5.1 D YNAMIC E XPOSURE OF THE P OPULATION
5.1.1 Exposure Map of the Population
The daily activity patterns for the working and non‐working population groups were derived from an Activity Diary from the household survey. The patterns of working and non‐working respondents are significantly different. The non‐working respondents spend much more time on home activities, while the majority of working respondents conduct their working activities outside. Further, there is also a slight difference between male and female respondents in the same activity group, wherein the female respondents spend a little more time with home and household activities. As shown in Figure 5‐1, the respondents with non‐working main activities have a higher proportion of home activities than respondents with working main activities, and so are female respondents compared to male ones. Furthermore, the female population has on average a higher proportion of non‐working people (housewives), and consequently carry out more activities at home in total. To derive daily activity patterns for the population groups of children and population engaged in education as their main activity, assumptions were made based on the occupancy rate of schools and universities as well as existing daily activity patterns of the working population group. This is due to the fact that the respondents were mostly household heads or spouses (see Table 4‐7) and consequently, these groups were underrepresented in the activity diary data.
68
Figure 5-1 Daily activity patterns of male and female population by main activities
Source: own analysis based on the UNU‐EHS Household Survey 2008, also presented in Setiadi et al., 2010
After linking the population groups, their daily activity patterns, and the building uses, the population distribution was calculated. At the end, the female population, children, and the elderly (including people in hospitals) were summarized as the more vulnerable population group due to their limited physical capability to conduct evacuation. Using the “Last‐Mile” hazard map as a basis for the potentially tsunami inundated areas, the exposure of the population in the morning, afternoon, and night time was calculated at the building level and in a grid of 100mx100m as well as in a grid of 250mx250m. Here, the grid 250mx250m was displayed to fit the grid size of the evacuation traffic modelling.
The morning time indicates clearly that the highest number of people in the hazard zones, about 181,000, are located in the potentially affected areas of the tsunami modelling scenario of the Last‐
Mile project, whereof 118.000 of them (65%) are women, children, and elderly. The night time indicates the lowest number of exposed people of about 133,000 and a similar proportion of people with lower evacuation capability. In general, the coastal areas represent densely populated and built areas and therefore the level of exposure is high. By comparing the exposure at different daytimes, the concentration of people in the coastal areas and city centre during the working hours can be identified (Figure 5‐2). The higher total number of exposed people in the morning time is due to additional working population living outside the exposed city areas, which conduct their working activities in the workplaces located inside the exposed areas. In the morning time, it is indicated that
Male - Working
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
12-6 am
6-9 am
9-12 am
12-3 pm
3-6 pm
6-9 pm
9-12 pm
12-6 am
Male - Others
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
12-6 am
6-9 am
9-12 am
12-3 pm
3-6 pm
6-9 pm
9-12 pm
12-6 am Female - Working
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
12-6 am
6-9 am
9-12 am
12-3 pm
3-6 pm
6-9 pm
9-12 pm
12-6 am
Female - Others
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
12-6 am
6-9 am
9-12 am
12-3 pm
3-6 pm
6-9 pm
9-12 pm
12-6 am home activities work activities maintaining household activities social activities
69
population density outside the exposed areas decreases (more “dark green” area), and consequently population density in the city centre and commercial areas along the main roads increases (“orange”
and “red” areas). In contrast, the population density at night time is more equally distributed.
Figure 5-2 Dynamic exposure of population by daytime
Source: own analysis
The following Table 5‐1 summarizes the estimated amount of potentially affected people calculated.
Table 5-1 Estimation of potentially affected population based on daytime and building use Settlement Workplaces and
activity centres
Schools Special facilities:
hospitals
Total
Men (15‐64
years)
Women, children, elderly
Men (15‐64 years)
Women, children, elderly
Men (15‐64 years)
Women, children, elderly
morning 19.196 47.541 32.628 30.473 11.367 39.365 600 181.170 afternoon 24.892 60.979 24.740 22.308 6.869 15.624 588 156.000
night 40.010 79.517 6.583 7.179 257 4 413 133.963
Source: own analysis
In order to validate the results of the analysis, the number of the total population and population groups (at night) was compared with the population data from the statistics for the year 2005 (last available ones during the data collection period). A detailed analysis of the number of people by building or building block as well as for the population during the day was not possible since there
Number of people per grid 250
tsunami hazard_buffer100 1-500
500-1000 1000-1500 1500-2000
>2000
Morning Afternoon Night
Number of people Grid 250mx250m
70
was no available basic data in the local statistics that could be used to validate the results of this analysis.
The total population number of the overall study area in the statistics was 428,452, of which 275,845 were women, children, and elderly citizens, while the calculation produced a total population of 387,450, of which 257,682 were women, children, and the elderly. A difference in the absolute population of about 10% (underestimation) exists while the ratio of people with less or more evacuation physical capability remains similar. Nevertheless, the number can give a rough estimation of the magnitude of potentially affected people by social groups, while it is more important to see the relative spatial difference of distribution in various city clusters and evacuation zones. Further, this result was used as an input parameter for the traffic modelling of the evacuation. Within the modelling, a buffer zone of 500 m beyond the actual inundation line was applied as a safety factor and it was assumed in this study that this could compensate the underestimated amount of exposed people.