velocity; (b): resulting velocity vt of a particle within the vertical flow in a device vd (c): particles of different sizes in a device with horizontal flow velocity vd.
Figure 1. (a): Behavior of a spherical particle in a stationary fluid according to Stokes. FB: buoyant force, FF: force of friction, FG: force of gravity, ρfl, ρp: densities of the fluid and the particle, Vp, dp: volume and diameter of the spherical particle, g: gravitational acceleration, η: dynamic viscosity, vs: terminal sinking velocity; (b): resulting velocity vt of a particle within the vertical flow in a device vd (c): particles of different sizes in a device with horizontal flow velocity vd.
Water 17 01152 g001
Water 17 01152 g002
Figure 2. Flow inside the sedimentation box. Red indicates areas with high flow velocity, and blue indicates areas with low flow velocity. Copyright by F. Schwertfirm (KM Turbulenz, Munich, 2023).
Figure 2. Flow inside the sedimentation box. Red indicates areas with high flow velocity, and blue indicates areas with low flow velocity. Copyright by F. Schwertfirm (KM Turbulenz, Munich, 2023).
Water 17 01152 g002
Water 17 01152 g003
Figure 3. Simulated distribution of SPM in sedimentation box for a single particle size. Red dots indicate particles moving with a high velocity; the darker the blue, the slower the movement of a particle until it settles. Copyright by F. Schwertfirm (KM Turbulenz, Munich, 2023).
Figure 3. Simulated distribution of SPM in sedimentation box for a single particle size. Red dots indicate particles moving with a high velocity; the darker the blue, the slower the movement of a particle until it settles. Copyright by F. Schwertfirm (KM Turbulenz, Munich, 2023).
Water 17 01152 g003
Water 17 01152 g004
Figure 4. Retention rates of laboratory experiments per PE and PS particle size and different flows.
Figure 4. Retention rates of laboratory experiments per PE and PS particle size and different flows.
Water 17 01152 g004
Water 17 01152 g005
Figure 5. PS concentration as retained by the sedimentation box and estimated concentration in the river. Displayed are the mean values of the two sampling runs and the three parallel boxes each. Error bars indicate the standard error. (a): Elbe (Cumlosen), (b): Teltowkanal (Kleinmachnow).
Figure 5. PS concentration as retained by the sedimentation box and estimated concentration in the river. Displayed are the mean values of the two sampling runs and the three parallel boxes each. Error bars indicate the standard error. (a): Elbe (Cumlosen), (b): Teltowkanal (Kleinmachnow).
Water 17 01152 g005
Table 1. Sedimentation box samples, Qavg: average discharge, fraction: mass share of the size fraction, ✗: the polymer has not been detected; SBR serves as a marker for TRWP (* samples from two different days; ** samples from two different days and three different boxes each. For these samples, the mass share, TOC, and polymer concentrations are displayed as mean values ± standard deviation; *** highly variable discharge).
Table 1. Sedimentation box samples, Qavg: average discharge, fraction: mass share of the size fraction, ✗: the polymer has not been detected; SBR serves as a marker for TRWP (* samples from two different days; ** samples from two different days and three different boxes each. For these samples, the mass share, TOC, and polymer concentrations are displayed as mean values ± standard deviation; *** highly variable discharge).
LocationCountryRiverMethodQavg
[m3/s]		Size
[µm]		Fraction [%]TOC
[%]		PE [µg/mg]PP [µg/mg]PS [µg/mg]SBR [µg/mg]
Böfinger HaldeAustriaDanubeIn situ124100–50019.5NA(9.92)0.270.580.47
<10080.52.1(0.73)✗0.030.15
Kloster-neuburgAustriaDanubeIn situ1920100–50067.71.0(0.49)✗0.003✗
<10032.30.4(0.16)✗✗✗
HainburgAustriaDanubeIn situ2050100–50029.93.1(1.32)✗✗✗
<10070.10.5(0.17)✗✗✗
BrnoCzech RepublicSvratkaIn situ7100–500NA7.7(2.52)0.270.470.69
<100NA1.8(0.25)✗0.06✗
PohanskoCzech RepublicThayaIn situ34100–50023.511.8(10.63)0.26✗✗
<10076.53.1(1.65)✗0.05✗
Lanžhot *Czech RepublicMoravaIn situ54100–50025.85.5(2.31)✗0.050.14
<10074.21.8(0.80)✗0.070.10
Budapest MOHungaryDanubeIn situ2336100–50026.34.6(2.97)✗1.37✗
<10073.71.0(0.46)✗0.008✗
PančevoSerbiaDanubeIn situ5320100–5008.28.8(9.14)0.380.220.17
<10091.87.7(0.53)✗0.04✗
BelgradeSerbiaSavaIn situ1700100–50010.411.4(6.01)0.160.481.03
<10089.61.9(0.52)✗✗0.18
RuseBulgariaDanubeIn situ6001100–50021.74.4(7.15)✗0.190.37
<10078.32.0(0.42)✗✗✗
UzhUkraineTiszaIn situ29100–50057.12.4(1.01)✗0.02✗
<10042.92.9(1.41)✗✗✗
Kleinmachnow **GermanyTeltow-kanalPump8 ***>5000.9 ± 0.003NA(7.82 ± 2.23)2.81 ± 2.571.89 ± 3.620.19 ± 0.03
100–50020.8 ± 0.0513.7 ± 0.63(9.70 ± 3.41)0.55 ± 0.177.90 ± 7.230.69 ± 0.16
50–10010.5 ± 0.0413.0 ± 0.11(10.55 ± 5.27)✗1.03 ± 0.530.93 ± 0.28
10–5030.6 ± 0.02NA(4.12 ± 0.27)0.14 ± 0.040.49 ± 0.092.61 ± 0.28
<1037.3 ± 0.02NA(0.33 ± 0.11)✗0.05 ± 0.010.12 ± 0.03
Cumlosen **GermanyElbePump408>5003.9 ± 0.02NA(4.44 ± 1.37)0.14 ± 0.180.22 ± 0.180.08 ± 0.09
100–50017.3 ± 0.0411.3 ± 0.68(7.53 ± 1.68)✗0.19 ± 0.040.15 ±
0.05
50–10014.3 ± 0.037.9 ± 0.43(6.46 ± 1.59)✗0.22 ± 0.070.25 ±
0.05
10–5025.7 ± 0.02NA(2.83 ± 0.05)✗0.23 ± 0.090.17 ± 0.02
<1038.6 ± 0.06NA(1.04 ± 0.29)✗0.13 ± 0.070.04 ± 0.02
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Saravia, C.J.; Ricking, M.; Grathwohl, P.; Bannick, C.G.; Obermaier, N. Surface Water Monitoring with Sedimentation Boxes: Assessing the Sampling Performance and Its Effect on Microplastic Concentration. Water 2025, 17, 1152. https://doi.org/10.3390/w17081152

AMA Style

Saravia CJ, Ricking M, Grathwohl P, Bannick CG, Obermaier N. Surface Water Monitoring with Sedimentation Boxes: Assessing the Sampling Performance and Its Effect on Microplastic Concentration. Water. 2025; 17(8):1152. https://doi.org/10.3390/w17081152

Chicago/Turabian Style

Saravia, Cristina Julieta, Mathias Ricking, Peter Grathwohl, Claus Gerhard Bannick, and Nathan Obermaier. 2025. "Surface Water Monitoring with Sedimentation Boxes: Assessing the Sampling Performance and Its Effect on Microplastic Concentration" Water 17, no. 8: 1152. https://doi.org/10.3390/w17081152

APA Style

Saravia, C. J., Ricking, M., Grathwohl, P., Bannick, C. G., & Obermaier, N. (2025). Surface Water Monitoring with Sedimentation Boxes: Assessing the Sampling Performance and Its Effect on Microplastic Concentration. Water, 17(8), 1152. https://doi.org/10.3390/w17081152

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop