the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 16 Jun 2025
Evaluating data quality and reference instrument robustness: insights from 12 years DI magnetometer comparisons in the Geomagnetic Network of China
Abstract. A statistical analysis was conducted on 12 years of geomagnetic instrument comparison data from the Chinese Geomagnetic Network (GNC). The study reveals that when applying a 90 % threshold, the corresponding instrument deviation thresholds are 0.21 ′ (D component) and 0.11′ (I component), which can serve as evaluation criteria at the network level. By integrating multi-source uncertainty decomposition (including instrument errors, operator-dependent errors, and pillar correction errors) and weighted ensemble analysis, systematic deviations between reference fluxgate theodolites and test instruments were quantified. Results demonstrate that reference instruments exhibit high stability and reliability, with mean deviations of -0.004 ′ (D) and 0.022 ′ (I), both within the 95 % confidence interval, and no long-term drift was observed. Operator-dependent errors were successfully isolated, with 0.13 ′ (D) and 0.06 ′ (I), consistent with observed experimental findings, confirming that operator-dependent errors constitute the primary contributor to instrument deviations. Notably, operator-dependent errors in D are significantly higher than I due to the complexity of azimuth alignment. These findings highlight the critical role of instrument comparisons in effectively monitoring equipment performance and assessing observational quality across stations, while validating the feasibility of standardized instrument evaluation methods and the long-term stability of reference instruments. Future efforts should integrate sensors and automation technologies to minimize human errors, thereby providing a higher-quality data foundation for geophysical studies.
- Preprint
(926 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-2406', Anonymous Referee #1, 05 Aug 2025
Comments
The subject of geomagnetic absolute instrument comparisons across national and international observatory networks is an important issue worthy of the exploration presented in this paper. The paper presents useful introductory information and explanation on the theory and methods used for the analysis together with specific details on the observatories and instrumentation from the Chinese network used in the study. Methods of calculation and statistical analysis of results from the multi-year data set are presented effectively in diagrams and discussion.
Corrections:
Include the word "of" in the title "... from 12 years of DI magnetometer..."
Line 69: replace "..A of the marker given," with "... of the marker is known,..."
Line 70: "(with the vertical circle maintained at ...")
Line 75: "Inclination measurements follows analogous procedures in the vertical plane, omitting azimuth mark referencing and using the magnetic meridian derived from the preceding declination measurements"
Line 80: "The declination measurement protocol is preceded and followed by sensor up and down azimuth mark readings and then involves four configurations...."
Line 83 : "D'=(Dz + D2 + D3 + D4)/4 and substituting the value D' into equation (1)
Line 92 ; The meaning of the word "integration" in this context is not clear to me.
Line 96: "...systematic verification of inter-instrument differences across..."
Line 107: "Where, W0(i:j) ,,,,"
Line 110: "... of inter-pillar differences..."
Line 122: Table 1:
A map of the observatory locations would be interesting, but obviously not necessary for the arguments presented in the paper.
Line 125: Table 2: Correct the spelling of "Instrument" in table header
I suggest a more precise description of the instruments would be useful in Table 2, including both the theodolite make/model and also the fluxgate make/model.
Line 152: "susceptible to operator error compared to ...."
Another significant source of possible error in declination readings, which is not present in inclination readings, is the accuracy of setting (90/270) on the vertical circle.
Line 158: remove "its" ..."necessary to analyses long term stability ...."
Line 163: "... 12 years of intercomparison ..."
Line 166: "...using multi-year..."
Line 167: "...integrating multi-year..."
Line 169: "... and multi-source uncertainty..."
Line 175: A reference would be helpful for Type B (line 175 ) and Type A (line 190) uncertainties
Line 180: Table 3: Is it possible to relate the "Class" column in Table 3 to instrument types listed in Table 2?
Line 193: "analyzing inter-station variances"
Line 226: "The multi-year mean..."
Line 240: "However, mean differences for both D and ..."
Line 258: "...azimuth mark alignment step, and the accuracy the vertical circle setting (90/270), required for declination measurements
Line 277: "Furthermore, multi-year data analysis revealed the mean differences and inter-annual...."
Line 293: the use of the phrase "... and so on" seems inappropriate
The words "discrepancy" (lines 96, 114, 100) "difference" (line 116), "deviation" (figure 2), are used in different places throughout the text. I suggest consistent use of "difference" would be better.
Citation: https://doi.org/10.5194/egusphere-2025-2406-RC1 -
RC2: 'Comment on egusphere-2025-2406', Anonymous Referee #2, 13 Aug 2025
General Comments
The paper analyses and discusses the results of DIM inter-comparison campaigns conducted between 2010 and 2024, involving 46 GNC-operated geomagnetic observatories in China. Following a concise introduction to the instrumentation and measurement principles, the authors enumerate the sources of error. The introduction has some shortcomings and needs to be complemented as detailed below.
Although the focus of the present work is on error analysis, the authors did not utilise the error information that can be derived from the absolute measurement sequences recorded in the measurement protocol sheets, such as sensor misalignment errors or electronic offset-related errors (see, e.g., Csontos and Sugar, 2024). These parameters can only be inferred from the measurement sequences (e.g., D1, D2, D3, D4) but not from the derived baseline values that underpin this study. This may explain their omission in the discussion, given that the protocols were not available. It is strongly recommended that the authors extend the error analysis to define and evaluate these parameters in future work. Additionally, there are some unclear points in the paper and missing information that require clarification.
Csontos, D. Sugar (2024), Dataset of geomagnetic absolute measurements performed by Declination and Inclination Magnetometer (DIM) and nuclear magnetometer during the joint Croatian-Hungarian repeat station campaign in Adriatic region, Data in Brief 54,110276, doi.org/10.1016/j.dib.2024.110276.
Specific comments
l 9 12 years: between 2010 and 2024
l10 applying a 90% threshold: threshold for what?
l18 due to the complexity of azimuth alignment > due to the complexity of azimuth alignment and levelling
l27 discrepancies > differences
l40 missing spaces following commas
l59 fluxgate sensor, mounted coaxially with > fluxgate sensor, mounted parallel to (the sensor cannot be mounted coaxially with the optical axis). This positioning is also a potential source of error not mentioned in the paper. It implies that the observations are made
l60 it generates zero output: assuming zero offset
l65 vertical > (magnetic) meridional: Inclination measurements are carried aligning the instrument with the magnetic meridian determined through declination measurements.
l67 Two observations are needed to find the true north direction. One with sensor up and another with sensor down to eliminate errors associated with the optical misalignment of the theodolite.
l75 vertical > (magnetic) meridional
l75 omitting azimuth marker: the vertical reference is provided by the gravity field through the suspension system of the theodolite.
l79 followed > follows
l80 four configurations > two azimuth readings and declination observations in four different positions to eliminate errors associated with theodolite optics, sensor misalignment and electronics offset.
l83 Eq. (2) misses a +/-90° term.
l88 Two distinct: start a new paragraph here
l92 Integration of declination: start a new paragraph here
l96 discrepancies > differences [not only here but several times later]
l100 under stable operation > be more specific about the stable operational conditions (temperature, magnetic cleanliness, etc.)
l107 Were, > , where
l107-108 Clarify the relation between minutes i-j and k.
l109 across distinct pillars > on different pillars
l106 and 111: It is a bit confusing that the argument of W_B is time in Eq. (4) but location in Eq. (5). Be consistent.
l112 Where, > , where
l113 Some notes on how the inter-pillar difference is derived would be beneficial.
l115 cross observatory fluxgate theodolite comparisons: Mentioning observatories in this context is a bit confusing to me. Would not it be better to say simply „cross-comparisons of fluxgate theodolites”?
Table 1: Some information on the location of the observatories would be beneficial.
Table 2: Some more detailed information on the instruments (type and angular resolution of the theodolites, type of the magnetic sensor/electronics) would be beneficial if this is available to the authors. There is not any information on the observers. One can only assume that all instruments were operated by different individuals, and the same person across different years. However, this is not necessarily the case.
l127 deviations > differences
l130 scaled according to the legend on the right: There are dots in the figure obviously larger than the largest shown in the referred scale.
Figure 2 Units are missing both from the y-axis labels and the scale shown in the legend.
Figure 2 There are 46 categories along the x-axis. This is equal to the number of observatories but not the number of the instruments. The entire paper focuses on instrumental differences, yet this crucial figure combines and merges the various instruments. This needs to be corrected.
l137 centring errors: What do you mean on centring errors? Positioning accuracy of the theodolite on the pillar? This has effect primarily on the declination baseline differences but small if any on inclination baseline differences. Have you checked this to find the reason of the differences?
l136-140 There are also large dots with large central values. Some more detailed analysis of the obtained differences would be beneficial here. Some conclusion, e.g. on the accuracy of various instrument types. Typical sources of error, etc.
l140 dispersion of multiple dots corresponding to the same station: This is where various instruments belonging to the same stations are mixed up. This needs to be corrected or the interpretation, statements and conclusions need to be corrected.
l141 Frequent personnel changes: This obviously has a great effect on the results. Information on observers are totally missing. (They could be identified e.g., by two numbers: 1st for the observatory, 2nd for the individual). Without having this information some of the statements (e.g., „This graphical approach thus effectively monitors instrument performance”) must be refined.
l146-47 „means of 0.00′ and 0.02′” AND „indicating excellent consistency among network fluxgate instruments”: These values depend on the choice of the reference instrument. Please clarify how the choice as made.
l151 additional azimuth marker alignment > uncertainties resulting from the additional azimuth marker alignment and theodolite levelling
l163 12 years > 12 inter-comparison campaigns [or similar, the comparisons cover 15 years from 2010-2024]
Table 3 Define instrument classes (codes).
l178 Type B standard uncertainty: What are Type A and B standard uncertainties?
l180 What is Delat in Eq. (6)? Provide a reference.
l184 Provide a reference.
l187 There are further instrumental uncertainties, e.g. magnetic contamination of the theodolite body.
Eqs. (10) and (11): x is not defined.
l198 station specific > pillar specific?
l217 operator induced > ,the operator induced
l256 Orange and green dots: there are no dots in the figure!
l258 additional azimuth marker alignment: and levelling
Figure 5 Are there any trends in the human errors? Difficult to see.
Citation: https://doi.org/10.5194/egusphere-2025-2406-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 375 | 30 | 11 | 416 | 34 | 34 |
- HTML: 375
- PDF: 30
- XML: 11
- Total: 416
- BibTeX: 34
- EndNote: 34
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1