-
Notifications
You must be signed in to change notification settings - Fork 6
/
ISSUES
24 lines (15 loc) · 951 Bytes
/
ISSUES
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
# Wrong initial source depth
If the initial source depth is very wrong, e.g. 500 km vs. 10 km,
then two problems arise:
- The picking based on the "predicted" pick arrival may
work but the resulting pick is too far off the prediction and the
pick will be skipped due to the too-large time difference.
- The predicted pick time of an early origin with wrong depth is not
updated with consolidating source depth. Even a very good pick
will therefore be rejected and there is no way out.
Test-Event e.g. gfz2022fpfe.
Solution: The client needs to update the predicted arrival time for
each and every origin.
# Too many stations processed at once
Processing of too many stations can slow down processing because of the SeedLink acquisition that waits until all time windows are filled.
Possible solution: Frequently pass intermediate datasets to the picker. E.g. every minute dump all currently complete streams and initiate the picking.