Data Description
Campaign microgravity surveys have been conducted at Kīlauea, Hawai‘i (USA), since 1975 (Dzurisin and others, 1980) and, when combined with deformation measurements, enable insights into mass change within the volcano (Jachens and Eaton, 1980; Johnson, 1992; Kauahikaua and Miklius, 2003; Johnson and others, 2010; Bagnardi and others, 2014; Poland and others 2019). For example, microgravity surveys between 1975-2008 measured residual gravity increases of up to 0.450 mGal across the volcano’s summit and have been interpreted as filling of void space by magma (Johnson and others, 2010). In March 2008 a new long-lived eruption began within Kīlauea’s Halema‘uma‘u crater (Wilson and others, 2008) which continued for 10 years until the volcano’s caldera collapse in 2018 (Neal and others, 2019). In 2009, to increase the Hawaiian Volcano Observatory’s (HVO) microgravity monitoring capabilities, HVO acquired two Scintrex CG-5 Autograv gravity meters (serial numbers 90940579 and 90940578). Since 2009, these two relative gravimeters have been used to conduct campaign microgravity surveys at a network of 33 to 55 benchmarks across the volcano’s summit region, the Kīlauea Campaign Gravity Network (KCGN).
This data release comprises raw (as recorded by the gravimeters) and processed gravity data from campaign microgravity surveys of the KCGN between 2009-2017 using HVO’s Scintrex CG-5 relative gravimeters. Gravity data were collected using two gravimeters, following a daily double-loop procedure consisting of three occupations of a reference station (at the start and end of each loop) and two occupations of selected benchmarks (once during each loop). All gravity measurements, unless otherwise noted, are relative to a reference station, benchmark “P1”, located 4 km northwest of the center of the caldera. This benchmark is assumed to be far enough away from active volcanic and (or) magmatic processes as to be considered gravitationally stable.
Data Processing
Gravity data were processed to obtain relative differences between the reference station and measurement benchmarks using the gTOOLS MATLAB package (Battaglia and others, 2022). If needed gTOOLS, was used to first re-correct gravity measurements for solid Earth tides (see discussion of surveys in 2009, 2011, and 2012, below), and then for ocean loading and linear instrument drift, before computing the weighted least squares-adjusted gravity values and their standard deviations. While the Scintrex CG-5 gravimeters can provide solid Earth-tide corrected measurements using the approach by Longman (1959), if needed gTOOLS can recalculate tides during data reduction. gTOOLS improves upon the tidal corrections employed by Scintrex by also a) computing the Moon and Sun longitudes with the original equations by Bartels (1957), b) employing updated values from U.S. Naval Observatory for the astronomical constants and c) considering anelastic-Earth effects from tides (Agnew, 2007). The ocean loading correction is computed using the rewritten version of the Fortran code HARDISP by Petit and Luzum (2010). Ocean loading harmonic coefficients are from Bos and Scherneck's (2013) ocean loading provider (http://holt.oso.chalmers.se/loading/) using the TPXO9-Atlas model (Egbert and Erofeeva, 2002) and a STW105 viscoelastic Green’s function (Kistowski and others, 2008).
No systematic kinematic Global Navigation Satellite Systems (GNSS) positions were taken consistently of benchmark locations prior to the 2018 caldera collapse. Free-air gravity effects due to vertical deformation between survey dates have therefore previously been calculated using ascending and descending Interferometric Synthetic Aperture Radar (InSAR) orbital passes (Bagnardi and others, 2014; Koymans and others, 2022) but are not provided in this data release. Similarly, gravity measurements have not been corrected for effects from fluctuations in the level of the 2008-2018 lava lake at the summit of Kīlauea, which can be significant (Carbone et al., 2013; Bagnardi and others, 2014). There are also no corrections applied due to possible variations in the height of the water table, which was assumed to range from 430-830 m above sea level before the 2018 caldera collapse (Kauahikaua, 1993), and not a significant source of gravity changes during this time (Bagnardi and others, 2014). For example, benchmark HVO47, near the center of the caldera, was at an elevation of ~1079 m above sea level prior to the 2018 caldera collapse. Additionally, Koymans and others (2022) identified 14 suspected intra-survey sudden offsets in gravity readings (tares) ranging between 0.020 to 0.130 mGal. We have not corrected for possible tares in this data release.
Campaign microgravity surveys in 2009 (meter 90940578), 2011 (both meters), and the two surveys in 2012 (both meters), had time zone errors in the instrument system time which led to incorrect instrument-provided solid Earth-tide corrections. These errors have been corrected using gTOOLS, and the incorrect gravity and tide corrections have been replaced in the provided raw and processed data (see the “surveys” table description). Additionally, for logistical reasons, for the one survey day on the caldera floor (benchmarks: BM82-501, HVO44, 110YY, HVO45, 109YY, HVO46, HVO47, HVO48, and HVO49) during the 2011 and two 2012 surveys, gravity was measured relative to benchmark HVO41. We have corrected for the offset between HVO41 and reference station, benchmark P1, in the processed data but not in the raw data.
Location information (latitude, longitude, elevation) for benchmarks was taken from the Hawaiian Volcano Observatory’s database of benchmarks and are primarily from pre-2018 measurements using a consumer-grade handheld Global Position System receiver and may be affected by datum transformations between Old Hawaiian and World Geodetic System (WGS) 84. These locations are provided for general reference only and not geodetic-precise elevations.
Database Description
Data are provided in a Structured Query Language (SQL) database self-contained file (“.sql”). The SQL database was constructed using MySQL, is relational, and contains a single schema comprised of eight data tables, described at the end of this summary. The database requires the use of an SQL server and can be reconstructed graphically using MySQL Workbench or any other compatible SQL relational data management platform (system dependent). Table names; field names, types, and precisions; and database relationships, are documented in the included diagram “hvo_kcgn_2009_2017_model.pdf”.
Tables
meters; serial numbers and short names for gravimeters in the database. For example, “CG5-0579” for the Scintrex CG-5 Autograv Gravity Meter with serial number 90940579.
meter_calibrations; calibration dates and values for the gravimeters in the database, calculated from occupations at Mauna Kea (Hawai‘i), Mount Hood (Oregon), and Mount Hamilton (California).
stations; benchmark (station) names and locations.
networks; network short names (“network_id”), for example, “KCGN” for Kīlauea Campaign Gravity Network, and the number of stations in the post-2018 network design.
network_makeup; station names, network short names they belong to, and a flag (“current”) indicating whether the station is used in the current (as of this data release) post-2018 network.
surveys; survey metdata including network short names, meter short names, survey begin and end dates (Hawaii Standard Time), gravimeter operator, standard Scintrex gravimeter metadata (Scintrex 2012), and whether solid Earth-tides were recalculated for the survey.
raw_data; standard Scintrex gravimeter data (Scintrex 2012). The field “date_HST” indicates the survey day in Hawaii Standard Time and can parsed to extract a single double-loop survey day. Double loops were always completed in one survey day but given the time difference between HST and Coordinated Universal Time (UTC), days often span two UTC dates. Positional information; latitude (“meter_lat”), longitude (“meter_lon”) and elevation (“meter_alt”) are from the instruments’ built-in GNSS receiver and may be different than the position in the “stations” table due to positional accuracies. “The “ref_station” flag denotes whether the survey day used a reference station other than benchmark “P1”. Null fields correspond to data used in the Scintrex CG-6 but not CG-5 Autograv gravity meters (Scintrex 2012; 2019). Gravity data are in mGals. For more information on raw data field descriptions please refer to the Scintrex CG-5 and CG-6 Autograv System Operation Manual l (Scintrex, 2012; 2019)
processed_data; Scintrex gravimeter data processed using the gTOOLS MATLAB package. Gravity data are in mGals.
NOTE: Following the 2018 Lower East Rift Zone eruption and caldera collapse several benchmarks were destroyed and (or) no longer easily accessible and the KCGN was modified to omit these locations and new benchmarks were added. HVO has continued post-2018 campaign microgravity surveys, occupying the network in 2018, 2019 and 2022. Data from post-2018 will be released as an update to this database in the future. To facilitate this update, this data release contains benchmarks from both the pre and post-2018 networks. Similarly, to facilitate HVO’s transition to Scintrex CG6 Autograv gravity meters, this database contains several currently unused fields specific to those instruments (see the “raw_data” table description).
Database Query Examples
We do not provide information on how to use SQL and (or) detailed information on how to access tables and data. However, we have included a selection of useful SQL query examples;
1. To access the ranges of survey dates available in the database, their corresponding survey identification numbers, and the gravity meter used;
SELECT id, date_start_HST, date_stop_HST, meter_id FROM surveys;
2. To access raw data from a single survey, using the survey identification number (for example id=5);
SELECT * FROM raw_data WHERE survey_id=5;
3. To access raw_data from a single-day within a survey (for example 2011-03-25), from a specific gravity meter (for example serial number 90940578);
SELECT * FROM raw_data WHERE date_HST="2011-03-25" AND meter_id="CG5-0578";
4. To access processed data for a single-station (for example HVO31) from a specific gravity meter (for example serial number 90940578);
SELECT * FROM processed_data WHERE station_id="HVO31" AND meter_id="CG5-0578";
5. To access all processed data with corresponding station location information (from table stations);
SELECT * FROM processed_data INNER JOIN stations ON stations.id=processed_data.station_id;
References
Agnew, D.C., 2012. SPOTL: Some programs for ocean-tide loading.
Bagnardi, M., Poland, M.P., Carbone, D., Baker, S., Battaglia, M. and Amelung, F., 2014. Gravity changes and deformation at Kīlauea Volcano, Hawaii, associated with summit eruptive activity, 2009–2012. Journal of Geophysical Research: Solid Earth, 119(9), pp.7288-7305.
Bartels, J., 1957. Gezeitenkräfte. In Geophysik II/Geophysics II. Springer, Berlin, Heidelberg.
Battaglia, M., Calahorrano-Di Patre, A. and Flinders, A.F., 2022. gTOOLS, an open-source MATLAB program for processing high precision, relative gravity data for time-lapse gravity monitoring. Computers & Geosciences, p.105028.
Bos, M.S. and Scherneck, H.G., 2013. Computation of Green’s functions for ocean tide loading. In Sciences of Geodesy-II (pp. 1-52). Springer, Berlin, Heidelberg.
Dzurisin, D., Anderson, L.A., Eaton, G.P., Koyanagi, R.Y., Lipman, P.W., Lockwood, J.P., Okamura, R.T., Puniwai, G.S., Sako, M.K. and Yamashita, K.M., 1980. Geophysical observations of Kilauea Volcano, Hawaii, 2. Constraints on the magma supply during November 1975–September 1977. Journal of Volcanology and Geothermal Research, 7(3-4), pp.241-269.
Egbert, G.D. and Erofeeva, S.Y., 2002. Efficient inverse modeling of barotropic ocean tides. Journal of Atmospheric and Oceanic technology, 19(2), pp.183-204.
Jachens, R.C. and Eaton, G.P., 1980. Geophysical observations of Kilauea volcano, Hawaii, 1. Temporal gravity variations related to the 29 November, 1975, M= 7.2 earthquake and associated summit collapse. Journal of Volcanology and Geothermal Research, 7(3-4), pp.225-240.
Johnson, D.J., 1992. Dynamics of magma storage in the summit reservoir of Kilauea Volcano, Hawaii. Journal of Geophysical Research: Solid Earth, 97(B2), pp.1807-1820.
Johnson, D.J., Eggers, A.A., Bagnardi, M., Battaglia, M., Poland, M.P. and Miklius, A., 2010. Shallow magma accumulation at Kīlauea Volcano, Hawai ‘i, revealed by microgravity surveys. Geology, 38(12), pp.1139-1142.
Kauahikaua, J., 1993. Geophysical characteristics of the hydrothermal systems of Kilauea Volcano, Hawaii. Geothermics, 22(4), pp.271-299.
Kauahikaua, J. and Miklius, A., 2003. Long-Term Trends in Microgravity at Kīlauea's Summit During the Pu'u “Õʻō-Kūpaianaha Eruption. US Geological Survey Professional Paper, 1676(1676), p.165.
Koymans, M.R., de Zeeuw-van Dalfsen, E., Evers, L.G., and Poland, M.P., 2022. Microgravity change during the 2008-2018 Kīlauea summit eruption: a decade of subsurface mass accumulation. Journal of Geophysical Research: Solid Earth, submitted.
Longman, I.M., 1959. Formulas for computing the tidal accelerations due to the moon and the sun. Journal of Geophysical Research, 64(12), pp.2351-2355.
Neal, C.A., Brantley, S.R., Antolik, L., Babb, J.L., Burgess, M., Calles, K., Cappos, M., Chang, J.C., Conway, S., Desmither, L. and Dotray, P., 2019. The 2018 rift eruption and summit collapse of Kīlauea Volcano. Science, 363(6425), pp.367-374.
Petit, G. and Luzum, B., 2010. IERS 2010 conventions. IERS Technical Note, (36), p.179.
Poland, M.P., de Zeeuw‐van Dalfsen, E., Bagnardi, M. and Johanson, I.A., 2019. Post‐collapse gravity increase at the summit of Kīlauea volcano, Hawaiʻi. Geophysical Research Letters, 46(24), pp.14430-14439.
Scintrex, 2012. CG-5 Scintrex Autograv System Operation Manual: Revision 8. Scintrex Limited, Canada.
Scintrex, 2019. CG-6 Scintrex Autograv System Operation Manual: Revision C. Scintrex Limited, Canada.
Wilson, D., Elias, T., Orr, T., Patrick, M., Sutton, J. and Swanson, D., 2008. Small explosion from new vent at Kilauea's summit. EOS, 89(27), p.203.
