Adding riverbanks to beaches citizen science monitoring for macrolitter, meso- and microplastics
Monitoring efforts for stranded debris have mostly focused on beaches (Serra-Gonçalves et al., 2019). Riverbanks are constantly supplied with plastic debris from the rivers, driving the need for more research and management of marine debris. Riverbanks were poorly investigated, with generally very few numbers of studied sites per river (Bruge et al., 2018; Rech et al., 2015). Only one study involved a large number of sites in German rivers, by involving citizen science with schoolchildren (Kiessling et al., 2019).
Citizen science monitoring provides a baseline understanding of debris composition, concentration and sources, and helps inform policies to reduce environmental impacts of plastic debris (Nelms et al., 2022). Numerous initiatives exist all around the world (Kawabe et al., 2022), but this study provides the first citizen science initiative for a comparison between debris found on riverbanks and beaches. This baseline study presents the application of debris citizen science monitoring called “Plastique à la loupe” to establish the first large-scale and long-term debris dataset for France, making it accessible to facilitate cost effective research efforts. In this study, conscientious collection by 3,113 schoolchildren from 149 classes removed a total of 48,023 macrolitters on riverbanks (n = 81 sites) and beaches (n = 66 sites) in two years. This labor-intensive monitoring effort would not have been feasible by a group of scientists, thus underlying the power of the Plastique à la loupe citizen science initiative for a French national survey. Since 2022, this dataset is used as complementary data in national assessments of aquatic litter pollution conducted by Cedre for French authorities in the context of the MSFD or other international monitoring programs.
Data quality controls
Participants were provided with support documents and visioconferences twice a year (by groups of 10 to 20 classes), allowing to gain confidence in their data-collection skills which was critical. The support document tool kit for teachers included (i) a support guide to explain the general concepts and objectives of the Plastique à la loupe initiative together with answer to frequently asked questions (FAQ), (ii) an easy and straightforward protocol guide slightly adapted from the OSPAR beach litter monitoring form (OSPAR, 2010), (iii) a photoguide for the macrolitter identification and (iv) a video guide for in situ training. The reliability of the sampling area chosen by the teachers was also verified by the scientists for each class. In addition, at the beginning of the schoolyear, the teams of teachers involved benefited from a one day formation to the project in the presence of the educational team of the Tara Ocean Foundation.
A main concern regarding citizen-science studies is whether the collected data are reliable and comparable to professional studies. In order to test the reliability of the sampling, sorting and data acquisition, 8 sampling sites (6 on riverbanks and 2 on beaches) were first analyzed by scientists (without removing plastics) before the on-site visit of schoolchildren and comparison showed no or very little difference for mesoplastics and macrolitter (data not shown). No significant difference with results gathered by experienced scientists was found in other citizen science studies performing similar data quality control (Thiel et al., 2013). However, more errors were found by the schoolchildren for the microplastics with non-plastic particles representing around 7% of the total number of microplastics on riverbanks and beaches. As previously observed, it was found that glass shards for example had been misidentified as small plastic debris (Hidalgo-Ruz & Thiel, 2015). This error can be easily corrected by the FTIR analysis that helps to detect non-plastic particles, which mitigates the impact of such error on the results.
Once the sampling site validation step was performed by scientists, only one site was excluded from the analysis for the length of the section was missing, thus underlying the high levels of coordination and personal motivation. Here, we underlined the importance of several steps including encouraging schoolchildren and teachers to describe any uncertainties to researchers, data auto-evaluation and communication of results as a concluding activity to enhance their commitment to the activity.
Distribution and composition of all debris on riverbanks and beaches
We observed that around 55% of all debris collected on riverbanks for 100 linear meters were plastic, which was much lower than on beaches (around 80% for 100 linear meters). This result is consistent with another study in Chili showing that plastics were the prevailing litter items and were more frequently found on beaches than on riversides (Rech et al., 2015). Another national study on German riverbanks found similar proportion of plastics among all debris (51%, including 20% of cigarette butts) (Kiessling et al., 2019). Other studies at local or regional scales found much higher proportions of plastics among all debris in the Adour riverbank (94%) and closed beaches (95%) (Bruge et al., 2018) or in the riverbanks of the Dutch Rhine-Meuse delta (85%) (Van Emmerik & Schwarz, 2020). Such discrepancy may be explained by local or regional disparities on the number of other types of debris (glass, metal, ceramics, paper, wood, rubber and textile) and on the modest sampling effort. In our case, a significant percentage of all debris on riverbanks were made of glass and metals, thus explaining the higher weight of all debris on riverbanks compared to beaches (median of 10 kg and 5 kg for 100 linear meters, respectively). These non-buoyant litter items are frequently attributed to non-riverine sources like direct litter dumping (Bravo et al., 2009), by opposition to the high abundance of plastic items that in addition can be transported by rivers and deposited on riverbanks due to their buoyancy and extreme persistence (Derraik, 2002; Moore, 2008).
Detailed plastic litters analysis in relation to their origin
Single-use plastics together with packaging (bags and wrappers) dominated most of the riverbanks (around 44.4%), in a higher proportion than on beaches (around 32.9%). In particular, food-related items dominated the top 10 single-use plastics. It was dominated by caps (mainly from plastic bottles) and thin wrapper on both riverbanks and beaches. Drink containers, shopping bags and food containers were found in higher proportions on riverbanks, despite they were also present on beaches. Most of these items are typically used by individuals and are classically found on riverbanks (Al-Zawaidah et al., 2021) and beaches (Lacroix et al., 2022). Either thrown away because of incivility (close to “take-away” restaurants), involuntary loss, or mismanagement (discarded during collection operations or transport by local authorities), they are ending up on city grounds, pushed away by the wind and runoff to rainwater collection systems which take them either straight to the closest river or to the next Waste Water Treatment Plant (WWTP) (Bruge et al., 2018). Cigarette butts, lollipop sticks and cotton swabs were found in higher proportions on beaches compared to riverbanks, probably due to incivility. Indeed, it has been shown that cigarette butts may not be considered littering by many smokers (Rath et al., 2012). As for the former three items, marine activities-related items (rope, buoys, floats, lures/lines, packaging straps) were much more present on beaches (24.9 ± 21.7%) than on riverbanks (4.5 ± 9.8%), probably reflecting the importance of higher losses from professional and recreative fishing activities in the marine environment in France. Together with fishing gears lost at sea during storms, discarding damaged nets is a common practice that results in debris accumulation on beaches or seafloor, close to zones of high fishing activity such as the north and south-west of the Gulf of Lion, and in the South Brittany region (Galgani et al., 2000). Here, we observed that around 87% of marine litter originated from land-based uses, which is consistent with classically found at a global scale (GRID-Arendal, 2016; Conservancy, 2017).
Together with the numerous broken glass and sharp metal objects, sanitary and medical litters represented a smaller portion of all the riverbanks litters (around 5%), but higher than uncounted on beaches (around 0.8%). They represent potentially dangerous items to human health, together with other items that were found less frequently such as decomposing food leftovers (which could attract disease-carrying animals or harm small children upon accidental ingestion) and litter items containing chemicals (e.g. aerosol cans, batteries, paint containers) (Kiessling et al., 2019). A specific awareness was given in the support guide, in the protocol guide and in the photoguide of Plastique à la Loupe initiative, to prevent risks for schoolchildren participants during sampling and sorting.
Litter types classified as “others” represented a significant proportion of all debris (10.5 ± 16.9 and 10.8 ± 17.9% on riverbanks and beaches, respectively). They included car parts, electronics, oil drums, batteries, etc. Attribution to this category is part of the OSPAR data collections scheme (OSPAR, 2020) and we decided to retain these data in our analyses. It diminished our ability to identify the source of litters, and we recognize that there are challenges regarding the source allocations for this category; yet, it gives information on macrolitters fragmentation, since the corresponding items are still recognizable. Photographs could have been used to go deeper in one specific item, but it is time consuming.
Interestingly, macroplastic fragments (> 2.5 cm) was the second dominant plastic type collected on riverbanks (23.2 ± 24.8%) despite it dominated the plastic debris on beaches (28.7 ± 24.5%). Fragmented plastic is a direct result of weathering and photodegradation, resulting in surface embrittlement and microcracking, yielding particles that are carried into the closest river or the next WWTP by wind and runoff to rainwater collection systems and also by wind and wave action when transported to beaches (Andrady, 2011). They mainly consist of foam, hard and soft fragments, of which their original item identity remains unknown.
Overall the detailed litter analysis provided more information to identify specific sources of (plastic) litter, and support policy-makers to implement prevention measures targeted at specific items.
Macro-, meso- and microplastics
To date, studies on microplastics mainly concerned ones floating at sea, while land-based studies of the stranded plastic litters on riverbanks and beaches focused more on macro- and mesoplastics (Vriend et al., 2020). Very few data exist on the comparison of all plastic sizes, despite a growing interest on understanding the “plastic cycle” (Hoellein & Rochman, 2021). The Plastique à la loupe initiative offers the possibility of tracking the different plastic sizes in a large set of riverbanks and beaches data. Tracing the source of plastics was possible only for a small proportion of the numerous collected items, mainly for identifiable macroplastics (25.8 ± 29.7% of all plastic size on riverbanks and 11.7 ± 19.0% on beaches) and microplastic pellets (13.1 ± 22.4% on riverbanks and 13.3 ± 19.5% on beaches). Most of the plastic items were non identifiable, resulting from the fragmentation of macroplastics into meso- and microplastics by breaking down in smaller size after exposure to ultraviolet light or mechanical forces once lost in the environment (Weinstein et al., 2016). Mesoplastics, originating from macroplastics fragmentation, represented a lower proportion of total plastic items on riverbanks than on beaches (21.7 ± 25.9% and 35.9 ± 22.0% respectively). However, it was difficult to conclude on any relation between the abundance of fragmented plastic litters and the distance to the sea from our current dataset, because of the lack of sufficient number of sites per river. We observed that abundances of meso- and micro-plastics were the most strongly correlated in both riverbanks and beaches.
Numbers of macro- and microplastics, and meso- and microplastics were positively correlated on both riverbanks and beaches. On beaches, there was a higher correlation between the abundances of meso- and microplastics than between macro- and microplastics (rs = 0.4, p-value = 6.2 x 10− 3; rs = 0.7, p-value = 7.7 x 10− 9), which is congruent with previous studies (Lee et al. 2013). The evaluation of the number of mesoplastics was proposed to serve as a better proxy of microplastic pollution than macroplastics, thus helping easier surveys to identify hot spots of microplastic pollution in large geographical areas with limited resources (Lee et al. 2013). That was not the case on riverbanks, where correlations between meso- and microplastics gave the same values than between macro- and microplastics (rs = 0.51, p-value = 1.1 x 10− 5 for both tests).
Microplastics represented a major part of the number of plastics found on both riverbanks (47.0 ± 34.2% of all plastic debris) and beaches (45.7 ± 25.5%). On riverbanks, a large proportion of microplastics were made of polystyrene (43%), which is congruent with previous results showing that such floating plastics tend to beach sooner and accumulate on riverbanks or lake beaches due to wind effects (Corcoran, 2015). On marine beaches, polyethylene dominated the microplastics (61.1%), as classically found in seawaters (Erni-Cassola et al., 2019). Interestingly, we observed on both riverbanks and beaches that a quarter of the microplastics were made of industrial pellets (primary microplastics, also known as virgin pellets or nurdles, recognized by their regular shape, usually cylindrical or ovoid), which form the feedstock of the plastics industry. These pellets enter the environment when they are spilled accidentally, either on land or at sea. Previous observation mentioned the clear link between the presence of industrial pellets and the vicinity to urban-industrial centers or in relation to accidental leakage during transport (Ryan et al., 2018).