View all hidden authors and organizations
Edited by Arizona State University BL Turner, approved on October 6, 2021 (reviewed on June 22, 2021)
Plastic waste is harmful to marine life and has become a major global environmental problem. The recent COVID-19 pandemic has led to an increase in the demand for single-use plastics, exacerbating the pressure on this already out of control problem. This work shows that more than 8 million tons of pandemic-related plastic waste have been generated globally, of which more than 25,000 tons have entered the global ocean. Most of the plastic comes from medical waste generated in hospitals, which dwarfs the contribution of personal protective equipment and online shopping packaging materials. This poses a long-term problem for the marine environment, mainly accumulating on beaches and coastal sediments. We call for better management of medical waste in the epicenter of an epidemic, especially in developing countries.
The COVID-19 pandemic has led to an increase in the demand for single-use plastics, which has exacerbated the pressure of the already out of control global plastic waste problem. Although it is suspected to be large, the scale and fate of this pandemic-related mismanagement of plastic waste is unclear. Here, we use our MITgcm marine plastic model to quantify the impact of the pandemic on plastic emissions. We show that as of August 23, 2021, 193 countries have produced 8.4 ± 1.4 million tons of pandemic-related plastic waste, of which 25.9 ± 380,000 tons were released into the global oceans, accounting for 1.5% of the total global river plastic emissions. ± 0.2%. The model predicts that the spatial distribution of global ocean emissions will change rapidly within 3 years, and a considerable amount of plastic debris will subsequently land on beaches and seabeds, and the Arctic will form a circumpolar plastic accumulation area. We found that hospital waste accounts for most of the global emissions (73%), and most of the global emissions come from Asia (72%), which requires developing countries to better manage medical waste.
Plastic has an excellent strength-to-weight ratio, durability, and low price, making it the material of choice for most disposable medical tools, equipment, and packaging (1, 2). The COVID-19 pandemic proves the indispensable role of plastics in the healthcare sector and public health safety (2). As of August 23, 2021, approximately 212 million people worldwide have been infected with the COVID-19 virus, with the most confirmed cases in the Americas (47.6%) and Asia (31.22%), followed by Europe (17.26%) (3). The surge in hospital admissions and virus testing have greatly increased the amount of plastic medical waste (4). In order to maintain the huge demand for personal protective equipment (PPE, including masks, gloves, and face shields), many single-use plastic (SUP) legislation has been revoked or postponed (2). In addition, blockades, social distancing, and restrictions on public gatherings have increased reliance on online shopping at an unprecedented rate, and packaging materials often contain plastic (5, 6).
Unfortunately, the disposal of plastic waste cannot keep up with the ever-increasing demand for plastic products. Pandemic centers are particularly difficult to dispose of waste (7), and not all used personal protective equipment and packaging materials are processed or recycled (8, 9). This poorly managed plastic waste (MMPW) is then discharged into the environment, and part of it reaches the ocean (10). The released plastic can be transported long distances in the ocean, encountering marine wildlife, and may cause injury or even death (11⇓ ⇓ –14). For example, a recent report estimated that 1.56 million masks will enter the ocean in 2020 (15). Early research also raised the potential problems of COVID-19 plastic pollution and its impact on marine life (16⇓ –18). Some cases of marine life entanglement, capture and ingestion of COVID-19 waste, and even deaths have been reported (19, 20). Plastic fragments may also promote species invasion and the migration of pollutants, including the COVID-19 virus (21⇓ –23). Despite the potential impact, the total amount of plastic waste associated with the pandemic and its impact on the environment and health are largely unknown. Here, we estimate the amount of excess plastic released into the global oceans during the pandemic and its long-term fate and potential ecological risks.
As of August 23, 2021, the total amount of excess MMPW generated during the pandemic is calculated to be between 4.4 and 15.1 million tons (Figure 1). We use the average of scenarios with different assumptions as our best estimate (method), which is approximately 8.4 ± 1.4 million tons. The main part (87.4%) of this excess waste comes from hospitals, which is estimated based on the number of COVID-19 hospitalized patients in each country (24) and the amount of medical waste generated by each patient (25). Personal use of PPE only accounts for 7.6% of the total excess waste. Interestingly, we found that the surge in online shopping has led to an increase in demand for packaging materials. However, we found that packaging and test kits are secondary sources of plastic waste, accounting for only 4.7% and 0.3%, respectively.
Due to the COVID-19 pandemic, poorly managed plastics from different sources (hospital medical waste, test kits, PPE, and online packaging) have been generated globally. Each source considers high-yield and low-yield scenarios (methods).
Table 1 shows the distribution of COVID-19 cases in different continents (Asia, Europe, North America, South America, Oceania and Africa). Approximately 70% of COVID-19 cases occur in North America, South America, and Asia (Table 1). We found that the generation of MMPW does not follow the case distribution because most MMPW is produced in Asia (46%), followed by Europe (24%), and finally North and South America (22%) (Table 1 and Figure 2E)). This reflects that the level of medical waste treatment in many developing countries such as India, Brazil and China (low-end and high-end estimated to be between 11.5% and 76%) is lower than that of developed countries with more cases in North America and Europe (such as the United States and the United States). Spain) (0 to 5%) (Figure 2A). Due to the large population wearing masks, MMPW produced by individual PPE is more likely to be in Asia (Figure 2C and SI appendix, Table S1) (26). Similarly, the MMPW generated by online shopping packaging is the highest in Asia (Figure 2D). For example, the top three countries in the global share of the express delivery industry are China (58%), the United States (14.9%) and Japan (10.3%), followed by the United Kingdom (4%) and Germany (4%) (27).
Percentage of confirmed COVID-19 cases (as of August 23, 2021), pandemic-related MMPW that eventually enters the environment, and pandemic-related MMPW transported to estuaries on different continents
Poorly managed plastics associated with the pandemic are accumulated in rivers and discharged into the global ocean. The panel is used for (A) hospital medical waste, (B) COVID-19 virus detection kit, (C) PPE, (D) online shopping packaging materials and (E) emissions caused by the sum of them. The background color represents the MMPW generated by each watershed, and the size of the blue circle represents the discharge of the estuary.
Based on each country’s MMPW production and hydrological models (28), we calculated the total pandemic-related emissions to be 25.9 ± 3.8 (12.3 for microplastics [< 5 mm], 13.6 for large plastics [> 5 mm]) Plastic flows into the global oceans from 369 major rivers and their basins (Figure 2E). We believe that the 369 rivers considered here (accounting for 91% of the plastic emissions from global rivers to the ocean) include most of the plastic emissions associated with the global pandemic. The top three rivers in terms of plastic waste emissions related to the pandemic are the Arabian River (5200 tons in Asia), the Indus River (4000 tons in Asia) and the Yangtze River (3700 tons in Asia), followed by the Yarlung Zangbo River Ganges (2400 tons in Asia), The Danube (1,700 tons, Europe) and the Amur (1,200 tons, Asia). These findings highlight hot rivers and watersheds that require special attention in plastic waste management.
Overall, the top 10 rivers accounted for 79% of pandemic plastic emissions, the top 20 accounted for 91%, and the top 100 accounted for 99%. About 73% of the flow comes from Asian rivers, followed by Europe (11%), and other continents have a smaller contribution (Table 1). This model is different from the production model of MMPW (Table 1), because the ability of rivers to export plastic loads to the ocean is different, which is measured by the yield ratio (defined as the ratio of plastic emissions from the estuary) and the total amount of MMPW produced in the basin). The yield ratio is affected by factors such as the distribution of plastic released along the river and the physical conditions of the river (for example, water runoff and velocity) (28). The top five rivers with the highest yields are the Yangtze River (0.9%), the Indus (0.5%), the Yellow River (0.5%), the Nile (0.4%) and the Ganges (0.4%). These rivers either have high population density, large runoff, and fast water flow near their mouths, or a combination of both. The combination of high pandemic-related MMPW generations and production ratios in Asian rivers has led to their discharge of high MMPW into the ocean.
We simulated the transportation and fate of 25,900 ± 3,800 tons of pandemic-related plastic waste through the Nanjing University MITgcm-Plastic model (NJU-MP) to assess its impact on the marine environment. This model takes into account the main processes that plastics undergo in seawater: stranding, drifting, settling, biofouling/decontamination, abrasion and crushing (29). The model shows that most of the plastic discharged by rivers is transferred from the surface ocean to the beach and seabed within 3 years (Figure 3). By the end of 2021, the mass fraction of plastics in sea water, ocean floor, and beaches will be modeled as 13%, 16%, and 71%, respectively. Approximately 3.8% of plastics are present in the surface ocean, and the global average concentration is 9.1 kg/km2. Our model also shows that the released pandemic-related plastics are mainly distributed in marine areas relatively close to their sources, such as low- and mid-latitude rivers in East and South Asia, South Africa and the Caribbean (Figure 4) and SI Appendix, Figure S2) . The sediment flux is mainly distributed near the main estuary (Figure 4 and SI appendix, Figure S2). This shows that the short-term effects of plastics associated with the pandemic are fairly limited to the coastal environment.
Prediction of the fate of pandemic-related plastics (including microplastics and macroplastics) discharged into the global oceans. (A) The mass fraction and average concentration of the surface ocean. (B) The mass scores of sea water, seabed and beaches.
The mass concentration model spatial distribution of COVID-19-related plastics on the ocean surface (AC, JL), beach (DF, MO), and seabed (GI, P-) will be in 2021, 2025, and 2100, respectively. The black boxes on the top panel represent the five subtropical ocean circulations (North Pacific circulation, North Atlantic circulation, South Pacific circulation, South Atlantic circulation, and Indian circulation). Panel AI is used for microplastics, while JR is used for large plastics.
The model suggests that the impact may extend to the high seas in 3 to 4 years. It is expected that the mass fraction of plastic in seawater will decrease in the future, while the mass fraction of plastic in the seabed and beaches will gradually increase. By the end of 2022, the proportion of MMPW discharged by rivers and related to the pandemic in seawater, seabed and beaches was modeled as 5%, 19%, and 76%, respectively, and the average surface ocean concentration dropped sharply to 3.1kg/km2. In 2025, 5 garbage patches in the center of the subtropical circulation merge, including 4 garbage patches in the North Atlantic, South Atlantic and Pacific Oceans and 1 garbage patch in the Indian Ocean (Figure 4 and SI appendix, Figure S2). The deposition flux hotspots of the North Atlantic and Arctic Oceans at high latitudes in 2025 were also simulated (Figure 4 and SI Appendix, Figure S2), reflecting the large-scale vertical movement of seawater (SI Appendix, Figure 2). S3).
We have discovered the long-term effects of pandemic-related waste releases in the global oceans. At the end of the century, the model indicates that almost all plastics associated with the pandemic will eventually enter the seabed (28.8%) or beaches (70.5%), which may damage the benthic ecosystem. It is estimated that by 2100, the global average ocean surface plastic concentration associated with the pandemic will drop to 0.3 kg/km2, accounting for 0.03% of total plastic emissions. However, two garbage dumps were still simulated in the northeast Pacific and southeast Indian Ocean, which brought continuous risks to the ecosystem there. The fate of microplastics is similar to that of large plastics, but due to the lower mobility of large plastics, they end up at a higher proportion of the beach (Figure 4 and SI appendix, Figure S1).
Due to the northern branch of the thermohaline circulation, the Arctic Ocean seems to be a dead end for the transportation of plastic debris (30). Approximately 80% of the plastic fragments discharged into the Arctic Ocean will sink rapidly, and by 2025 will form a circumpolar plastic accumulation zone. This year, the Arctic seabed accounts for 13% of the global plastic deposition flux, but this proportion will increase to 17% by 2100. Due to the harsh environment and high sensitivity to climate change, Arctic ecosystems are considered particularly vulnerable (31, 32), which makes the potential ecological impact of exposure to the expected accumulation of special Arctic plastics worrying.
It is speculated that the epidemic will not be fully controlled within a few years, and many containment policies will continue to be implemented (33). By the end of 2021, it is conservatively estimated that the number of confirmed cases will reach 280 million (34). The total amount of pandemic-related MMPW generated will reach 11 million tons, resulting in the discharge of 34,000 tons of global rivers into the ocean. Due to the record number of confirmed cases in India, the production and emissions of MMPW are expected to be more biased towards Asia (3). Given the linear relationship between emissions and the quality of ocean plastics, the fate and transportation of newly generated plastic emissions can be inferred from our current results.
Due to the lack of accurate data (for example, the number of used masks and online shopping packages and the proportion of poorly managed waste in the case of overcapacity), our estimates of pandemic-related MMPW releases are highly uncertain. For example, our estimates of emissions from mask use are much lower than those of Chowdhury et al. (35), suppose that a person uses a mask every day, and we assume that a mask lasts for 6 days based on the survey data (methods). Therefore, we consider multiple scenarios to limit the actual situation (methods). The estimated MMPW change as hospital medical waste is ±53%, while the estimated MMPW change from packaging and PPE is ±25% and approximately 3.5 times, respectively. The estimates of river MMPW emissions to the ocean are also uncertain because they are based on coarse-resolution (ie, watershed) hydrological models (28). In addition, factors such as the crushing, abrasion and stranding rate of the plastic in NJU-MP also have a significant impact on the simulation results (29). Despite these uncertainties, the spatial pattern of releases related to the pandemic and their relative fate in different regions of the ocean are more robust.
Marine plastic emissions related to the pandemic accounted for 1.5 ± 0.2% of total river plastic emissions (28, 36). A large part of the emissions is medical waste, which will also increase potential ecological and health risks (37) and even the spread of the COVID-19 virus (38). This provides a lesson in the structural changes required for waste management. Lifting or postponing the ban on SUP after a pandemic may complicate plastic waste control. There is a need to increase the global public's awareness of the environmental impact of PPE and other plastic products. Innovative technologies need to be promoted to better collect, sort, process and recycle plastic waste, and to develop more environmentally friendly materials (15, 39). In the epicenter, especially in developing countries, better management of medical waste is necessary.
We have developed an inventory of excess plastic waste generated by the COVID-19 pandemic. We considered four types of sources: medical waste generated by hospitals, virus detection kits, PPE used by residents, and online shopping packages.
For the medical waste generated in the hospital, we estimate the amount based on the number of hospitalized patients (nH) and the medical waste generation rate (HCWGR) per patient. nH is estimated based on the number of COVID-19 infections (nI) and the global average hospitalization rate (HR) of the disease: nH=nI×HR. [1]
The nI and HR data are based on statistics from the World Health Organization (3). The HCWGR of COVID-19 patients is approximately twice that of ordinary patients (40), which is calculated based on the life expectancy (LE) and carbon dioxide emissions (CDE) of Minoglou et al. (25): HCWGR=2×(0.014LE 0.31CDE). [2]
This relationship is developed based on statistical data from 42 countries around the world, and can explain 85% of the variability of HWGR data (25). LE data comes from Roser et al. (41), CDE data comes from Worldometer (42).
The medical waste generated by the virus detection kit is estimated based on the number of tests performed and the amount of waste generated per test. The previous data comes from Ritchie et al. (43) The latter comes from Cheon (44) and ShineGene (45). According to the specifications of the test kit, the waste generated in each test ranges from 21 to 28 grams per test.
For personal protective equipment used by residents, we only consider masks, because other items such as gloves and face masks are not commonly used. We use two methods to estimate the number of masks used: consumption-based and production-based. For the former method, we first assume that an ideal condition is that everyone uses a new mask every 6 days (46). We assume that the actual mask usage rate is 25% to 75% in this case. The population data comes from the United Nations (26). For the latter method, we assume that all masks produced are used up. Global production (PW) is estimated based on the production of masks in China (PC), which is the world's largest producer of masks (54% to 72%) (47): PW=PC÷p, [3] where p is China The share of masks produced (47). We also considered two scenarios for the quality of waste generated by each mask (surgical mask or N95 mask).
The online shopping package (np) in this study refers to the extra part caused by lifestyle changes during the pandemic compared with the normal situation (no COVID-19 pandemic) (nno-covid): np=nactual – nno-covid, [ 4] Where nactual is the actual online package usage from 2020 to the first quarter of 2021, estimated based on the financial reports of the world's top six e-commerce companies (Taobao, Tmall, Amazon, JD, eBay, and Walmart)) (48 ⇓ ⇓ ⇓ –52). nno-covid is calculated based on the number of packages in 2019 and the average annual growth rate in recent years (53). Then estimate the mass (m) of plastic waste generated based on the average mass of plastic in the packaging material (mp) (54): m=np×mp. [5]
The amount of MMPW (i) of each source can be calculated based on the waste generation rate (Rw) of the above four sources (Rw), the proportion of plastic waste to the total waste (Pp), and the proportion of poorly managed waste to the total waste (Pm): MMPW=∑i=14Rwi×Ppi×Pmi. [6]
We believe that the first two source categories are medical waste, while the latter two source categories are municipal waste. The Pm of each country is specified according to the type of waste. The Pm of municipal waste is based on Schmidt et al. (28). There is no reliable data on the Pm of medical waste. We use the data of Caniato et al. (55) As a function of economic conditions (56) and the level of waste treatment and disposal in individual countries. The data set includes two scenarios, and we considered an additional scenario that is 50% lower than the lower scenario to resolve the uncertainty in this part.
We estimated the river discharge of MMPW into the ocean related to the pandemic based on the basin model developed by Schmidt et al. (28) Calculated the ratio of the plastic emissions from the estuary to the total MMPW produced in the entire corresponding watershed. We assume that the ratio of plastic waste and other waste related to the pandemic is the same. In this study, we considered a total of 369 major rivers and their basins. We split country-specific pandemic-related MMPW data into each watershed based on the number of regular MMPWs (28).
The resolution of the NJU-MP model is 2° latitude × 2.5° longitude, with 22 vertical levels, and is driven by ocean physics from the integrated global system model, with a time step of 4 hours (29). The model considers five types of plastics with different chemical compositions, and pre-determines the density of each type: polyethylene (PE, 950 kg/m-3), polypropylene (PP, 900 kg/m-3), polychloride Ethylene (PVC, 1,410 kg/m-3), polyurethane (PU, 550 kg/m-3) and others (1,050 kg/m-3). The density of plastic is modeled as increasing during biological contamination, but decreasing during decontamination (57). When low-density polymers float, their density determines their buoyancy, while high-density polymers sink into the sediments (58, 59). Each category has six size classifications: four belong to microplastics: <0.0781 mm, 0.0781 to 0.3125 mm, 0.3125 to 1.25 mm, and 1.25 to 5 mm, and two belong to large plastics: 5 to 50 mm and >50 mm. So there are a total of 60 plastic tracers in the model. For simplicity, we assume that all plastic fragments are spheres. For large plastics, half of the pandemic-related MMPW emissions from rivers are 5 to 10 mm, and half are> 50 mm, while microplastics are the largest bins (ie 1.25 to 5 mm). After being discharged into the ocean, the plastic is cleared by the beach interception (57) and sinks into the deeper ocean and eventually sinks to the seafloor. Biofouling of lightweight plastic types (PE and PP) was modeled according to the method of Kooi et al. (60) But adjusted for more realistic scenarios. Three plastics with different degrees of biological attachment are considered. In addition, the model also considers removal processes including UV degradation, chipping, and abrasion.
MMPW generation and river discharge data for all countries are provided in the Environmental Biogeochemical Modeling Group (EBMG), https://www.ebmg.online/plastics (61). All research data is included in the article and/or SI appendix.
This research was awarded by the National Natural Science Foundation of China (42177349, 41875148), the Special Fund for Fundamental Scientific Research Expenses of Central Universities (0207-14380168), the Frontier Science Center of Geomaterials Cycle, Jiangsu Innovation and Entrepreneurship Talent Program, and Jiangsu Climate Change Collaborative Innovation Center. Thanks to the High Performance Computing Center of Nanjing University for the numerical calculation of its blade cluster system.
↵1Y.P. and PW made the same contribution to this work.
Author contributions: ATS and YZ design research; YP and PW conducted research; YP and PW analysis data; YP, ATS and YZ wrote this paper.
The author declares no competing interests.
This article is directly contributed by PNAS.
This article contains online support information https://www.pnas.org/lookup/suppl/doi:10.1073/pnas.2111530118/-/DCSupplemental.
This open access article is distributed under the Creative Commons Attribution-Non-Commercial-No Derivative License 4.0 (CC BY-NC-ND).
Thank you for your interest in advertising on PNAS.
Note: We only ask you to provide your email address so that the people you recommend the page to know that you want them to see it, and that it is not spam. We do not capture any email addresses.
Feedback privacy/legal
Copyright © 2021 National Academy of Sciences. Online ISSN 1091-6490. PNAS is a partner of CHORUS, COPE, CrossRef, ORCID and Research4Life.