Research ArticleENVIRONMENTAL STUDIES

Adaptation required to preserve future high-end river flood risk at present levels

See allHide authors and affiliations

Science Advances  10 Jan 2018:
Vol. 4, no. 1, eaao1914
DOI: 10.1126/sciadv.aao1914
  • Fig. 1 Mean daily discharge (historic period, 1971 to 2004).

    Median over the ensemble of all 50 combinations of 10 hydrological and 5 climate models. Their runoff output was routed by the river routing model CaMa-Flood (12) to derive their mean daily discharge in the historic period; plotted here is the median of all model combinations. Very dry cells (discharge, <0.1 mm/day) are masked (white); this mask is also used for the other figures.

  • Fig. 2 Relative increase in affected people without adaptation measure (realization ensemble median).

    Increase is given as the multiple of change between future and historic periods relative to the historic period. A value of 2 means that three times as many people are at risk of high-end river flooding during 2035 to 2044 compared to 1971 to 2004. Regions with affected population in the future period, but none in the historic one, are marked “new” (black). Subfigures show regional foci on the (A) United States, (B) Europe, (C) Africa, and (D) Southeast and East Asia. Decrease in affected population is cut off to 0. Pop. density, population density.

  • Fig. 3 Increase in the regional flood protection level required to preserve the current high-end flood risk for the period 2035 to 2044 (realization ensemble median).

    Additional protection is given in levels, starting with 0 for regions without adaptation need. Level boundaries are 0, 1, 2, 4, 8, 16, 32, 64, 128, 256, 512, and 1000 years return period. Numbers shown are absolute difference in level numbers to current protection per subnational region in the FLOPROS database (6). Subfigures show regional foci on the (A) United States, (B) Europe, (C) Africa, and (D) Southeast and East Asia.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/4/1/eaao1914/DC1

    fig. S1. Increase in the regional flood protection level required to preserve the current high-end flood risk for the period 2035 to 2044 (realization 16.7 percentile, lower likely range).

    fig. S2. Increase in the regional flood protection level required to preserve the current high-end flood risk for the period 2035 to 2044 (realization 83.3 percentile, upper likely range).

    fig. S3. Required adaptation relative to current protection to preserve the current high-end flood risk for the period 2035 to 2044 (realization ensemble median).

    fig. S4. Affected people in the historic period.

    fig. S5. Affected people in the future period.

    fig. S6. Absolute increase in high-end flood risk.

    fig. S7. Climate model agreement (historic period).

    fig. S8. Climate model agreement (future period).

    fig. S9. Hydrological model agreement (historic period).

    fig. S10. Hydrological model agreement (future period).

    fig. S11. Example histogram of affected people (in India).

    fig. S12. Example histogram of affected people (in Egypt).

    fig. S13. Zoomed-in views of selected metropolitan areas; increase in the regional flood protection level required to preserve the current high-end flood risk for the period 2035 to 2044.

    fig. S14. Schematic of the method to yield the affected population from discharge.

    fig. S15. Probability plot correlation coefficient for the preindustrial control run of 439 years as a goodness of (GEV) fit measure.

    fig. S16. Probability density functions for the fitted GEV distribution at four representative grid cells (hot/cold and wet/dry).

    fig. S17. Increase in the regional flood protection level required to preserve the current high-end flood risk for the period 2035 to 2044 (realization ensemble median) using the Gumbel distribution for the extreme value fit (cf. Fig. 3 for GEV fit).

    table S1. Main characteristics of the GHMs as used in this study, based on the study of Warszawski et al. (7).

    CSV (comma-separated-values) file of the raw data

  • Supplementary Materials

    This PDF file includes:

    • fig. S1. Increase in the regional flood protection level required to preserve the current high-end flood risk for the period 2035 to 2044 (realization 16.7 percentile, lower likely range).
    • fig. S2. Increase in the regional flood protection level required to preserve the current high-end flood risk for the period 2035 to 2044 (realization 83.3 percentile, upper likely range).
    • fig. S3. Required adaptation relative to current protection to preserve the current high-end flood risk for the period 2035 to 2044 (realization ensemble median).
    • fig. S4. Affected people in the historic period.
    • fig. S5. Affected people in the future period.
    • fig. S6. Absolute increase in high-end flood risk.
    • fig. S7. Climate model agreement (historic period).
    • fig. S8. Climate model agreement (future period).
    • fig. S9. Hydrological model agreement (historic period).
    • fig. S10. Hydrological model agreement (future period).
    • fig. S11. Example histogram of affected people (in India).
    • fig. S12. Example histogram of affected people (in Egypt).
    • fig. S13. Zoomed-in views of selected metropolitan areas; increase in the regional flood protection level required to preserve the current high-end flood risk for the period 2035 to 2044.
    • fig. S14. Schematic of the method to yield the affected population from discharge.
    • fig. S15. Probability plot correlation coefficient for the preindustrial control run of 439 years as a goodness of (GEV) fit measure.
    • fig. S16. Probability density functions for the fitted GEV distribution at four representative grid cells (hot/cold and wet/dry).
    • fig. S17. Increase in the regional flood protection level required to preserve the current high-end flood risk for the period 2035 to 2044 (realization ensemble median) using the Gumbel distribution for the extreme value fit (cf. Fig. 3 for GEV fit).
    • table S1. Main characteristics of the GHMs as used in this study, based on the study of Warszawski et al. (7).

    Download PDF

    Other Supplementary Material for this manuscript includes the following:

    • CSV (comma-separated-values) file of the raw data (Microsoft Excel format)

    Files in this Data Supplement: