The Iquique Local Network and PicArray

authors:Simone Cesca (GFZ German Research Centre for Geosciences, Potsdam, Germany), Monika Sobiesiak (Polish Academy of Sciences, Krakow, Poland), Carlos Tassara (Universidad Nacional Arturo Prat, Iquique, Chile), Manuel Olcay (Universidad Nacional Arturo Prat, Iquique, Chile), Erwin Günther (GFZ German Research Centre for Geosciences, Potsdam, Germany), Stefan Mikulla (GFZ German Research Centre for Geosciences, Potsdam, Germany), Torsten Dahm (GFZ German Research Centre for Geosciences, Potsdam, Germany)
Abstract:The Iquique Local Network (ILN), a temporal network of broadband and short period seismic stations has been operating in Northern Chile since 2009. The aim of this installation was to locally densify the permanent seismic installation of the Integrated Plate Boundary Observatory in Chile (IPOC), with the main goal to decrease the magnitude of detected earthquake, to improve the hypocentral location accuracy, to allow a more accurate investigation of seismic source parameters, and to analyse proposed seismogenic structures of the Northern Chile seismic gap. The network setup evolved with time, with different geometries at different installation phases, aiming to study different seismicity features. In the first phase, started in 2009 and operational since 2010 until autumn 2013, the network had a sparse configuration, targeting a broad region extending from 19.5° S in the North to approximately 21.3° S South of Iquique. In the following stage, operational until fall 2017, most broadband stations were rearranged into a small aperture seismic array (PicArray) close to the village of Pica, to monitor with array techniques the shallow seismicity at the plate interfacer, intermediate and deep focus seismicity.
citations:

Recommended citation for the data report:

Cesca, S., Sobiesiak, M., Tassara, C., Olcay, M., Günther, E., Mikulla, S., & Dahm, T. (2018). The Iquique Local Network and PicArray. Scientific Technical Report STR - Data; 18/02. GFZ German Research Centre for Geosciences. https://doi.org/10.2312/gfz.b103-18022

If you use the dataset described in this report, please use the following citation:

Cesca, S., Sobiesiak, M., Tassara, A., Olcay, M., Günther, E., Mikulla, S., & Dahm, T. (2009). The Iquique Local Network and PicArray. GFZ Data Services. https://doi.org/10.14470/vd070092

Introduction

The Iquique Local Network (ILN) as one component of the International Plate boundary Observatory Chile (IPOC; GFZ and CNRS 2006) is a seismological network installed around the North Chilean harbour city of Iquique. With its average station spacing of approximately 25 km, the network aims at recording small to moderate seismic events to verify the crustal structure and seismic behaviour of the subduction interface in this region. Since the general task of the IPOC is to monitor the deformation signal before a large earthquake in a brought frequency range, the data of the ILN should provide insight in processes along material boundaries and geometrical features to investigate such an ‘earthquake preparation phase’ on a local scale. The ILN was originally built as a long term temporal deployment, also supporting hazard analysis for Iquique and the surrounding region. It hosts the nearest station to the coast which is especially important for investigating tsunamigenic earthquakes and their related hazard. An important scientific aim of the ILN was to verify the assumed hypothesis that seismogenic features in the area of the network identified by pronounced gravity anomalies suggested that the area could play a major role in a future earthquake. This earthquake occurred on April 1st, 2014 with a magnitude of Mw 8.2. Although the rupture was located off-shore, the ILN had the exact N-S extension of the rupture length parallel to the off-shore rupture (Schaller et al., 2015).

The PicArray was deployed in November 2013 in the vicinity of the village of Pica. The site was chosen as a central spot with respect to the former Northern Chile seismic gap, the availability of a very quiet station from the ILN, the lack of significant noise sources, and the ground conformation, well suited to host a small scale array. The seismic array, originally composed of 9 broadband stations, had an aperture of about 3 km. The PicArray installation aimed to enhance the seismological target of the former ILN configuration, allowing the application of beamforming and array based techniques to analyse the shallow seismicity at the plate interface, the intermediate and deep focus seismicity below Northern Chile and neighbouring regions and to analyse microseismicity and seismicity in the vicinity of the Pica village.

In November 2017, the PicArray setup was further modified, redeploying some of the PicArray installation to build a small network at about Lat 21°S, aiming to better monitor the slab at the southern edge of the 2014 Mw 8.2 Iquique earthquake rupture area.

Data Acquisition

Experimental Design and Schedule

The installation of ILN and PicArray comprised different deployment stages.

Iquique Local Network, stage 1 (2009-2013)

The first stage took place in the period 2009-2013. The installation developed with time, due to availability of the instruments and advancement of construction work at each station site. The very first stage just had two GURALP stations installed, one in a mining gallery (NEUQ) and the second one at the basement of a building of the Arturo Prat University in Iquique (UNAP). This first stage also had an antenna configuration with four REFTEK stations around the French PBO site at Humberstone. During the following deployment phases more GURALP stations were installed in their vaults. The REFTEK stations changed their locations several times to fill voids in the deployment scheme until filled by GURALP stations on important network sites. The ILN layout at April 27, 2011, covered a total area of 150 km in N-S and about 120 km in E-W direction (Fig. 1).

_images/map-monika.jpg

Fig. 1 Instrumentation and field campaigns in the area of the Iquique Local Network (ILN). The colored triangles represent the respective station sites, for color code see legend in the upper right corner. The blue diamond at the southern end of the network marks the creepmeter station on the Chomache fault and the magenta diamond denotes the long base line tiltmeter in the mine of Santa Rosa. The small black crosses capture the tracks and stations of the gravity field campaign in 2011 and small black dots show GPS sites either measured as campaign data or at permanent sites. The small inlay in the right lower corner gives an overview on the entire seismo-tectonic situation in the North Chilean Seismic Gap (NCSG) with coarse approximations for the rupture planes of the Mw 8.1, 1995, Antofagasta (green), the Mw 7.8, 2007, Tocopilla and Mw 8.2, 2014, Iquique

Iquique Local Network and PicArray, stage 2 (2013-2017)

A full station redeployment was performed in November 2013. All seismic broadband stations of the former ILN sites were removed, except stations UNAP, PATA and CHOM. Some of the former installation sites were later reinstrumented by the National Seismological Centre of Chile (CSN). At the time of the redeployment, many sensors and acquisition systems were not working correctly, a few instruments were damaged, also due to vandalism to the station sites. Using the available working instrumentation, the PicArray could be built.

The PicArray was deployed in November 2013, close to the village of Pica. Originally composed of 9 broadband stations, it was planned and built around the station PICL. The array location was chosen due to the low noise of the site, its accessibility and the geographical locations, which is optimal to target the central part of the former Northern Chile seismic gap, as well as local seismicity clusters. The array has a slight elongation along an EW direction, due to the topography of the site, and an aperture of about 3 km. The array was instrumented with broadband stations, and only at a later stage one additional short period seismic sensor was installed (station PICLA)

The station distribution at January 1, 2017, is shown in Fig. 5, and Table 1 summarises the most important information about each station.

Recent configuration and future plans

In May 2017, upon the failure of some sensors and dataloggers, and the stolen equipement at PICL8, the PicArray configuration was reduced to 5 stations. The remaining working instrumentation was used to build 3 new broadband installations, named PATS, SECO and SALG. The main motivation for the new setup was to better monitor the slab at the southern edge of the Iquique earthquake rupture region. The same region has been target of a denser GPS installation within IPOC.

Site Descriptions

With the exception of the UNAP (Iquique city) and AERO (at the base of the Iquique Airport tower) installations, station sites are remotely located in unpopulated areas, with no relevant anthropogenic seismic noise sources (e.g. as at station PATA, Fig. 2). As for the station installation, the sites have been prepared at different times and with different installation setups. For all stations equipped with GURALP CMG-3EPSC broadband seismometers, underground vaults were constructed where the seismometers were installed on concrete basis reaching ~ 1m into the ground (Fig. 3). This concept fostered a better coupling in the overall sandy soil conditions in the network area and resulted in excellent signal to noise ratios. The concrete basis was also used at station APACa, PicArray stations and later installations, although it was not possible to construct an underground vault (Fig. 4). Stations equipped with short period sensors were installed in holes in the ground. All installation included insulation and protection of the equipment against strong temperature variations and weather conditions.

_images/PB200483.JPG

Fig. 2 View of station PATA during maintenance in November 2013 (Photo S. Cesca). The station is located in a remote site, far from any anthropogenic source of seismic noise. The installation vault is protected by a fence and a trap-door.

_images/PB200484_2018-03-20T11:26:28.985932.JPG

Fig. 3 Vault at station PATA, hosting a broadband seismometer, acquisition system and battery (Photo S. Cesca, November 2013).

_images/PB140217.JPG

Fig. 4 View of PICL4 site, during the installation phase (November 2013, Photo S. Cesca). Sensors and acquisition system were protected and insulated within half metal barrels, digged into the ground at about 1 m depth. A thin layer of stones and concrete was built within the barrels, to provide a flat and rigid basement for the seismic sensor.

Instrumentation and sensor orientation

The ILN and PicArray operated up to 20 seismological stations. Out of these, 14 corresponded to broadband stations, equipped with GURALP instruments, each station including a CMG-3ESPC seismometer with 60s to 50 Hz frequency response, a CMG-DM24 S3 digitiser with 3 24-bit channels, 64Mb flash memory and GPS antenna, and a CMG-DCM Data Communications Module with removable data storage device. The remaining 7 stations were equipped with REFTEK stations connected to a 1Hz Mark L 3D seismometer. As power supply we used solar panels and car batteries (in most cases 35 Ah), which were checked, recharged or exchanged on a regular schedule approximately every year. During station maintenance the stations were in most cases recording, to reduce as possible data gaps. Sensor orientation has been adjusted using a magnetic compass during installation.

Data Description

Data Completeness

Data cover a long installation period and different installation setups. Data are incomplete, due to discontinuous station maintenance, vandalism episodes (e.g. removing solar panels and/or batteries), temporally reduced power supply (e.g. sand partially covering the solar panels), as well as degradation of power supply (i.e. batteries).

Fig. 7 shows the uptime of each stations.

Data Processing

Available data is raw data in miniSEED format (recorded in GCF format and converted to miniSEED). The miniSEED data on hand is in units of counts, not filtered and not resampled. Sampling is either 100 or 200 Hz, varying for different station sites and installation phases. Station metadata, including restitution information and characteristics of the sensors, are provided in stationXML format. Data and metadata can be downloaded from the GEOFON repository.

Data quality and Noise Estimation

Antropogenic noise is low at most stations, since sites were chosen in unpopulated areas and far from any infrastructure. Some, spatially distributed mining sites are present in the region. Most relevant natural noise sources can be wind, affecting high frequency records, and the ocean, affecting the amplitude of microseismic noise, higher for coastal stations. Transient, human noise sources can be present at the PicArray due to the vicinity of a small rubbish dump site and the construction of a small road, completed between 2015 and 2017.

Fig. 6 shows noise probability density functions for all channels.

Timing Accuracy

Seismic data consist of digital data with a sample rate of 100 or 200 samples per second.

Data Access

File format and access tools

The data are stored in the GEOFON database, and selected time windows can be requested by EIDA access tools as documented on http://geofon.gfz-potsdam.de/waveform/. Normally the data are delivered in miniseed format. The current data access possibilities can always be found by resolving the DOI of the dataset.

Availability

These data are freely available under the Creative Commons Attribution 4.0 International Licence (CC BY 4.0). http://creativecommons.org/licenses/by/4.0

When using the data please cite: Cesca, Simone; Sobiesiak, Monika; Tassara, Arturo; Olcay, Manuel; Günther, Erwin; Mikulla, Stefan; Dahm, Torsten (2009): The Iquique Local Network and PicArray. GFZ Data Services. Other/Seismic Network. DOI: http://doi.org/10.14470/VD070092

Conclusions and recommendations

The data of the network densified the IPOC permanent installation in Northern Chile with different configurations at different installation stages.

The seismic installation started in 2009 and seismic data include records of the Mw 9.0 Maule Earthquake in 2010, the large Mw 8.2 Iquique earthquake of April 1, 2014 (Schurr et al. 2014, Cesca et al. 2016), as well as its foreshocks and aftershocks sequences.

Future installation in the region may rely on the low anthropogenic noise, except in the vicinity of towns, mines and a few major communication roads. The presence of strong winds and sandy soil can temporally affect the functionality of solar panels and reduce the power supply.

Acknowledgments

We are thankful to UNAP Iquique and the Pica municipality to help installation and maintenance operations and to the Departamento de Geofísica, University of Chile, Santiago, for technical and administrative support. The Plate Boundary Project Iquique Local Network and PicArray are part of the IPOC Integrated Plate boundary Observatory Chile seismic network (GFZ et al., 2006)

References

Cesca, S., Grigoli, F., Heimann, S., Dahm, T., Kriegerowski, M., Sobiesiak, M., Tassara, C., Olcay, M. (2016): The Mw 8.1 2014 Iquique, Chile, seismic sequence: a tale of foreshocks and aftershocks . - Geophysical Journal International, 204, 3, pp. 1766-1780. DOI: http://doi.org/10.1093/gji/ggv544

GFZ German Research Centre for Geosciences; Institut des Sciences de l’Univers-Centre National de la Recherche CNRS-INSU (2006): IPOC Seismic Network. Integrated Plate boundary Observatory Chile - IPOC. Other/Seismic Network DOI: http://doi.org/10.14470/PK615318.

Schaller, T. Andersen, J., Götze, H.-J., Koproch, N., Schmidt, S., Sobiesiak, M., Splettstößer, S. (2015): Segmentation of the Andean margin by isostatic models and gradients. - Journal of South American Earth Sciences, 59, pp. 69-85. DOI: http://doi.org/10.1016/j.jsames.2015.01.008

Schurr, B., Asch, G., Hainzl, S., Bedford, J., Hoechner, A., Palo, M., Wang, R., Moreno, M., Bartsch, M., Zhang, Y., Oncken, O., Tilmann, F., Dahm, T., Victor, P., Barrientos, S., Vilotte, J. (2014): Gradual unlocking of plate boundary controlled initiation of the 2014 Iquique earthquake. - Nature, 512, pp. 299-302. DOI: http://doi.org/10.1038/nature13681

Table 1 Station table. Note that start and end times represent the maximum validity of the corresponding configurations, not the actual data availability or time in the field. Azi: Azimuth of north or ‘1’ component.
Label Lat Lon Ele Azi Rate Sensor ID Logger Id Start End Channels
AERO -20.53991 -70.17818 20 90 100 CMG-3ESP/60 T35739 DM24 A1388 2010-11-24 2011-04-20 HHZ HHN HHE
APAC -19.81656 -69.39541 2196 90 100 CMG-3ESP/60 xxxx DM24 yyyy 2011-11-25 2013-11-07 HHZ HHN HHE
BUEN -19.95618 -69.90733 1120 90 100 CMG-3ESP/60 T34776 DM24 C622 2010-12-09   HHZ HHN HHE
CERR -20.62221 -69.66251 987 90 100 L4-3D 4956 REFTEK-72A 7692 2010-12-07   HHZ HHN HHE
CHOM -21.09404 -70.0102 1200 90 100 CMG-3ESP/60 xxxx DM24 C623 2011-04-27 2013-11-19 HHZ HHN HHE
CHOM -21.09404 -70.0102 1200 90 200 CMG-3ESP/60 xxxx DM24 C623 2013-11-20   HHZ HHN HHE
CORZ -19.4945 -69.79348 0 90 100 L4-3D xxxx REFTEK-72A yyyy 2010-01-01   HHZ HHN HHE
HUAR -19.93203 -69.72611 1135 90 100 CMG-3ESP/60 T35732 DM24 C620 2009-12-03   HHZ HHN HHE
MAM1 -20.13347 -69.45546 1426 90 100 CMG-3ESP/60 T35808 DM24 A1389 2010-12-03   HHZ HHN HHE
MAM2 -20.09217 -69.395 1426 90 100 L4-3D 4954 REFTEK-72A 7682 2010-12-06   HHZ HHN HHE
NEUQ -20.1722 -70.07323 1043 90 100 CMG-3ESP/60 T34639 DM24 C617 2009-05-17   HHZ HHN HHE
PATA -20.6918 -70.00257 780 90 100 CMG-3ESP/60 T35738 DM24 A1387 2010-11-24 2013-11-19 HHZ HHN HHE
PATA -20.6918 -70.00257 780 90 200 CMG-3ESP/60 T35738 DM24 A1387 2013-11-20   HHZ HHN HHE
PICL -20.50209 -69.26111 1773 90 100 CMG-3ESP/60 T34777 DM24 C618 2009-12-01 2013-11-07 HHZ HHN HHE
PICL -20.50209 -69.26111 1773 90 200 CMG-3ESP/60 T34777 DM24 C618 2013-11-08   HHZ HHN HHE
PICL2 -20.4995 -69.26623 1744 90 200 CMG-3ESP/60 T34776 DM24 C622 2013-11-12   HHZ HHN HHE
PICL3 -20.5015 -69.27381 1708 90 200 CMG-3ESP/60 T35808 DM24 1389 2013-11-13   HHZ HHN HHE
PICL4 -20.50238 -69.28196 1654 90 200 CMG-3ESP/60   DM24 C617 2013-11-13   HHZ HHN HHE
PICL5 -20.4897 -69.28003 1656 90 200 CMG-3ESP/60 T35467 DM24 C624 2013-11-14   HHZ HHN HHE
PICL6 -20.48641 -69.26605 1748 90 200 CMG-3ESP/60 T35461 DM24 C614 2013-11-14   HHZ HHN HHE
PICL7 -20.49121 -69.25018 1827 90 200 CMG-3ESP/60 T35733 DM24 1384 2013-11-15   HHZ HHN HHE
PICL8 -20.51208 -69.27737 1612 90 200 CMG-3ESP/60 T35732 DM24 1385 2013-11-15   HHZ HHN HHE
PICL9 -20.49152 -69.26118 1749 90 200 CMG-3ESP/60 T34782 DM24 C620 2013-11-14   HHZ HHN HHE
PICLA -20.50773 -69.25383 1947 90 200 L4-3D 4955 DM24 A2386 2014-11-19   HHZ HHN HHE
PINT -20.76133 -69.39726 1146 90 100 CMG-3ESP/60 T34782 DM24 A1386 2009-12-05   HHZ HHN HHE
POZO -20.25246 -69.75943 1024 90 100 CMG-3ESP/60 T35733 DM24 A1384 2009-12-04   HHZ HHN HHE
SOG1 -19.67028 -69.61834 1590 90 100 CMG-3ESP/60 T35467 DM24 C624 2010-11-29   HHZ HHN HHE
SOG2 -19.52838 -69.37863 2722 90 100 CMG-3ESP/60 T35461 DM24 C614 2011-04-26   HHZ HHN HHE
TIRA -20.32991 -69.57166 1020 90 100 CMG-3ESP/60 T35734 DM24 A1385 2009-12-05   HHZ HHN HHE
UNAP -20.24393 -70.14041 0 90 100 CMG-3ESP/60 T34622 DM24 A1383 2009-05-14   HHZ HHN HHE

Fig. 5 Station distribution in experiment (red symbols). If present, white-filled symbols show permanent stations and other temporary experiments archived at EIDA or IRIS-DMC, whose activity period overlapped at least partially with the time of the experiment. If present, open symbols show station sites which were no longer active at the time of the experiment, e.g. prior temporary experiments.

HHZ HHN HHE
_images/AERO-HHZ.png _images/AERO-HHN.png _images/AERO-HHE.png
_images/APAC-HHZ.png _images/APAC-HHN.png _images/APAC-HHE.png
_images/BUEN-HHZ.png _images/BUEN-HHN.png _images/BUEN-HHE.png
_images/CERR-HHZ.png _images/CERR-HHN.png _images/CERR-HHE.png
_images/CHOM-HHZ-1.png _images/CHOM-HHN-1.png _images/CHOM-HHE-1.png
_images/CHOM-HHZ-2.png _images/CHOM-HHN-2.png _images/CHOM-HHE-2.png
CORZ-HHZ.png CORZ-HHN.png CORZ-HHE.png
_images/HUAR-HHZ.png _images/HUAR-HHN.png _images/HUAR-HHE.png
_images/MAM1-HHZ.png _images/MAM1-HHN.png _images/MAM1-HHE.png
_images/MAM2-HHZ.png _images/MAM2-HHN.png _images/MAM2-HHE.png
_images/NEUQ-HHZ.png _images/NEUQ-HHN.png _images/NEUQ-HHE.png
_images/PATA-HHZ-1.png _images/PATA-HHN-1.png _images/PATA-HHE-1.png
_images/PATA-HHZ-2.png _images/PATA-HHN-2.png _images/PATA-HHE-2.png
_images/PICL-HHZ-1.png _images/PICL-HHN-1.png _images/PICL-HHE-1.png
_images/PICL-HHZ-2.png _images/PICL-HHN-2.png _images/PICL-HHE-2.png
PICL2-HHZ.png PICL2-HHN.png PICL2-HHE.png
_images/PICL3-HHZ.png _images/PICL3-HHN.png _images/PICL3-HHE.png
_images/PICL4-HHZ.png _images/PICL4-HHN.png _images/PICL4-HHE.png
_images/PICL5-HHZ.png _images/PICL5-HHN.png _images/PICL5-HHE.png
_images/PICL6-HHZ.png _images/PICL6-HHN.png _images/PICL6-HHE.png
_images/PICL7-HHZ.png _images/PICL7-HHN.png _images/PICL7-HHE.png
_images/PICL8-HHZ.png _images/PICL8-HHN.png _images/PICL8-HHE.png
_images/PICL9-HHZ.npz.png _images/PICL9-HHN.png _images/PICL9-HHE.png
_images/PICLA-HHZ.npz.png _images/PICLA-HHN.npz.png _images/PICLA-HHE.npz.png
_images/PINT-HHZ.png _images/PINT-HHN.png _images/PINT-HHE.png
_images/POZO-HHZ.png _images/POZO-HHN.png _images/POZO-HHE.png
_images/SOG1-HHZ.png _images/SOG1-HHN.png _images/SOG1-HHE.png
_images/SOG2-HHZ.png _images/SOG2-HHN.png _images/SOG2-HHE.png
_images/TIRA-HHZ.png _images/TIRA-HHN.png _images/TIRA-HHE.png
_images/UNAP-HHZ.png _images/UNAP-HHN.png _images/UNAP-HHE.png

Fig. 6 Noise probability density functions for all stations for database holdings

_images/iq_out.png

Fig. 7 Overview of uptimes of all stations generated with obspy-scan