THESIS
2016
xix, 298 pages : illustrations, maps ; 30 cm
Abstract
The 5.12 Wenchuan earthquake occurred in Sichuan Province on 12 May 2008 caused
great losses to both human lives and economics. This research comprehensively and minutely
recorded, analyzed and presented the fatalities caused by the earthquake at town level, including
affected areas in Gansu Province and Shaanxi Province. Modified Mercalli Intensity
(MMI) scale, population density, per capita GDP and topography are selected as four primary
parameters controlling the fatality rate. Specifically, Beichuan County, Dujiangyan City and
Lushan are selected as representative site-specific study sites for deriving an empirical formula
to describe the relationship between fatality rate and the four primary parameters. Both
multiple regression analysis and calibrated VIS model are applied...[
Read more ]
The 5.12 Wenchuan earthquake occurred in Sichuan Province on 12 May 2008 caused
great losses to both human lives and economics. This research comprehensively and minutely
recorded, analyzed and presented the fatalities caused by the earthquake at town level, including
affected areas in Gansu Province and Shaanxi Province. Modified Mercalli Intensity
(MMI) scale, population density, per capita GDP and topography are selected as four primary
parameters controlling the fatality rate. Specifically, Beichuan County, Dujiangyan City and
Lushan are selected as representative site-specific study sites for deriving an empirical formula
to describe the relationship between fatality rate and the four primary parameters. Both
multiple regression analysis and calibrated VIS model are applied and the goodness-of-fit of
their results is compared. In addition, for the multiple regression analysis, different combinations
of the site-specific data sets are examined and compared in the derivation of both linear
and log form empirical formulas.
After the site-specific studies, a global fatality database covering the whole Wenchuan
earthquake is established. The database contains information of 159 counties, 4,534 towns
and 87,000 fatalities. It comprehensively documents the MMI scale, the population density at
the end of 2007, the per capita GDP at the end of 2007, the topography and any other valuable
information at town level. If a town crosses multiple MMI scales, it is further divided into
subordinated areas with each containing only one integer MMI scale. For towns without
known casualties, an empirical formula derived from the site-specific analysis is used to estimate
the fatality rates and allocate the county level or city level casualty to town level or subordinated
area level. The database is a valuable asset for future studies.
Global earthquake vulnerability analyses are also performed and vulnerability curves
are constructed for the whole quake zone. The fatality rate is expressed with respect to MMI
scale, population density at the end of 2007, per capita GDP at the end of 2007 and topography.
Also, MMI is held constant to further examine how the fatality rate varies with population
density, per capita GDP and topography. Moreover, special regional studies, e.g. urban
versus rural areas and north versus south areas, are also conducted by categorizing the data
sets in the global analysis into urban/rural areas and north/south areas, respectively.
GIS is used throughout this research in the division of towns crossing multiple MMI
scales into subordinated areas by a geo-referencing function, presentation of fatality rate for
the whole quake zone and presentation of fatality rate for the core quake zones.
Post a Comment