干旱区生态环境知识资源中心

An exceptional solar activity event occurred at the end of October 2003. On October 29th, seven groups of sunspots were visible on the Sun's surface and the geomagnetic index K, reached the extreme value of 9, leading to beautiful auroras, among other effects. Similar events also occurred in November 2003. These events have been an exceptional opportunity to estimate effects of the thermosphere total density perturbations in extreme conditions on a Low Earth Orbit (LEO orbit corresponding in this study at altitudes ranging from 800 km up to 1400 km). Specifically, we study the best way to get reliable geodetic products even during such solar events, and how well the thermosphere models (DTM-78, DTM-94, DTM-2000, and MSIS-86 models) allow us to accomplish this and to predict the observed perturbations on the orbital elements. Thus our analysis is twofold. First, we have computed orbits of satellites equipped with the on-board DORIS tracking system, at an altitude ranging from 800 to 1330 km and for periods of time including these exceptional events (October 29-31 and November 20, 2003). We have computed 30-hour arc orbits, estimating the drag coefficient parameter very frequently (every minute) using a tight random-walk constraint or using a white noise reset in the Gipsy/Oasis software. We show that significant errors are obtained for the considered thermosphere models, but can be greatly improved using a more sophisticated data processing. We also investigated how these proposed processing strategies affect the quality of the DORIS geodetic products. Significant improvements were found for weekly stations coordinates estimations as well as for polar motion determination.

In addition, we have investigated the global quality of the modeling of the variations of the mean orbital elements in using thermosphere models over long periods of time including such severe geomagnetic events. This original approach is based on averaging techniques implemented in the CODIOR software. We have analyzed the continuous long-term evolution of the semi-major axis of the geodetic satellites Starlette, Stella and Ajisai tracked by the Satellite Laser Ranging (SLR) network. In this second part, we give: (i) a single global empirical coefficient per satellite, to validate the models over the whole duration of the orbital arc, including the exceptional solar activity events, and (ii) one coefficient per a period of about 2 months to quantify the seasonal differences between the models and the observations.

As a result of both investigations, we compare the capability of recent models of thermosphere to allow us to get reliable geodetic products and mean orbital elements variations over short and long periods of time as well as for the recent exceptional geomagnetic events. Different altitudes are considered: around 800-900 km for the Stella and Starlette satellites tracked by laser ranging systems and SPOT tracked by DORIS; around also 1300-1400 km for the Ajisai satellite tracked by laser techniques and TOPEX and Jason also tracked by DORIS. It appears in our study that laser and DORIS data are complementary to probe the thermosphere, and to test the quality of thermosphere models in specific conditions for getting accurate geodetic products. (c) 2005 COSPAR. Published by Elsevier Ltd. All rights reserved.