Estimating Intensities and/or Strong Motion Parameters Using Civilian Monitoring Videos: The May 12, 2008, Wenchuan Earthquake

Abstract

One of the important issues in macroseismology and engineering seismology is how to get as much intensity and/or strong motion data as possible. We collected and studied several cases in the May 12, 2008, Wenchuan earthquake, exploring the possibility of estimating intensities and/or strong ground motion parameters using civilian monitoring videos which were deployed originally for security purposes. We used 53 video recordings in different places to determine the intensity distribution of the earthquake, which is shown to be consistent with the intensity distribution mapped by field investigation, and even better than that given by the Community Internet Intensity Map. In some of the videos, the seismic wave propagation is clearly visible, and can be measured with the reference of some artificial objects such as cars and/or trucks. By measuring the propagating wave, strong motion parameters can be roughly but quantitatively estimated. As a demonstration of this 'propagating-wave method', we used a series of civilian videos recorded in different parts of Sichuan and Shaanxi and estimated the local PGAs. The estimate is compared with the measurement reported by strong motion instruments. The result shows that civilian monitoring video provide a practical way of collecting and estimating intensity and/or strong motion parameters, having the advantage of being dynamic, and being able to be played back for further analysis, reflecting a new trend for macroseismology in our digital era.

Seismic hazard maps based on Neo-deterministic Seismic Hazard Assessment for China Seismic Experimental Site and adjacent areas

“…Using the relationships between peak ground accelerations and Modified Mercalli Intensity (MMI) proposed by Wald et al (1999), we obtained the distribution of intensity (MMI) from Fig. 17 Lekkas, 2010;Yang et al, 2011), meaning that one could have a variety of choices for the selection of a benchmark. However, in general, the purpose of modelling is not to be in total agreement with all existing results, but basically to match the main features.…”