Gisola v1.2.2

This release solves an issue when retrieving Magnitude Information due to inconsistencies found on various FDSNWS-event services.

Issue

• When Gisola, retrieves information for an event, it tries to associate the information with the respective origin which is a type of "Hypocenter". Some FDSNWS-event services do not provide this info.

Solution

• As a last resort, it uses the preferred magnitude found (which might be of a different type).
• Additionally, for compatibility issues, the includeallarivals FDSNWS-event attribute is removed because some FDSNWS-event services do not provide it.

"To Whom Does This Concern?"

• If you already use this software calculating MTs using event source different than the FDSNWS-event, you do not need to do anything. If you retrieve event information from a particular FDSNWS-event service that does not trouble you, again you do not need to do anything. This concerns use cases where Gisola invokes error when specific FDSNWS-event service is specified.

Full Changelog: v.1.2.1...v.1.2.2

Gisola v1.2.1

This release solves an issue found when passing events to Gisola.

Issue

• The problem was when feeding Gisola with an event (e.g. --event-xml) that already contains a Focal Mechanism (FM) that was previously calculated from another source than Gisola. The reason is that particular attributes which are given to the FM from Gisola might be missing from another source.

Solution

• Add the reading attributes mechanism only when revision is happening.

"To Whom Does This Concern?"

• You might experience that issue if Gisola is fed by events (using e.g. --event-xml, FDSNWS-event, etc.) that ALREADY contain Focal Mechanisms (FM)s that were previously calculated from ANOTHER source than Gisola itself. The actual problem is that if Gisola finds an FM in the event XML file, it will try to find the tag: ns0:components. Since other tools that calculate FMs might not provide this feature, it can trigger this error. The version 1.2.1 with this bugfix is supposed to solve this. UNLESS there is a "reviewed" FM in the event XML file, then it will trigger the error (which makes sense, because Gisola can only review, its own solutions).

Full Changelog: v.1.2.0...v.1.2.1

Gisola v1.2.0

This release solves issues found when calculating MTs for small events and small distances (e.g. microseismicity).

Issue

• The problem stemmed from the calculation of dt in variance reduction computation, where the precision of 2 digits proved insufficient. It was necessary to modify the precision to 4 digits to address the issue.
• A bug in the map plotting was identified, arising from the combination of small plot distances with the combination of the latitude and longitude values.

Solution

• Change in Variance reduction computation the rounding to 4 digits
• Update the basemap library along with dependencies

"To Whom Does This Concern?"

• If you already use this software calculating MTs for regular magnitudes and involved distances, you do not need this update (apart from corner cases that may fall the the above description of the issue).
• If that involves you:
  1. download the latest Gisola code under this release/tag v.1.2.0
  2. remove old conda environment (or the one you use): conda env remove --name gisola
  3. re-apply the new one as described in Installation in Wiki: conda env create -f environment.yml

Full Changelog: v.1.1.0...v.1.2.0

Gisola v1.1.0

Major bug fix when same TYPE records found

Before the fix, Gisola could use e.g. NET.STA.LOC.HH as NET.STA.LOC.BH and vice versa. Theoretically, since both types of traces passed the filters, both are equivalent and qualified for the process. The MT solution will not change, but streams could be displayed wrong in terms only of channel type (at the level of station).

In the cases, where a station has more than one channel (e.g. HH and BH), the bug could affect the plots.
Specifically,

1. it could plot only the stream's label NET.STA.LOC.HH* as NET.STA.LOC.BH* (and vice versa) in misfit.png
2. it could plot the other channel in streams.png
3. it could append the other channel in QuakeML

1,2,3 would happen altogether (not some of them) or not at all, without an error message. Only through log file, someone can observe if the bug affected his plots.

This is the major change/commit: 25273e6

Gisola v1.0.1

New release after fixing FMVAR nan error

Gisola v1.0.0

This is the first release of Gisola!
v1.0.0