Dear Taizoh-san and Keystone Project Team,
I would like to thank you and the Keystone Project Team for a most enjoyable visit. I was very impressed by the facilities here at CRL and at Kashima. The Keystone Project is unique in the world and it seems that it has the potential to make great contributions to our understanding of the Earth and modern geodesy. The VLBI system in the Keystone Project is the only current system running real-time and on such short baselines. The SLR system is also unique in the proximity of the systems and its collocation with VLBI and GPS. As such, you have a unique opportunity to address a number of critical issues in Earth sciences. To me, one of the most important issues to address is the "stability" of the Earth. While our representation of the motions of points on the Earth surface is through linear motions with occasional episodic displacements and post-seismic motions, there is mounting evidence, mainly from GPS, that the actual behavior of the Earth may be more complicated. The deviations from secular motion are generally less than 10 mm in horizontal components and seem to be of large scale (i.e., correlated over may hundreds of kilometers). But so far, there is little evidence that two different geodetic systems see the same deviations. In this regard, the collocated Keystone systems are unique in a tectonically active region. Because the non-secular motions appear correlated over large distances, the expected signal sizes on a small network are likely to be only a few millimeters. This is the case in California where the RMS scatter of daily positions estimates for many of the GPS stations separated by <100 km is between 1 and 2 mm for horizontal components and 5-10 mm in height. Of course in California there is only a GPS network and it is not possible to compare GPS to other space geodetic signals.
For the Keystone Project I see several tasks that would greatly increase our understanding of geodetic systems and the Earth. With this system, it should be possible to have millimeter accuracy results available from all three geodetic systems and the detailed comparison between them should reveal much about these systems and the Earth. For VLBI, I think you should investigate using phase delays which should improve the delay precision by a factor of ten or so; it will be interesting to see how much the station position estimates improve with this increased data quality. For GPS, you could potentially improve the phase center models for the antennas through the comparison of the position estimates from VLBI, GPS and SLR, and through detailed studies of the phase residuals. One interesting approach here would be to use the PRESTAR system with its clock shared with a nearby GPS receiver using an omnidirectional antenna. Such as system is being developed in the US for in-situ calibration of GPS antennas. One of the largest uncertainties in GPS measurements arises from the phase center variations of the GPS antennas. SLR has, in principle, a major advantage over GPS and VLBI due to its much lower sensitivity of atmospheric water vapor. There is also an intrinsic strength in making a direct range measurement rather than the biased range measurements implicitly make by GPS and VLBI. The challenge in SLR is to realize these advantages; at the moment it appears that SLR generates geodetic results which are not of higher quality than GPS and VLBI even in the height component. Understanding why this is so, and improving the performance of SLR seems to be a high priority. The Keystone system is again unique in being able to address these fundamental issues.
For all systems, it seems that the availability of the JMA 10 km resolution atmospheric data sets will allow detailed studies of the effects of atmospheric propagation on geodetic systems. Establishing how large the effects of inhomogeneity in the atmosphere are is an important task. Of great utility to the whole geodetic community would be the development a parameterization of these inhomogeneities so that they could be estimated from the geodetic data themselves thus extending the currently used zenith delay and gradient estimation strategies. If the VLBI/SLR and GPS systems do not generate the same results between collocated stations then CRL is in a unique position to understand and correct the deficiencies in each of these systems. Ultimately, if the systems can be shown to agree at the 1 mm level, at least in the changes of the positions that they measure, then again the Keystone Project is in a unique position to improve the accuracy of all three systems to better than 1 mm. Ultimately it is not clear what will set the final accuracy achievable with any of these systems: it could be the atmosphere or the solid Earth itself. The operation of these systems and the redundancy of the measurements systems should definitively establish the effects to be expected before, during, and after earthquakes. Motions after earthquakes should establish the rheological properties of the Earth and the forces acting on it after an earthquake. Detection of a reliable pre-seismic signals, measured by three independent systems, would be of fundamental importance in the study of earthquakes. Of course, these signals may be too small to detect currently, but the Keystone Project is in a unique position to be able to address this issue with multiple techniques.
Again I thank you for a wonderful visit, and I look forward to seeing some great results coming out of the Keystone Project.