Home Articles Maintenance of geodetic reference points using GNSS

Maintenance of geodetic reference points using GNSS

Geospatial Information Authority of Japan (GSI) is the governing institute on geodetic survey and mapping and manages about 130,000 geodetic reference points in Japan and their survey results. Recent development of GNSS technology led to additional GNSS-based control stations as new geodetic reference points. Concurrently with this arrangement, GSI adjusted the reference frame for geodetic survey in Japan to world geodetic system to enable public survey with GNSS, and published a set of re-calculated survey results named JGD2000 compliant with International Terrestrial Reference Frame 1994 (ITRF94).

On the other hand, Japan is located on the plate boundary and is frequently suffered from large earthquakes such as the 2011 off the Pacific coast of Tohoku Earthquake. Owing to accumulated strains and crustal deformations caused by plate subduction, some areas have had considerable gaps between static JGD2000 and actual coordinates estimated by GNSS observations. Thus, GSI introduced semi-dynamic correction method in 2010 into a part of GSI”s basic surveys and other public surveys in Japan to solve major problem with revision of survey results, that is, how to manage relation between static geodetic datum and actually moving positions of reference points and how to reduce the total cost for revision.

In this paper, GSI”s activity on maintenance of survey results of geodetic reference points using GNSS observation network and semi-dynamic correction is presented.

Japanese geodetic reference points network

(1) Passive control points
As mentioned above, GSI maintains about 130,000 geodetic reference points. Most of all are passive control points with traditional stone monuments on ground. Passive control points are divided into roughly two major networks, triangulation point network for horizontal datum and bench mark network for vertical datum. These control points had been planned and constructed since late 19th century and were classified from 1st order to 4th order according to the average interval of the network. The number of points in each order is shown in Table 1. The 1st order triangulation survey was finished in 1913, whose results were the earliest set of survey results in whole area of Japan.

Table 1. Control Points in Japan (maintained by GSI)

The survey results were re-calculated and released in 2002 as the coordinates at the epoch of 1997 in the ITRF94 system (Geographical Survey Institute, 2003), except for those in East Japan areas which were revised after the 2011 off the Pacific coast of Tohoku Earthquake. The set of survey results was named Japan Geodetic Datum 2000 (JGD2000) which was recently updated to JGD2011 with revised coordinates of geodetic control points in East Japan after the 2011 off the Pacific coast of Tohoku Earthquake.

(2) GNSS-based control stations
GSI”s GPS observation started in 1993 as a Continuous Strain Monitoring System with GPS (COSMOS-G2) with 110 GPS stations in Pacific coast of central Japan area (Sagiya et al., 1995). The network was expanded nationwide in 1994 and integrated into a network named GPS Earth Observation Network System (GEONET) started in 1996 (Miyazaki et al., 1998). GEONET keeps observation at more than 1200 stations called GNSS-based Control Stations in real time with 1Hz sampling and provides three kinds of solutions; Quick Solution, Rapid Solution, and Final Solution (Yamagiwa et al., 2006). These results are used for crust monitoring and other research purposes like meteorology and ionospheric research. In 2002 revised Japanese Survey Act was enforced, which enables surveyors to use GPS data for public surveys, and after then GEONET stations have also been used for public surveys.

Management of survey results
Japan is located on the tectonically active area and there can be some gaps between geodetic datum and actually moving positions, in other words, between survey results and actual GNSS observation results. In old days, these effects were negligible because public survey projects focused on local surveys using small networks whose baselines were around 4 km. Recently network RTK-GPS with long baselines of over 30km has been used for public surveys, which required correction of errors derived from accumulated strain.

Because it cost too much to re-survey at all geodetic control points, GSI decided to use semi-dynamic correction to correct errors by steady motion, and to combine re-survey at selected points and re-calculation at other points to correct errors by unsteady motion.

(1) Semi-dynamic correction
Major causes of change in positions of geodetic reference points in Japan are plate motion, which is steady motion and can be modeled, and local crustal deformation due to earthquakes. Tanaka et al. (2007) developed the crustal movement models based on the method by Grant and Blick (1998) used for correction of geodetic datum in New Zealand and applied them to coordinate corrections in Japan. The correction parameters are calculated at each 5km-spacing grid from GEONET”s Final Solution and GNSS observation results at selected passive control points (high order control points) (Hiyama et al. (2010)).

The semi-dynamic correction has been adopted since 2010 in the case of public surveys with long baselines (Hiyama et al. (2010)). GSI provides the correction software called SemiDynaEXE and correction parameter files at the GSI”s web site (https://www.gsi.go.jp/). The correction parameters are usually updated once per year.

(2) Revision of survey results
As for local crustal deformations, it is hard to make models completely. GSI has corrected the survey results at the points affected by the large earthquakes by the combination of GNSS observation at selected points and re-calculation at other points using correction parameters generated from the re-survey results by Kriging method. Summary of recent revisions of survey results in Japan due to large earthquakes is shown in Table 2.

Table 2. Recent revisions of survey results due to earthquakes

There are two key points to be taken into account for revision of survey results: In the case of inter-plate earthquakes, crust motions are comparatively well-modeled, whereas intra-plate earthquakes cause more complicated crustal deformations and need many re-surveys around epicenters. It is known that post-seismic deformation still continues after the big main shock of an earthquake (Heki et al., 1997). We have to monitor daily solutions from GEONET stations until post-seismic slip calms down to avoid duplicated re-surveys.

The results of re-survey are, in usual case, obtained through network adjustment under the condition that the coordinates of the points at boundary of re-survey area are fixed. The revision due to the 2011 off the Pacific coast of Tohoku Earthquake was the except which the revision area was too large to use conventional method for revision. GSI used Very Long Baseline Interferometry (VLBI) with GEONET”s solution to calculate new coordinates of GNSS-based Control Stations on newest International Terrestrial Reference Frame (ITRF2008), and then the coordinates of triangulation stations were calculated through re-survey and re-calculation using correction parameters (Hiyama et al., 2011). The new set of revised survey results together with existing results in un-revised areas are now called Japan Geodetic Datum 2011 (JGD2011).


  • Grant D., and G. Blick (1998): A new geocentric datum for New Zealand, New Zealand Surveyor, No. 288.
  • Geographical Survey Institute (2003): The New Geodetic Reference System of Japan -Its adoption and application to our products-, Bulletin of the GSI, Vol.50
  • Heki K., S. Miyazaki and H. Tsuji (1997): Silent fault slip following an interplate thrust earthquake at the Japan Trench, Nature, 386.
  • Hiyama Y., Y. Morishita, H. Yamao, T. Yutsudo, K. Ochi and M. Iwata (2010): Towards the Introduction of Semi-Dynamic Correction, Journal of GSI, Vol. 120 (Japanese)
  • Hiyama Y., A. Yamagiwa, T. Kawahara, M. Iwata, Y. Fukuzaki, Y. Shouji, Y. Sato, T. Yutsudo, T. Sasaki, H. Shigematsu, H. Yamao, T. Inukai, M. Ohtaki, K. Kokado, S. Kurihara, I. Kimura, T. Tsutsumi, T. Yahagi, Y. Furuya, I. Kageyama, S. Kawamoto, K. Yamaguchi, H. Tsuji and S. Matsumura (2011): Revision of Survey Results of Control Points after the 2011 off the Pacific Coast of Tohoku Earthquake, Bulletin of the GSI, Vol. 59
  • Miyazaki, S., Y. Hatanaka, T. Sagiya, and T. Tada (1998): The Nationwide GPS Array as an Earth Observation System, Bulletin of the GSI, Vol.44
  • Sagiya, T., A. Yoshimura, E. Iwata, K. Abe, I. Kimura, K. Uemura, and T. Tada (1995): Establishment of Permanent GPS Observation Network and Crustal Deformation Monitoring in the Southern Kanto and Tokai Areas, Bulletin of GSI, Vol.41
  • Tanaka Y., H. Saita, J. Sugawara, K. Iwata, T. Toyoda, H. Hirai, T. Kawaguchi, S. Matsuzaka, Y. Hatanaka, M. Tobita, Y. Kuroishi and T. Imakiire (2007): Efficient Maintenance of the Japanese Geodetic Datum 2000 Using Crustal Deformation Models -PatchJGD & Semi-Dynamic Datum, Bulletin of GSI, Vol. 54
  • Yamagiwa, A., Y. Hatanaka, T. Yutsudo and B. Miyahara (2006): Real-time capability of GEONET system and its application to crust monitoring, Bulletin of GSI, Vol. 53

Note: *The papers in Bulletin of GSI (after Vol.49) are available online at following web site. https://www.gsi.go.jp/REPORT/BULLETIN-bul-home.html