|
|
|
|
| Re: You Mean Close Hits? |
Yes, you're right about close hits vs close misses. But using this method, its something that is known before hand so it wouldn't really be a miss if it hit within these parameters. Actually, this was a rough draft and naturally it would have to be something that everyone would be agreeable to. I noticed that on my example I messed up on my location parameters. Here is an updated one with the parameters tightened a little and a 70 point value added. time frame: magnitude location With your 3.0 example, it still would get a hit using the above but how many of these would have to hit before it became statistically significant? I would think quite a few. But then it would depend on the size of the window. Hopefully you did understand the problem with the scenario that I showed. Adding the above to each prediction I think could also be handled by Alans method of calculating probability as well. He'd probably have to run it thru a few times to get the different probabilities, but it is doable. I see Canie but this other subject up about evacuating the wrong location. That is why I would prefer predicting for a larger area. That way a panic situation could be avoided for any one particular location and in addition, rescue units would be on standby in outside locations from where the quake actually hit where the rescue units in the quake zone would be hard pressed to fend for themselves anyway. A couple of other things. It was mentioned that different faults were involved in two different locations. I don't really see how that matters. No prediction method can say what fault a quake will be on when hundreds of faults are in a region. All you know is that one is coming. Heck, even a new fault could be starting out that wasn't there before. It happens all the time. Faults die out and new ones form. It all depends how on stress is being distributed. Just my 2 cents. Dennis |
|
|
|
|
