Earthquake prediction has been a challenge for those studying hazard events. Other events are more easily predicted but not trembling. This is because tremors occur worldwide, magnitudes that are recorded over a vast spectrum between M2.5-M10. Depths ranging from shallow to under 700 km, as in the Fiji Islands. However, if we sift through the data, it is not chaotic. Some locations are more likely to occur with deep earthquakes. Other sites generate more aftershocks or foreshocks than others. So, our initial problem is to have each of these regions to be defined through coordinates and studied over long periods.
Let's explain in more detail, let's define our input data to work with Earth only,
a) location (determine your coordinates)
b) tectonics in the region you are studying. Are you in a convergence, divergence, or transform? Are there subduction zones? Define each of them and use position maps. (Google Maps)
c) magnitude variation M2.5.-?
d) Depths occurrence- Some places in the West of the United States deep earthquake occurrences do not reach 300km.
This is our input data and is in various catalogs like USGS, IRIS, EMSC.
e) regression on time. We take the data collected over long periods. However, for small earthquakes < 4.5, the catalogs are useful in a range of 50 years back. If we collect all the information above it is a possibility to predict some more significant events. Also, to determine which location will generate some aftershocks stronger, or which area is going to have a significant event examining the intensity of the foreshocks.
That is the first step. My research work a step farther, it examines the possible interactions with Sun- Moon on Earth.
2000 Florida Ave. NW Washington, DC 20009 (202) 462 6900
© American Geophysical Union. All rights reserved.