Short Review of True-North Alignment Method on the Field Demonstrating the benefits of the use of an Optical Gyrocompass

EGU21-16524 - 28/04/2021

Pierrick Auregan, Elliot de Toldi, Theo Laudat, Laurent Mattio, Frederic Guattari

Summary

• One slide for 2min teasing

• Stations misorientations using a magnetic compass

• Modern seismology needs better orientation

• Post-processing station orientation accuracy

• Optical gyrocompass needed improvments

• Tools to transfer the North line

• Relative alignement performances

• References and links

Annex A : Gyrocompass specification undestanding (seclat and RMS)

Annex B : Magnetic North motion impact on declination error in time

Annex C : Total budget error of the station orientation

Short Review of True-North Alignment Method on the Field Demonstrating the benefits of the use of an Optical Gyrocompass

Modern seismic data

analysis needs good

station orientation

“ Better than 1° ”

Do you know why

gyrocompass

specifications are in

“degree seclat” ?

Did you know that

worldwide stations installed

with a magnetic compass

have a standard deviation

orientation of ~6° ?

Magnetic North is

moving at

55km/year, so

what is the

orientation error

using one year old

declination table ?

EKSTROM & BUSBY. SRL, 2008

What is the

misorientation due to

the transfer of the

North line from the

gyrocompass to the

instrument ? Is optical gyrocompass still

expensive, heavy, and export

limited ?

Stations misorientations using a magnetic compass

All seismic networks (Europe, Asia, US etc…) are subject to large station misorientations

o 32% of Spanish Broadband National Network > 15° misorientation

(Rueda and Mezcua (2015))

o Misorientation of more than 100 stations of EOBSarray has σ = 11,6° (cf ref [1])

o 41.3% of ANSS stations have misorientations > 4° (cf ref [2])

o 11/30 stations of AlpArray in Austria have misorientations > 7° (cf ref [5]),

and 12/26 stations of AlpArray in Hungary have misorientations > 6° (cf ref [6])

o 24% of 473 stations of USArray have misorientations >4°, and 10% > 7° (cf ref [10])

o 270/800 = 34% of CNDSN have misorientations > 8° (cf ref [12])

Worldwide station misorientation sigma ≈ 6°

Ref [10]

The use of a magnetic compass is the reason why there is such a discrepancy

o Improper tool to transfer North line and rough visual alignment

o Mistakes on declination calculation (including the fact that North pole is moving years after years)

o Local magnetic disturbances

“ The determination of the northward direction has been routinely performed for years with the use of standard compass,

with the best accuracy being 5° in the case of no magnetic disturbances in the nearby surroundings.

However, such accuracy is no longer sufficient.” (from Ref [4])

Modern seismology needs better orientation

5° was not enough => 0.5° is the new target

Figure 2.14 : Array-derived rotation uncertainty as a

function of varying sensors’ misalignment

“In short, the two most influential sources of uncertainty are the instrumental phase response and the stations’ misalignment.

For those two sources of uncertainty, the standard guidelines are insufficient when applied to array-derived rotation.” (From ref [14])

”The amount of uncertainty associated to array-derived rotation decrease by one order of

magnitude if the gyrocompass is used instead of the compass […]. Furthermore, for a chosen

amount of uncertainty, the normalized wavelength band is increased and encompasses to a

greater extent the long-period normalized wavelengths” (From ref [14])

Modern seismic data analysis needs good orientation

o Receiver functions, amplitudes analysis, direction of propagations [1]

o Moment tensor inversion: 10° leading to up to 40% error [3]

o Modern 3-Component observations [4]

o Seismic wave separation [12]

o Array Derivative Rotation [14]

Ref [14]

Post-processing station orientation accuracy

A lot of work for a limited accuracy

Different techniques are described, depending mainly of events used, but all of them show an accuracy > 2°

o In ref [11], (NIU 2011), even the analys of more than 20 events cannot decrease the error bar under 3°

o In ref [1], (XU 2018), short and long period signals are compared but ther error bar cannot be lowered under +-5°

Ref [11] Ref [1]

Optical gyrocompass needed improvments

When good performance was not enough to be adopted

Optical gyrocompass is the most reliable and efficient tool to orientate a station, but it has drawbacks limiting its use as a standard

tool for all networks

o Expensive [1][2][4][9][11]

o Heavy [1][11]

o Export restriction [2][9]

.

So iXblue developed a dedicated

optical gyrocompass for station orientation : Seistans

o Cheaper, Lightest, Free of export

o GNSS denied compatible (for basement use)

o Optimized for static use

o Wide range of accessories to :

➢ Display heading

➢ Transfer the North line

As mentioned in the conclusion of ref [14], the smaller the array, the greater is the impact of misorientation of the station.

So, as stated by ref [14], ”Aligning all stations with a chosen station of reference, can be done (Donner et al., 2017)” to improve even

more the computation of the data array.

An optical gyrocompass as Seistans can be used to do this. Using either a contact plate or the laserline. The use of a dedicated contact

plate to match the same instrument used at each stations of the array will bring the higher relative accuracy.

The relative accuracy Seistans can offer is 0.2° max (and NO seclat here!!!)

For a distance < 30m, 180° rotation maximum, and duration < 10minutes, between two positions

Relative alignement performances

When instruments orientation are compared directly without reference to True North

Heading (°)

position 1 position 2

5 minutes 314,994 45,177

10 minutes 315,136 45,275

15 minutes 315,123 45,209

20 minutes 315,147 45,236

30 minutes 315,164 45,248

Tools to transfer the North line

+ nothing +grooves +contact +laserline

baseplate housing onto blueSeis-3A Reference plate Contact plate LaserLine

+-0.12° +-0.30° +-0.03° +-0.04° +-0.08° +-0.25°

Raw' contact with

Precision pinsWithout precision pins

Precision pins to transfer from the mechanical reference of the gyrocompass to the instrument

Tools to transfer the North line : Focus on laserline toolkit

Laserline is usually prefered on the

field:

o Versatile

o Easy to use

o No need for a handy operator

o Contact less

o Contained uncertainties

o Factory alignement of the laser

line with the precision pins of

the interface plate

Seistans Hands on Vidéo- Part 1: Find the True North using a gyrocompass

- Part 2: Transfer the North Line to an instrument

AGU2020 posterS012-0009 - True-North Alignment on the Field: From a

Compass to an Optical Gyrocompass

• [1] XU, Hongrui, LUO, Yinhe, TANG, Chi‐Chia, et al. Systemic Comparison of Seismometer Horizontal Orientations

Based on Teleseismic Earthquakes and Ambient‐Noise DataSystemic Comparison of Seismometer Horizontal

Orientations. Bulletin of the Seismological Society of America, 2018, vol. 108, no 6, p. 3576-3589.

• [2] RINGLER, Adam T., HUTT, Charles R., PERSEFIELD, Kyle, et al. Seismic station installation orientation errors

at ANSS and IRIS/USGS stations. Seismological Research Letters, 2013, vol. 84, no 6, p. 926-931.

• [3] ZAHRADNÍK, Jiří et CUSTÓDIO, Susana. Moment tensor resolvability: Application to southwest Iberia. Bulletin

of the Seismological Society of America, 2012, vol. 102, no 3, p. 1235-1254.

• [4] VECSEY, Luděk, PLOMEROVÁ, Jaroslava, JEDLIČKA, Petr, et al. Data quality control and tools in passive

seismic experiments exemplified on the Czech broadband seismic pool MOBNET in the AlpArray collaborative

project. Geoscientific Instrumentation, Methods and Data Systems, 2017, vol. 6, no 2, p. 505.

• [5] FUCHS, Florian, KOLÍNSKÝ, Petr, GRÖSCHL, Gidera, et al. AlpArray in Austria and Slovakia: technical

realization, site description and noise characterization. Advances in Geosciences, 2016, vol. 43.

• [6] GRÁCZER, Zoltán, SZANYI, Gyöngyvér, BONDÁR, István, et al. AlpArray in Hungary: temporary and

permanent seismological networks in the transition zone between the Eastern Alps and the Pannonian basin. Acta

Geodaetica et Geophysica, 2018, vol. 53, no 2, p. 221-245.

• [7] MOLINARI, Irene, CLINTON, John, KISSLING, Edi, et al. Swiss-AlpArray temporary broadband seismic

stations deployment and noise characterization. Advances in Geosciences, 2016, vol. 43, p. 15-29.

• [8] COCHRAN, Elizabeth S., WOLIN, Emily, MCNAMARA, Daniel E., et al. The US Geological Survey’s rapid

seismic array deployment for the 2019 Ridgecrest earthquake sequence. Seismological Research Letters, 2020.

• [9] Davis, P., and L. Gee (2009). GSN network operator sensor orientation best practices,

http://www.iris.edu/hq/files/programs/gsn/gsnqual/GSN_alignment_practices_v1.0-1.pdf

• [10] EKSTRÖM, Göran et BUSBY, Robert W. Measurements of seismometer orientation at USArray transportable

array and backbone stations. Seismological Research Letters, 2008, vol. 79, no 4, p. 554-561.

• [11] WANG, Xin, CHEN, Qi‐Fu, LI, Juan, et al. Seismic sensor misorientation measurement using P‐wave particle

motion: An application to the NECsaids Array. Seismological Research Letters, 2016, vol. 87, no 4, p. 901-911.

• [12] NIU, Fenglin et LI, Juan. Component azimuths of the CEArray stations estimated from P-wave particle motion.

Earthquake Science, 2011, vol. 24, no 1, p. 3-13.

• [13] http://nnsn.geo.uib.no/eworkshop/index.php?n=Main.Orientation

• [14] Roxanne RUSCH. Array-derived rotations carried out using the LSBB seismic array: quantification and graphical representation of the uncertainty. Sciences de la Terre. Université Côte d’Azur, 2020. Français. NNT:2020COAZ4054 https://tel.archives-ouvertes.fr/tel-03177660

User manual SeistansDatasheet Seistans

References

Product documentation

References and links

Review of station orientation.pdf

Detailled review of quoted papers

True North precision depends on latitude So all specification of True North finder are

always expressed in seclat

=> Precision(Lat°) = specification / cosine(Lat°)

12

Heading

accuracy (°)

for 0.23° seclat

North Pole 90 360,0

Spitzbergen 80 1,32

Igloolik 70 0,67

Alaska 60 0,46

Canada 50 0,36

New-York 40 0,30

Texas 30 0,27

Hawaï 20 0,24

Costa-Rica 10 0,23

Equator 0 0,23

Latitude (°)

Annex A: Gyrocompass specification undestanding (seclat and RMS)

Specified heading accuracy =0,23 ° seclat (RMS) = 0,69 ° seclat (Max)

σ

When no extra indication, gyrocompass specification at always expressed in RMS

Magnetic North is travelling at ~55km/year

The worst case corresponds to an observation of the magnetic north at the onligtude perpendicular to Magnetic North pole motion

At this longitude, the effect of 55km/year variation corresponds to an error of 0.55° seclat when the position of previous year is used instead of current one.

It can be confirmed by a direct observation of the graphic. At 85° latitude, 0.55° seclat corresponds to ~6° heading error, which corresponds well to the heading displacement of the magnetic North at this latitude.

Annex B: Magnetic North motion impact on declination error in time

Here is the good way to estimate the maximum error of a single heading measurement of True North orientation

=𝑅𝑀𝑆 𝑠𝑝𝑒𝑐𝑖𝑓𝑖𝑐𝑎𝑡𝑖𝑜𝑛 . 3

𝐶𝑜𝑠𝑖𝑛𝑒 (𝐿𝑎𝑡𝑖𝑡𝑢𝑑𝑒)+ 𝑇𝑟𝑎𝑛𝑠𝑓𝑒𝑟 𝑜𝑓 𝑁𝑜𝑟𝑡ℎ𝑙𝑖𝑛𝑒𝑢𝑛𝑐𝑒𝑟𝑡𝑎𝑖𝑛𝑡𝑦

Annex C: Total budget error of the station orientation

Here is the good way to estimate the maximum error of a singleheading measurement of relative orientation using Seistans*

= 0.2 + 2 . 𝑇𝑟𝑎𝑛𝑠𝑓𝑒𝑟 𝑜𝑓 𝑁𝑜𝑟𝑡ℎ𝑙𝑖𝑛𝑒𝑢𝑛𝑐𝑒𝑟𝑡𝑎𝑖𝑛𝑡𝑦

*Under motion condition described slide 8

At high latitude and/or with small

scale array, it can bebetter to use relative orientation instead of

True North