Satellite operations depend on being able to generate accurate predictions of a spacecraft ephemeris in a very short period of time. This ephemeris is used by the mission controllers to plan operations such as instrument pointing and orbit adjust maneuvers. This paper examines the methods and parameterizations necessary to achieve an accurate ephemeris quickly in an operational environment. The specific application described is for NASA's ICESat (ice Cloud and Land Elevation Satellite) mission. The primary tracking data for ICESat is GPS pseudorange and carrier phase. Near real time orbit determination accuracy with this data is better than 2m RMS. Because of stringent pointing requirements it is necessary to predict the satellite ephemeris to an accuracy of better than 400m in track after 48 hours. This requirement has been achieved despite the significant drag perturbations on the 600km altitude orbit. This paper presents the orbit determination and prediction techniques used in the ICESat operational system as well as examples of the orbit determination and prediction accuracy achieved.

A few months after launch of the TOPEX/POSEIDON (T/P) spacecraft, during the verification phase of the mission, the Precision Orbit Determination Team discovered residual along–track accelerations which were unexpected and of unknown origin. This paper describes the analysis of radiation forces acting on the T/P spacecraft for the purpose of understanding and providing an explanation for the anomalous accelerations. The radiation forces acting on the solar array, which experiences warping due to temperature differences between the front and back surfaces, are analysed and the resulting along–track accelerations are determined. This analysis led to the development of a new radiation force model which includes (1) warping of the solar array, (2) solar array deployment deflections as large as 2°, (3) necessary adjustments to certain surface property parameters, and (4) an adjustment to the coefficient of drag. As a result of the new model, estimates of the empirical constant along–track acceleration are significantly reduced in magnitude and are less dependent upon the location of the Sun relative to the orbit plane.

The reduced dynamic technique (RDT) has become popular for removing unmodeled dynamic error. In particular the technique is useful for removing errors caused by poor atmospheric modeling. This paper examines the ability of the technique to remove drag error at a range of low altitudes where the error was simulated by perturbing the drag coefficient. The orbit determination was performed using simulated GPS flight receiver data with the GIPSY-OASIS II software package. The RDT sigma's for minimizing orbit error in the radial, cross track, and along track directions, and for minimizing the three dimensional position error are presented for a variety of low altitude, sun synchronous orbits. In an autonomous application, RDT removes drag as a significant error source at altitudes above about 350 kilometers. Also, for this altitude regime, the optimum sigma is nearly the same whether or not drag error is included in the solution.

High-accuracy sea surface positioning is required for sea floor geodesy, satellite altimeter verification, and the study of tides and sea level. The Colorado Center for Astrodynamics Research (CCAR) at the University of Colorado has been engaged in GPS buoys technology development since 1986. Over the past decade a dozen or so experiments have been conducted using various ocean buoys equipped with GPS receivers. Early ground-based experiments were conducted with an oscillating antenna mechanism and TI 4100 (Texas Instruments) receivers. Later experiments used Trimble, Rogue and Turbo Rogue receivers. Various buoy designs were experimented with including spar buoys, wave-rider buoys, and other conventional ocean buoys. This paper will briefly review some of the GPS buoys experiments and their results.

Space-based radio navigation systems such as the Global Positioning System (GPS) can provide us with a unique opportunity to study the effect of the ionosphere as the signals propagate from the satellites to the GPS receivers. Based on a modified version of the University of New Brunswick's (UNB) DIPOP software package, the authors enhanced the algorithm to model ionospheric total electron content (TEC) using dual frequency GPS observations from stations of the International GPS Service (IGS). The algorithm uses a spatial linear approximation of the vertical TEC above each IGS station using stochastic parameters in a Kalman filter estimation to describe the local time and geomagnetic latitude dependence of the TEC. The authors demonstrate that the enhanced UNB algorithm in conjunction with an IRI-95 update procedure is capable of modeling the diurnal variation of TEC even during a solar storm period in the low latitude region using a higher order surface estimation technique for each GPS station processed

The Geoscience Laser Altimeter System (GLAS) mission is designed to measure changes in the volume and mass of both the Greenland and Antarctic ice sheets. The ICESat satellite will carry the GLAS altimeter, and will have a nominal orbit altitude of 600 km, eccentricity of 0.0013, and orbit inclination of 94 degrees. The groundtrack repeat period is 182 days and will be maintained to within 1 km at the equator via routine orbit adjustments. Science requirements for the GLAS mission demand that the laser altimeter be pointed to within 50 meters, crosstrack, of a defined reference groundtrack. As the actual ICESat groundtrack drifts away from the reference groundtrack, the attitude must be controlled such that the altimeter boresight is pointed at the reference groundtrack, This orientation may be described by a rotation, theta, about the instantaneous local horizontal direction vector, which lies in the orbit plane and is pointed in the direction of motion of the satellite. This paper describes the attitude algorithm required to maintain reference groundtrack pointing. The algorithm was tested for two cases: (I) a maximum equatorial offset of 1 km and (2) a maximum inclination error of 0.03 degrees in addition to the 1 km equatorial offset. Results show that the rotation angle, theta, ranges between +/- 350 arcsecs during a given day for case (1) and +/- 1200 arcsecs during a given day for case (2). The effect of topography on computation of the rotation angle was also studied. Neglecting topography results in rotation angle errors that range between -2 arcsecs and +3 arcsecs for case (1) and -8 arcsecs (23 m crosstrack) to +10 arcsecs (29 m crosstrack) for case (2). Finally, an overview of an automated attitude control system, which includes the corrections due to topography (JGP95E), is discussed.

A study of Legeckis eddies in the equatorial Pacific during seasonal periods of high zonal shear between the Southern Equatorial Current and the Northern Equatorial Counter Current is undertaken. We use both Geosat and TOPEX altimetry data, in combination with advanced very high resolution radiometer (AVHRR) temperature fields and numerical results generated by the Parallel Ocean Climate Model (POCM). Geosat data analyzed using the Marquardt‐Levenberg Fourier series algorithm [Dayyani et al., 1996], exhibit eddy features in good qualitative agreement with both linear stability theory of the zonal barotropic flow and with other independent field and satellite observations. A comparison of sea surface height (SSH) and sea surface temperature (SST) anomaly fields suggests that the phase relation between them may be explained in a first approximation by simple cat's‐eye kinematics [Musman, 1989]. This is explored further using the POCM, TOPEX, and AVHRR data. The POCM time series of SST and SSH anomalies obtained during a time of quasi‐steady eddy propagation clearly show that the phase ϕ between their respective minima varies continuously with latitude, ranging from near zero at latitudes well below the separatrix of the cat's eye to values approaching π radians at latitudes near the critical layer of the cat's eye. The unsteady wave systems observed in both the Geosat and TOPEX SSH anomaly data show the birth and death of individual eddies, with a fairly clear trend that births occur 1°N of the zonal wave train while deaths occur only in the latitude of the mature wave train.

One year of global TOPEX altimeter sea‐level and correction data were analyzed to consider the effect on the statistics of altimeter range corrections due to (1) seasonal variability. (2) removal of shallow water, (3) area and data‐density weighting, and 14) separate consideration of the equatorial region. Seasonal analysis was done to provide a better understanding of the temporal behavior of the statistics. The effects of shallow water were removed, since shallow water has different dynamics than deep water and could skew some resuts. Area and data‐density weighting was performed to keep the disproportionate amount of data available at the turning latitude from biasing the results. Finally, the equatorial region was considered separately, since equatorial conditions and dynamics are different than those at mid and high latitudes. It was determined that the statistics of each of the corrections was affected by one or more of these factors. Knowledge of these statistics should be useful in applying the corrections to future altimeter measurements. Overall, this effort should provide a better basis for using advanced filter techniques to improve altimeter sea‐level heights.

An automated software system has been developed to ride low cost, high-accuracy GPS-based orbit determination for low Earth orbiting spacecraft. The software system consists of a PERL shell script that controls the execution of the MicroCosm(R) precision orbit and geodetic parameter determination software system. The script is executed from command line inputs and is capable of providing end-to-end orbit determination for a low Earth satellite on a standard UMIX workstation or PC. The script is designed to automatically retrieve ground-based GPS tracking data from the IGS, solar flux and geomagnetic indices from NOAA, and earth orientation parameters from the USNO. The script then executes the processing modules required to generate the spacecraft ephemeris.The various levels of orbit accuracy that can be obtained for GPS-tracked spacecraft are assessed by computing orbits for the TOPEX/POSEIDON (T/P) and RADCAL satellites. The T/P spacecraft occupies a 1330 km altitude orbit and carries a dual-frequency P-code GPS receiver. T/P orbits computed with the automated software system are compared with a precise orbit ephemeris computed from laser and doppler tracking. Given the 3 cm radial orbit accuracy of the T/P POE, such comparisons help to define the current limits on GPS-based orbit determination using a variety of data types. The RADCAL satellite occupies an 800 km altitude orbit and carries a single-frequency C/A code GPS receiver. Such a combination of orbit altitude and GPS receiver is representative of many missions scheduled for launch within the next ten years. Processing the RADCAL data therefore provides some insight on the accuracy that can be obtained for satellites with altitudes lower than that of T/P and that are tracked with lower cost, reduced-capability GPS receivers.

This study is a preliminary analysis of the accuracy of various ionosphere models to correct single frequency altimeter height measurements for ionospheric path delay. In particular, research focused on adjusting empirical and parameterized ionosphere models in the parameterized real-time ionospheric specification model (PRISM) 1.2 using total electron content (TEC) data from the Global Positioning System (GPS). The types of GPS data used to adjust PRISM included GPS line-of-sight (LOS) TEC data mapped to the vertical, and a grid of GPS derived TEC data in a Sun-fixed longitude frame. The adjusted PRISM TEC values, as well as predictions by IRI-90, a climatological model, were compared to TOPEX/Poseidon (T/P) TEC measurements from the dual-frequency altimeter for a number of TIP tracks. When adjusted with GPS LOS data, the PRISM empirical model predicted TEC over 24 1 h data sets for a given local time to within a global error of 8.60 TECU rms during a midnight centered ionosphere and 9.74 TECU rms during a noon centered ionosphere. Using GPS derived sun-fixed TEC data, the PRISM parameterized model predicted TEC within an error of 8.47 TECU rms centered at midnight and 12.83 TECU rms centered at noon. From these best results, it is clear that the proposed requirement of 3-4 TECU global rms for TOPEX/Poseidon Follow-On will be very difficult to meet, even with a substantial increase in the number of GPS ground stations, with any realizable combination of the aforementioned models or data assimilation schemes.

A new method for wave motion detection from satellite altimetric measurements of sea surface height is presented. The essence of the approach is to construct a two‐dimensional traveling‐wave Fourier series representation of the amplitude field within a prespecified oceanic region. The method employs an iterative, nonlinear least‐squares technique based on the Marquardt‐Levenberg algorithm to solve for model parameters describing characteristic features of the evolving wave system. The Marquardt‐Levenberg Fourier series (MLFS) algorithm was applied to Kelvin waves active during the 1986–1987 El Nino event in the equatorial Pacific ocean using GEOSAT Exact Repeat Mission altimetry data. Characteristics of the wave system were found to be in essential agreement with earlier field measurements and the observations of Cheney and Miller (1987) obtained using time series developed from GEOSAT data. The advantage of the present detection scheme lies in its speed and ability to determine a wave system's dispersion relation over a finite range of wavenumbers, and hence the group velocity of that system.

The authors discuss the development of space geodetic systems-important tools in an attempt to understand the evolution of the atmosphere and oceans-particularly as they pertain to the problem of global warming. The science of time and frequency metrology has made possible the development of space geodetic systems that are being applied to the high accuracy tracking of new space-based instruments. These new instrument sets and methodologies have been designed to be appropriate for many years to decades of operation in order to measure the open ocean circulation and sea level as an indicator of global warming and will be a vital means for assessing the effectiveness of future remedial actions. The issues of time and frequency are tied into all of these space geodetic techniques, from the cesium and rubidium oscillators aboard global positioning system (GPS) satellites, to the epoch time and time interval measurements of satellite laser ranging (SLR) and hydrogen maser references used in very long baseline interferometry (VLBI). >

We examine here a novel yet simple technique for monitoring changes in global mean sea level over periods of weeks to years with an accuracy of a few centimeters using altimeter data from a well‐tracked satellite. The technique is based on the following argument. A satellite's orbit period and, hence, the mean semimajor axis of the orbit, are accurately determined by tracking systems located on the earth's surface. With laser or radiometric tracking systems, the mean semimajor axis can be determined with an accuracy of better than a centimeter. If the satellite carries an accurate altimeter, the difference between the mean semimajor axis and the global mean height measured by the altimeter is mean sea level. Any temporal change in the measurement of sea level must be due to drift in the altimeter or to change in the volume of the sea. To demonstrate the application of this technique, the fully corrected altimeter measurements of sea level obtained by Seasat during a 24‐day period when the satellite was in an exactly repeating orbit were separated into eight 3‐day segments and averaged over each of these periods. The calculated mean sea level relative to the Goddard Earth Model‐10B geoid varied over a range of ±7 cm during these eight periods. The variation is due primarily to errors in the vertical component of the satellite's ephemeris, which had a standard deviation of approximately 1.3 m. The results have important consequences. The proposed Topex/Poseidon altimetric satellite mission should have an ephemeris error of around 10 cm, and the satellite's ground track will repeat every 10 days. Thus, we expect that data from this satellite could be used to monitor variations in mean sea level with an accuracy of 1 or 2 cm during the 3‐ to 5‐year duration of the mission. Errors in the observations will be due primarily to either drift in the altimeter or orbit error. If an alternate technique is available to monitor the altimeter drift, the measurements can be used to monitor mean sea level with an accuracy which may be sufficient to detect global climate change.

Accurate knowledge of the position of an altimetric satellite is required for the altimeter range data to be effective in measuring ocean topography. This study addresses the use of high-precision altimeter data from NASA's TOPEX Mission in reducing the radial component of the orbit error of the U.S. Navy's N-ROSS satellite. Simulated altimetric crossing arc residuals between the TOPEX and N-ROSS orbits are minimized using both geometric and dynamic techniques in an effort to reduce the N-ROSS radial error to a level comparable to that of TOPEX. Tracking of N-ROSS by the Navy's NAVSPASUR system is simulated, and crossover residuals are created from the TOPEX and N-ROSS orbits. A simple geometric fit is shown to reduce the radial component of the NAVSPASUR N-ROSS orbit error from 350 m RMS to below 10 m RMS. In comparison, the dynamic approach of estimating the initial conditions of the N-ROSS orbit using a twentieth degree and order gravity field and a combined data set of tracking and altimeter crossover data yields a 6 m RMS residual error. Sub-meter accuracy can be attained by geometrically fitting these residuals to remove long wavelength orbit error.

Radar altimeters of great precision (≈10 cm), such as the one that flew on the SEASAT satellite, are capable of measuring the small oceanic height variations associated with geostrophic ocean currents. An experiment was conducted in the Kuroshio Current east of Japan, verifying this capability. Air‐expendable bathythermographs (AXBT's) were dropped to coincide with the SEASAT subtrack during flights on September 25 and October 5 and 13, 1978. Changes in surface dynamic height between flights were inferred from the AXBT data. They agreed generally to within ±10 cm of height changes observed in the altimeter data.

The SEASAT project, which demonstrated the feasibility of microwave oceanographic remote sensing, was sponsored by the National Aeronautics and Space Administration and managed by the Jet Propulsion Laboratory. The evaluation of the measurement system (elements of the satellite, the sensors, the data handling, and data processing subsystems) was a key activity of the SEASAT project. This paper summarizes the primary achievements and highlights the generic limitations of the evaluation process. Other papers in this issue present details of the geophysical evaluation process carried out under the auspices of the SEASAT project.

The first two steps in the evaluation of the performance of the first dedicated oceanographic satellite have been completed. The engineering assessment and sensor evaluation activities have thoroughly examined the performance of the instrument, and the first level of data processing algorithms, with encouraging results.

During some 3 months of orbital operations, Seasat collected a unique set of global synoptic data on ocean winds, waves, temperature, and topography. All indications from a preliminary analysis of these data are that most of the mission's proof-of-concept objective—the demonstration of nearly all-weather microwave surveillance of the world's oceans—will be met.

Preliminary analysis of radar altimeter data indicates that the instrument has met its specifications for measuring spacecraft height above the ocean surface (+/- 10 centimeters) and significant wave height (+/- 0.5 meter). There is ample evidence that the radar altimeter, having undergone development through three earth orbit missions [Skylab, Geodynamics Experimental Ocean Satellite 3 (GEOS-3), and Seasat], has reached a level of precision that now makes possible its use for important quantitative oceanographic investigations and practical applications.