Docstoc

earthquakes

Document Sample
earthquakes Powered By Docstoc
					The Spatial Patterns Of
Earthquake Casualties
(Damages) And Social
     Vulnerability

         Zahra Golshani
 Natural Resource & Environmental Science
            University of Illinois
               Feb,17,2009
                  Introduction
   Natural hazard Loss reduction through
    mitigation, preparedness and recovery programs

   Social factors play significant role in determining
    population vulnerability to hazard (not just
    physical nature of hazard)

   Increasing disparities in wealth and socio-
    economic status increases the potential loss for
    greater human loss
         Introduction (continue)
   Risk modeling has been limited to physical
    aspect

   Developing integrated models of risk assessment
    that would incorporate the social and economic
    consequences of earthquakes

   Incorporating indirect/non-structural loss to
    current models
        Objective/Questions
 An attempt to assess vulnerability in
  spatial term (including both social
  vulnerability and physical damage)
 Testing damage & social vulnerability
  relationship
 Does it confirm the literature?
 Creating social risk map
                 Conceptual model
     Cutter-1996

                         Geographic
  Risk                     Context    Biophysical
                                      Vulnerability


             Hazard
                                                      Place Vulnerability
             potential


                            Social
Mitigation                               Social
                            Fabric    Vulnerability
     Defining Social Vulnerability
Some fundamental factors that influence
  social vulnerability includes:
 Lack of access to resources, including
  information and knowledge
 Limited access to political power and
  representation
 Weak building or weak individuals
(Blaiki et al.1994;cutter 1997;Mileti 1999)
      Measures Of Social Vulnerable
               Population
Characteristics                        Variable
   Differential access to resources     Vulnerable-POP
    & information/greater              (by Age)
    susceptibility due to physical      Female population
    weakness
                                        Non-white (Race)
                                        Low education



   Wealth or poverty                     Per capita Income
                                          Median house value
                Data Description
   Social Vulnerability Data : all was obtained from
    census manipulations were done to obtain the
    rates
Vulnerable-POP (people under 14 & over 70)
Female population
Non-white (Race)
Low education (no school through 6 grade population)
Per capita Income
Median house value
   Average dollar value of residential damage in
    zipcode-89
                 Data Description-2
some change due to data issues

   Casualty (only 33 casualty in Northridge so it was not
    possible to do meaningful analysis with that data.

The Turkey casualty data--- only 5% avaliable (census
  data are aggregated)

Therefore damage data for Northridge earthquake
  was used: Average estimated damage for single and
  multi family at zipcode level (Obtained from CPS Report: California
    Policy Seminar )
               Procedures
 Creating damage map (Total damage for
  single and multi family houses as the
  Physical vulnerability indicator)
 Social vulnerability maps (using different
  variables)
 Compare the results
 Aggregate the social map and compare
  the results
                   Method
   Using GIS to manipulate and join data
    from different sources

   Using GIS to create single and integrated
    maps and compare the results
                  Case Study

   Northridge earthquake (modest
    earthquake)

   Jan 17,1994
         Physical Damage Map
       (LA County-zipcode level)
 $ value
 Distribution of data is
  not normal
 Skewed to the right
 287 out of 312
  zipcode had less than
  $1 million estimated
  damage
Damage & Poverty-Rate Maps
   A closer look

ZIP    TotalDam PovertyPCT
 90013      26,385   58.12%
 90021      22,500   56.38%
 90017      15,602   50.45%
 90058      11,000   47.77%
 90813      26,000   45.63%
Damage & Non_White Rate Maps
     A closer look

ZIP    TotalDam NonWhitePCT
 90305     5,895      95.23%
 90008     8,490      93.88%
 90043     6,204      91.59%
 90047     5,356      91.57%
 90746       400      86.84%
Damage & Household size Maps
      Closer look

ZIP    TotalDam HHSize
 90262        -      4.70
 90011     10,507    4.65
 91733        -      4.60
 91744     22,286    4.59
 90221      2,000    4.57
Damage & Female population Maps
     A closer look

ZIP    TotalDam POPFemale
 90201        -     52,334
 90650     62,889   52,145
 90011     10,507   49,331
 90280        -     48,548
 90250      1,667   48,297
Damage & Low Education
   population Maps
      A closer look

ZIP    TotalDam    LowEdu
 90011      10,507    20,892
 90201         -      16,680
 91331     193,814    16,284
 90280         -      14,606
 90255      13,556    14,090
       Household Median value and
       Vulnerable population results

ZIP    TotalDam HouseMedValue        ZIP    TotalDam    POP_VUL
 90014     10,133       35,600        90201         -      37,762
 93523        -         45,800        90011      10,507    36,949
 93591        -         77,300        90650      62,889    34,649
 92301        -         78,800        90280         -      32,891
 93560        -         85,400        91331     193,814    32,776

                      ZIP TotalDam IncomePerCap
                       90058  11,000       7,359
                       90813  26,000       7,567
                       90001   5,775       7,632
                       90033  11,685       7,775
                       90003   5,075       7,804
Comparison of social &physical
     vulnerability maps
              Social vulnerability & damage
               relationship (significant at 10%)
        Regression Statistics
Multiple R              0.100613275
R Square                0.010123031
Adjusted R Square        0.00685611
Standard Error          0.066538826
Observations                    305

ANOVA
                           df            SS           MS           F     Significance F
Regression                        1   0.013718993 0.013718993 3.098646142 0.079366325
Residual                        303   1.341506846 0.004427415
Total                           304   1.355225839

                      Coefficients Standard Error  t Stat     P-value   Lower 95% Upper 95% Lower 95.0% Upper 95.0%
Intercept              0.032705178 0.012492977 2.617885149 0.009292053 0.008121198 0.057289159 0.008121198 0.057289159
Socio_Vul_Rank        -0.041997263 0.023858053 -1.760297174 0.079366325 -0.088945714 0.004951187 -0.088945714 0.004951187
    Observations/implications
 Modeling the spatial pattern of social
  vulnerability with GIS does not contradict
  the literature
 Implications on natural hazard planning
 Identifying high risk locations (physical
  damage) -mitigation
 Targeting social vulnerable groups in
  response, relief and recovery phases
           Future direction
 Running different kind of regressions
  including spatial regression on data
 Using other physical vulnerability
  indicators such as distance to epicenter,
  distance to Fault, peak ground
  acceleration
 Using other variables such as injury
               conclusion
 The benefit of integration
 Improving risk assessment models
 Better and more efficient natural hazard
  planning
Questions & Comments??


Thanks for your time
have a great afternoon

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:5
posted:10/23/2011
language:English
pages:30