The program I use to calculate the probability of an earthquake follows a method devised by Dr. Alan Jones of the State University of New York at Binghamton. He created it to eliminate or greatly reduce the effect of clumping in the dataset which is common in earthquake histories since large earthquakes typically have many aftershocks which bias the frequency count.
The program expects to read a quake file taken from a NEIC search in "Compressed" format. It accepts your input for the first and last year, the quake file (*.txt only) and window size in days, hours and minutes.
It takes the given time span, divides it into consecutive windows of the size you requested and sorts the quakes into their respective windows.
Then it counts the number of windows which contain quakes and divides by the total number of windows to get the probability that a window chosen at random would contain at least one quake.
This is an observed probability and is dependant on the dataset, so different time periods will give different answers as seismicity changes with time.
The program runs under DOS or in a DOS window under Windows. It can be downloaded from the download page.
Your questions and comments are welcome. |
