Work Description

Date: Jan. 1, 2020

Dataset Title:
2015 Transient Induced Current (TIC) Events

Dataset Creators:
McCuen, Brett A.

Dataset Contact:
bmccuen@umich.edu

Funding:
National Science Foundation (NSF)

Research Overview:
Severe space weather events like geomagnetic storms and substorms cause large disturbances
of the surface magnetic field that generate geomagnetically induced currents (GIC) in
electrically conducting material on Earth. Large GICs capable of damaging trans-formers and
causing large-scale power grid failure generally have timescales of minutes to tens of minutes
and short-timescale (< 1 minute) induced currents have not been considered a substantial threat.
However, recent evidence suggests that transient induced currents (TIC) caused by
second-timescale surface magnetic field perturbations are a potential hazard to technological
infrastructure and may have an alternate coupling mechanism than typical GIC. In this study,
we identify these TIC events in ground magnetometer data from the Magnetometer Array for Cusp
and Cleft Studies (MACCS) through-out 2015. We characterize a set of these large-amplitude,
short-timescale (< 1 minute) surface magnetic field disturbances and investigate their
association to space weather events in order to better understand their impact on electrical
systems on Earth.

Methodology:
The dB/dt search algorithm requires MATLAB R2017a or later. It can be run using the
dBdt_main.m file while the accompanying scripts called by dBdt_main.m (clean_maccs.m,
simple_dbdt.m, extremes1.m, newdbdt.m) are in the same folder. The first ten lines of the
dBdt_main.m script should be specified to the station, minimum value for duration of magnetic
disturbance (LT = ‘ ‘), maximum value for duration of disturbance (HT = ‘ ‘), minimum B value
(dt = ‘ ‘) and minimum absolute magnitude of disturbance (ldbdt = ‘ ‘). The window size for
smoothing (n = ‘ ‘) can also be specified, the current working status of the algorithm has
the smoothing turned off by commenting out the last line in the clean_maccs.m script, turn
smoothing on by removing the ‘%’ preceding this line if you want to incorporate smoothing of
the magnetic field data before searching for disturbances.

The dBdt_main.m script operates on the MACCS data or any other magnetic field data if
formatted correctly to pass through the functions. The data should be separated in arrays as
date_timex, date_timey, date_timez (in MATLAB datenum format) and Bx, By, Bz in numerical
format (units of nT). The final product returned from the routine is a 7 column matrix, each
row gives values for the start and end time of the event, start and end B value, the time
elapsed of the event, dt, the magnetic field change of the event, dB, and finally the dB/dt.
Depending on the timescale and amplitude of the disturbances being searched as well as the
incorporation of and window size of moving mean smoothing, the algorithm may return events
that are the result of magnetometer noise as a result of either instrument malfunction or
magnetic deviation from interference in the vicinity of the magnetometer. As a result, the
events should be manually inspected to determine if they are of physical nature of a result
of noise.

Instrument and/or Software specifications:
he MACCS magnetometers collects 8 samples per second in three axes, then averages and records
the data at two samples per second. The half-second sampling rate and high sensitivity
(a noiselevel of 0.01 nT) of the MACCS magnetometers is sufficient to detect shorter period
Pc 1 and 2pulsations. The magnetometers are aligned with the magnetic field so that the
x-component isin the north-south direction. All available data from these stations throughout
the entire year of 2015 were collected for analysis.

Files contained here:
2015_AllEvents.csv - Compiled events, showing the number of event, station at which event occurred,
start and end time of each dB/dt, start and end B value, dB, dt, dB/dt. The final column is
the local magnetic time (hour fraction). Please note two events removed on 04/23/2021 from previous
version because they were determined to be events due to noise.
2015_AllEvents_MIEandStormRelation.csv - the substorm onset delay (from SuperMAG substorm event
list), the relation to nighttime magnetic impulse events (MIE, from Engebretson et al., 2019), and
the relation to geomagnetic storms (from SuperMAG SYM-H index).
dBdt_main.m - main file to run algorithm
clean_maccs.m - cleans data
simple_dbdt.m - calculates the dB/dt between each and every data point
extremes1.m - determines when the sign of the slope of each dB/dt changes and filters for the
time thresholds and minimum dB/dt value
newdbdt.m - recalculates dB/dt between each point where the slope changes direction
20191127-15-56-substorms.csv - downloaded table from SuperMAG substorm event list with the start
time of every substorm onset during 2015
onset_delay.m - matlab script to run to determine the substorm onset delay of each TIC event

Related publication(s):
Engebretson, M. J., Pilipenko, V. A., Ahmed, L. Y., Posch, J. L., Steinmetz, E. S., Moldwin, M. B.,
… Vorobev, A. V. (2019). Nighttime Magnetic Perturbation Events Observed in Arctic Canada: 1.
Survey and Statistical Analysis. Journal of Geophysical Research: Space Physics, 124(9), 7442–7458.
https://doi.org/10.1029/2019JA026794

Use and Access:
This data set is made available under a Creative Commons Attribution-NonCommercial 4.0 International
license (CC BY-NC 4.0).

To Cite Data:
McCuen, B. 2015 Transient-Large-Amplitude (TLA) Geomagnetic Perturbation Events [Data set]. University of Michigan - Deep Blue.
https://doi.org/10.7302/9t46-0092

Curation Notes:
On Feb. 15 and 18, 2021, title, metadata and readme instances of "Transient Induced Current (TIC)" were updated
to "Transient-Large-Amplitude (TLA)" based on manuscript review feedback.

On May 3, 2021, Previous files 2015_AllEvents.csv and 2015_AllEvents_MIEandStormRelation.csv were deprecated
and final versions (with two events removed) were added. These two events were removed from the
event list because they were determined to be due to error in the magnetometer data.

On June 6, 2021, previous file 2015_AllEvents_MIEandStormRelation_Final.csv was updated with the inclusion of the PGG stations for comparison to MIE events, rather than just the three stations used previously. This edit was based on manuscript review feedback.