the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Rescue, Integration, and Analytical Application of historical data from eight pioneering geomagnetic observatories in China
Abstract. Decades to centuries of continuous geomagnetic observation data have extensive scientific research and practical application value, especially in revealing the long-term variation rules of the geomagnetic field, which is irreplaceable. During the International Geophysical Year (1957–1958), China established geomagnetic observatories in Beijing, Lhasa, Lanzhou, Wuhan, Guangzhou, Changchun, and Urumqi, forming the initial structure of China's geomagnetic observation network together with the Shanghai Observatory. These observatories have continuously observed despite facing many challenges since their establishment, accumulating a large amount of valuable observational data, making significant contributions to the progress of geomagnetic scientific research and development. However, the scattered storage state of these historical data and the potential risk of damage pose a threat to the integrity and reliability of the data. This study conducted a rescue integration of the historical observational data from eight pioneering geomagnetic observatories in China, significantly improving data quality and facilitating long-term preservation and use of the data. This article introduces the basic conditions of eight observatories, including their locations, changes in location, observation environments, magnetic rooms, the magnetism of building materials, the layout of building facilities, measuring instruments, etc. These are the main prerequisites and foundations for ensuring the quality of observation data. Then, it introduces the integration and processing of historical data, including data collection, digitization, unification of formats, anomaly detection, and processing. Subsequently, the processed data were validated, including assessments of daily variations accuracy and long-term stability. The results show that the quality of the integrated historical data has been significantly improved. These datasets are of great value for improving historical geomagnetic field models, studying variable fields, main geomagnetic fields, and their long-term variations. Finally, we applied the data to the analysis and research of Sq and geomagnetic jerks, exploring the spatiotemporal variation characteristics of Sq and jerk in the China. Sq is mainly a daytime phenomenon, and its variation pattern in the middle and low latitude regions is mainly characterized by its dependence on latitude and local time. The geomagnetic jerk phenomenon exhibits significant regional differences and asynchronous occurrence times of jerks. Jerk events in 1969, 1979, 1991, 2003, and 2019 were observed at all observatories and had distinct jerk variation characteristics. Other jerks were only observed at some observatories or individual observatories. The maximum time difference for the occurrence of the same jerk event at different observatories was 2 years. This study aims to provide these precious datasets to the scientific community and the public so that these data can be integrated with data from other sources, thereby further exploring the spatiotemporal evolution and physical mechanisms of the geomagnetic field. The historical datasets of the eight geomagnetic observatories that have been integrated and quality controlled are available at https://doi.org/10.5281/zenodo.14560950 (Zhang et al., 2024b).
- Preprint
(1637 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 17 May 2025)
-
RC1: 'Comment on essd-2025-6', Anonymous Referee #1, 17 Mar 2025
reply
It is obvious that the authors of this work have accomplished a very valuable and meaningful task. Scientists working on related research can utilize results of this work for in-depth studies. I strongly recommend publishing it as soon as possible.
But for publication, the manuscript needs several minor modifications. The specific suggestions are as follows:
1. The titles of Section 2.1.7 & 2.1.8 not match the content, and need to be modified.
2. Why was the data after 2022 not processed, and what’s the reason? Please briefly explain in the article.
3. In Figure 6 on page 19, the unit of Local Time should be labeled.Citation: https://doi.org/10.5194/essd-2025-6-RC1 -
AC1: 'Reply on RC1', Suqin Zhang, 20 Mar 2025
reply
Dear Reviewer,
Thank you very much for your positive feedback and constructive suggestions on our manuscript. We greatly appreciate your time and effort in reviewing our work. We have carefully addressed your comments and made the necessary revisions to improve the manuscript. Below are our responses to your specific suggestions:
- **Regarding the titles of Section 2.1.7 & 2.1.8**:
We agree with your observation that the titles did not fully match the content. We have revised the titles to better reflect the content of these sections.
- **Regarding the data after 2022**:
We apologize for not clarifying this point in the original manuscript. The data after 2022 was not included because it was incomplete at the time of our analysis. We have added a brief explanation in Section 2.3 to address this issue.
- **Regarding Figure 6 on page 19**:
We thank you for pointing out the missing unit for Local Time in Figure 6. We have now labeled the unit clearly in the revised figure to avoid any confusion.
Thank you again for your valuable feedback.
Sincerely,
Suqin Zhang
On behalf of all authors
Citation: https://doi.org/10.5194/essd-2025-6-AC1
-
AC1: 'Reply on RC1', Suqin Zhang, 20 Mar 2025
reply
-
CC1: 'Comment on essd-2025-6', Xin Changjiang, 01 Apr 2025
reply
The continuous, complete, accurate and reliable geomagnetic field data highlights its value. With the development of technology, the observation data of some prestigious geomagnetic observatories have almost experienced the process from analog to digital observation, and the processing and storage of the data have taken a different way, especially for some data stored in paper and floppy disk, there is a serious risk of loss as time goes by. This paper introduces the observation of China's eight pioneering geomagnetic observatories, describes detailedly the complete process of data collection, digitization and standardization of analog data, data processing and quality control, and verifies the reliability of data by comparing and correcting the daily variation accuracy and secular variation stability, and applies the data to the analysis of different geomagnetic phenomena. It further embodies the important significance of rescuing and integrating the historical data of geomagnetic observatories. The examples in this paper are appropriate and clear, the words, formulas, symbols are accurate, and the ICONS are standard. The research method is feasible, the idea is clear and the conclusion is reliable, which has important guiding significance for the data rescue, data processing, quality control and analysis application of geomagnetic observatories.
Citation: https://doi.org/10.5194/essd-2025-6-CC1 -
AC2: 'Reply on CC1', Suqin Zhang, 09 Apr 2025
reply
Thank you for your positive feedback on our paper. We are pleased to hear that our work on the geomagnetic observatories' data rescue and integration is valued.
Citation: https://doi.org/10.5194/essd-2025-6-AC2
-
AC2: 'Reply on CC1', Suqin Zhang, 09 Apr 2025
reply
-
CC2: 'Comment on essd-2025-6', Z.H. He, 16 Apr 2025
reply
Line 213: The author has misplaced the Section 2.1.7 Lhasa Observatory (IAGA code LSA) to the Line 203 Section 2.1.6 Lhasa Observatory (IAGA code LSA).
Line 223: And the similar mistake in 2.1.8 Lhasa Observatory (IAGA code LSA).
They should be corrected.
Citation: https://doi.org/10.5194/essd-2025-6-CC2 -
AC3: 'Reply on CC2', Suqin Zhang, 22 Apr 2025
reply
Thank you for your comment. We have revised the titles to better reflect the content of these sections.
Citation: https://doi.org/10.5194/essd-2025-6-AC3
-
AC3: 'Reply on CC2', Suqin Zhang, 22 Apr 2025
reply
-
CC3: 'Comment on essd-2025-6', Z.H. He, 18 Apr 2025
reply
Line 315:Obtained in (2)-->step (2)
Line 330: Take the corrected signal in (7)-->step (7)
The authors use some different coordination systems. Most of them, such as DHZ and DHI, have been defined in the manuscript, but the XYZ system is not defined in the paper. Please describe it in detail.
Citation: https://doi.org/10.5194/essd-2025-6-CC3 -
AC4: 'Reply on CC3', Suqin Zhang, 22 Apr 2025
reply
Thank you for your suggestion. We have added descriptions about X and Y in line 298 of the revised manuscript.
Citation: https://doi.org/10.5194/essd-2025-6-AC4
-
AC4: 'Reply on CC3', Suqin Zhang, 22 Apr 2025
reply
-
CC4: 'Comment on essd-2025-6', Junjie Sun, 24 Apr 2025
reply
This study presents a method to enhance data quality and support long-term preservation by conducting a rescue integration of historical observational data from eight pioneering geomagnetic observatories in China. The fundamental conditions of these observatories are detailed, including their original locations and changes, observation environments, magnetic rooms, the magnetism of building materials, the layout of facilities, and measuring instruments. These factors serve as the primary prerequisites and foundations for ensuring the quality of observational data. The results demonstrate a significant improvement in the quality of the integrated historical data, which has also been applied to the analysis of geomagnetic jerks. The maximum time difference for the occurrence of the same jerk event across different observatories was interpreted to be 2 years. Below are several suggestions and interesting topics for discussion.
1. The sentences in lines 537-540 appear somewhat repetitive.
2. Regarding the long-term changes depicted in Figure 7, are there corresponding physical background changes or events driving these shifts?
3. The variations observed at different stations in Figure 6 show asynchronous behavior over time; what causes this, and are there underlying physical factors influencing it?
4. It is suggested to fill the blanks in Table 1 and Table 3 with slashes or other symbols and clarify the meaning of the symbols in the table titles.Citation: https://doi.org/10.5194/essd-2025-6-CC4 -
RC2: 'Comment on essd-2025-6', Anonymous Referee #2, 06 May 2025
reply
The manuscript presents digitized data from eight long-term geomagnetic observatories in China. Digitizing, preserving and archiving such data and making them available for research is a highly important task to ensure that previous scientific efforts and valuable sources of information about the long-term evolution of natural phenomena (in this case the geomagnetic field and its variations) does not get lost and remains available for present and future research. The manuscript provides a generally good description of the data sources and the presented data and is complemented by quality checks of the digitized data and two examples for applications. The dataset and manuscript certainly deserve publication. I have several requests for additional information or clarification that in my opinion would make the manuscript even more useful as documentation of the newly available data. My main concern is the first point below, as it can be quite concerning and confusing for potential users of the data if data that should be the same but are obtained from different sources differ. The data are available as described and easy to read and use, but I also have some requests for clarification on it.
- a) It should be clearly stated which data have been newly digitized, and for the overlapping part the data should be carefully checked against the hourly mean values already held at the World Data Center for Geomagnetism in Edinburgh (https://wdc.bgs.ac.uk/data.html , https://wdc-dataportal.bgs.ac.uk/) . The updated compilation from the eight observatories as presented here should be sent to the WDC so that that database also is up to date, and only one version of the data is around. I didn’t check for all eight observatories, but, e.g., BJI data are available at the WDC from 1957 – 1994 with a short gap around 1879, and LZH data are available for several years between 1980 and 2019, so some of the data presented here in a more complete form are already available there.
- b) The description of the observatories and their history is very valuable. For an easier assessment of the potential influence of re-locations, it would be useful to give the approximate distance between old and new location in km in addition to the coordinates. Is it true that in none of the cases of re-location observations at both locations to determine the resulting “jumps” in the data series have been done (it is not mentioned in any of the cases)? If this has been done it should be mentioned.
- c) The first sentence of section 2.1.5 is confusing – the mentioned construction start in 1874 was not in Sheshan, but in Xujiahu, right? What was probably meant to say with the first sentence is that Sheshan (with predecessors) has the longest uninterrupted data series? Please re-formulate for clarity.
- d) I suggest to put the sentence mentioning Table 1 and the table itself at the beginning of section 2.1, i.e. before 2.1.1. At the moment the sentence seems to belong to section 2.1.8, although it refers to all the observatories described in 2.1. It could also be informative to expand the table to include information about the available data, i.e. start of the data series and gaps.
- e) The description of the used instruments in section 2.2 is rather vague. For example, I assume much of the old relative recordings before it became digital were photographic recordings, but that is never mentioned (p9, l.261-262). Was the CJ6 theodolite really used to measure horizontal intensity, not inclination? There are a lot of abbreviations/model names (such as CZM, DTZ, GSM-19T). Can some more information such as manufacturer be given? In particular if the manufacturer still exists, this should be done. To my knowledge, the two mentioned GSM instruments are made by GEM Systems (https://www.gemsys.ca/) . I don’t know about many of the others, but if possible, such information (references or websites) should be given. Is the instrumentation perhaps documented in the observatory yearbooks? Then this should also be mentioned with references to the yearbooks.
- f) Somewhat related to the previous point: Did all observatories become digital at the same time? As digital data should be available from that time on (probably even 1 min data, not just hourly values? – it would also be useful to mention somewhere which of the observatories are in INTERMAGNET now) this point in time should be clearly given for each observatory (perhaps it could also be included in an extended Table 1?)
- g) Somewhat related to point e): The presentation of the data is the central point of the manuscript. Section 2.3 deserves some more details, so that the reader understands which parts of the data have been compiled in which way: On p10, in line 284 annual reports are mentioned and later for the digitization process “digital images” of the data are mentioned. I assume that the continuous recordings before digital instruments were installed were recorded photographically. It should be stated more clearly that data were digitized from published numbers, not directly from the original recordings. When scanning these pages, did you try to use characters (number) recognition, or did you indeed just scan images? If I understand the sentence in line 289/290 correctly, character recognition did not work and all data were digitized manually?
- h) Regarding the detection and estimation of jumps, section 3: have all jumps related to observatory re-location been estimated in this way (including the time of the jump, which is known in these cases)? Or have there been comparative measurements at some of the sites so that jump values had been documented?
- i) p16, l. 417: I don’t understand how increasing interference in the observatory environment would cause a long-term trend. Wouldn’t one rather expect stronger scatter, rather than a continuous systematic drift in that case? But indeed I also don’t have a suggestion for an interpretation of the drift, as an instrument drift would normally be captured by the absolute measurements. Might there be some slow change of the absolute pillar?
- j) p17, discussion of the differences: I think one of the main factors of short-term variations in the differences are remaining contributions from external magnetic fields in the data. If you only do a data selection for night times, but not additionally for geomagnetic activity, the data will still contain influences from geomagnetic storms and I would expect to see variations in the differences with the solar cycle. This seems included in your second reason, but I think it is one of the main reasons and should be stated more clearly. For the other aspect, this would be that the IGRF doesn’t describe the secular variation accurately, and the five-year piecewise linear secular variation is not the best representation of the continuous secular variation. This could also be mentioned in addition to general IGRF quality considerations. I don’t really understand what is meant by “participation of observation data” in l. 438/439.
- k) p22, lines 541 and 543: From the first sentence I understand that the 1949 event was detected elsewhere in the world, so why would it only be a national or even more isolated regional event?
- l) In Table 3, it would be helpful to distinguish between “no data available” and “data available, but no jerk detected” instead of having blank spaces for both. I suggest to use to different symbols (or blanks for no data, and a “-“ symbol for data, but no jerk) or something like that to distinguish the two cases.
The quality/resolution of all figures is rather low in the downloaded pdf. It should be checked if this is just a problem of the review version of the manuscript and improved if necessary.
Data set:
- It would be more convenient for the user to have the three components, i.e. full magnetic field vector information, per time stamp for one observatory in one file. For easy compatibility with other observatories’ data it would be best to follow the internationally adopted IAGA2002 format suggested by the International Association of Geomagnetism and Aeronomy (see publication by IAGA Working Group V-DAT at https://www.ncei.noaa.gov/services/world-data-system/v-dat-working-group/iaga-2002-data-exchange-format ).
- The content of the files doesn’t seem to be described. This should be done even if it is obvious that the first columns are date and time, and the last column is the magnetic component, but it is not clear to me what the number in the column in between is, it seems like just a counter that starts with 0 and I don’t understand what it is useful for.
- And as already stated above, I would strongly suggest to also submit the dataset to the World Data Center for Geomagnetism in Edinburgh, where hourly mean values from many geomagnetic observatories around the world is held.Minor details of wording etc. in the manuscript:
-p1, l. 17: delete the word “rules”
-p2, l.70: delete the word “observation” before “observatories”
-p4, l.120: “observations” (plural)
-p5, l.158: Mention in the first sentence of the paragraph when this problem occurred
-p6, l.173: “observations” instead of “observational”
-p7,l.206: “resumed” instead of “resuming”
-p7, l.208: “It is located” instead of “The is located”
- the titles of sections 2.1.7 and 2.1.8 have wrong observatory names
-p8, l.241: I assume the sentence should say “observatories’ yearbooks”, or “in each observatory’s yearbook”
-p12, l.350: I suggest to use “contributions” instead of “components” here
-p13, l.378: “each observatory” (singular)
-p14 give a reference for the IGRF (e.g., the latest publication which currently is for IGRF 13th generation)
-p14, l.389: The IGRF describes the core or main field as detectable at Earth’s surface, not just its long-wavelength components (shorter wavelengths do not play a role for practical purposes or any of the applications of the IGRF, because they are not measurable either due to their distance from the source or due to masking by the lithospheric field)
-Fig 6: The figure caption should contain more details of what is shown in the upper and lower panels, in particular the lower panel histograms should be briefly explained.
-Fig 7: Similarly, I suggest to include the information that now is given in each panel in the legend only once in the figure caption. I don’t see the pink arrows mentioned in the legends.
- p22, l. 538: “may have occurred”Citation: https://doi.org/10.5194/essd-2025-6-RC2
Data sets
An integrated and quality-controlled historical datasets of eight pioneering geomagnetic observatories in China S. Q. Zhang et al. https://doi.org/10.5281/zenodo.14560950
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
256 | 49 | 16 | 321 | 10 | 10 |
- HTML: 256
- PDF: 49
- XML: 16
- Total: 321
- BibTeX: 10
- EndNote: 10
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1