Web Analytics and Web Statistics by NextSTAT
Skip Navigation Links
+ The CentreExpand + The Centre
+ Island ProfilesExpand + Island Profiles
+ Research ActivitiesExpand + Research Activities
+ NewsExpand + News
+ Photo Gallery
+ Education & OutreachExpand + Education & Outreach
+ DownloadsExpand + Downloads
Skip Navigation Links
+ EarthquakesExpand + Earthquakes
+ VolcanoesExpand + Volcanoes
+ TsunamisExpand + Tsunamis
DID YOU FEEL IT?
Tell us about your earthquake experiences.
ARE YOU PREPARED?
Here is your earthquake safety checklist.
ASK THE EXPERTS
Our experts are here to answer your questions
Skip Navigation Links
RESEARCH ACTIVITIES
Seismology

Tobago and Earthquakes

Application of Distributed Computing Techniques in earthquake and volcano monitoring

Seismic Hazard Assessment

Strong motion network

Chaos associated with seismic swarms

Earthquake Classifying Neural Networks Trained with Random Dynamic Neighborhood PSOs

Tobago and Earthquakes This is a PhD project registered with The University of the West Indies. Three (3) themes are under investigation:
1. Forecasting of the strongest 1997 earthquake near Tobago using b-value analysis and other indicators –  this theme reports on the b-value analysis conducted during the activity of 1997, which led to the projection that a stronger earthquake was imminent; examine the data for other short-term indicators of stronger activity.
2. Projected seismicity near Tobago based on past seismicity near and off to the east of Tobago – this theme examines the entire database of significant earthquakes near, and in the far-field off the east of Tobago, as far as the Vema Fracture Transform, with a view to improving the understanding of the tectonics at work in the generation of earthquakes near Tobago. The possibility of faults being pushed closer to failure by remote earthquakes is becoming of more interest in the seismological community, and this particular chapter is currently being tailored for journal submission at an early date.
3. Hydrological effects of the April 22, 1997 Tobago earthquake – this theme documents, and analyses the groundwater impact of the strongest shock of the 1997 series, by collecting information from government agencies and property owners. A possible model consistent with the observations is used to determine the change in the size of crack apertures, which gave rise to the large volume of observed groundwater discharge.
Principal Investigator: Joan Latchman

Top

Application of Distributed Computing Techniques in earthquake and volcano monitoring This is a PhD project registered at the University of the West Indies. . This research project is primarily concerned with the integration of Information and Telecommunication technology (ITC) into the regional seismic data acquisition/monitoring network. Investigations continued in the following three areas.
1. Use of broadband Internet access technology (Asymmetrical Digital Subscriber Line) and digital radio telemetry (Wireless Ethernet and Spread Spectrum radios) to improve the “last mile” segment of the seismic data communication network.
2. Use of Watchdog Timer and other techniques to improve the availability and reliability of the remote data acquisition/monitoring systems.
3. Development of a middleware software package that is suitable computer-networking infrastructure in the Eastern Caribbean. Two areas of development are being undertaken. The first employs traditional HTTP and FTP methods while the second method uses message queuing techniques. The Indian Ocean tsunami of December 26, 2004 re-stimulated the drive to establish a regional Tsunami Warning System and strengthened the prospects for the establishment of a regional VSAT network to facilitate data and message communication. In anticipation of this we have started to investigate how the regional earthquake and volcano monitoring network infrastructure can be adapted to use satellite communication technology.
Principal Investigator: Lloyd Lynch

Top

 Seismic Hazard Assessment Using unclipped broadband recordings of strong earthquakes from three or more stations across the region we are studying ground motion attenuation patterns in order to gain better insights of the best attenuation relationship to use in future seismic hazard assessments. The results of this investigation is also being used to validate the last (2003) revision of regional hazard maps which uses attenuation functions that was selected based on clues derived from recordings of a few strong earthquakes in the Trinidad and Tobago area.
Principal Investigator: Lloyd Lynch

Top

 Strong motion network Investigation of the data obtained from an existing 4-station strong motion network on the island of Trinidad. These stations have been recording data intermittently for over four years since the initial installation (2000). Being in its infancy, there have been many teething problems arising in the collection and analysis of the data that is obtained from the network. The major limitation of the network is the sparse density distribution of the stations. The data has been successfully analysed and documented for the period January 2000 to November 2004. The next stage of development involves correlations of the magnitude and PGA of specific felt events with intensity maps, soil type and epicentral distance. Some of the results of this study have been presented to a team of international geophysicists and seismologists at the GFZ, Potsdam in September 2005. The culmination of this exercise will lead to new attenuation relationships which can be used in creating updated seismic hazard maps for Trinidad.
Principal Investigator: Rosemarie Mohais

Top

 Chaos associated with seismic swarms The phenomenon of chaotic time dependence of natural events has prompted some interest in the examination of seismic earthquake swarms associated with the island of Dominica. This research is concerned with possible chaotic attractors associated with swarm activity observed over the long term period 1960-2003 as well as any attractors associated with individual swarm sequences especially over the period 2000-2003. Principal Investigator: Rosemarie Mohais

Top

 Earthquake Classifying Neural Networks Trained with Random Dynamic Neighborhood PSOs Research is being done in collaboration with the Department of Math & Computer Science (Arvind Mohais) in the area of Artificial Intelligence. Using the extensive database of signals collected from the strong ground motion monitoring network, artificial neural networks developed by Arvind Mohais were first trained to discriminate the earthquakes from the non-earthquakes using a sorted dataset. Subsequent unsorted datasets were sorted by the trained neural networks with 97% accuracy. The efficient use of artificial intelligence in this field will significantly reduce the time spent on sorting through the raw data collected at strong motion seismic stations. The results of this collaboration were published at the GECCO 2007 Conference.
Principal Investigator: Rosemarie Mohais

Top