Lessons learned from reconstruction following a disaster : enhancement of regional seismic safety attained after the 1976 Tangshan, China earthquake

Comparing an estimated death toll in the reconstructed environment of Tangshan in 1991 with the actual deaths in the 1976 earthquake, we evaluated the enhancements in seismic safety attained during the recontruction. It was found that while the most severely affected area was remarkably improved, the less severely affected ara was not rebuilt to the standards of the most severely affected area. The increased safety of the most severely affected area can be attributed to the reinforced concrete elements of the buildings. The hazardous situation in the less severely affected area is due the collapsible nature of the unreinforced masonry construction (AU)

Interpretation of published data of the 1976 Tangshan, China a earthquake for the determination of a fatality rate function

Analysing the seismological and damage data of the Tangshan earthquake, we derived a function correlating the percent loss of masonry buildings to that of human lives. We determined the relationship so that the spatial distribution of the death estimated on the basis of the relationship was the most consistent with that of the data. We employed the method of system identification in this determination. The relationship obtained was a downward convex, monotonously increasing function having a fatality rate of 30

at a collapse rate of 100

(AU)

Post - event rapid estimation of earthquake fatalities for the management of rescue activity

By integrating existing Knowledge in seismology, earthquake engineering, and epidemiology, we developed a computer model applicable to the post-event, rapid estimation of expected fatalities for any given earthquake. Our design purpose for this model was to provide adequate information to simplify the post-event management of rescue activities. The model requires very few input variables making it a potentially useful tool in the early phase of rescue mobilization. Imput data required in the estimation are earthquake magnitude, epicenter location, and three regional data as follows: population density, dominant building type, and seismic intensity increment due to local site effects. Once designed, we tested the model's performance accuracy using the data obtained from recent significant disasters