Pergamon A&. Space Res. Vol. 29, No. 6, pp. 961-913, 2002 0 2002 COSPAR. Published by Elsevier Science Ltd. All rights reserved Printed in Great Britain www.elsevier.com/locate/asr PII: SO273- I 177(02)00062-5 0273-l 177/02 $22.00 + 0.00 EVALUATION OF IR195 IONOSPHERE MODEL FOR RADAR ALTIMETER APPLICATIONS C. Zhao ’ , D. Bilitza’ , C. Shum ’ , S. Schaer ‘, G. Beutler ’ and S. Ge ’ ’ Department of Civil and Environmental Engineering and Geodetic Science. The Ohio State Univers@, 470 Hitch&ok Hall, 2070 Neil Avenue, Columbus. OH 43210-1275, USA. ’ National Space Science Data Center, Goddard Space Flight Center, Code 633/632, Greenbelt, MD 20895, USA jAstrorlomica1 Institute, University of Berne, Sidlerstrasse 5, CH-3012 Berne, Switzerland ABSTRACT Using the TOPEX dual-frequency (Ku- and C-band) radar altimeter measured nadir total electron contents (TECs) from an orbital altitude of 1354 km, the fidelity of the IRI95 ionosphere model and GPS-derived Global Ionosphere Map (GIM) have been evaluated. In particular, the models have been evaluated over the ocean for the time period 1993-1999, representing low, medium and high solar activity periods. In addition, the evaluation is also focused on the model performance for TEC variations due to local solar time, geographical locations, and long-term trend, or drift. The major spatial and temporal features of ionosphere depicted by IRI95 model agrees reasonably well when comparing with TOPEX ground-truth, except during high solar activity period in 1999. The error in IRI95 would translate into an apparent sea level change of 0.3-0.6 mm/yr for the majority (70%) of the global ocean. During high solar activities in 1999, the error reaches 1 mn~yr. Results indicate that GIM provides more accurate ionosphere delays during high solar activities. For the use of single-frequency altimeters (e.g., GFO) for sea level studies during high solar activity period, ionosphere models need further improvement. 0 2002 COSPAR. Published by Elscvier Science Ltd. All rights reserved. INTRODUCTION The measurement and modeling of the electron content in the Earth’s ionosphere which produce delays in phase group propagation of radio waves is important for Space Physics studies and for a number of applications such as navigation and radar remote-sensing. One important application is to enable the correction or removal of ionosphere delays for spaceborne altimeters equipped with single-frequency (Ku-band, 13.6 GHz) radars, for accurate measurements of the sea surface heights (e.g., Urban et al., 1997, Urban, 2000, Shum et al., 2000). At present, the only spaceborne altimeter designed to use the dispersive nature of the ionosphere refraction to correct for the delay with a dual-frequency altimeter (13.6 and 5.6 GHz) is the NASA/CNES TOPEX/POSEIDON (T/l’) altimeter (e.g., Imel, 1994). With abundant historic and current single-frequency altimeters (Seasat, Geosat, ERS-1, ERS-2, and GFO), the use of physical or empirical ionosphere models for the correction of altimetric range delays to produce sea level measurements, remains the only option. A number of ionosphere models have been used for this purpose. The physical or assimilated models include the Bent model (Llewellyn and Bent, 1973), the Klobuchar model (Klobuchar, 1987), the Parameterized Real-Time Ionospheric Specification Model (PRISM) and the PIM physical model (Daniel1 et al., 1993), and the International Reference Ionosphere (IRI) models, IRI90 and IRI95 (Bilitza, 1997). The empirical models include the DORIS model (Imel, 1994), the PRARE model (Wilems and Reigber, 1987), and Global Ionosphere Maps (GIMs) using GPS tomography or mapping functions from Ll/L2 frequency limb-sounding (e.g., Hajj et al., 1994, Wilson et al., 1995, Schaer et al., 1996, Schaer, 1999, and Rius et al., 1997). Among these models, the IRI95 model (Bilitza, 1997) is widely use d for altimeter applications. The nadir total electron content (TEC) data collected by the T/P dual-frequency radar altimeter since 1992, when properly calibrated (Imel, 1994, Ruffni et al., 1998), provides an independent data source to assess the performance of ionosphere models over the ocean. Another data source is the GPS Ionosphere Maps (GIMs) generated by analysis of data from global GPS network (Hajj et al., 1994, Schaer et al., 1996). Urban et al. (1997) studied the perfomlance of the IRI90 and IRI95 models for altimetric applications during low and medium solar active 967