1
Title: Predicted Growth in Plastic Waste Exceeds Efforts to Mitigate Plastic Pollution
Authors: Stephanie B. Borrelle1,2,3*, Jeremy Ringma4†,5,6, Kara Lavender Law7, Cole C.
Monnahan8, Laurent Lebreton9,10, Alexis McGivern11, Erin Murphy12,13, Jenna Jambeck3, George H. Leonard14, Michelle A. Hilleary15, Marcus Eriksen16, Hugh P. Possingham17,18, Hannah De
5
Frond1, Leah R. Gerber12,13, Beth Polidoro13,19, Akbar Tahir20,21, Miranda Bernard12,13, Nicholas Mallos14, Megan Barnes6,22, Chelsea M. Rochman1.
Affiliations:
1Department of Ecology and Evolutionary Biology, University of Toronto, Toronto, ON, Canada.
2 College of Engineering, University of Georgia, Athens, GA, USA.
10
3David H. Smith Conservation Research Program, Society for Conservation Biology, Washington DC, USA.
4 School of Biological Sciences, The University of Western Australia, Crawley, Western Australia, Australia.
5 Centre for Biodiversity and Conservation Science, University of Queensland, St Lucia, Queensland, 15
Australia.
6 Department of Natural Resources and Environmental Management, University of Hawai‘i at Mānoa, NREM, Honolulu, HI, 1902 East West Way, Honolulu, Hawaiʻi, 96816, USA.
7 Sea Education Association, Woods Hole, MA, USA.
8 Status of Stocks and Multispecies Assessments program, Resource Ecology and Fisheries Management, 20
Alaska Fisheries Science Center, National Oceanic and Atmospheric Administration, 7600 Sand Point Way NE, Seattle, WA 98115, USA.
9 The Ocean Cleanup Foundation, Rotterdam, The Netherlands.
10 The Modelling House, Raglan, New Zealand.
11 School of Geography and the Environment, University of Oxford, UK.
25
12 Center for Biodiversity Outcomes, Arizona State University, Tempe, AZ, USA.
13 School of Life Sciences, Arizona State University, Tempe, AZ, USA.
14 Ocean Conservancy, Washington, D.C., USA.
15 Center for Leadership in Global Sustainability, Virginia Polytechnic Institute and State University, Virginia, USA.
30
16 5 Gyres Institute, Los Angeles, California, USA.
17 School of Biological Sciences, The University of Queensland, Brisbane, Queensland, 4072, Australia.
18 The Nature Conservancy, Arlington, Virginia, USA.
19 School Mathematics and Natural Sciences, Arizona State University, Glendale, AZ, USA.
20 Department of Marine Science, Faculty Marine and Fisheries Sciences, Universitas Hasanuddin, 35
Makassar, Indonesia.
21 Research Center for Natural Heritage, Biodiversity and Climate Change, Universitas Hasanuddin, Makassar, Indonesia.
2
22 Centre for Environmental Economics and Policy, The University of Western Australia, Crawley, Western Australia, Australia, 6009
*Correspondence to: [email protected]
Abstract: Plastic pollution is a planetary threat, affecting nearly every marine and freshwater
5
ecosystem globally. In response, multi-level mitigation strategies are being adopted, but with a lack of quantitative assessments of how such strategies reduce plastic emissions. We assess the impact of three broad management strategies (plastic waste reduction, waste management and environmental recovery) at different levels of effort to estimate plastic emissions to 2030 for 173 countries. We estimate that 19.3 – 23.4 Mt, or 11% of plastic waste generated globally in 2016,
10
entered aquatic ecosystems. Considering ambitious commitments currently set by governments, annual emissions may reach up to 53 Mt yr-1 by 2030. To reduce emissions to a level well below this prediction, extraordinary efforts to transform the global plastics economy are needed.
One Sentence Summary: Despite global commitments to address plastic pollution, growth in plastic waste continues to outpace reduction efforts.
15
Main Text: Countries around the world are struggling to manage current volumes of plastic waste and ubiquitous plastic pollution (1, 2). From the poles to the deep ocean basins, marine and freshwater ecosystems are accumulating the world’s plastic debris (3–5). Simultaneously, the petrochemical industry announced over $204USD billion in investment, driven by the shale gas boom – leading to a projected acceleration in virgin plastic production (6).
20
As plastic production surges, multi-scale commitments aim to reduce plastic emissions into the environment (e.g. Addressing Single-Use Plastic Products Pollution (Resolution EA.4/L9) and United Nations Environment Assembly Resolutions Marine Litter and Microplastics, 1, 7; Goal 14.1 of the United Nations Sustainable Development Goals, 8). Communities, NGOs and businesses are cleaning beaches and promoting zero-waste lifestyles (9). Governments are
25
banning and placing levies on single-use consumer plastic products and, with the private sector, are investing in plastic waste management including integration into a circular economy (10–12).
A recent amendment to the Basel Convention targets marine plastic pollution by tracking the global trade of plastic waste to address issues of oversupply to countries that lack the capacity to manage it (13). However, all commitments to date lack a quantitative model that connects these
30
actions to a measurable reduction in plastic emissions.
Here, we present a mechanistic model to evaluate how different levels of effort would reduce plastic emissions into the world’s freshwater and marine ecosystems, that is major rivers, lakes, and oceans (hereafter ‘aquatic ecosystems’) by 2030. For 173 countries, representing ~97% of the world’s population, we estimate the amount of inadequately managed plastic waste entering
35
aquatic ecosystems annually from 2016 to 2030 for three scenarios: business as usual (BAU) where plastic production and waste generation follow current trajectories, an ambitious scenario that draws upon existing global commitments to reduce plastic emissions (1, 9, 10, 14, 15), and a target scenario to reduce annual plastic emissions. Because an environmentally acceptable threshold has yet to be defined, we set the target scenario to 8 Mt – the estimated global
40
emissions in 2010 to the oceans (16; a subset of aquatic ecosystems considered here) that
3
galvanized global action on plastic pollution by a variety of stakeholders (7). Scenarios demonstrating the level of effort required to achieve lower targets can be found in the Supplementary Materials.
We predict plastic emissions entering aquatic ecosystems to 2030 by integrating expected population growth (17), annual waste generation per capita (2), proportion of plastic in waste (2;
5
incorporating an increase in plastic materials associated with predicted production increases), and the proportion of inadequately managed waste by country (2, 16, 18; Supplementary
Materials; Fig. S1). For 173 countries with available data, we calculate annual plastic emissions entering aquatic ecosystems using a distance-based probability function. This function estimates the proportion of inadequately managed waste to reach the nearest major river, lake or ocean
10
based on spatially-explicit waste generation and downhill flow accumulation (18, 19;
Supplementary Materials; Figs. S1 & S2). That is, the closer to an aquatic ecosystem that waste is generated and inadequately managed, the greater the probability it will enter that aquatic ecosystem.
We adjusted variables for each country based upon their socio-economic status, as defined by the
15
World Bank (17): High Income (HI), Upper-Middle Income (UMI), Lower-Middle Income (LMI) and Low Income (LI) to account for the differences in plastic waste generation rates and waste management infrastructure among economies (2; Table 1; Supplementary Materials).
Across the three scenarios, we model three types of mitigation strategies over time: reducing waste generation (e.g., bans on single-use plastics), improving waste management (capture and
20
containment of plastic waste), and environmental recovery (e.g., clean up). A list of example actions that could be taken to achieve each type of strategy can be found in the Supplementary Materials (Table S2). We use a Monte Carlo simulation to propagate uncertainty of input parameters and scenarios (Supplementary Materials).
We estimate that approximately 19.3 – 23.4 Mt, or 11% of plastic waste generated globally in
25
2016, entered aquatic ecosystems (Fig. 1; Supplementary Materials; Table S4). This is consistent with an estimate of annual river emissions to the global oceans (0.8 – 2.7 Mt; 20) that is
calibrated with field observations. Our estimate is larger because it includes the amount that accumulates in lakes and rivers, in addition to the plastic that escapes to the ocean. Under BAU, we predict that the amount of plastic waste entering the world’s aquatic ecosystems could reach
30
90 Mt yr-1 by 2030 if waste generation trends continue as expected, with no improvements in waste management (Fig. 1A; Supplementary Materials; Table S4).
Under the ambitious scenario, we predict between 19.8 – 53.0 Mt yr-1 of plastic emissions to aquatic ecosystems by 2030 – remaining at or exceeding 2016 levels despite tremendous reduction efforts by the global community (Fig. 1A; Table 1; Supplementary Materials; Table
35
S4). The ambitious scenario to reduce plastic emissions is informed by global commitments from the G7 Plastics Charter, the European Union Strategy, the United Nations Environment
Programme, Clean Seas, and the Our Oceans conferences. Because these commitments generally lack specific numerical targets, and not all countries have made commitments, we apply
reduction targets to all countries within an income status based upon existing commitments made
40
by individual countries (Supplementary Materials). The ambitious scenario includes: 1) plastic
4
waste generation reduced from predicted trends by 10% in HI, 5% in UMI, 5% in LMI, and no change from 2016 in LI countries; 2) an increase in the proportion of managed waste, where HI countries reach a minimum of 90% managed waste (compared to a 2016 mean of 63%), UMI countries reach 70% (2016 mean of 40%), LMI countries reach 50% (2016 mean of 21%), and LI countries reach 30% (2016 mean of 6%), and; 3) recovery of annual global plastic emissions
5
from aquatic environments of up to 10% by 2030 in all countries (Table 1; Supplementary Materials; 21).
For the third scenario, we use our model to estimate the effort necessary to achieve a specified plastic emissions target by 2030 (< 8 Mt yr-1). We first focused on each intervention strategy (plastic reduction, waste management and environmental recovery) independently while holding
10
the others at the ambitious scenario levels. If additional actions were to solely focus on
reduction, plastic waste generation would need to be reduced by 85% across all income levels. If additional actions were to solely focus on waste management, every country would have to make exceptional efforts to properly manage ≥ 99% of its plastic waste. If additional actions were to solely focus on recovery, 85% of annual global emissions would have to be recovered from the
15
environment by 2030 (Supplementary Materials; Table S3). Although many stakeholders heavily promote only one of these strategies as the “best one”, these results demonstrate that drastic reductions in future plastic emissions cannot be achieved with any one strategy independently (Table 1).
Next, we systematically increased the level of effort for all three strategies simultaneously until
20
the target was reached in 2030 (mean global emissions of < 8 Mt; Fig. 1A; Supplementary Materials; Fig. S3; Tables S3). This requires plastic waste generation to be reduced by 40% in HI, 35% in UMI and LMI, and 25% in LI countries compared to the BAU trajectory. Levels of managed waste must reach 99% in HI and UMI countries, 80% in LMI countries, and 60% in LI countries. Recovery of 40% of annual global emissions by 2030 is needed (Fig. 1A; Table 1).
25
Considering all three strategies combined, the effort required to meet a reduction target of even 8 Mt far exceeds the existing and highly ambitious commitments to date from governments,
industries, NGOs and communities combined (e.g., 1, 9, 10, 14, 15).
It is important to note that these values may be an underestimate of plastic emissions. Across all scenarios, UMI and LMI countries contribute the most plastic waste emissions compared to HI
30
and LI countries (Fig. 1B; Table 1; Supplementary Materials; Appendix 3). However, the trade of plastic waste was not accounted for in the current model (Supplementary Materials). Waste shipped predominantly from HI to UMI, LMI and LI countries for processing may enter into a country with no formal waste management system or one that is less tractable, therefore
misrepresenting HI countries contributions to plastic emissions (22). Other factors may also lead
35
to uncertainties in our results. Global scale data for plastic waste generation, collection, and disposal are often lacking or unreliable because of inconsistencies in reporting among countries, differences in methodologies and units used in reporting, and omitted values (2, 18). We do not include primary microplastics, microplastics produced from the wear products still in use or microplastics entering the environment via wastewater – although these are likely comparatively
40
small in mass. We also do not include abandoned, lost or discarded fishing gear (ALDFG), which are important sources of plastic waste especially in marine ecosystems (23). We do not account for the unregulated burning of inadequately managed plastic waste, which may decrease
5
plastic emissions. Finally, there is a lack of data for most countries representing the efficacy of the informal waste management sector (2). One study in India estimated that 50% – 80% of generated plastic waste is recovered by the informal sectors (garbage collectors, waste pickers, waste dealers), and thus kept out of the environment (24; Supplementary Materials). The creation of a long-term standardized global monitoring program and open-access data for plastics placed
5
on the market, waste generation and management, the international trade of waste, environmental emissions, and transport in the environment will improve our ability to quantify both plastic emission pathways and the efficacy of mitigation strategies.
Our results show that the efforts required to meaningfully reduce plastic emissions by 2030 are extraordinary (Fig. 1; Table 1). Increased waste management capacity alone cannot keep pace
10
with projected growth in plastic waste generation. Further, without major technological
innovation, it is inconceivable that efforts to recover plastic waste from the environment could reach even 10% of annual emissions (~2 – 5.3 Mt in 2030), while our model identifies 40%
recovery is required to reduce emissions below 8 Mt (Table 1). These findings emphasize that unless growth in plastic production and use is halted, a fundamental transformation of the plastic
15
economy to a circular framework is essential, where end-of-life plastic products are valued rather than becoming waste.
Importantly, increasing global efforts to manage plastic waste must consider plastic pollution as a multi-dimensional issue. This includes evaluating the financial and social costs of
implementing (or not implementing) mitigation strategies, and also the impacts of different
20
mitigation strategies on economies, social justice, and human and environmental health, to achieve global sustainable development goals. For example, waste-to-energy processing (i.e., incineration) reduces plastic waste volumes, but may cause human health impacts from hazardous byproducts, create social justice issues, and increase greenhouse gas emissions (25, 26). Without such considerations, we risk creating perverse outcomes from the transformational
25
shifts needed to address plastic pollution.
Plastic pollution is a burgeoning threat to the sustainability of our planet (7, 8, 27). The world is responding at an already impressive scale, with grassroots action, national-level product bans, public-private partnerships for investment in waste management infrastructure, innovative alternatives to leakage-prone plastic products, and greater transparency in the trade of plastic
30
waste (7, 10, 13). Still, our results show that achieving substantial reductions in global plastic emissions to the environment requires an urgent transformative change. Key policies to achieve such a transition include: reducing or eliminating the use of unnecessary plastics; setting global limits for virgin plastic production; creating globally aligned standards for commodity plastics to be practically recoverable and recyclable by design; and developing and scaling plastic
35
processing and recycling technologies. Such harmonized policies can enable plastics to remain a valuable and useful commodity (10, 12). Further, some plastics will inevitably be emitted to the environment. Thus, recovery of plastic waste has to be a sustained priority to minimise adverse impacts on species and ecosystems (28) and limit harmful waste management practices such as open burning (25). Without this transformation, we risk continuing to invest large amounts of
40
human capital and financial resources with little to no hope of reducing plastic pollution in the world’s lakes, rivers, and oceans.
6
References and Notes:
1. United Nations Environment Assembly, “UNEA Resolutions Marine Litter and Microplastics (1/6, 2/11, 3/7)” (United Nations, Nairobi, Kenya, 2019), (available at https://papersmart.unon.org/resolution/second-adhoc-oeeg).
5
2. S. Kaza, L. C. Yao, P. Bhada-Tata, F. Van Woerden, “What a Waste 2.0 : A Global Snapshot of Solid Waste Management to 2050.” (World Bank, Washington, DC, 2018), (available at https://openknowledge.worldbank.org/handle/10986/30317).
3. M. Wagner, C. Scherer, D. Alvarez-Muñoz, N. Brennholt, X. Bourrain, S. Buchinger, E.
Fries, C. Grosbois, J. Klasmeier, T. Marti, Microplastics in freshwater ecosystems: what we
10
know and what we need to know. Environmental Sciences Europe. 26, 1–9 (2014).
4. R. W. Obbard, S. Sadri, Y. Q. Wong, A. A. Khitun, I. Baker, R. C. Thompson, Global warming releases microplastic legacy frozen in Arctic Sea ice. Earth’s Future. 2, 2014EF000240 (2014).
5. L. C. Woodall, L. F. Robinson, A. D. Rogers, B. E. Narayanaswamy, G. L. Paterson, Deep
15
sea litter: A comparison of seamounts, banks and a ridge in the Atlantic and Indian Oceans reveals both environmental and anthropogenic factors impact accumulation and
composition. Name: Frontiers in Marine Science. 2 (2015).
6. America Chemistry Council, “U.S. Chemical investment linked to shale gas: $204 billion and counting” (Fact sheet, 2019), (available at
20
https://www.americanchemistry.com/Policy/Energy/Shale-Gas/Fact-Sheet-US-Chemical- Investment-Linked-to-Shale-Gas.pdf).
7. United Nations Environment Assembly, “Addressing Single-use Plastic Products Pollution”
(UNEP/EA.4/L.10, Nairobi, Kenya, 2019), (available at
https://papersmart.unon.org/resolution/uploads/k1900861.pdf#overlay-context=node/271).
25
8. United Nations, “Sustainable Development Goal 14” (2018), (available at https://sustainabledevelopment.un.org/sdg14).
9. Ministry of Marine Affairs and Fisheries Republic of Indonesia, “Our Ocean
Commitments” (2018), (available at https://ourocean2018.org/?l=our-ocean-commitments).
10. European Union, EU Directives on plastic pollution (2017;
30
http://ec.europa.eu/environment/waste/packaging/legis.htm).
11. Y. Geng, J. Sarkis, R. Bleischwitz, How to globalize the circular economy. Nature (2019).
12. World Economic Forum, “The new plastics economy: rethinking the future of plastics.”
(2017), (available at
http://www3.weforum.org/docs/WEF_The_New_Plastics_Economy.pdf.).
35
7
13. Conference of the Parties (COP), Basel Convention on the Control of Transboundary Movements of Hazardous Wastes and Their Disposal BC-14/12 (1995;
http://www.basel.int/Portals/4/download.aspx?d=UNEP-CHW-IMPL- CONVTEXT.English.pdf).
14. Ocean Plastics Charter, “Ocean Plastics Charter.” (2018), (available at g7.gc.ca/wp-
5
content/uploads/2018/06/OceanPlasticsCharter.pdf).
15. UNEP, Clean Seas (2017; https://oceanconference.un.org/commitments/?id=13900).
16. J. R. Jambeck, R. Geyer, C. Wilcox, T. R. Siegler, M. Perryman, A. Andrady, R. Narayan, K. L. Law, Plastic waste inputs from land into the ocean. Science. 347, 768–771 (2015).
17. World Bank, “Data Catalog” (2019), (available at
10
https://datacatalog.worldbank.org/dataset/population-estimates-and-projections).
18. L. Lebreton, A. Andrady, Future scenarios of global plastic waste generation and disposal.
Palgrave Communications. 5, 6 (2019).
19. B. Lehner, K. Verdin, A. Jarvis, New global hydrography derived from spaceborne elevation data. Eos, Transactions American Geophysical Union. 89, 93–94 (2008).
15
20. L. J. J. Meijer, T. van Emmerik, L. Lebreton, C. Schmidt, R. van der Ent, “Over 1000 rivers accountable for 80% of global riverine plastic emissions into the ocean” (preprint,
EarthArXiv, 2019).
21. M. Cordier, T. Uehara, “Will innovation solve the global plastic contamination: how much innovation is needed for that?” (2167–9843, PeerJ Preprints, 2018).
20
22. A. L. Brooks, S. Wang, J. R. Jambeck, The Chinese import ban and its impact on global plastic waste trade. Science Advances. 4, eaat0131 (2018).
23. G. Macfadyen, T. Huntington, R. Cappell, Abandoned, lost or otherwise discarded fishing gear. (Food and Agriculture Organization of the United Nations (FAO), 2009).
24. B. Nandy, G. Sharma, S. Garg, S. Kumari, T. George, Y. Sunanda, B. Sinha, Recovery of
25
consumer waste in India–A mass flow analysis for paper, plastic and glass and the contribution of households and the informal sector. Resources, Conservation and Recycling. 101, 167–181 (2015).
25. Center for International Environmental Law, “Plastics and Health: The Hidden Costs of a Plastic Planet.” (CIEL, Washington DC, 2019), (available at https://www.ciel.org/wp-
30
content/uploads/2019/02/Plastic-and-Health-The-Hidden-Costs-of-a-Plastic-Planet- February-2019.pdf).
26. C. Wiedinmyer, R. J. Yokelson, B. K. Gullett, Global emissions of trace gases, particulate matter, and hazardous air pollutants from open burning of domestic waste. Environmental science & technology. 48, 9523–9530 (2014).
35
8
27. P. Villarrubia-Gómez, S. E. Cornell, J. Fabres, Marine plastic pollution as a planetary boundary threat–The drifting piece in the sustainability puzzle. Marine Policy. 96, 213–220 (2018).
28. K. Bucci, M. Tulio, C. Rochman, What is known and unknown about the effects of plastic pollution: A meta‐analysis and systematic review. Ecological Applications (2019).
5
29. R Core Team, R: A language and environment for statistical computing (Vienna, Austria, 2013).
30. C. Schmidt, T. Krauth, S. Wagner, Export of plastic debris by rivers into the sea.
Environmental science & technology. 51, 12246–12253 (2017).
31. R. Geyer, J. R. Jambeck, K. L. Law, Production, use, and fate of all plastics ever made.
10
Science Advances. 3, e1700782 (2017).
32. UN Environment, “Governments agree landmark decisions to protect people and planet from hazardous chemicals and waste, including plastic waste” (Geneva, Switzerland, 2019), (available at http://www.brsmeas.org/?tabid=8005).
33. J. H. Brand, K. L. Spencer, F. T. O’shea, J. E. Lindsay, Potential pollution risks of historic
15
landfills on low‐lying coasts and estuaries. Wiley Interdisciplinary Reviews: Water. 5, e1264 (2018).
34. S. Gündoğdu, C. Çevik, B. Ayat, B. Aydoğan, S. Karaca, How microplastics quantities increase with flood events? An example from Mersin Bay NE Levantine coast of Turkey.
Environmental Pollution. 239, 342–350 (2018).
20
35. C. C. Murray, N. Maximenko, S. Lippiatt, The influx of marine debris from the Great Japan Tsunami of 2011 to North American shorelines. Marine pollution bulletin. 132, 26–32 (2018).
Acknowledgments: Thanks to H. Savelli, K. Ingeman and E.S. Darling for input and comments on earlier drafts. Thanks to five anonymous reviewers for their suggestions for improving this
25
manuscript. Funding: This work was supported by the National Socio-Environmental Synthesis Center (SESYNC) under funding received from the National Science Foundation DBI-1639145.
SBB was supported by the David H. Smith Postdoctoral Research Fellowship; Author
contributions: S.B. Borrelle: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing - original draft, Writing - review &
30
editing, Project administration, Funding acquisition.J. Ringma: Conceptualization,
Methodology, Software, Formal analysis, Investigation, Writing - review & editing. K. L. Law:
Methodology, Validation, Formal analysis, Data curation, Writing - review & editing. C.C.
Monnahan: Methodology, Software, Validation, Formal analysis, Investigation, Writing - original draft. L. Lebreton: Methodology, Software, Validation, Formal analysis, Investigation,
35
Writing - original draft, Writing - review & editing. A. McGivern: Investigation, Writing - review & editing. E. Murphy: Investigation, Writing - review & editing. J. Jambeck:
Methodology, Investigation, Writing - review & editing. G. Leonard: Investigation, Writing -
9
review & editing. M.A. Hilleary: Investigation. M. Eriksen: Investigation. H.P. Possingham:
Methodology, Investigation, Writing - review & editing. L.R. Gerber: Investigation, Writing - review & editing. B. Polidoro: Investigation. A. Tahir: Investigation. H. De Frond:
Investigation, Writing - review & editing. M. Bernard: Investigation. N. Mallos: Writing - review & editing. M. Barnes: Conceptualization, Software, Investigation, Project
5
administration, Funding acquisition.C.M. Rochman: Conceptualization, Methodology, Formal analysis, Investigation, Resources, Writing - original draft, Writing - review & editing, Project administration, Funding acquisition.Competing interests: Authors declare no competing interests; and Data and materials availability: All data is available in the main text or the supplementary materials and in the GitHub Repository https://github.com/SBBorrelle/Borrelle-
10
et-al-Science-Plastic-Emissions.
Supplementary Materials:
Materials and Methods Supplementary Text
15
Figures S1-S5 Tables S1-S4 References (29-35)
20
10
Fig. 1. Annual global plastic emissions into aquatic ecosystems (includes major rivers, lakes, and the oceans) in million metric tonnes (Mt) from 2016 to 2030 (A), and for each income status (B) as defined by the World Bank (17): HI: High income; UMI: Upper-middle income; LMI: Lower-
5
middle income; LI: Lower income countries; showing the business as usual (BAU) (yellow), ambitious (blue), and target < 8 Mt (purple) scenarios. The shaded areas represent the 80%
credible intervals indicating the uncertainty in plastic waste generation and the scenario
implementation into the future. The orange horizontal line represents the target of < 8 Mt, which is a frequently cited statistic in global policy discussions as an unacceptable amount of plastic
10
emissions to the marine ecosystem alone (a subset of aquatic ecosystems considered here) (7).
11 Change in plastic
waste generation from predicted growth to 2030
% per capita
% Managed waste levels by 2030
Recovery by 2030
% of global annual emissions
2030 Income status emissions (Mt) 80% credible interval
2030 Global plastic emissions
(Mt) 80% credible
interval Business
as usual
Country level projections based on predicted trends
HI:
UMI:
LMI:
LI:
no change no change no change no change
All: 0
HI:
UMI:
LMI:
LI:
3.6 - 7.4 14.8 - 36.1 15.6 - 41.1 1.9 - 5.3
35.8 – 90.0
Ambitious HI:
UMI:
LMI:
LI:
-10 -5 -5 0
HI:
UMI:
LMI:
LI:
90 70 50 30
All: 10
HI:
UMI:
LMI:
LI:
1.9 - 4.1 7.5 - 21.6 9.1 - 24.1 1.2 - 3.5
19.8 – 53.0
Target (< 8 Mt)
HI:
UMI:
LMI:
LI:
-40 -35 -35 -25
HI:
UMI:
LMI:
LI:
99 99 80 60
All: 40
HI:
UMI:
LMI:
LI:
0.5 - 0.9 0.5 - 4.1 2.0 - 5.6 0.4- 1.4
3.4 – 12.0
Table 1. Mitigation strategy scenario values and projections of 2030 plastic emissions. Income status and global plastic emissions in 2030 and the levels of plastic waste reduction, waste
management improvement, and recovery of plastic waste under each scenario – business as usual
5
(BAU), ambitious and target (< 8 Mt). Specific actions that can be used to achieve reductions in plastic waste generation (e.g., product bans or taxes), waste management improvement (e.g., increased collection and controlled landfill), and recovery (e.g., beach clean ups) can be found in Table S2. In the ambitious and target scenarios, changes in plastic waste generation are
reductions implemented over time and fully achieved by 2030, and to the same level by countries
10
in the same income status as defined by the World Bank (17): HI: High income; UMI: Upper- middle income; LMI: Lower-middle income; LI: Lower income countries. “no change” indicates that 2016 proportions of inadequately managed plastic remain at 2016 values.
Submitted Manuscript: Confidential
1
Supplementary Materials for
Predicted Growth in Plastic Waste exceeds efforts to mitigate plastic pollution
Stephanie B. Borrelle, Jeremy Ringma, Kara Lavender Law, Cole C. Monnahan, Laurent Lebreton, Alexis McGivern, Erin Murphy, Jenna Jambeck, George H. Leonard, Michelle A.
Hilleary, Marcus Eriksen, Hugh P. Possingham, Leah R. Gerber, Beth Polidoro, Akbar Tahir, Miranda Bernard, Hannah De Frond, Nicholas Mallos, Megan Barnes, Chelsea M. Rochman.
Correspondence to: [email protected]
This PDF file includes:
Materials and Methods Supplementary Text Figs. S1 to S5 Tables S1 to S4
Captions for Data S1 to S4
Other Supplementary Materials for this manuscript include the following:
Data S1 to S4 1. Model inputs
2. Country Level Model Parameter Outputs 3. Economic Status Emissions Summary 4. Country Emissions Summary
Submitted Manuscript: Confidential
2 Materials and Methods
Overview
We forecast annual plastic emissions from 2016 through 2030 and compare multiple reduction scenarios to a business as usual (BAU) scenario. We used a mechanistic process flow model, programmed in the open source software R Version 3.5.0 (29) to estimate the proportion of plastic waste entering major rivers, lakes and oceans (hereafter referred to as “aquatic
ecosystem”) to 2030. The proportion for a country is based on the spatially-explicit generation of plastic waste on land combined with the likelihood for it to reach the nearest aquatic ecosystem (Eq. 1), the estimated proportion of inadequately managed waste (2, 16, 18), the expected population growth (17), and the expected change in plastic waste generation (per capita
municipal waste generation and proportion of plastic in municipal waste, 2). Sources of data are shown in Table S1. Next, we ran simulations using dynamic control variables set to change under different mitigation scenarios. These dynamic control variables are: reductions in plastic waste generation (a reduction in the amount of plastics that enter the waste management system);
increases in the proportion of managed waste (waste that is captured and securely contained);
and the recovery of plastic waste from the environment (e.g., capture and clean-up; Table S2).
Expected changes in fixed variables and scenario formulation for dynamic control variables were assigned based on the income status of each country; High income (HI), Upper-middle income (UMI), Lower-middle income (LMI), and Low income (LI), as per the World Bank’s definition (17). We grouped countries by income because of known trends in plastic use (2), similarities in waste management capacity, and feasibility of implementing different
interventions. Uncertainties in both fixed and dynamic variables were incorporated into model outputs through a Monte Carlo simulation with 1000 iterations, and expressed in figures as the mean output with 80% credibility intervals, and in the text as the range of 80% credible intervals (based on the 10th and 90th percentiles of modelled estimates).
Data
We used data to inform this baseline model for 173 countries, representing 96.6% of the global population in 2016 and decreasing to 96.2% in 2030 for predicted population growth (where data were available for the drainage and flow accumulation model, waste generation and predicted trends, and population projections), and generated scenarios about how they will change into the future. We focus on land-based sources of macroplastic emissions into the environment at the intervention points at the intersection of use and disposal of plastic products (waste), and actions taken once plastic waste has entered aquatic ecosystems. Our model does not include primary sources of microplastics, secondary microplastics from the degradation of macroplastic materials and wastewater treatment, or sea-based sources, such as abandoned, lost or otherwise discarded fishing gear (ALDFG), which are other important sources of plastic pollution in aquatic ecosystems (23) (Table S1).
Determining plastic emissions baseline
Plastic emissions, Ec,t,s, are computed annually for country c, in year t, and scenario s for each of 173 countries for years 2016-2030 according to:
𝐸𝑐,𝑡,𝑠 = 𝑃𝑐,𝑡 𝑊𝑐,𝑡 𝛽𝑐,𝑡𝑅𝑐,𝑡𝑀𝑐,𝑡,𝑠(1 − 𝜌𝑐,𝑡,𝑠)(1 − 𝜑𝑐,𝑡,𝑠) (1)
Submitted Manuscript: Confidential
3 where Pis population size, W is waste generation per capita (Fig. S1A), β is the proportion of waste that is plastic (Fig. S1B), Mis the proportion of inadequately managed waste (Fig. S1C), and Ris the emissions ratio (Fig. S1D; described below, using Eqs. 2 & 3) in county c and year t for scenario s. For each scenario s, one or more of the following interventions were applied:
1. Total plastic waste generated is reduced by 𝜌
2. The proportion of inadequately managed waste M is reduced; and 3. A proportion of plastic emissions 𝜑 is recovered.
Below we expound on how each variable was calculated and how we incorporated uncertainty into the projected emissions.
Population growth and per capita waste generation
Annual population size predicted to 2030 for each country is taken directly from the World Bank (17). Per capita waste generation rates and proportion of waste that is plastic for year 2016 are from Kaza et al. (2), and proportion of inadequately managed waste for each country is from Kaza et al. (2), Jambeck et al. (16), and Lebreton & Andrady (18).
Future per capita waste generation rates were computed from the annual expected change in waste generation for each country’s income status as per Kaza et al. (2), where the values to 2030 are a function of the per capita GDP and waste generation (computed from Fig. 2.6, page 27 in (2); GDP data (17)). The change in the proportion of plastic in the waste stream increases at a rate of 0 – 10%per annum from the 2016 proportion for a country and not exceeding 35% of the waste for any country (see also discussion on the Monte Carlo simulation, Appendix 1 for 2016 values, and Appendix 2 for country summaries of the change in β, 𝜑, 𝑀, and E from 2016 to 2030). If data were missing for waste generation per capita (n=2) or percent plastic (n=33) in the municipal solid waste for a country, the average value according to that country’s income status and region was used (Appendix 1).
Waste management categories defined in Kaza et al. (2) that we considered to be
inadequately managed include: “open dumping”, “waste disposed of in waterways or the marine environment”, “other”, and “unaccounted for”. The category “unspecified landfill” is ambiguous, and could reflect landfills ranging from properly managed (e.g., sanitary landfills) to open dumps (2). High proportions of waste in unspecified landfills were reported across the four income categories. Therefore, we assumed the proportion of inadequately managed waste in “unspecified landfill” varied according to economic status as: 0% for HI, 30% for UMI, 60% for LMI and 90% for LI countries. When values for inadequately managed waste for a country derived from Kaza et al.(2), were zero, greater than 100%, or missing we used data from Lebreton & Andrady (18) (Appendix 1). To account for unintentional or intentional littering, a minimum value for the proportion of mismanaged waste was set at 2% in 2016 (16). The maximum value for the
proportion of mismanaged waste was taken to be 99.9% in 2016.
Proportion of inadequately managed waste entering aquatic ecosystems
To estimate the emission ratio for a country Rc, the proportion of inadequately managed waste that enters aquatic ecosystems, we considered the distance from aquatic ecosystems to be mechanistically linked to the proportion of plastic waste that ultimately reaches an aquatic ecosystem. While Jambeck et al. (16) used a threshold based on the population living 50 km from the coastline, new spatial information (18) allowed us to calculate this proportion as a function of distance from aquatic ecosystems. For each country, we used the spatially-explicit distribution of per capita waste generation described in Lebreton & Andrady (18), and assumed
Submitted Manuscript: Confidential
4 that waste generation and waste management have the same spatial distribution. The data file of spatial waste generation used in our analysis can be found at
https://figshare.com/s/843a17a4995b4c9e8af6.
We then used a flow accumulation model to estimate the proportion of plastic emissions (mass of inadequately managed plastic waste) that enters an aquatic ecosystem. To calculate this, we used 30-second resolution gridded flow accumulation and drainage direction data distributed by HydroSHEDS (19). For each ~1 km2 cell on land we recorded downhill trajectories following dominant drainage direction until the trajectory reached an aquatic ecosystem, defined as a grid cell with flow accumulation across a surface area of 10,000 km2 (~10,000 cells) or greater, which we considered to be representative of large aquatic ecosystems (major river, lake and ocean) (Fig. S1; in HydroSHEDS, lakes are treated as a streamline that connects to the ocean, not accounting for dams and weirs, thus here, the ocean also includes lakes and major rivers). We continued to track the drainage of each large aquatic ecosystem to determine whether or not its ultimate endpoint (i.e., minimum elevation) was an inland depression or the ocean. If the aquatic ecosystem led to an inland depression, the cells where the flow originated were discarded (i.e., waste generated in these cells was assumed to remain on land). We assumed that it was unlikely for inadequately managed plastic to reach an aquatic ecosystem at a distance of more than 100 km from its origination. Thus, if cells where waste is generated were more than 100 cells (~100 km) from a downstream aquatic ecosystem, then they were also discarded. Finally, we assume that the probability of inadequately managed plastic in grid cell i to reach the downstream aquatic ecosystem decreases rapidly with distance from that aquatic ecosystem (reaching a value of 0 at 100 km):
𝛼𝑖 = 1 − 𝑈𝑙𝑜𝑔101(𝐷𝑖 + 1) (2)
where 𝛼𝑖 is the proportion of inadequately managed plastic waste originating in grid cell i that drains from a distance, Di, into an aquatic ecosystem; 𝑈 is a random variable drawn from a uniform distribution (interval 0.9 - 1.0, a range derived using an expert elicitation process from eight co-authors) that is included to reflect the uncertainty in this calculation. As part of a Monte Carlo simulation experiment (see below), 1000 replicates of Eq. 2 were generated for each grid cell.
We then calculated the emission ratio 𝑅𝑐 as the proportion of inadequately managed plastic waste entering aquatic ecosystems (Figs. S1D & S2) relative to the total produced in the country by summing across all cells in the country:
𝑅𝑐 =∑ 𝑊∑ 𝑊𝑖 𝑖𝛼𝑖
𝑖
𝑖 , (3)
where 𝑊𝑖 is the total waste generated for all grid cells, i, in country c. 𝑅𝑐 was assumed to be constant to the year 2030. Note that the calculation of R gives the proportion of total waste generated (W) that enters an aquatic ecosystem; because the spatial distribution of waste generation and waste management is assumed to be the same. The proportion of plastic in total waste and the inadequately managed fraction is then applied in the emissions calculation (Eq. 1).
The HydroSHEDS data extent is limited to 62°N leading to countries whose borders cross this latitude to have incomplete drainage and flow accumulation data. Countries where this occurred included the United States, Canada Russia and others, therefore to avoid excluding these
Submitted Manuscript: Confidential
5 countries from the analysis entirely we retained them and for areas that are north of 62°, we set the average emissions ratio to 0.25 (country average was 0.35).
Forecasting plastic emissions scenarios
The business as usual (BAU) scenario is a reference scenario to which we can compare the ambitious and target (< 8 Mt) scenarios described below. BAU assumes that population growth and waste generation per capita will follow trends published by the World Bank (17), and Kaza et al. (2), respectively (described above). The BAU scenario assumes that the proportion of inadequately managed waste (M) in Eq. 1) remains constant at 2016 levels (with variation due to the Monte Carlo simulations), and that there are no reductions in waste generation or recovery from the environment (𝜌𝑐,𝑡,𝑠 𝑎𝑛𝑑 𝜑𝑐,𝑡,𝑠 = 0 in Eq. 1; Appendix 1). The BAU assumes an
increase in the proportion of plastic in the waste stream associated with the estimated increase in plastic production (6), see ‘Including uncertainty with Monte Carlo simulation’ for
implementation details.
Mitigating plastics entering the environment can occur at various points along the plastic lifecycle. The reduction targets in our scenarios are driven by three broad categories of actions that would reduce plastic emissions:
1) reduced plastic waste generation; that is, either reduced plastics use (or consumption), or a reduction in plastics disposed and managed as waste (i.e., in favor of reclamation and reuse in a circular economy process) (11),
2) improved waste management that securely captures and contains plastic waste, and 3) recovery (capture or clean-up) of plastic waste once it has entered aquatic ecosystems.
These intervention strategies could be achieved using a suite of actions, which may include actions listed in Table S2, and others not listed. Across all scenarios, we set the levels of the dynamic control variables – reduction, improved waste management, and recovery – to be implemented over time to reach the final proportional reduction target in each scenario by 2030 for each income status (Table 1).
The ambitious scenario was informed by policy and infrastructure creation levers from the G7 Plastics Charter, the EU Strategy, the United Nations Environment Programme, and the Our Oceans conferences and the peer-reviewed literature (1, 9, 10, 14, 21). However, as few
commitments provide numerical targets, we developed the ambitious scenario by estimating the proportional contributions of a range of commitments. Further, we assumed that the same level of commitment will be adopted by all countries in the same income status. In this ambitious scenario, plastic waste generation is reduced from predicted waste generation levels at a linear rate to reach target reductions in 2030 of 10% in HI, 5% in UMI and LMI, and no increase in plastic waste generation in LI countries. The ambitious scenario strives for all HI countries to reach 90% managed waste by 2030 (2016 mean of 63%), all UMI countries to reach 70%
managed (2016 mean of 40%), LMI countries to reach 50% (2016 mean of 21%), and all LI countries to reach 30% managed waste (2016 mean of 6%) (Table 1 and Appendix 2 for country level trajectories of waste management implementation). In this ambitious scenario, we also assume that 10% of global annual emissions will be recovered from aquatic ecosystems by 2030, implemented at a linear rate.
The target (< 8 Mt) scenario was selected because it is often used in global policy
discussions as an unacceptably high annual plastic emissions rate to the oceans (1, 7). Examples of the level of effort required to achieve emissions lower than the 8 Mt target are given in Table S3.
Submitted Manuscript: Confidential
6 The resulting levels of interventions to reduce emissions to this level were determined through a series of model runs by systematically increasing each of the control variables (Figs.
S3, S4 & S5; Table S3). First, we tested how much we would have to increase each strategy independently, while holding the other interventions at the ambitious scenario level (Table S3).
Because the amount of increase was so large, and seemed implausible for any solution alone, we reduced waste generation, improved waste management, and increased recovery in parallel at 5%
increments until reaching a total emissions mean of < 8 Mt (Table S3; Fig. S3; and see Fig. S5 for Indonesia as an example). Income status and country emissions summaries are in Appendix 3
& 4.
Including uncertainty with Monte Carlo simulation
The values input into our projection model (Eq. 1) are not known precisely, so to reflect this we used a Monte Carlo simulation to incorporate uncertainty in our results. Monte Carlo
simulation entails assuming probability distributions for the variables in Eq. 1, drawing a random value from each distribution to calculate plastic emission projections, and repeating the process 1000 times to compute a mean and distribution around the mean. Here we describe the
assumptions and method used for the different variables.
The spatial emission ratios included a random variable from a uniform probability
distribution (Interval= 0.9, 1; variable U in Eq. 2; derived from an expert elicitation process of eight co-authors) in the calculation of the proportion of inadequately managed plastic waste that enters an aquatic ecosystem, or emissions ratio (variable R in Eq. 3). We also included
uncertainty in the projections of per capita waste generation (W), proportion of waste that is plastic (β), and the proportion of inadequately managed waste (M). Uncertainties for the mean value of these variables were drawn from a uniform probability distribution with intervals from the mean as follows; the annual increase in per capita waste generation W was between 0.54 – 1.45% per annum based on Kaza et al.’s (2) predictions; for β we assumed that the annual growth rate was between 0 – 10%, reaching a maximum of 35% of plastic in waste for any country; and for M we assumed that the annual rate of implementation for increased waste management was between 10 – 35% for the both the ambitious and target < 8 Mt scenario. That is, a country can improve its capacity to capture and contain plastic waste incrementally to approach the target level. This means some countries would reach the target level of managed waste prior to 2030, while some may be unable to reach the target level by 2030 (Appendix 2). No uncertainty was added for the implementation of reduction and recovery scenarios as all countries are presumed to achieve the target values at a linear rate (Table 1).
Thus, to calculate a single replicate of the results for each country we 1) generated an emissions ratio value, 2) generated slopes (trajectories for the growth of a variable from 2016 – 2030) from a uniform probability distribution, with intervals as described above for W, β, and M, then 3) calculated those trajectories based on the given scenario and income status, and 4)
applied Eq. 1. The result is a projection of plastic emissions per country and scenario
(summarized in Appendix 4). We use the Monte Carlo replicates (1000 times for each country and scenario combination) to characterize uncertainty about these results presenting the mean values and 80% credible intervals (10% and 90% percentiles).
Supplementary Text
Model comparison and limitations
Submitted Manuscript: Confidential
7 The difference in plastic emissions to the environment compared to previous estimates (16) stem from two key factors. First, we use a spatially explicit waste generation data (18) to
calculate the proportion of plastic waste that is inadequately managed that has the potential to enter aquatic ecosystems, marine and major freshwater lakes and rivers based on drainage and flow accumulation (19). This combines the calculation of plastic waste entering rivers, lakes and oceans further inland than 50km – the distance used in Jambeck et al (16), and rivers by Lebreton et al. (18) and Schmidt et al (30). Second, we calculate an increase in the proportion of plastic in the waste stream for each country, accounting for growth in plastic products being consumed (Appendix 2).
Our model uses the rate of historical plastic waste generation as a representation of future growth, but does not take into account for additional acceleration of plastic production driven by the expansion of shale gas production (6). The increase in plastic production may especially affect growth in plastic waste generation in LMI and LI countries, where access to plastic- packaged products has increased and will continue to increase at an unknown rate (31). Further, we did not account for the projected development of a country’s economic status (e.g., from LI to LMI), which would likely result in an increase in per capita waste generation.
The trade of plastic waste was not accounted for in the current model, which would affect the country and global totals of plastic waste emissions (22). This is because waste shipped from HI and UMI countries to LI and LMI countries for processing is included in the estimate for the origination country in our model when, in fact, this waste may enter into a country with no formal waste management system or one that is less tractable (22). We do not include this
because we know of no formal studies of the fate of imported waste. Further, the global recycling rate is on average 9% of all plastic waste (although it varies by country) (31). Of this 9%, about half of all plastic waste is exported. For example, in 2016 before import restrictions, 14.1 Mt of plastic scrap was exported by 123 countries, with China taking most of it, 7.35 Mt, from 43 different countries (22). Since exported plastic is only 4.5% of the plastic waste stream, on a global scale, it would not impact the model more than 4.5% by mass. However, we do
acknowledge that on a more local scale (like in the case of the majority of waste going to one country) it could have an impact. Thus, if individual countries are examined, plastic waste import data should be considered.
A recent amendment to the Basel Convention aims to make the transfer of plastic scrap among countries and its fate in the importing country more transparent (13, 32). This
amendment, agreed to by more than 180 countries, requires consent from an importing country before an exporting country can export plastic scrap, in an effort to specifically address marine plastic pollution. These data are crucial for improving our understanding of the fate of imported plastic waste materials.
We did not account for the effect of the informal recovery sector (IRS) on reducing plastic emissions. The World Bank estimates there are approximately 15 million people working in the IRS around the world; however quantitative estimates of the impact of this sector on reducing plastic pollution are lacking (2). Therefore, when examining individual country emissions, the impact of the IRS should be considered.
Additional limitations not directly discussed in the main article include uncertainty in transmission rates of inadequately managed waste from land into aquatic ecosystems. We do not consider climatic factors, including rainfall (dry versus wet countries) or seasonality that would affect transmission rates of plastic waste from land to aquatic ecosystems. Further work in this area will greatly improve our ability to more accurately quantify plastic emissions. We do not
Submitted Manuscript: Confidential
8 account for the release of stored plastic waste due to coastal erosion (33), extreme weather events (34), and natural disasters (35), which have the potential to release large amounts of plastic waste intermittently. Finally, we do not account for the unregulated burning of inadequately managed plastic waste, which may decrease plastic emissions but result in alternative and severe impacts, such as human health impacts, increased carbon emissions and reduced air quality (25, 26).
In terms of the scenarios, the estimates were set as proportional changes in plastic waste generation, improvements in waste management, and the recovery of plastic emissions from the environment. We assumed that all countries in the same income category would adopt the same level of reduction, and that implementation would occur at the same rate across all income status categories. These uncertainties also impact other estimations of aquatic plastic pollution rates and will not be resolved without significant investment in monitoring of waste generation and management.
Submitted Manuscript: Confidential
9 Fig. S1.
Country level summaries for (A) kilograms (kg) of waste generated per capita per day in 2016;
(B) proportion of plastic in waste in 2016; (C) proportion of inadequately managed waste in 2016, and (D) country average emissions ratio (Eq. 3). The countries that are in grey are those that are not included because of a lack of data. These missing countries collectively represent 4.4% of the global population in 2016, increasing to 4.8% for 2030 population predictions (17).
Submitted Manuscript: Confidential
10 Fig. S2.
To calculate an emissions baseline for each country (example here of Indonesia). (A) is the log of plastic waste generation (kg; methods described in Lebreton & Andrady (18), which is intersected with a 30 second resolution gridded flow accumulation and drainage direction map (19) (% drained) calculated using a spatial emissions function (Eq. 2), shown in (B). The
emissions ratio (C) is then calculated by dividing the proportion of inadequately managed waste that drains into an aquatic ecosystem by the total inadequately managed plastic for each country, which is based on data from 2016 (2). We assumed that this ratio is constant to the year 2030 (Eq. 3). Figures for all other countries included in the analysis can be found in the GitHub repository https://github.com/SBBorrelle/Borrelle-et-al-Science-Plastic-Emissions
Submitted Manuscript: Confidential
11 Fig. S3.
Mean (line) and 80% credible intervals (shaded area) of global plastic emissions for each of the trials to reach emissions for the target < 8 Mt scenario (orange horizontal line).
Target trial #7 Target trial #8 Target trial #9
Target trial #4 Target trial #5 Target trial #6
Target trial #1 Target trial #2 Target trial #3
2020 2025 2030 2020 2025 2030 2020 2025 2030
0 8 20 30 40
0 8 20 30 40
0 8 20 30 40
Year
Plastic emissions (Mt)
Submitted Manuscript: Confidential
12 Fig. S4.
Example Monte Carlo simulations (100 runs shown) of plastic emissions (in million metric tonnes; Mt) of the reduction scenario model runs for Indonesia to 2030. For each of the plastic waste reduction scenarios (top row), waste management was held constant at 50% and recovery was held at 10% of emissions. For waste management scenarios (middle row), plastic waste reduction was held constant at 5% and recovery was held at 10% of emissions. For recovery scenarios (bottom row), plastic waste reduction was held constant at 5% and waste management was held at 50%.
Submitted Manuscript: Confidential
13 Fig. S5.
Target scenario selection example for Indonesia. The figure shows the Monte Carlo simulations (100 runs shown) of plastic emissions (in million metric tonnes; Mt), for each of the trials including reduction in plastic waste generation, waste management improvements, and recovery to achieve the emissions target of < 8 Mt by 2030. Values and annual plastic emissions for each income status and trial are shown in Table S3.
Submitted Manuscript: Confidential
14 Table S1.
Data source summary (see Data Appendix 1).
Data Description Source Link
Income status
Countries were classified as per the World Bank income of GNI per capita
World Bank (17)
https://datacatalo g.worldbank.org/d ataset/country- profiles
Population growth
Country level population data for 2016 and projections to 2030
World Bank (17)
https://datacatalo g.worldbank.org/d ataset/population- estimates-and- projections Waste
generation rate per capita
Total municipal solid waste [Kg/ person/day]
Kaza at al. (2) – 2016 adjusted values used for model inputs that calibrate for the range of time that the data are from.
Annual rate of change in waste
generation per capita
Waste generation is expected to increase with GDP growth.
Kaza et al. (2) taken from Fig. 2.6 (page 27).
Percent plastic waste
Proportion of municipal solid
waste composed of plastics Kaza et al. (2) Annual rate of
change in percent plastic waste
Proportion of plastics in municipal solid waste is expected to increase annually.
Calculated with a Monte Carlo simulation starting from the 2016 proportion reported in Kaza et al. (2) and increasing at between 1-10%
per annum with a maximum proportion of 35% in any country.
Proportion of inadequately managed waste
The proportion of waste which is inadequately managed
Kaza et al. (2); Lebreton & Andrady (18), and Jambeck et al. (16)
Environmental recovery
The proportion of waste removed from rivers, lakes, and the oceans to reduce environmental
contamination.
For the Ambitious scenario we were informed by estimates from Cordier
& Uehara (21)
Submitted Manuscript: Confidential
15 Table S2.
The broad categories of interventions we evaluate in the plastic futures alternatives are reduction in plastic waste generation, waste management improvements (that result in a reduction in inadequately managed waste), and the recovery of plastic emissions from aquatic ecosystems (major rivers, lakes and oceans). Each of these interventions can be achieved through a variety of specific actions, some examples of which are listed below (non-exhaustive list).
Method Specific Actions
Reductions in plastic waste generation (ρ)
• Reducing virgin plastic production made from fossil feedstocks
• Education, public awareness campaigns leading to behavioral change and reductions in personal plastic waste generation
• Legislative level bans, levies or taxes on ‘single-use’,
‘disposable’ or ‘unnecessary’ products, such as thin film shopping bags
• Replacement with alternative feedstocks that are easily compostable
• Requirements for producers to report information
regarding quantity and types of designated products and packaging supplied
Waste management (M) • Increased availability of waste bins and waste collection
• Increased recycling sorting and processing capacity
• Improved collection
• Engaging informal waste sector using financial incentives/bounties for plastic recovery
• Return to point of sale collection systems, e.g. container deposit schemes
• Residential and multifamily dwelling collection systems
• Collection, reuse, and recycling performance targets
• Building engineered landfills (low/no plastic emissions)
• Reclamation and repurposing (e.g., using waste to make new products)
• Internationally harmonized recycling standards and/or environmental standards for plastic products to improve recyclability
Recovery (φ): the interception, and clean- up of plastic pollution
• River clean up, e.g., river booms, dams, trash wheels
• Beach clean-ups
• Trash boats
• Ocean based clean-up devices
Submitted Manuscript: Confidential
16
• Seabins or other litter capture devices
