Research ArticleENVIRONMENTAL SCIENCES

Increasing mitigation ambition to meet the Paris Agreement’s temperature goal avoids substantial heat-related mortality in U.S. cities

See allHide authors and affiliations

Science Advances  05 Jun 2019:
Vol. 5, no. 6, eaau4373
DOI: 10.1126/sciadv.aau4373
  • Fig. 1 Observed and projected changes in extreme high temperature over the contiguous United States.

    (A) Estimated trends of the annual mean TXx values between 1979 and 2013 from the EWEMBI dataset. (B) Difference in the decadal averages of TXx between the 2° and 1.5°C worlds. (C) Same as (B) but between the 3° and 1.5°C worlds. (D) Same as (C) but between the 3° and 2°C worlds. The differences in (B), (C), and (D) are averages across 30 bias-corrected ensemble members of Hadley Centre Atmospheric Model version 3P (HadAM3P). Stippling indicates regions where neither trends nor differences are significant at the 5% significance level using the two-sigma test and the Kolmogorov-Smirnov test, respectively. The markers indicate the locations of the cities included in this study.

  • Fig. 2 Differences in the number of hot days over a decade-long period between the 1.5° and 2°C scenarios and the 3°C baseline scenario over the studied U.S. cities.

    The box plots show the distribution of differences computed from 90 ensemble members, with the middle line showing the median, the box showing the interquartile range (IQR), the whiskers indicating values that are 1.5 IQR from the lower and upper quartiles, and the dots indicating any outliers. A box plot that does not include 0 indicates a significant difference in the number of hot days between the two scenarios.

  • Fig. 3 Avoidable fraction of heat-related deaths if the current trajectory warming of 3°C is brought down to the 1.5° or 2°C Paris Agreement thresholds.

    The value of each bar indicates the mean avoidable attributable fraction across 90 climate model ensemble members. The error bars show the 95% eCI that accounts for both the uncertainty arising from internal climate variability and the uncertainty associated with the estimated exposure-response relationship (gray shading in fig. S1). Confidence intervals that do not include 0 indicate statistically significant results at the 5% level.

  • Fig. 4 One-in-30-year heat-related mortality that is avoidable by stabilizing future warming at the 1.5° and 2°C Paris Agreement thresholds rather than 3°C.

    The point estimates show the mean 1-in-30-year mortality level across 101 plausible exposure-response relationships, whereas the error bars show the 95% eCI accounting for uncertainties from internal climate variability and the exposure-response relationship. All estimates assume constant population. Confidence intervals that do not include 0 (dotted line on each panel) indicate a statistically significant number of avoidable deaths. The size of each bubble on the central map is proportional to the square root of the city’s population in July 2016. The color of each bubble indicates the city’s projected population change between 2015 and 2040. Other return periods are given in fig. S3.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/5/6/eaau4373/DC1

    Fig. S1. Estimated exposure-response relationships between daily mean temperature and all-cause mortality over selected U.S. cities.

    Fig. S2. One-in-30-year heat-related mortality per 100,000 persons that is avoidable by stabilizing future warming at the 1.5° and 2°C Paris Agreement thresholds rather than 3°C.

    Fig. S3. Heat-related mortality return period curves in future stabilization scenarios of 1.5°, 2°, and 3°C.

    Fig. S4. Population-normalized heat-related mortality return period curves in future stabilization scenarios of 1.5°, 2°, and 3°C.

    Table S1. The MMT and its percentile rank in the 1987–2000 observations in each city.

    Table S2. Maximum observed and projected temperatures and the percentage of days on which the projected temperature exceeds the maximum observed temperature in each scenario and city.

  • Supplementary Materials

    This PDF file includes:

    • Fig. S1. Estimated exposure-response relationships between daily mean temperature and all-cause mortality over selected U.S. cities.
    • Fig. S2. One-in-30-year heat-related mortality per 100,000 persons that is avoidable by stabilizing future warming at the 1.5° and 2°C Paris Agreement thresholds rather than 3°C.
    • Fig. S3. Heat-related mortality return period curves in future stabilization scenarios of 1.5°, 2°, and 3°C.
    • Fig. S4. Population-normalized heat-related mortality return period curves in future stabilization scenarios of 1.5°, 2°, and 3°C.
    • Table S1. The MMT and its percentile rank in the 1987–2000 observations in each city.
    • Table S2. Maximum observed and projected temperatures and the percentage of days on which the projected temperature exceeds the maximum observed temperature in each scenario and city.

    Download PDF

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article