The downstream air pollution impacts of the transition from coal to natural gas in the United States

Abstract

The recent shift in the United States from coal to natural gas as a primary feedstock for the production of electric power has reduced the intensity of sectoral carbon dioxide emissions, but—due to gaps in monitoring—its downstream pollution-related effects have been less well understood. Here, I analyse old units that have been taken offline and new units that have come online to empirically link technology switches to observed aerosol and ozone changes and subsequent impacts on human health, crop yields and regional climate. Between 2005 and 2016 in the continental United States, decommissioning of a coal-fired unit was associated with reduced nearby pollution concentrations and subsequent reductions in mortality and increases in crop yield. In total during this period, the shutdown of coal-fired units saved an estimated 26,610 (5%–95% confidence intervals (CI), 2,725–49,680) lives and 570 million (249–878 million) bushels of corn, soybeans and wheat in their immediate vicinities; these estimates increase when pollution transport-related spillovers are included. Changes in primary and secondary aerosol burdens also altered regional atmospheric reflectivity, raising the average top of atmosphere instantaneous radiative forcing by 0.50 W m−2. Although there are considerable benefits of decommissioning older coal-fired units, the newer natural gas and coal-fired units that have supplanted them are not entirely benign.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Fig. 1: Locations of AMPD reporting fossil-fuel-burning electric power plants in the United States, by primary feedstock.
Fig. 2: The impacts of old and new coal- and natural-gas-fired units on ambient PM2.5 concentrations, as measured by a combination of satellite- and ground-based measurements.
Fig. 3: Changes in mortality rates and crop yields at the location and county level.
Fig. 4: Regional RF changes due to electric power sector changes, 2005–2016.

Data availability

All data used in these analyses are publicly available, as described above. Processed, compiled datasets to replicate these analyses are available at https://doi.org/10.7910/DVN/RIZQUN.

Code availability

Code to generate compiled data and to replicate all of the analyses here (results, figures, tables) is available at https://github.com/jaburney/naturalgastransition.

References

  1. 1.

    Air Markets Program Data (EPA, accessed 13 April 2018); https://ampd.epa.gov/ampd/

  2. 2.

    Schrag, D. P. Is shale gas good for climate change? Daedalus 141, 72–80 (2012).

  3. 3.

    Hughes, J. D. Energy: a reality check on the shale revolution. Nature 494, 307–308 (2013).

  4. 4.

    de Gouw, J. A., Parrish, D. D., Frost, G. J. & Trainer, M. Reduced emissions of CO2, NOx, and SO2 from U.S. power plants owing to switch from coal to natural gas with combined cycle technology. Earth’s Future 2, 75–82 (2014).

  5. 5.

    Rogelj, J. et al. Disentangling the effects of CO2 and short-lived climate forcer mitigation. Proc. Natl Acad. Sci. USA 111, 16325–16330 (2014).

  6. 6.

    Integrated Assessment of Black Carbon and Tropospheric Ozone: Summary for Decision Makers (UNEP and WMO, 2011).

  7. 7.

    Proctor, J., Hsiang, S., Burney, J., Burke, M. & Schlenker, W. Estimating global agricultural effects of geoengineering using volcanic eruptions. Nature 560, 480–483 (2018).

  8. 8.

    Shindell, D. T., Faluvegi, G., Rotstayn, L. & Milly, G. Spatial patterns of radiative forcing and surface temperature response. J. Geophys. Res. Atmos. 120, 5385–5403 (2015).

  9. 9.

    Persad, G. G. & Caldeira, K. Divergent global-scale temperature effects from identical aerosols emitted in different regions. Nat. Commun. 9, 3289 (2018).

  10. 10.

    2014 National Emissions Inventory Report (EPA, 2014); https://gispub.epa.gov/neireport/2014/

  11. 11.

    Clearinghouse for Inventories and Emissions Factors (CHIEF) (EPA, 2018); https://www.epa.gov/chief

  12. 12.

    Air Quality System (AQS) (EPA, 2019); https://www.epa.gov/aqs

  13. 13.

    Bell, M. L., Ebisu, K., Peng, R. D., Samet, J. M. & Dominici, F. Hospital admissions and chemical composition of fine particle air pollution. Am. J. Respir. Crit. Care Med. 179, 1115–1120 (2009).

  14. 14.

    Pope, C. A. III et al. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. JAMA 287, 1132–1141 (2002).

  15. 15.

    Thurston, G. D. et al. Ambient particulate matter air pollution exposure and mortality in the NIH-AARP diet and health cohort. Environ. Health Perspect. 124, 484–490 (2015).

  16. 16.

    Long, S. P., Ainsworth, E. A., Leakey, A. D. & Morgan, P. B. Global food insecurity. Treatment of major food crops with elevated carbon dioxide or ozone under large-scale fully open-air conditions suggests recent models may have overestimated future yields. Proc. R. Soc. B 360, 2011–2020 (2005).

  17. 17.

    Feng, Z. & Kobayashi, K. Assessing the impacts of current and future concentrations of surface ozone on crop yield with meta-analysis. Atmos. Environ. 43, 1510–1519 (2009).

  18. 18.

    McGrath, J. M. et al. An analysis of ozone damage to historical maize and soybean yields in the United States. Proc. Natl Acad. Sci. USA 112, 14390–14395 (2015).

  19. 19.

    Van Dingenen, R. et al. The global impact of ozone on agricultural crop yields under current and future air quality legislation. Atmos. Environ. 43, 604–618 (2009).

  20. 20.

    Cohen, A. J. et al. Estimates and 25-year trends of the global burden of disease attributable to ambient air pollution: an analysis of data from the global burden of diseases study 2015. Lancet 389, 1907–1918 (2017).

  21. 21.

    Shindell, D. et al. Simultaneously mitigating near-term climate change and improving human health and food security. Science 335, 183–189 (2012).

  22. 22.

    Lelieveld, J., Evans, J. S., Fnais, M., Giannadaki, D. & Pozzer, A. The contribution of outdoor air pollution sources to premature mortality on a global scale. Nature 525, 367–371 (2015).

  23. 23.

    Burnett, R. et al. Global estimates of mortality associated with long-term exposure to outdoor fine particulate matter. Proc. Natl Acad. Sci. USA 115, 9592–9597 (2018).

  24. 24.

    Lelieveld, J. et al. Effects of fossil fuel and total anthropogenic emission removal on public health and climate. Proc. Natl Acad. Sci. USA 116, 7192–7197 (2019).

  25. 25.

    Driscoll, C. T. et al. US power plant carbon standards and clean air and health co-benefits. Nat. Clim. Change 5, 535–540 (2015).

  26. 26.

    Buonocore, J. J. et al. Health and climate benefits of different energy-efficiency and renewable energy choices. Nat. Clim. Change 6, 100–105 (2016).

  27. 27.

    Prehoda, E. W. & Pearce, J. M. Potential lives saved by replacing coal with solar photovoltaic electricity production in the US. Renew. Sustain. Energy Rev. 80, 710–715 (2017).

  28. 28.

    Millstein, D., Wiser, R., Bolinger, M. & Barbose, G. The climate and air-quality benefits of wind and solar power in the United States. Nat. Energy 2, 17134 (2017).

  29. 29.

    Abel, D. W. et al. Air-quality-related health impacts from climate change and from adaptation of cooling demand for buildings in the eastern United States: an interdisciplinary modeling study. PLoS Med. 15, e1002599 (2018).

  30. 30.

    Abel, D. W. et al. Air quality-related health benefits of energy efficiency in the United States. Environ. Sci. Technol. 53, 3987–3998 (2019).

  31. 31.

    Currie, J., Davis, L., Greenstone, M. & Walker, R. Environmental health risks and housing values: evidence from 1,600 toxic plant openings and closings. Am. Econ. Rev. 105, 678–709 (2015).

  32. 32.

    Luechinger, S. Air pollution and infant mortality: a natural experiment from power plant desulfurization. J. Health Econ. 37, 219–231 (2014).

  33. 33.

    Yang, M. & Chou, S.-Y. The impact of environmental regulation on fetal health: evidence from the shutdown of a coal-fired power plant located upwind of New Jersey. J. Environ. Econ. Manage. 90, 269–293 (2018).

  34. 34.

    Severnini, E. Impacts of nuclear plant shutdown on coal-fired power generation and infant health in the Tennessee valley in the 1980s. Nat. Energy 2, 17051 (2017).

  35. 35.

    Casey, J. A. et al. Increase in fertility following coal and oil power plant retirements in California. Environ. Health 17, 44 (2018).

  36. 36.

    Casey, J. A. et al. Retirements of coal and oil power plants in California: association with reduced preterm birth among populations nearby. Am. J. Epidemiol. 187, 1586–1594 (2018).

  37. 37.

    QuickStats database (USDA, accessed 30 January 2018); https://www.nass.usda.gov/Quick_Stats/

  38. 38.

    Myhre, G. et al. in Climate Change 2013: The Physical Science Basis (eds Stocker, T. F. et al.) Ch. 8 (IPCC, Cambridge Univ. Press, 2013).

  39. 39.

    Haywood, J. & Shine, K. The effect of anthropogenic sulfate and soot aerosol on the clear sky planetary radiation budget. Geophys. Res. Lett. 22, 603–606 (1995).

  40. 40.

    Charlson, R. J. et al. Climate forcing by anthropogenic aerosols. Science 255, 423–430 (1992).

  41. 41.

    Chylek, P. & Wong, J. Effect of absorbing aerosols on global radiation budget. Geophys. Res. Lett. 22, 929–931 (1995).

  42. 42.

    Chay, K. Y. & Greenstone, M. The impact of air pollution on infant mortality: evidence from geographic variation in pollution shocks induced by a recession. Q. J. Econ. 118, 1121–1167 (2003).

  43. 43.

    Eftim, S. E., Samet, J. M., Janes, H., McDermott, A. & Dominici, F. Fine particulate matter and mortality: a comparison of the six cities and American cancer society cohorts with a medicare cohort. Epidemiology 19, 209–216 (2008).

  44. 44.

    Zeger, S. L., Dominici, F., McDermott, A. & Samet, J. M. Mortality in the medicare population and chronic exposure to fine particulate air pollution in urban centers (2000-2005). Environ. Health Perspect. 116, 1614–1619 (2008).

  45. 45.

    Knittel, C. R., Miller, D. L. & Sanders, N. J. Caution, drivers! children present: traffic, pollution, and infant health. Rev. Econ. Stat. 98, 350–366 (2016).

  46. 46.

    Bell, M. L., McDermott, A., Zeger, S. L., Samet, J. M. & Dominici, F. Ozone and short-term mortality in 95 US urban communities, 1987-2000. JAMA 292, 2372–2378 (2004).

  47. 47.

    Burney, J. & Ramanathan, V. Recent climate and air pollution impacts on Indian agriculture. Proc. Natl Acad. Sci. USA 111, 16319–16324 (2014).

  48. 48.

    Metaxoglou, K. & Smith, A. Productivity spillovers from pollution reduction: reducing coal use increases crop yields. Am. J. Agric. Econ. https://doi.org/10.1093/ajae/aaz044 (2019).

  49. 49.

    Brauer, M. et al. Ambient air pollution exposure estimation for the global burden of disease 2013. Environ. Sci. Technol. 50, 79–88 (2015).

  50. 50.

    Apte, J. S., Marshall, J. D., Cohen, A. J. & Brauer, M. Addressing global mortality from ambient PM2.5. Environ. Sci. Technol. 49, 8057–8066 (2015).

  51. 51.

    Vodonos, A., Awad, Y. A. & Schwartz, J. The concentration-response between long-term PM2.5 exposure and mortality; a meta-regression approach. Environ. Res. 166, 677–689 (2018).

  52. 52.

    Auffhammer, M., Ramanathan, V. & Vincent, J. R. Integrated model shows that atmospheric brown clouds and greenhouse gases have reduced rice harvests in India. Proc. Natl Acad. Sci. USA 103, 19668–19672 (2006).

  53. 53.

    Rollins, A. W. et al. Evidence for NOx control over nighttime SOA formation. Science 337, 1210–1212 (2012).

  54. 54.

    Zheng, B. et al. Heterogeneous chemistry: a mechanism missing in current models to explain secondary inorganic aerosol formation during the January 2013 haze episode in north China. Atmos. Chem. Phys. 15, 2031–2049 (2015).

  55. 55.

    Levy, J. I., Diez, D., Dou, Y., Barr, C. D. & Dominici, F. A meta-analysis and multisite time-series analysis of the differential toxicity of major fine particulate matter constituents. Am. J. Epidemiol. 175, 1091–1099 (2012).

  56. 56.

    Thurston, G. D. et al. Ischemic heart disease mortality and long-term exposure to source-related components of US fine particle air pollution. Environ. Health Perspect. 124, 785–794 (2015).

  57. 57.

    Bell, M. L. & Ebisu, K. Environmental inequality in exposures to airborne particulate matter components in the United States. Environ. Health Perspect. 120, 1699–1704 (2012).

  58. 58.

    Hsu, A., Reuben, A., Shindell, D., de Sherbinin, A. & Levy, M. Toward the next generation of air quality monitoring indicators. Atmos. Environ. 80, 561–570 (2013).

  59. 59.

    Ryerson, T. et al. Observations of ozone formation in power plant plumes and implications for ozone control strategies. Science 292, 719–723 (2001).

  60. 60.

    Shindell, D. & Faluvegi, G. The net climate impact of coal-fired power plant emissions. Atmos. Chem. Phys. 10, 3247–3260 (2010).

  61. 61.

    Burnham, A. et al. Life-cycle greenhouse gas emissions of shale gas, natural gas, coal, and petroleum. Environ. Sci. Technol. 46, 619–627 (2011).

  62. 62.

    Berntsen, T., Fuglestvedt, J., Myhre, G., Stordal, F. & Berglen, T. F. Abatement of greenhouse gases: does location matter? Climatic Change 74, 377–411 (2006).

  63. 63.

    Shindell, D. T. The social cost of atmospheric release. Climatic Change 130, 313–326 (2015).

  64. 64.

    Aakre, S., Kallbekken, S., Van Dingenen, R. & Victor, D. G. Incentives for small clubs of arctic countries to limit black carbon and methane emissions. Nat. Clim. Change 8, 85–90 (2018).

  65. 65.

    van Donkelaar, A. et al. Global estimates of fine particulate matter using a combined geophysical-statistical method with information from satellites, models, and monitors. Environ. Sci. Technol. 50, 3762–3772 (2016).

  66. 66.

    Air Quality Reports Input Data (EPA, accessed 9 September 2019); https://www.epa.gov/hesc/rsig-related-downloadable-data-files

  67. 67.

    Bhartia, P. K. OMI/Aura Ozone (O3) Total Column Daily L2 Global Gridded 0.25 degree x 0.25 degree V3 (Goddard Earth Sciences Data and Information Services Center, 2012); https://doi.org/10.5067/Aura/OMI/DATA2025

  68. 68.

    Krotkov, N. A., Li, C. & Leonard, P. OOMI/Aura Sulphur Dioxide (SO2) Total Column Daily L2 Global Gridded 0.125 degree x 0.125 degree V3 (Goddard Earth Sciences Data and Information Services Center, 2014); https://doi.org/10.5067/Aura/OMI/DATA2023

  69. 69.

    Boersma, K. F. et al. An improved retrieval of tropospheric NO2 columns from the Ozone Monitoring Instrument. Atmos. Meas. Tech. 4, 1905–1928 (2011).

  70. 70.

    Torres, O. O. OMAERUVd: OMI/Aura Near UV Aerosol Optical Depth and Single Scattering Albedo L3 1 day 1.0 degree x 1.0 degree V3 (NASA Goddard Space Flight Center, Goddard Earth Sciences Data and Information Services Center, 2008); https://doi.org/10.5067/Aura/OMI/DATA3003

  71. 71.

    Platnick, S. et al. MODIS Atmosphere L3 Monthly Product (NASA MODIS Adaptive Processing System, Goddard Space Flight Center, 2015); https://doi.org/10.5067/MODIS/MOD08_M3.006

  72. 72.

    AIRS Science Team/Teixeira, J. AIRS/Aqua L3 Monthly Standard Physical Retrieval (AIRS-only) 1 degree x 1 degree V006 (Goddard Earth Sciences Data and Information Services Center, 2013); https://doi.org/10.5067/Aqua/AIRS/DATA321

  73. 73.

    M2TMNXRAD: MERRA-2 tavgM_2d_rad_Nx: 2d,Monthly mean,Time-Averaged,Single-Level,Assimilation,Radiation Diagnostics V5.12.4 (GMAO, 2015); https://doi.org/10.5067/OU3HJDS973O0

  74. 74.

    Compressed Mortality File (US Centers for Disease Control (CDC) National Vital Statistics Service 2005–2016, accessed 1 February 2018); https://wonder.cdc.gov/controller/datarequest/D140

  75. 75.

    Coakley, G. & Chylek, P. The two-stream approximation in radiative transfer: including the angle of the incident radiation. J. Atmos. Sci. 32, 409–418 (1975).

  76. 76.

    Russell, P., Kinne, S. & Bergstrom, R. Aerosol climate effects: local radiative forcing and column closure experiments. J. Geophys. Res. Atmos. 102, 9397–9407 (1997).

  77. 77.

    Delene, D. J. & Ogren, J. A. Variability of aerosol optical properties at four North American surface monitoring sites. J. Atmos. Sci. 59, 1135–1150 (2002).

Download references

Acknowledgements

I thank the Policy Design and Evaluation Lab and Big Pixel Initiative (both at UC San Diego), and the National Science Foundation (CNH Award no. 1715557) for funding.

Author information

Correspondence to Jennifer A. Burney.

Ethics declarations

Competing interests

The author declares no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 Temporal and geographic distribution of old coal units taken offline, new natural gas units brought online, and new coal units brought online in the United States, 2005-2016.

These technological changes (closing and opening of new and old electric power generation units) occur at discrete moments in time, resulting in changes in emissions fluxes into the surrounding area. These changes are assumed to be as-if random in space and time, vis-à-vis each other, and discontinuities in ambient conditions are estimated across the sample for these natural experiments (see Methods).

Extended Data Fig. 2 Emissions of CO2, SO2, and NOx associated with shutdown of old coal-fired units, and start-up of new natural gas-fired and coal-fired units.

(Left Column) Coal-fired units approaching decommissioning are often ‘ramped down’ prior to shutdown, as reflected in decreasing gross load and emissions. (Centre and Right Columns) Conversely, as they ramp up after commissioning, new units may take some time to settle into steady-state. Further downstream impacts of a coal unit shut-down are thus likely to begin to manifest in the year prior to final closure, and impacts of new units may change over time. Boxes show the 25th-75th percentiles, with the median indicated by the bar, with whiskers indicating the 2.5 to 97.5 percentile confidence interval; values outside of this range are not shown. (Note the different scales for new coal unit generation and CO2 emissions, and for new natural gas generation and NOx and SO2 emissions, marked with asterisks).

Extended Data Fig. 3 Starting (2005) levels and trends over the study period for PM2.5, O3, SO2, and NO2.

Dots in the trends plots show locations of coal-fired units taken offline (red) and natural gas-fired units brought online (blue) during the study period. As shown in this analysis, part of these changes is attributable to shifts in electric power production feedstock, but other policies and regulations (for example fuel efficiency standards) and technology changes (for example emissions controls technologies) have contributed as well. Coal and diesel combustion are responsible for most SO2 emissions, while NO2 (a portion of NOx) comes from transportation as well as combustion of coal and natural gas. NO2 concentrations are more tightly associated with urban areas and transportation corridors. Ozone production is nonlinear, based on reactions of NOx and volatile organic compounds in the presence of sunlight. Particulate Matter includes aerosols from many sources, including primary carbonaceous aerosols, sulfates (from SO2), nitrates (in part from NOx), dust, and sea salt (see Supplementary Fig. 1).

Extended Data Fig. 4 Pollution surrounding power plant locations.

(a,c,e) As in Fig. 2b: Average near-surface (Planetary Boundary Layer) SO2, tropospheric NO2, and surface O3 surrounding operating electric power plants, by fuel type. SO2 and NO2 decrease radially around plants. Although SO2 is not a main byproduct of natural gas combustion, some plants have a combination of gas and coal-fired units, and others may use different types of fuels. Ozone dynamics are more complicated around an emissions source, consistent with previous studies. (b,d,f) As in Fig. 2a: Raw average changes in ambient O3, SO2, and NO2 in the time leading up to, and after, a coal-fired unit shutdown. Estimates include location-level fixed effects (that is concentrations for each location are de-meaned to show changes from baseline). Error bars show the 5th-95th% confidence interval, based on standard errors clustered at the location level.

Extended Data Fig. 5 Comparison of surface ozone impacts of power generation units.

As in Fig. 2c. Comparison of different models relating a change in number of units of a given feedstock within a given radius of a county, and average levels of O3 in that county. Addition of a natural gas-fired unit (and to a lesser extent, a coal-fired unit) is associated with increased ozone levels (likely via increased NOx production).

Extended Data Fig. 6 One example location.

Data from a coal-fired unit shut down in Georgia, showing the changes in ambient PM2.5, O3, SO2, and NO2. The thick blue line shows power generation (gross load), with the shutdown marked by the grey bar. Black lines show pollutant concentrations: the solid line shows concentration of each pollutant in the immediate region around the power plant, with dashed lines out to a 100km radius. This is an individual instance of the aggregate averages presented in the main text (Fig. 2, as well as Figures ED2 and ED5, and all Supplementary Tables).

Extended Data Fig. 7 Instrumental variables impact estimates.

As in Fig. 3a,b, only using an instrumental variables approach to estimate the effect of a 1 µg m-3 increase in PM2.5 on mortality and crop yields. In this approach, coal unit shutdowns are first related to PM2.5 concentrations; those predicted PM2.5 values are then related to mortality and crop yields. This approach strips out the variation in aerosol PM2.5 and other covarying pollutants not associated with unit shutdown. Central estimates are similar to Fig. 3a,b (but with larger error bars) indicating robustness of the approach of relating unit shutdowns directly to downstream outcomes. However, results should be interpreted as the impact of all pollutants covarying with PM2.5, and not PM2.5 alone.

Extended Data Fig. 8 A summary of impacts results estimated from different models at the county level.

The top row shows reduced form results for pollution, mortality, and crop impacts for 3 county-based models. The 25km and 200km coal models are shown in Fig. 2 and Fig. 3 in the main text. The third (top set of points) model includes natural-gas fired units. Red dots indicate coal unit impacts, and blue dots indicate natural gas unit impacts. The bottom row shows a comparison of instrumental variables (IV) results, whereby the number of units within a given radius is first related to changes in ambient pollution; those changes in pollution are then related in a second step to outcomes. Although results are cast as per μg m-3, they should be interpreted as the impacts of all pollution that covaries with PM2.5. The robustness of these IV results across models highlights the need for more causally-identified impacts studies and provides evidence that natural gas-fired units are not benign. Error bars show the 5th-95th percentile confidence interval; all estimates include county and year fixed effects, with standard errors clustered at the county level. Small grey bars show the average of the three models for each outcome.

Extended Data Fig. 9 Total impacts of coal-fired fleet.

The left column shows the results presented in Fig. 3c-f, with mortality and crop yield impacts integrated over the study period for plants within (a) a 200km radius, and (b) a 25km radius from each county. Right column shows the calculation described for impacts of remaining coal-fired units still operating, assuming that their impacts are the same as those that have been decommissioned.

Extended Data Fig. 10 Comparison between mortality results from this study and other literature.

Central mortality estimates in this study are similar to previous empirical exposure studies, for both total mortality and infant mortality. The Thurston et al, Eftim et al, and Zeger et al studies all focused on adults; Chay & Greenstone and Knitell et al focus on infant mortality. GBD results (2005-2013) are derived by combining PM2.5 reduction estimates from and pollution mortality from the GBD web interface. Apte et al results are for the lowest quartile (U.S. in that category), cast as percentages, and Burnett et al are estimated from the GEMM total mortality curve provided in the paper. Although not statistically significant (error bars show 5th-95th percentile confidence interval) the instrumental variables estimates from this analysis nevertheless highlight the importance of future causally-identified observational studies, as well as the critical role more comprehensive monitoring may play in reducing measurement errors (see Figure ED8).

Supplementary information

Supplementary Information

Supplementary methods, discussion, Figs. 1–5, Tables 1–9 and references.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Burney, J.A. The downstream air pollution impacts of the transition from coal to natural gas in the United States. Nat Sustain (2020) doi:10.1038/s41893-019-0453-5

Download citation