Research ArticleSOCIAL SCIENCES

Urbanization affects peak timing, prevalence, and bimodality of influenza pandemics in Australia: Results of a census-calibrated model

See allHide authors and affiliations

Science Advances  12 Dec 2018:
Vol. 4, no. 12, eaau5294
DOI: 10.1126/sciadv.aau5294
  • Fig. 1 The ensemble average of incidence of new infection, prevalence of infected agents, and cumulative infection temporal dynamics for simulated influenza in 2006, 2011, and 2016.

    Comparison of simulation results for (A) incidence, (B) prevalence, (C) accumulated incidence (attack rate), and (D) trend in dynamics of peak day and prevalence; error bars (± SEM) designate standard error of the mean.

  • Fig. 2 Maps and histograms demonstrating geographic and temporal bimodality in epidemic spread.

    The histograms represent the number of statistical areas (SA2) experiencing peak disease prevalence on a given day. The colors correspond to heuristic classification; green bars indicate the first wave, yellow bars are undetermined, and red bars indicate the second wave. The colors on the map correspond to those in the histogram and demonstrate the geographic distribution of each pandemic wave.

  • Fig. 3 Analysis of bimodality in disease prevalence.

    (A) The state-level prevalence for New South Wales (NSW) and South Australia (SA), comparing prevalence curves for 2006 and 2016. (B) National prevalence for 2016 and the two Gaussian curves used to fit the data. (C) Increasing interpeak separation, increasing aspect ratio of the first mode, and decreasing aspect ratio of the second mode (error bars: ±SEM).

  • Fig. 4 Variation in population growth rates across years for different levels of region remoteness.

    (A) Population growth by remoteness. The urban population has been increasing steadily, while the rural population declines in both relative (all nonurban areas) and absolute (remote regions) terms. (B to D) Time series representations of urban (B), nonurban (C), and remote (D) populations since 2008. (E and F) Time series of relative urban population fractions computed against nonurban (E) and remote (F) populations.

  • Fig. 5 The role of seeding.

    (A and B) Raw prevalence curves demonstrating the effect of applying the seeding conditions of 2016 on the topologies of 2006 (A) and 2011 (B). (C) The residual between prevalence in 2016 and prevalence in 2006 (black curve) and 2011 (orange curve) when seeded with 2016 airport traffic. The dashed vertical lines in (C) indicate the maxima of Gaussian fits for the second epidemic wave.

  • Table 1 Average daily incoming international air traffic.
    AirportStateYear
    200620112016
    SydneyNew South Wales13,21415,99519,991
    MelbourneVictoria5,9238,55712,802
    BrisbaneQueensland5,0535,9467,299
    PerthWestern Australia2,7664,5125,906
    Gold CoastQueensland2851,0441,435
    AdelaideSouth Australia4927661,170
    CairnsQueensland1,186707824
    DarwinNorthern Territory160356355
    TownsvilleQueensland01139

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/4/12/eaau5294/DC1

    Movie S1. The color mapped, temporal evolution of the average simulated prevalence for 2006, using the SA2 partition.

    Movie S2. The color mapped, temporal evolution of the average simulated prevalence for 2011, using the SA2 partition.

    Movie S3. The color mapped, temporal evolution of the average simulated prevalence for 2016, using the SA2 partition.

    Fig. S1. Incidence proportion for various Ro values.

    Fig. S2. Prevalence proportion for various Ro values.

    Fig. S3. Attack rate for various Ro values.

    Model description

  • Supplementary Materials

    The PDF file includes:

    • Legends for movies S1 to S3
    • Fig. S1. Incidence proportion for various Ro values.
    • Fig. S2. Prevalence proportion for various Ro values.
    • Fig. S3. Attack rate for various Ro values.
    • Model description

    Download PDF

    Other Supplementary Material for this manuscript includes the following:

    • Movie S1 (.avi format). The color mapped, temporal evolution of the average simulated prevalence for 2006, using the SA2 partition. The right panels depict the total prevalence curve and three major Australian cities with a 50 km ring to display the seeding regions centered on international airports.
    • Movie S2 (.avi format). The color mapped, temporal evolution of the average simulated prevalence for 2011, using the SA2 partition. The right panels depict the total prevalence curve and three major Australian cities with a 50 km ring to display the seeding regions centered on international airports.
    • Movie S3 (.avi format). The color mapped, temporal evolution of the average simulated prevalence for 2016, using the SA2 partition. The right panels depict the total prevalence curve and three major Australian cities with a 50 km ring to display the seeding regions centered on international airports.

     

    Correction (11 April 2019): Minor typographical errors have been corrected, and the two columns listing abundance error in Table 4 have been incorporated into a single column.

    The original movie 2 is accessible here and movie 3 is accessible here.

    Files in this Data Supplement:

Navigate This Article