Abstract
Methods/Materials
MMI reports were collected from 5 major earthquakes (7 < M < 7.9) that occurred in different global regions. 10 reports were collected from each earthquake. I used a Google Earth anthropogenic biome map, and MultiSpec, a satellite image analysis program, to find the anthropogenic distribution surrounding all 50 response sites. I normalized the data and ran multiple regression analyses to establish a relationship between MMI and anthropogenic biomes. I created a "General Model" by running an analysis on all 50 MMI reports, and five "Specific Models" by running analyses on MMI reports specific to their corresponding earthquake. All models were compared to the equation used by the USGS for shakemaps.
Results
The Village anthrome is directly related to MMI in the General Model, where F(x)=4.803-0.003(Distance)+0.028(Village). The General Model is directly related to and dependent on the village anthrome, while the Specific Models are inversely related to and dependent on a variety of anthromes. In the residual model for slope analysis, the General Model, Specific Models, and USGS Model had slopes within a 0.005 range of each other. Finally, a one-sample T-test comparing the General Model to the USGS model produced a P-value=0.002, suggesting that MMI reports are not significantly inflated.
Conclusions/Discussion
The T-test and slope analysis suggest that humans can accurately estimate ground shaking. Because the General Model is created from a globally-diverse pool of data, it may be used to estimate MMI from around the world; the nuances of a single global region will not greatly affect estimated MMI. The synergy between Villages high population and total area may be why is it the most influential anthrome
This project aims to determine how our perception of seismic shaking is influenced by the anthropogenic biome, or man-made environment, in which we live.
Science Fair Project done By Dorothy L. Silverman