Estimating population risk for coastal disasters using spatial models with global data

Yuri Gorokhovich, Shannon Doocy

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Coastal areas present high risk in case of tsunami, hurricanes or floods due to the higher population densities. Traditional physical models or risk maps provide limited help since disaster spatial extent can not be available immediately for the emergency management. This impairs post- disaster response; more fatalities can be expected due to the uneven distribution of medical supplies, food or equipment. Geographic Information Systems analysis with global datasets on terrain and population provides new venue for the post-disaster response in the form of immediate (within 24 - 96 hours) model of affected population and geographical extent of disaster. Presented case study shows such example for the Northern Sumatra affected by tsunami of 2004. The results of presented modeling were compared with population data collected from the post- tsunami field survey. Obtained regression is statistically meaningful (R2 =0.58) and indicates that presented methodology can be a useful tool during the post-disaster management. Copyright ASCE 2008.

Original languageEnglish (US)
Title of host publicationSolutions to Coastal Disasters Congress 2008 - Proceedings of the Solutions to Coastal Disasters Congress 2008
Pages308-317
Number of pages10
StatePublished - Dec 22 2008
EventSolutions to Coastal Disasters Congress 2008 - Oahu, HI, United States
Duration: Apr 13 2008Apr 16 2008

Publication series

NameSolutions to Coastal Disasters Congress 2008 - Proceedings of the Solutions to Coastal Disasters Congress 2008
Volume312

Other

OtherSolutions to Coastal Disasters Congress 2008
CountryUnited States
CityOahu, HI
Period4/13/084/16/08

ASJC Scopus subject areas

  • Ocean Engineering
  • Safety, Risk, Reliability and Quality

Fingerprint Dive into the research topics of 'Estimating population risk for coastal disasters using spatial models with global data'. Together they form a unique fingerprint.

Cite this