TemplateBank¶
A TemplateBank is a thin wrapper on an obsplus EventBank. TemplateBanks handle database tasks for storing, selecting and reading in catalogues of templates from/on disk. The idea is to have a suite of templates for all possible events of interest; when something interesting happens this database can be queried to provide the relevant templates for that trigger. Because TemplateBanks subclass EventBank, all
the look-up methods for EventBank are available.
Lets have a look at how we might generate a TemplateBank from scratch for a day of the New Zealand national catalogue.
[1]:
from rt_eqcorrscan.database import TemplateBank
from obspy.clients.fdsn import Client
from obspy import UTCDateTime
bank = TemplateBank(base_path="./template_bank")
client = Client("GEONET")
catalog = client.get_events(
starttime=UTCDateTime(2019, 6, 21),
endtime=UTCDateTime(2019, 6, 23),
latitude=-38.8, longitude=175.8, maxradius=0.2)
print(catalog)
7 Event(s) in Catalog:
2019-06-21T03:46:25.054642Z | -38.824, +175.865 | 2.178845358 M | manual
2019-06-21T03:52:00.142063Z | -38.826, +175.855 | 1.230595585 M | manual
2019-06-21T08:28:16.143864Z | -38.827, +175.866 | 1.364494767 M | manual
2019-06-21T08:29:23.083982Z | -38.827, +175.878 | 2.066853896 M | manual
2019-06-21T09:07:54.730731Z | -38.819, +175.904 | 1.420446524 M | manual
2019-06-21T15:34:49.878441Z | -38.764, +175.615 | 1.530488164 M | manual
2019-06-22T09:35:27.682975Z | -38.819, +175.866 | 1.880393466 M | manual
First we will put the events into the database.
[2]:
bank.put_events(catalog)
bank.read_index()
[2]:
| time | latitude | longitude | depth | magnitude | event_description | associated_phase_count | azimuthal_gap | event_id | horizontal_uncertainty | ... | standard_error | used_phase_count | station_count | vertical_uncertainty | updated | author | agency_id | creation_time | version | path | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 2019-06-21 03:46:25.054642 | -38.823746 | 175.864609 | 5000.000000 | 2.178845 | Taupo | 46.0 | 67.671886 | smi:nz.org.geonet/2019p462586 | 3811.082605 | ... | 0.279893 | 25.0 | 39.0 | 0.000000 | 2020-04-29 04:26:43.090682624 | scevent@kseqp02.geonet.org.nz | WEL(GNS_Primary) | 2019-06-21 03:46:42.845267 | /2019/06/21/2019-06-21T03-46-25_62586.xml | |
| 1 | 2019-06-21 03:52:00.142063 | -38.826271 | 175.855438 | 5000.000000 | 1.230596 | Taupo | 21.0 | 163.628212 | smi:nz.org.geonet/2019p462596 | 6603.341442 | ... | 0.229197 | 15.0 | 15.0 | 0.000000 | 2020-04-29 04:26:43.098682624 | scevent@kseqp02.geonet.org.nz | WEL(GNS_Primary) | 2019-06-21 03:52:21.600176 | /2019/06/21/2019-06-21T03-52-00_62596.xml | |
| 2 | 2019-06-21 08:28:16.143864 | -38.827057 | 175.866028 | 7828.764915 | 1.364495 | Taupo | 26.0 | 91.194962 | smi:nz.org.geonet/2019p463116 | 5572.845242 | ... | 0.255632 | 16.0 | 19.0 | 8385.635050 | 2020-04-29 04:26:43.106682624 | scevent@kseqp02.geonet.org.nz | WEL(GNS_Primary) | 2019-06-21 08:28:36.574978 | /2019/06/21/2019-06-21T08-28-16_63116.xml | |
| 3 | 2019-06-21 08:29:23.083982 | -38.827339 | 175.877670 | 5000.000000 | 2.066854 | Taupo | 61.0 | 40.233490 | smi:nz.org.geonet/2019p463118 | 3076.595177 | ... | 0.350769 | 37.0 | 53.0 | 0.000000 | 2020-04-29 04:26:43.130682368 | scevent@kseqp02.geonet.org.nz | WEL(GNS_Primary) | 2019-06-21 08:29:41.755653 | /2019/06/21/2019-06-21T08-29-23_63118.xml | |
| 4 | 2019-06-21 09:07:54.730731 | -38.819088 | 175.904159 | 5000.000000 | 1.420447 | Taupo | 32.0 | 76.976382 | smi:nz.org.geonet/2019p463190 | 4347.700016 | ... | 0.206139 | 18.0 | 28.0 | 0.000000 | 2020-04-29 04:26:43.142682624 | scevent@kseqp02.geonet.org.nz | WEL(GNS_Primary) | 2019-06-21 09:08:14.577520 | /2019/06/21/2019-06-21T09-07-54_63190.xml | |
| 5 | 2019-06-21 15:34:49.878441 | -38.764454 | 175.615082 | 113987.533600 | 1.530488 | Taupo | 21.0 | 265.354118 | smi:nz.org.geonet/2019p463916 | 11605.852290 | ... | 0.406860 | 21.0 | 14.0 | 13899.074190 | 2020-04-29 04:26:43.146682624 | scevent@kseqp02.geonet.org.nz | WEL(GNS_Primary) | 2019-06-21 15:35:31.012134 | /2019/06/21/2019-06-21T15-34-49_63916.xml | |
| 6 | 2019-06-22 09:35:27.682975 | -38.818531 | 175.866409 | 13595.767970 | 1.880393 | Taupo | 39.0 | 56.649483 | smi:nz.org.geonet/2019p465944 | 5168.447127 | ... | 0.478102 | 26.0 | 34.0 | 7484.634962 | 2020-04-29 04:26:43.162682368 | scevent@kseqp02.geonet.org.nz | WEL(GNS_Primary) | 2019-06-22 09:35:48.159662 | /2019/06/22/2019-06-22T09-35-27_65944.xml |
7 rows × 28 columns
Template_bank.make_templates.[3]:
bank.make_templates(
catalog=catalog, client=client, lowcut=2., highcut=15.,
samp_rate=50., filt_order=4, prepick=0.5, length=3, swin="all")
No data downloaded for NZ.NNVZ.10.EHT
No data downloaded for NZ.BKZ.10.HHR
No data downloaded for NZ.BKZ.10.HHT
No data downloaded for NZ.KWHZ.10.EHR
No data downloaded for NZ.MTHZ.10.EHR
No data downloaded for NZ.HIZ.10.HHT
No data downloaded for NZ.HIZ.10.HHT
No data downloaded for NZ.TSZ.10.HHR
No data downloaded for NZ.RITZ.10.EHT
No data downloaded for NZ.WATZ.10.EHT
No data downloaded for NZ.WHTZ.10.EHT
No data downloaded for NZ.KATZ.10.EHT
No data downloaded for NZ.TMVZ.10.HHT
No data downloaded for NZ.ETVZ.10.HHR
No data downloaded for NZ.ETVZ.10.HHR
No data downloaded for NZ.NNVZ.10.EHT
No data downloaded for NZ.OTVZ.10.HHT
No data downloaded for NZ.SNVZ.10.EHT
No data downloaded for NZ.MRHZ.10.EHR
Request would result in too much data. Denied by the datacenter. Split the request in smaller parts
Detailed response of server:
/home/chambeca/miniconda3/envs/eqcorrscan/lib/python3.7/site-packages/obspy/io/mseed/core.py:790: UserWarning: The encoding specified in trace.stats.mseed.encoding does not match the dtype of the data.
A suitable encoding will be chosen.
warnings.warn(msg, UserWarning)
[3]:
Tribe of 7 templates
This both returns the Tribe of templates that are created, and writes them to disk for later retrieval. Note that lots of warnings are raised by this due to missing links between picks. These can be ignored for this tutorial.
[4]:
import warnings
with warnings.catch_warnings():
warnings.simplefilter("ignore")
tribe = bank.get_templates()
print(tribe)
Tribe of 7 templates
get_templates supports filtering using event queries:
[5]:
bank.get_templates(starttime=UTCDateTime(2019, 6, 21, 9))
[5]:
Tribe of 3 templates
Usage in RT-EQcorrscan¶
A TemplateBank is required to use the CatalogListener and Reactor classes. Both these classes will listen to a remote client service for new events and add those events to the TemplateBank. The Reactor class will also react to any events that meet user-defined thresholds, collect the appropriate temlates using the TemplateBank, and run the real-time matched-filter process.
Usage outside of RT-EQcorrscan¶
Because TemplateBanks return Tribes, TemplateBanks represent an efficient means of storing Tribes on disk. Tribes can be used offline for matched-filter detection.