Strategies for rapid global earthquake impact estimation: the Prompt Assessment of Global Earthquakes for Response (PAGER) system - Seismic Risk Analysis and Management of Civil Infrastructure Systems - page 847

Mapping Project (GRUMP v1.0, see http://sedac.ciesin.columbia.edu/gpw/)
dataset, and the ShakeMap shaking intensity estimate. We use three time
intervals to defi ne time of day (day, night, and transit). For each earth-
quake's time of day, we estimate the total indoor population among three
broad occupancy categories, i.e., residential, non-residential and outdoor
population, using a global workforce database (Jaiswal and Wald, 2010).
Similarly, for vulnerability assessment, we collated the data specifi c to col-
lapse probability by construction classes from different countries, using
both empirical data as well as expert judgment surveys, and then developed
a set of collapse probability functions (Jaiswal et al. , 2011).
By combining Hazus level 4 injury rates for the complete damage state
with collapse (NIBS-FEMA, 2009) and fatality rates by structure type for
the D5 damage grade proposed by Spence for the European LessLoss study
(Spence, 2007b), we developed fatality rates specifi c to PAGER structure
types (PAGER-STR, Jaiswal and Wald, 2008). Loss computations are per-
formed at the grid-level as shown in Fig. 31.4 in which the grid-specifi c

Pop.
trap.in
Type-A

Deaths in
Type-A

Pop.
trap.in
Type-B
Pop.
trap.in
Type...

Pop. in
Type-A

Pop. in
Type-B
Pop. in
Type...

Res.

Day Nonres.

Night Outdoor

Transit

Grid
cell
'

Urban/
rural

i '

31.4 Overview of grid-based population exposure and hazard
information used to compute earthquake damage and losses within
PAGER's semi-empirical and analytical loss models.