search

usgs / swarm

a Java application designed to display and analyze seismic waveforms in real-time.
http://volcanoes.usgs.gov/software/swarm
Other
36 stars 16 forks source link

S-arrival time in hypo71 #294

Closed alebrovelli closed 4 years ago

alebrovelli commented 4 years ago

As mentioned in the other thread, I believe the S-arrival times are not correctly considered in hypo71. I made an example to illustrate the problem. Here is the output from hypo71 out1

The picked arrival times were (from the QuakeML) xml

The P-arrival time is the same in both files, whereas (in this case) the S-time listed by hypo71 is identical to the P-arrival time. In other cases, I found that the P- and S-arrivals in the hypo71 output are different, but the S- is still incorrect when compared to the QuakeML. It seems to me that if the difference in arrival time is smaller than 1sec, the P- and S- arrivals are riported as identical to hypo71, otherwise the S-arrival is calculated as the P-time shifted by a certain number of seconds (maybe the number of seconds between P and S, minus 1?).

As for the S- arrivals that cross the minute compared to P-arrival, my interpretation of the manual is that one should add 60seconds (or a multiple of) to the S, i.e. P-arrival 10:50 21.49 S-arrival at 10:51:12.00 should be listed as 72.00 in the hypo71 input, that is, the S-arrival occurs 72.00 seconds after the HR:MIN. Not sure either that my interpretation is correct.

I attach also the xml and output file of my example, in case you need them. SwarmTest.zip

Alessandro

dlnorgaard commented 4 years ago

swarm-2.8.14-SNAPSHOT-bin.zip

@alebrovelli Would you mind trying the attached Swarm to see if the issue is fixed? I think have confirmed it but would like to double check.

dlnorgaard commented 4 years ago

Hi. I believe this issue is now resolved (to go into next release) so I am closing the ticket. However, if you are able to test and find that it is not the case, please let me know and I will reopen.

alebrovelli commented 4 years ago

Hi Diana, I was finally able to do some testing. I confirm the snapshot you shared fixes the problem. I will test again once the new release is out. Thank you very much! Alessandro