www.sciencemag.

org/content/347/6223/768/suppl/DC1

Supplementary Materials for

Plastic waste inputs from land into the ocean
Jenna R. Jambeck,* Roland Geyer, Chris Wilcox, Theodore R. Siegler, Miriam
Perryman, Anthony Andrady, Ramani Narayan, Kara Lavender Law

*Corresponding author. E-mail: jjambeck@uga.edu

Published 13 February 2015, Science 347, 768 (2015)

DOI: 10.1126/science.1260352

This PDF file includes:

Materials and Methods

Supplementary Text
Fig. S1
Tables S1 to S6
Full Reference List

Other Supplementary Material for this manuscript includes the following:

(available at www.sciencemag.org/content/347/6223/768/suppl/DC1)

Data S1 (Excel file)

Methods

Estimating per capita waste generation rates and percentage of plastic in the waste stream
in 2010

The World Bank generated the most recent and most comprehensive estimates of per capita
waste generation rates and percentage of plastic waste for 145 countries in the year 2005 (5). Of
the 192 coastal countries in our analysis, waste generation rates were reported for 128 countries,
and percent plastic waste for 73 countries. To estimate these quantities for the remaining
countries, we applied average values for each economic classification defined by the World Bank
(HIC = high income; UMI = upper middle income; LMI = lower middle income; LI = low
income) based upon 2010 gross national income per capita (GNI; from
http://unstats.un.org/unsd/pocketbook/World_Statistics_Pocketbook_2013_edition.pdf). One
exception is China, for which the 2010 value from a more recent World Bank study was used
(23). This study reported a lower waste generation rate (1.1 kg/person/day) than would have
been assigned using China’s 2010 economic classification (1.2 kg/person/day). Waste generation
rates likely increased from 2005 to 2010, thus our estimates are conservative.

To project the trend of plastic in the waste stream from 2005 onwards, we developed a model to
predict the annual growth rate of the percent plastic in the waste stream using measured
percentage of plastic in the municipal solid waste stream in the United States from 1960 (0.4%)
through 2012 (12.7%), reported by the U.S. Environmental Protection Agency (24) (Fig S1).
This proportional growth reflects increased plastic use due, in part, to the substitution of plastic
for heavier materials (i.e., glass, metal). We fit three linear models (constant, first order and
second order) to the curve of annual change in percent plastic versus time in the United States.
The constant rate of increase (0.19% per year, standard error 0.0623%) was the best fit as
determined by the lowest Akaike Information Criterion (AIC) score. This fractional increase,
applied annually to each country in the study from 2005 onwards, is conservative compared to
the growth in global plastic resin production (average ~5% per year from 1960 to 2011; 3).

Estimating percentage of waste that is mismanaged

To quantify the percentage of mismanaged waste, we considered inadequate waste management

practices separately from littering. We classified waste management practices for 81 coastal
countries in which disposal methods were reported (5); we considered waste managed in landfills
(high- and middle-income countries only) and in composting, recycling, and waste-to-energy
programs to be “adequately managed”. Dumps and landfills in low-income countries are
described by the World Bank as, “Low-technology sites usually open dumping of wastes. High
polluting to nearby aquifers, water bodies, settlements” (5). In addition, first-hand study of solid
waste management in 14 developing countries by one of the authors (T. R. Siegler) supports the
assertion that landfills in low-income countries are not adequately managed. Therefore, we
considered landfills in low-income countries and all dumps to be “inadequately managed”. The
results were not substantially different if landfills in low-income counties were considered
adequately managed or if those data were removed altogether.
We developed a logistic regression model to estimate the percentage of waste that is
inadequately managed in each country. We modeled the ratio of adequate to inadequate waste
management using data on waste disposal methods, economic classification and geographic
region (as defined by the World Bank) for 81 countries for which we had complete data (5). We
explored the effect of 2010 GNI and geographic region on the probability of inadequate
management. We were also concerned about the variation in knowledge across the reporting
countries on the fate of waste. In some cases the “Other” category of waste disposal methods
accounted for as much as 94% of the total reported fates, although the median share of the
reported fates in the Other category was 0.015%. We accounted for this by using the ratio of
waste in the Other category to the total waste as a weight for the data in the regression, thus
down-weighting data where there was significant uncertainty with respect to fate. Based on AIC
scores, the best model used both GNI and region (Table S4). As expected, the probability of
inadequate disposal of waste decreased with increasing income. Four of the regions had
significantly different disposal behavior; two regions (Europe and Central Asia (ECA) and Latin
America and the Caribbean (LCR)) had a lower inadequate management fraction than expected
based on income alone, while two regions (East Asia and Pacific (EAP) and Middle East and
North Africa (MENA)) had a higher inadequate management fraction than expected (Table S5).
Using this fitted relationship we predicted the mean percentage of inadequately managed waste
for the remaining countries, including a standard error.

Litter studies are difficult to synthesize because they are typically designed to evaluate counts of
particular items and rarely report mass, and they vary substantially in methodology, which limits
comparison between studies. We estimated percentage of waste littered using the only available
national estimate of litter mass (25), which reported 4.17 million MT of litter generated in the
United States in 2008, equivalent to approximately 2% of national waste generation (24). For
each country we estimated 2% of the mass of total waste generated is littered. Although littering
is ill-defined in the absence of formal waste management, in countries where waste management
infrastructure is robust, litter can have a measurable impact (e.g., the United States and countries
in the European Union).

Estimating the input of mismanaged plastic waste to the ocean

Some percentage of the total mismanaged plastic waste (inadequately managed plus litter) enters
the ocean and becomes marine debris. To our knowledge no direct estimates of this conversion
rate exist. The percent of mismanaged waste entering the ocean is highly variable and dependent
on local factors such as weather conditions (e.g., rain storms flushing debris from waterways),
topography and vegetation, and infrastructure that removes or traps mismanaged waste before it
reaches the ocean, such as municipal street sweeping, beach cleaning and stormwater catchment
devices.

To loosely bound the estimate of the mass of plastic waste that enters the ocean we used
municipal water quality data from the San Francisco Bay (California) watershed. In the context
of water quality assessment, litter and “trash” have been identified as contaminants of concern
(26), driving initiatives to quantify capture rates by infrastructure at municipal or county levels.
Total Maximum Daily Loads (TMDLs; the maximum quantity of a pollutant that can enter a
waterway while still allowing the waterway to meet its water quality standards (Section 303(d) of
the Clean Water Act)) are developed for impaired waterways with water quality below
applicable standards. Trash TMDLs have been developed for, or are under development for, 73
waterways in California, the Anacostia River watershed in Washington, DC and Maryland, and
the Duck Creek in Mendenhall Valley, Alaska (27-29). The Trash TMDL, where defined and
typically for trash greater than 5 mm in size, is set at zero.

Baseline and monitoring data were collected in 71 municipalities in the San Francisco Bay
watershed to evaluate the effectiveness of measures designed to meet the zero trash TMDL (note
that no such data exist for the Washington, DC/Maryland and Alaska regions) (29). The baseline
trash loading rate (gallons), the quantity of trash (gallons) collected by street sweeping, storm
drain catchment, and pump station cleaning, and the trash loading rate (gallons/year, defined as
baseline minus the trash collected) were documented from each report. For each municipality,
the percentage of trash that was not collected by street sweeping or catchment was also reported.
The minimum, maximum and mean, computed over 71 municipalities, of the quantity of trash
collected by street sweeping, catchments or pump stations, and the uncaptured residual, are given
in Table S6. From these data an estimated 61% of trash (all materials littered in the watershed),
was uncaptured by street sweeping or catchments, and thus available to enter waterways that
ultimately drain to the Pacific Ocean. In our study, we assumed a more conservative range of
conversion rates (15%, 25%, 40%) of mismanaged plastic waste to marine debris in order to
estimate the mass of plastic that entered the ocean from land-based waste.

Projections from 2010 to 2025

To extend our estimates of the mass of mismanaged plastic waste to the year 2025, we utilized
population projections for each country for 2015, 2020 and 2025 (13). We held 2010 per capita
waste generation rates constant until 2025 when projected rates (given for 128 countries (5) and
using averages by economic category for the remainder) were applied. We projected the
percentage of plastic waste using the method described above, and used a business-as-usual
approach assuming no improvements in waste management infrastructure (i.e., mismanaged
waste fractions were constant). We chose this approach because of the inability to predict future
infrastructure development, and because it provided a framework to examine the effect of
potential mitigation strategies such as a reduction in mismanaged waste through infrastructure
development.

To determine the size of coastal populations, gridded population density raster data was
downloaded for use in ArcMap 10.1® for 2010 and 2015 (30). A 50 km buffer was drawn
around the world’s coastlines, and the gridded population raster data was clipped to this buffer.
This allowed us to calculate a coastal (within 50 km of the coastline) population for each
country. To project the coastal populations forward from 2015 we assumed that the coastal
populations would increase at a rate equal to the total projected population increase for each
country.

Because the fraction of inadequately managed waste and percent plastic in the waste stream were
derived from predictive models, as described above, we used the standard error associated with
these fits to generate error bars on the 2025 projections of the mass of mismanaged waste
available to become marine debris. For each pentad with population growth data we randomly
generated 1000 values of both the mismanagement fraction and the plastic percentage from
normal distributions with the mean and standard deviation defined using the mean and standard
error associated with the respective predictive model. The error bars in Figure 1 describe the
minimum and maximum value (from the 1000 scenarios) of the mass of mismanaged waste for a
particular year.

Supplementary text

Comparison of global plastic input from mismanaged waste to ocean estimates of floating plastic
debris

Cozar et al. (16) estimated the mass of floating plastic debris (7,000 – 35,000 tons) from data
collected using surface-towed plankton nets. Plastic debris collected in these nets is typically
microplastics, 0.33 mm – 5 mm in size. Eriksen et al. (17) reported 35,540 tons of floating
microplastics from plankton net data, and 233,400 tons of “larger plastic items” (> 20 cm in size)
from shipboard visual survey data. Both estimates of the mass of net-collected plastic debris,
and the combined estimate from net plus visual survey data, are orders of magnitude smaller than
our estimate of 4.8 to 12.7 million tonnes (5.3 to 14.0 million tons) of plastic entering the ocean
in 2010 from land-based waste. Our estimate includes all plastic materials (including those that
sink) in all size classes, whereas the published ocean estimates only compute the mass of floating
plastic in a particular size class (or classes). In addition, we estimate the input of plastic waste in
a single year (2010), while the ocean estimates represent an accumulation of floating plastic
debris over an unknown time period (in part because the fragmentation and degradation rates of
plastic in the ocean, and therefore the “age” of debris collected, are unknown).

Future projections

Our results indicate China had the largest mass of mismanaged plastic waste in 2010, similar to
previously reported trends (20,23). By 2025, South Asia (e.g., India) is predicted to have a large
increase in the mass of mismanaged plastic waste. In addition, two African countries (Nigeria
and Senegal) showed large population growth and, therefore, increased mismanaged waste.
Following projected trends through 2100 of large population growth, urbanization and increased
waste generation, the forethought to develop infrastructure to adequately manage waste in
African countries could mitigate increasing future inputs of plastic into the marine environment.
Figure S1: Annual change of percent plastic in municipal solid waste in the United States as a
function of year, 1960 – 2012 (24), illustrating a mean annual increase of 0.19%.
Table S1: Annual and cumulative quantities (millions of metric tons (MMT)) of mismanaged
plastic waste and plastic marine debris (assuming three different conversion rates) for 2010-2025.

Mismanaged 15% marine 25% marine 40% marine

Year plastic waste debris debris debris
[MMT/year] (MMT) (MMT) (MMT)
2010 31.9 4.8 8.0 12.7
2015 36.5 5.5 9.1 14.6
2020 41.3 6.2 10.3 16.5
2025 69.9 10.5 17.5 28.0
Cumulative 618.7 92.8 154.7 247.5
Table S2. Top 20 countries ranked by mass of mismanaged plastic waste in 2010 and 2025, with
percent increase in coastal population from 2010 to 2025. MMT, million metric tons

Year 2010 Year 2025

% pop.
Mismanaged Mismanaged
Country change
Rank plastic waste Country plastic waste
since
[MMT/year] [MMT/year]
2010
1 China 8.82 China 17.81 3.7%
2 Indonesia 3.22 Indonesia 7.42 11.9%
3 Philippines 1.88 Philippines 5.09 26.0%
4 Vietnam 1.83 Vietnam 4.17 13.3%
5 Sri Lanka 1.59 India 2.88 18.7%
6 Thailand 1.03 Nigeria 2.48 45.1%
7 Egypt 0.97 Bangladesh 2.21 18.5%
8 Malaysia 0.94 Thailand 2.18 5.4%
9 Nigeria 0.85 Egypt 1.94 25.0%
10 Bangladesh 0.79 Sri Lanka 1.92 9.0%
11 South Africa 0.63 Malaysia 1.77 23.6%
12 India 0.60 Pakistan 1.22 26.6%
13 Algeria 0.52 Burma 1.15 11.1%
14 Turkey 0.49 Algeria 1.02 18.4%
15 Pakistan 0.48 Brazil 0.95 10.6%
16 Brazil 0.47 South Africa 0.84 7.2%
17 Burma 0.46 Turkey 0.79 16.2%
18 Morocco 0.31 Senegal 0.74 44.3%
19 Korea, North 0.30 Morocco 0.71 14.1%
20 United States 0.28 North Korea 0.61 5.0%
Table S3: The effect of a variety of mitigation strategies on the amount of mismanaged plastic
waste generated and the amount of plastic waste entering the ocean as marine debris in 2025
assuming three different conversion rates. MMT, million metric tons

15% 25% 40%

Mismanaged
marine marine marine
plastic waste
debris debris debris
[MMT/year]
Mitigation strategy Reduction (MMT) (MMT) (MMT)
No intervention 0% 69.1 10.4 17.3 27.7
1 Reduce mismanaged waste by 50% in Top 20 41% 41.0 6.2 10.3 16.4
2 Reduce mismanaged waste by 50% in Top 10 34% 45.7 6.9 11.4 18.3
3 Reduce mismanaged waste by 50% in Top 5 26% 50.9 7.6 12.7 20.4
4 Reduce mismanaged waste by 85% in Top 35 75% 17.4 2.6 4.4 7.0
5 Cap at 1.7 kg/person/day and 11% plastic 26% 51.5 7.7 12.9 20.6
6 Top 10 = 0% combined with Strategy 5 77% 15.9 2.4 4.0 6.4
Table S4: Comparison of model quality, using the Akaike Information Criterion (AIC) scores,
to predict the probability of inadequate waste management.

Rank Models AICs

4 Intercept only 5647.7
3 Intercept + GNI2010 3067.1
5 Intercept + Region 10,403.2
2 Intercept + GNI2010 + Region 2800.1
1 Intercept + GNI2010 + Region1 2344.5
1
Full model with observations weighted for uncertainty
Table S5: Terms and significance for the best fit model for the probability of inadequate waste
management. Coefficients correspond to the response variable on the logit scale.

Term Estimate Std. Error z value Pr(>|z|)

Intercept 1.7400 0.1233 14.1110 < 2e-16
GNI2010 -0.0002 0.0000 -18.1870 < 2e-16
Region EAP 0.3267 0.2885 1.1320 0.2575
Region ECA -1.1300 0.1515 -7.4570 0.0000
Region LCR -1.7130 0.1360 -12.6000 < 2e-16
Region MENA -0.4626 0.1435 -3.2230 0.0013
Region OECD -16.8900 337.5000 -0.0500 0.9601
Regions defined by World Bank: EAP = East Asia and Pacific; ECA = Europe and Central Asia;
LCR = Latin America and the Caribbean; MENA = Middle East and North Africa; OECD = The
Organization for Economic Co-operation and Development.
Table S6: Percentage of trash collected by infrastructure in the San Francisco Bay watershed
(29), and the residual uncollected percentage that is available to enter the ocean as marine debris.
% total trash
% total trash collected in % total trash
n = 71 collected by street stormwater collected in pump % total trash
municipalities sweeping catchments stations uncollected
Minimum 0% 1.2% 0% 36%
Maximum 61% 5.0% 16.5% 95%
Mean 34% 3.2% 1.5% 61%
References and Notes
1. National Research Council (U.S.) Study Panel on Assessing Potential Ocean Pollutants,
Assessing Potential Ocean Pollutants: A Report of the Study Panel on Assessing
Potential Ocean Pollutants to the Ocean Affairs Board, Commission on Natural
Resources, National Research Council (National Academy of Sciences, Washington, DC,
1975).
2. International Maritime Organization, “International Convention for the Prevention of Pollution
from Ships (MARPOL), annex V prevention of pollution by garbage from ships”
(International Maritime Organization, London, 1988);
www.imo.org/Environment/mainframe.asp?topic_id=297.
3. “Plastics – the facts 2013” (PlasticsEurope, Brussels, Belgium, 2013);
www.plasticseurope.org/Document/plastics-the-facts-2013.aspx?FolID=2.
4. “Municipal solid waste generation, recycling, and disposal in the United States: Facts and
figures for 2010” [U.S. Environmental Protection Agency (EPA), Washington, DC,
2011]; www.epa.gov/solidwaste/nonhaz/municipal/pubs/msw_2010_rev_factsheet.pdf
5. D. Hoornweg, P. Bhada-Tata, “What a waste: A global review of solid waste management”
(The World Bank, Washington, DC, 2012);
https://openknowledge.worldbank.org/handle/10986/17388
6. R. C. Thompson, C. J. Moore, F. S. vom Saal, S. H. Swan, Plastics, the environment and
human health: Current consensus and future trends. Philos. Trans. R. Soc. London Ser. B
364, 2153–2166 (2009). Medline
7. D. K. A. Barnes, F. Galgani, R. C. Thompson, M. Barlaz, Accumulation and fragmentation of
plastic debris in global environments. Philos. Trans. R. Soc. London Ser. B 364, 1985–
1998 (2009). Medline doi:10.1098/rstb.2008.0205
8. 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, 315–320
(2014). doi:10.1002/2014EF000240
9. M. C. Goldstein, D. S. Goodwin, Gooseneck barnacles (Lepas spp.) ingest microplastic debris
in the North Pacific Subtropical Gyre. PeerJ 1, e184 (2013). Medline
10. H. S. Carson, M. R. Lamson, D. Nakashima, D. Toloumu, J. Hafner, N. Maximenko, K. J.
McDermid, Tracking the sources and sinks of local marine debris in Hawai’i. Mar.
Environ. Res. 84, 76–83 (2013). Medline doi:10.1016/j.marenvres.2012.12.002
11. A. Lechner H. Keckeis, F. Lumesberger-Loisl, B. Zens, R. Krusch, M. Tritthart, M. Glas, E.
Schludermann, The Danube so colourful: A potpourri of plastic litter outnumbers fish
larvae in Europe’s second largest river. Environ. Pollut. 188, 177–181 (2014). Medline
doi:10.1016/j.envpol.2014.02.006
12. Materials and methods are available as supplementary materials on Science Online.
13. “Probabilistic projections of total population: Median and confidence intervals,” (United
Nations, Department of Economic and Social Affairs, New York, 2012);
http://esa.un.org/unpd/ppp/Data-Output/UN_PPP2010_output-data.htm.
14. K. L. Law, S. Morét-Ferguson, N. A. Maximenko, G. Proskurowski, E. E. Peacock, J.
Hafner, C. M. Reddy, Plastic accumulation in the North Atlantic subtropical gyre.
Science 329, 1185–1188 (2010). Medline doi:10.1126/science.1192321
15. K. L. Law, S. E. Morét-Ferguson, D. S. Goodwin, E. R. Zettler, E. Deforce, T. Kukulka, G.
Proskurowski, Distribution of surface plastic debris in the eastern Pacific Ocean from an
11-year data set. Environ. Sci. Technol. 48, 4732–4738 (2014). Medline
doi:10.1021/es4053076
16. A. Cózar, F. Echevarría, J. I. González-Gordillo, X. Irigoien, B. Ubeda, S. Hernández-León,
A. T. Palma, S. Navarro, J. García-de-Lomas, A. Ruiz, M. L. Fernández-de-Puelles, C.
M. Duarte, Plastic debris in the open ocean. Proc. Natl. Acad. Sci. U.S.A. 111, 10239–
10244 (2014). Medline doi:10.1073/pnas.1314705111
17. M. Eriksen, L. C. Lebreton, H. S. Carson, M. Thiel, C. J. Moore, J. C. Borerro, F. Galgani, P.
G. Ryan, J. Reisser, Plastic pollution in the world’s oceans: More than 5 trillion plastic
pieces weighing over 250,000 tons afloat at sea. PLOS ONE 9, e111913 (2014). Medline
doi:10.1371/journal.pone.0111913
18. “2013 resin review” (American Chemistry Council, Washington, DC, 2013).
19. R. E. Marshall, K. Farahbakhsh, Systems approaches to integrated solid waste management
in developing countries. Waste Manag. 33, 988–1003 (2013). Medline
doi:10.1016/j.wasman.2012.12.023
20. D. Hoornweg, P. Bhada-Tata, C. Kennedy, Environment: Waste production must peak this
century. Nature 502, 615–617 (2013). Medline doi:10.1038/502615a
21. M. Braungart, Upcycle to eliminate waste. Nature 494, 174–175 (2013).
doi:10.1038/494174b
22. T. Lindhqvist, K. Lidgren, in Ministry of the Environment, From the Cradle to the Grave -
Six Studies of the Environmental Impact of Products (Ministry of the Environment,
Stockholm, Sweden, 1990), pp. 7–44.
23. D. Hoornweg, P. Lam, M. Chaudhry, “Waste management in China: Issues and
recommendations” (The World Bank, Washington, DC, 2005).
24. “Municipal solid waste generation, recycling, and disposal in the United States, tables and
figures for 2012” (EPA, Washington, DC, 2014);
www.epa.gov/waste/nonhaz/municipal/pubs/2012_msw_dat_tbls.pdf.
25. Mid-Atlantic Solid Waste Consultants, “2009 national visible litter survey and litter cost
study” (Keep America Beautiful, Washington, DC, 2009).
26. N. Armitage, A. Rooseboom, The removal of urban litter from stormwater conduits and
streams: Paper 1 - The quantities involved and catchment litter management options.
Water S.A. 26, 181–187 (2000).
27. “Total maximum daily load (TMDL) for debris in the waters of Duck Creek in Mendenhall
Valley, Alaska” (EPA, Seattle, 2000);
http://dec.alaska.gov/water/tmdl/pdfs/duckcreekdebris.pdf.
28. “Decision rationale: Total maximum daily loads of trash for the Anacostia River watershed,
Montgomery and Prince George’s Counties, Maryland and the District of Columbia,”
(EPA, Washington, DC, 2010);
www.epa.gov/reg3wapd/pdf/AnacostiaTMDLPortfolio.pdf.
29. “Preliminary baseline trash generation rates for San Francisco Bay Area MS4s, technical
memorandum” (Bay Area Stormwater Management Agencies Association, Oakland, CA,
2012).
30. Gridded Population of the World (GPW), v3 (2014);
http://sedac.ciesin.columbia.edu/data/set/gpw-v3-population-count-future-estimates/data-
download.