Gloval Exposure Model

GED
GLOBAL EXPOSURE MODEL/DATABASE
GAR2013

Basics
* Probabilistic risk analysis (PRA) results will be presented by country, for this reason each exposure dataset should be develop view in this way. Some particular categories assignation/disaggregation will be included in order to disaggregate annual average loss (AAL) risk results, i.e. fiscal/public risk
* The blue table means thatinformation contained on it came from WDI-GDF database by World Bank.
* The red table means that information contained on it came from estimates or indicators build from derivate information of blue tables.
* The green table means that information contained on it is the exposure dataset records that correspond to an exposed element (certain structural building type into a certain urbanarea with an assigned exposed value). The green table belongs to a specific country (in this case Colombia).
* The dataset records should be on shape file format where each record represents a certain structural building type into a certain urban area with an assigned exposed value. In this case there will be many records as urban areas by structural types by sectors.
Exposure model miss outs* Appling indicators of labor force, income level, health and education services straight ahead from national level to sub-national regardless complexity level of each urban area/settlement.
* Structural type distribution percentages are for people who occupy each structural type, not the building distribution per structural type.
* The capital stock value to distribute across thecountry surely contains the infrastructure values of public services networks, transportation and so on, which are exposure elements that are not included in the current exposure model, and neither in the structural types (and vulnerability functions).
1 step: Classify a country by country development
The Word Bank has a criterion to classify countries/economies based on gross national income (GNI)per capita and has a four income groups (Low, Middle Low, Middle High and High) with an specific GNI per capita limit (<$1.025, $1.026-$4.035,$4.036-$12.475, >$12.476 respectively). In the case of Colombia results in a “Middle High” category this is shown on Table 1 based on previous limits.
Table 1: Latin-American countries development category
Country | GNI per capita, Atlas method(current US$) | Country development |
… | … | … |
COL | 4,070 | Middle High |
ARG | 6,050 | Middle High |
PER | 3,370 | Middle Low |
… | … | … |
… | … | … |
2 step: Extract the urban population count
Population count global coverage to be used for the GED will be the 2005 LandScan product updated to 2010 with the population growth indicator by country

Figure 1: Population countcoverage

Population count coverage
Adjusted urban mask coverage
Urban population

Figure 2: Population urban count process
The urban population count is extracted from national population coverage intersected with adjusted urban areas mask. The cited mask is conformed from MODIS 500m artificial areas coverage (Schneider, 2010) in addition of contiguous cells from population count coverage havingmore than 2,000 inhabitants.
3 step: Classify each one of the urban areas by complex type
Ones the urban population count is extracted, a categorization of complexity are performed based on the number of people (settlement size according to Satterthwaite (2006)) in the specific patch of adjusted urban areas mask as follows:
Table 2: Classification criteria according to Satterthwaite (2006)Complexity Type | Index | Description |
Major Urban | 1 | >=20,000 inhabitants |
Minor Urban | 2 | <20,000 and >2,000 inhabitants |
Rural | 3 | <= 2,000 inhabitants |
Ones the complexity type is assign an aggregation of population data is performed in in a cell space of 5x5 pixels, in order to reduce the database size looking to improve the time to process the information an...