Methodology:

A 2.2 km length of Linden SPE-7196 single-mode fibre optic cable was deployed into a shallow (~30 cm deep) trench dug into snow, using a bespoke trenching sled towed behind a snowmobile. The cable was deployed in a zig-zag geometry (see cable coordinates and related datasets) through an array of nodal seismic receivers: the deployment facilitates data measurement along the long- and short-axis of the array, and along its diagonal. Installation was completed on 2024-02-05: snow was shovelled over the trench to reduce wind noise, and the cable was left to sinter into the snow to improve coupling, ahead of seismic shooting which commenced on 23:46 UTC on 2024-02-06.

The cable was monitored using a Febus Optics A1-R interrogator (a Rayleigh backscatter instrument which measures the strain rate in the cable, thus the cable becomes a string of pseudo-geophones). The total length of cable connected into the interrogator is 3.4 km, comprising in order i) a 1 km launch lead (a buffering length of cable to ensure faithful amplitude recording along the length of the fibre under test), ii) the main 2.2 km of fibre under test, and iii) 0.2 km of redundant cable which remained on a spool.

The A1-R interrogator records in a passive mode (i.e., continuous recording) while 212 active-source seismic shots were taken. These were taken at 35 locations along the length of the cable, using a 12 lb sledgehammer striking a 20 x 20 x 2 cm impact plate. Up to 10 repeat shots were taken at each location to facilitate stacking and increase signal to noise ratio, but the data are provided as unstacked records within the larger passive-source data volume. Documentation includes the quality of each shot as documented by the field crew, and any evident environmental / ambient noise source that was present while shooting was taking place (e.g., nearby snowmobile activity).

Data collection:

Data recorded with FEBUS Optics A1 Interrogator, connected to Linden SPE-7196 single-mode fibre optic cable. Controlled seismic sources were generated by a 12 lb sledgehammer striking a 20 x 20 x 2 cm resin impact plate placed at the snow surface.

Total cable length: 3400 m, of which

0-1000 m constitutes the launch cable,

1000-3200 m constitutes the main fibre under test,

3200-3400 m constitutes 200 m of redundant cable which remains on a spool, and

Seismic shots at a 47 m interval along the cable, in the intervals 1000-2230 m and 2820-3200 m.

Data quality:

The location of shots is accurate to ~0.1 m. Energy from shots in raw data can be observed to offsets up to ~150 m from the shot location, which should improve when data are stacked. The shooting log indicates the quality of each hammer impact, as reported by the operator and ranked between 1-3 (3 being the highest quality): the aim was to record at least 5 shots of rank 3 at each location. Occasional double bounces are reported (numeric value 4), in which the hammer strikes twice in quick succession having not been caught after its initial impact: these should be rejected during data analysis as the waveform would complicate interpretation. Nonetheless, the files are still included in this archive to prevent gaps in the passive record. Other acquisition notes include comment on ambient noise in the data, notably where a nearby snowmobile engine was running at the time of shooting.

The impact time of seismic shots is recorded to the nearest minute, such that they can be identified in the passive record. Data filenames inherently record the corresponding date and UTC time, taking the format EWG_ACTIVE_yyyy-mm-dd_hh-mm-ss_UTC.h5.