Development of Hazard Damaged Buildings Model by ChiChi

Development of Hazard Damaged Buildings Model by Chi-Chi Earthquake Data B. J. Lee College of Construction and Development Feng Chia University Nov. 17, 2003

Contents Introduction Overview of Building Damage Estimation of Strong Motion Fragility Curves for the Building Damage Building Seismic Risk Analysis Conclusions

Introduction Chichi Earthquake & Chelungpu Fault The Chelungpu fault cross Guangfu elementary school, Wufeng , Taichung County Earthquake Occurred Date September 21, 1999 (GMT: 12. 6 sec. 17: 47, Sep. 20) Epicenter location 23. 87° NL, 120. 75° EL Focus Depth 7~10 Km Scale ML=7. 3 (CWB) Totally or Partially Collapsed Buildings More than 20, 000 buildings

Introduction Building Damage Survey Report 8, 773 buildings were surveyed in preliminary report

Introduction Building Damage Ratio = = incomplete (Buildings) No suitable database • Although the survey report is very useful, the inventory of buildings was not possible to associated with data. • To construct the fragility curve to be used for damage assessments, however, building damage data associated by inventory are necessary.

Introduction Building Damage Database Partially & Totally Collapsed Building Name Type of Structure Address Floor Area Construction Period List of earthquake building subsidy Chichi Earthquake Building Damage Data Warehouse …… Database of Building Tax (inventory of buildings)

Overview of Building Damage in the Chichi Earthquake Use Base Area to Measure Building Damage One Building Base area

Overview of Building Damage in the Chichi Earthquake Structural Type Distribution of Totally Collapsed Buildings

Overview of Building Damage in the Chichi Earthquake Collapsed Ratio for Each Structural Type

Overview of Building Damage in the Chichi Earthquake Collapsed Ratio of Reinforced Concrete Buildings in Different Construction Periods

Estimation of Strong Ground Motion Distribution of Accelerometer Sites Recover region : (Miaoli、Taichung、 Nantou、Jiayi、Yunlin and Changhua) Number of accelerometer sites : 103 Min=21. 05 gal Max=989. 22 gal Avg. =214. 62 gal St. Dev. =177. 492 gal

Estimation of Strong Ground Motion Attenuation Model - Campbell Form Coefficients of different strong motion factors

Estimation of Strong Ground Motion Correction Factor for Campbell Form (Tsai, 2001) = correction factor = number of observation (103 sites) = data of Chichi earthquake = data of Campell form

Estimation of Strong Ground Motion Campbell Form Including Correction Factor Chichi earthquake records (103 stations)

Estimation of Strong Ground Motion Geostatistic Theory Trend Residual Equation of horizontal attenuation relationship Kriging

Estimation of Strong Ground Motion Ordinary Kriging System (Semi-variogram) • Spherical Model Data interval : 2500 M h Distance (M)

Estimation of Strong Ground Motion Estimated PGA Distribution in Recover Region PGA(gal) Taichung City Taichung County Nantou County Che-long-pu fault

Fragility Curve for Building Damage Fragility Curves A lot of seismologists hypothesize lognormal distribution between strong motion and building damage (Shinozuka, 1999; Yamaguchi & Yamazaki, 1999; Osamu Murao, 1999) After Determining the mean and standard deviation, we can construct the fragility curve

Fragility Curve for Building Damage Study region – 15 Km buffer of Chelungpu fault line

Fragility Curve for Building Damage Fragility curves of the different structural types – totally collapsed

Fragility Curve for Building Damage Fragility curves of RC building for three different construction periods

Building Seismic Risk Analysis Topographical Classification of Nantou County Near-Fault Area Hanging Wall Foot Wall Puli Basin

Building Seismic Risk Analysis CDF of estimated PGA for different topographic conditions

Building Seismic Risk Analysis Topographic Condition - - 5. 99 - - Before 1982 6. 76 1983~1989 Near-fault line Basin 5. 88 6. 11 6. 07 0. 20 0. 22 0. 21 0. 05 0. 48 7. 29% 4. 85% 11. 05% 7. 84% 6. 87 0. 45 3. 83% 2. 37% 6. 38% 3. 87% After 1990 7. 06 0. 55 3. 53% 2. 33% 5. 45% 3. 72% Reinforced Masonry 6. 72 0. 44 6. 79% 4. 36% 10. 73% 7. 17% Mud-Brick 6. 19 0. 33 30. 50% 21. 23% 41. 88% 35. 72% Masonry 6. 48 0. 40 13. 78% 9. 13% 20. 59% 15. 19% RC Hanging wall Foot wall

Building Seismic Risk Analysis Risk probability The damage ratio of building = the degree of building resist the strong motion index (R) < PGA(S) Risk probability of building damage refers to the probability when resistance of the building against strong motion index ( ) is smaller than significance of strong motion ( )

Building Seismic Risk Analysis Risk probability The stands for the cumulative probability distribution and is also called reliability index.

Building Seismic Risk Analysis Risk probability Topographic Condition - - 5. 99 - - Before 1982 6. 76 1983~1989 Near-fault line Basin 5. 88 6. 11 6. 07 0. 20 0. 22 0. 21 0. 05 0. 48 7. 29% 4. 85% 11. 05% 7. 84% 6. 87 0. 45 3. 83% 2. 37% 6. 38% 3. 87% After 1990 7. 06 0. 55 3. 53% 2. 33% 5. 45% 3. 72% Reinforced Masonry 6. 72 0. 44 6. 79% 4. 36% 10. 73% 7. 17% Mud-Brick 6. 19 0. 33 30. 50% 21. 23% 41. 88% 35. 72% Masonry 6. 48 0. 40 13. 78% 9. 13% 20. 59% 15. 19% RC Hanging wall Foot wall

Building Seismic Risk Analysis n Murao et al. (1999) Pi = w N k åk=1 kl m k = building category l = indicates the subsurface soil condition Nk = ratio of buildings with category k w kl = risk probability of building category k on soil condition l