The geostationary weather satellites are a critical component of the National Weather Service operations and on going nation wide modernization program. Geostationary satellites, because of their ability to constantly image the Earth, are important tools for observing severe weather such as hurricanes, severe thunderstorms, flash floods and winter storms. When satellite data are combined with other observing technologies such as weather radars, the operational forecaster has significantly increased their capabilities to produce timely, specific and very accurate short-term forecasts. The new generation of geostationary weather satellites are producing new products such as more frequent images over the United States, high density wind fields in the vicinity of hurricanes, a more comprehensive look at the phenomenon known as the Lake Effect Snow Storm, and the formation and dissipation of local fog areas. Atmospheric temperature and moisture profiles are now possible operationally with independent imager and sounder instruments. The sounder instrument is providing important information on the flow of low-level moisture from the Gulf of Mexico, an important source of energy for severe weather over the eastern part of the country.

Two successful launches of NOAA's Geostationary Operational Environmental Satellites (GOES) provide continuous images and atmosphere soundings from positions over the Pacific Ocean and the Atlantic Ocean. GOES-8, launched April 13, 1994 and GOES-9, launched on May 23, 1995 are the first three axis stabilized meteorological satellites ever placed in orbit. Their unique capabilities have provided over two years of opportunities for observing the atmosphere with 1 km resolution visible and 4 and 8 km infrared channels at 3.9, 6.7, 10.7, and 12.0 micrometers. In addition, the 19- channel atmospheric sounders provide soundings throughout the atmosphere over vast portions of the two oceans. This paper describes the status of current GOES satellites and provides a look into the future for coming spacecraft and instruments.

Future weather satellites for NOAA at geosynchronous orbit may be smaller, less costly, and developed by a different process than is currently done. This path is sometimes called the 'smaller, cheaper and faster' process being pursued by NASA. We believe in the future there will be less money, a focus on using the right technology and the desire to get the most value for the resources invested in space missions. In this paper we give an update on our progress to define future GOES. It will include our efforts to trade on user requirement early, to use evolutionary technology, and to consider new cost reduction and program management techniques.

By using current technology, it appears possible to build and launch a prototype for an advanced geosynchronous imager on a small within a few years. This could be done independently of upgrading the other GOES mission functions of atmospheric sounding, communications, and space- and solar-monitoring. At NASA-GSFC, we are engaged in a feasibility study for a Geosynchronous Advanced Technology Environmental System (GATES). GATES is envisioned as a high performance imager on a small dedicated satellite with a complete ground system. GATES could fly in the era of the Earth Observing System and serve as a prototype for NOAA's next generation of operational satellites. In addition to carrying all the channels identified for NOAA's GOES-R Imager, GATES is being designed to carry all the broadband channels specified for NASA's Moderate-resolution Imaging Spectro-radiometer instrument on the EOS platforms. This channel complement gives GATES the capability to fill in the space-time gaps in cloud observations from polar-orbiting satellites and to serve as a cross-reference between polar radiometers. Multispectral rapid-imaging requirements are met by using several recently developed technologies: large detector arrays with active cooling, star-tracking and gyroscopic attitude-determination, a small and rigid spacecraft, a heat-resistent telescope, a phased-array Ka- band downlink, realtime digital image rectification, and Internet data distribution on the ground. The GATES design is so small and agile that it could use the momentum wheels to scan the entire spacecraft back-and-forth across the Earth.

The new series GOES-8 and -9 launched in 1994 and 1995 provide more flexible instrument coverage and higher resolution than previous GOES spinners. This added flexibility and the 3-axis stabilized operating mode however resulted in a more complex satellite system with independent imager and sounder, and on-board image navigation systems requiring more daily commanding. Nearly 5000 realtime commands are currently sent each day to each GOES-8 and -9 spacecraft compared to 200 commands per day for GOES-7. These new technological advancements in spacecraft design presented new challenges for the NASA operations support personnel. In order to prepare for launch, post-launch test, and on-orbit operations, a rigorous mission planning scheme was developed to assure safe commanding of the flight system and monitoring of its state of health. This paper overviews the key mission operations approaches and philosophies developed for the GOES I-M missions and key operations tools that were developed to aid operations personnel in performing complex routine and special operations tasks.

The successful launch of GOES-8 in 1994 introduced an enhanced capability for diurnal monitoring of subpixel fire activity and aerosol transport in the Western Hemisphere. The higher spatial and temporal resolution, greater radiometric sensitivity, and improved navigation of GOES-8 offer many advantages for monitoring fires and smoke in North, Central, and South America. In South America the GOES-8 automated biomass burning algorithm (ABBA) is being used to continue monitoring trends in biomass burning associated with agricultural practices and deforestation activities as well as documenting the extent and transport of associated aerosols. GOES-8 ABBA results obtained during the 1995 biomass burning season indicate a strong diurnal cycle in fire activity and associated aerosol transport regimes extending over millions of km². Examples of GOES-8 diurnal monitoring of fire intensity and size in the United States, Canada, Mexico, Guatemala and Belize show the utility of using GOES-8 as an early warning mechanism for identifying and monitoring wildfires in these regions. The success of the GOES-8 ABBA in the Western Hemisphere suggests the utility of initiating a global geostationary fire monitoring effort.

Automated procedures for deriving cloud-motion vectors from a series of geostationary images have been developed by the Cooperative Institute for Meteorological Satellite Studies (CIMSS) and have been operational in NOAA since 1993. Since that time, CIMSS has continued work to improve the processing techniques, upgrade the quality of the product, and add further capabilities. Scientific improvements include the addition of the water-vapor intercept technique for assigning heights to semi-transparent clouds, a new version of the automated quality control algorithm which reduces the large mean slow bias observed in the initial system, and improved selection of suitable tracers. These improvements have been successfully applied to the 'Day 1' GOES-8/9 operational winds system in NOAA. Research has already begun to define 'Day 2' operations. Improved resolution and signal of the GOES-8/9 imager has made automated water-vapor motion vector production possible for the first time, enabling the measurement of wind velocities in clear air. Work has begun on an automated algorithm for imagery registration quality control and the optimal density for wind products as a function of coverage and computing power is being investigated.

Marine statiform clouds (MSC) cover large areas of the globe that are visible to GOES. The operational satellite cloud retrieval algorithms are prone to biases when analyzing MSC, due to the often sub-pixel size cloud elements and radiative temperatures close to that of the underlying ocean. For example, the relatively large pixel size and calibration drifts in GOES-7 imagery have made it difficult to extract unbiased MSC properties using thermal threshold techniques. Here, we apply a novel retrieval approach to the two important MSC regimes which can be monitored well from the GOES-8 satellite: the Pacific Ocean just west of California/Baja and Peru/Chile. MSC cloud parameters for these areas are retrieved together with surface temperature and column water vapor in a temporally and spatially consistent manner that is insensitive to sensor resolution and calibration errors. Semi-operational analysis of GOES-8 imagery began in December 1995. So, the main focus is on assessing the diurnal variability of MSC. Following a brief description of the retrieval technique, we present initial results describing the full diurnal cycle of MSC fractional cloud cover and cloud top temperature, monitored using the single-channel version of the algorithm. In addition, we address the daytime variability of other important cloud parameters using a bispectral extension of the retrieval scheme. The results are also compared with other pertinent MSC analyses.

Geostationary orbit provides a wonderful viewpoint for collecting movies of meteorological processes. Because GOES- 8 images are large and frequent, high-performance workstations and user-friendly software are required to review the current pools of GOES weather imagery at full resolution. During the post-launch checkouts, the GOES-8 and GOES-9 satellites were exercised in rapid-scan imaging modes that challenge most display systems. At NASA-GSFC, the Interactive Imaging Spread Sheet (IISS) software/hardware was used to roan and zoom through gigabytes of GOES image data collected in 1994-95. We use the IISS to examine exciting meteorological features within the hurricanes of 1995 and during typical episodes of severe weather that occur somewhere over the United States every day. The IISS demonstration shows how weather forecaster could make good use of the full resolution and depth of GOES-8 imagery, adapting to the meteorological even as it develops, especially when synchronized with corresponding weather radar.

The new generation GOES, with significantly higher spatial resolution and signal to noise ratio than its predecessors for the split window channel, makes it feasible for the first time to derive SST from a geostationary platform. The GOES data are well calibrated and navigated. The temporal information from GOES can be used to identify residual cloud contamination and detect diurnal variation of SST. All these features are valuable assets for monitoring the earth environment from space.

A method has been developed to derive both height and motion from satellite images using a purely geometrical technique. This is a combination of using stereo to measure height by viewing cloud from two view point and using several images in time to measure motion of clouds. The method requires that the cloud be viewed from two or more different perspectives with at least two of the views being at different times. With the new method time coincidence is a disadvantage, and higher accuracy is obtained when all views are at different times. For height verification, improvements have been made in the cloud height by shadow method. Accuracies of better than 1 km height and 0.5 m/sec are typical of the results.

A major problem with multi-spectral satellite imagery is that images in many of the spectral channels contain redundant information about the atmosphere. By using principal component (PC) analysis to transform multi-channel satellite images, this information redundancy can be reduced. PC image (PCI) analysis finds the information that is common among the various channel images and puts that information into the PCIs in descending order of significance. The first PCI contains common information, leaving other information for higher-ordered PCIs. A second PCI is then formed which contains information common to the channel images other than that explained by the first PCI. The process continues until the number of PCIs is equal to the number of channel images being transformed. If none of the original channel images contains redundant information, then a PCI transformation is not needed. However, this is not the case with most satellite images. Because of channel image redundancy, the number of useful PCIs is often less than the number of channel images being transformed. The highest-order PCIs may contain only noise or slight differences among some of the channel images, however, these differences are of importance, especially for less obvious meteorological features in the atmosphere. Many interesting examples of PCIs created from GOES-8/9 imager and sounder data are possible. For the 5-channel GOES imager, the interpretation of the PCIs is fairly predictable when they are created on a large spatial scale. The main difference in interpretation occurs between day and night when the presence or lack of visible radiation is an important factor. For the GOES sounder, the interpretation of the PCIs is more complex, considering that input can consist of up to 19 channel images. Different subsets of the sounder channel images result in different products. With proper selection of channel images, emphasis can be placed upon either temperature or water vapor features in the atmosphere, or on features of the ground surface or clouds.

Estimation of the atmospheric wind field based on cloud tracking using a time sequence of satellite imagery is an extremely challenging problem due to the complex dynamics of the imaging instruments and the underlying non-linear phenomena of cloud formation and weather. Cloud motion may involve both partial fluid motion and partial solid motion, which we model as semi-fluid motion. Motion algorithm with subpixel accuracy using differential geometry invariants of surfaces was developed to track clouds. The motion model is general enough to include both physical and geometrical constraints. Typically, a polynomial displacement function is used to model the local deformation behavior of a surface patch undergoing semi-fluid motion. The cloud tracking algorithm recovers local cloud surface deformations using a sequence of dense depth maps and corresponding intensity imagery, that captures the time evolution of cloud-top heights. Either intensity or depth information can be used by the semi-fluid motion analysis algorithm. A dense disparity or depth map that can be related to cloud-top heights is provided by the Goddard Automatic Stereo Analysis module for input to the motion analysis module. The results of the automatic cloud tracking algorithm are extremely promising with errors comparable to manually tracked winds. Experiments were performed on GOES images of Hurricanes Frederic, Gilbert and Luis, and a temporally dense 1.5 minute time interval thunderstorm sequence covering Florida region. Future work involves using multispectral information, incorporating robustness, cloud motion segmentation and adaptive searching for improving operational cloud-tracking performance.

The new generation of Geostationary Operational Environmental Satellites (GOES) have an imager instrument with five multispectral bands of high spatial resolution,and very high dynamic range radiance measurements with 10-bit precision. A wide variety of environmental processes can be observed at unprecedented time scales using the new imager instrument. Quality assurance and feedback to the GOES project office is performed using rapid animation at high magnification, examining differences between successive frames, and applying radiometric and geometric correction algorithms. Missing or corrupted scanline data occur unpredictably due to noise in the ground based receiving system. Smooth high resolution noise-free animations can be recovered using automatic techniques even from scanline scratches affecting more than 25 percent of the dataset. Radiometric correction using the local solar zenith angle was applied to the visible channel to compensate for time- of-day illumination variations to produce gain-compensated movies that appear well-lit from dawn to dusk and extend the interval of useful image observations by more than two hours. A time series of brightness histograms displays some subtle quality control problems in the GOES channels related to rebinning of the radiance measurements. The human visual system is sensitive to only about half of the measured 10- bit dynamic range in intensity variations, at a given point in a monochrome image. In order to effectively use the additional bits of precision and handle the high data rate, new enhancement techniques and visualization tools were developed. We have implemented interactive image enhancement techniques to selectively emphasize different subranges of the 10-bits of intensity levels. Improving navigational accuracy using registration techniques and geometric correction of scanline interleaving errors is a more difficult problem that is currently being investigated.

Desktop computers have evolved to the point where they can process GOES variable-length (GVAR) data blocks at 1 Gbyte/hour, despite the complexity of the data stream and the varied demands of realtime users. Because of the lack of off-the-shelf GVAR processing packages before the launch of GOES-I in 1994, we assembled a custom open system at NASA- GSFC to automatically ingest, process and disseminate full- resolution GOES-8 and GOES-9 imagery 24 hours per day. This data service is now widely used by the internet community. The hardware-software package described here can be copied freely. Similar systems are now available from commercial vendors.

This paper addresses imager evolution from the early geostationary imagers to the present and includes changes in requirements which drove instrument development and design. The GOES missions are compare with emphasis on development, performance differences, and design challenges. The imager overview section focuses on the current system design and contains a system description, instrument characteristics, operational performance, a summary of the test program, and the launch schedule. The key instrument design areas discussed are the optical design, scan control and operation, thermal control design, in-flight calibration, and operational data distribution. Technical advancements and differences between the current GOES imager and its immediate predecessors are addressed. Plans for follow-on GOES missions beyond the GOES I-M series, in the context of the evolution of GOES imager instruments, provide a view into technological developments in the near-future.

The GOES Sounder is similar in appearance, size, and design to the GOES Imager. It provides independent radiometric sounding of the atmosphere from GOES-East and GOES-West, which was accomplished on previous GOES by a single instrument with responsibility split between the functions of imaging and sounding. With 19 channels ranging from 0.70 micrometers to 14.71 micrometers , the Sounder probes the atmosphere to measure radiance at different depths while also monitoring surface and atmospheric temperatures, ozone, water vapor content, and cloud cover by means of the calibration of radiance data at selected wavelengths. Like the Imager, the Sounder scans the full earth, and can be commanded to scan local regions of interest. The scan mirror steps across the disk of the earth in synchronization with a dead zone on a filter wheel. The filters define 18 of the channels as they rotate on the wheel in the path of light split into three beams traveling to detectors divided into longwave, midwave, and shortwave arrays. For visible sounding of cloud cover, energy is split off ahead of the filter wheel and passed to the 19th channel and also to star sensing detectors used for instrument navigation.

Measured infrared reflectance and transmittance data, as a function of wavelength and incidence angle, are presented for polarization inducing components on a typical Geostationary Operational Environmental Satellite (GOES) atmospheric sounding instrument. The data are then used to construct Mueller matrices from which instrument polarization sensitivity, not only to incoming radiance from an observed scene, but also to the instrument's own radiant signature. A comparison is made between the polarizing characteristics of a typical filter-type infrared sounding radiometer and a proposed Micheson-type FOurier transform alternative.

Geostationary earth observation satellite applications almost always require high pointing accuracy. The ability to meet very high pointing accuracy, at or below microradian levels, is limited by our ability to measure and correct pointing errors in the laboratory environment. Certain types of pointing errors are invariant over time and appear as fixed pattern over the field of view. These errors may not even be sensed by the servo. Such errors can be measured by other means and corrected through the servo in an open loop manner. This paper describes the method developed to measure the fixed pattern errors (FPE) associated with a servo controlled, two degrees of freedom mirror arrangement by autocollimation of a theodolite. This system has been developed for GOES-8 and above S/C instruments. Several interesting optical phenomena were observed during measurement of the FPE. Mathematical models have been developed to help understand and explain these phenomena. The paper also describes how several numerical techniques were used to simplify the laborious testing by minimizing the number of measurements without compromising pointing accuracy. Fourier analysis of the test data show that the FPS's are reduced to within +/- 4.5 microradians. The measurement accuracy obtained in these tests are believed to be at least two times better than the measurement accuracy reported so far using theodolites or other similar instruments.

Post launch tests on the GOES satellite cameras called for determination of the positions of various imaging detectors. The GOES spacecraft's two cameras, the imager and sounder, are both scanners that use focal plane arrays. The below method determines their detectors' relative positions by recording detector outputs as the edge of the moon drifts past their instantaneous fields of view. While this method works well with simulated data, it is still experimental and is still being tested with flight data.

The GOES program seeks a reliable method to measure the imager's modulation transfer function (MTF) once the instrument is in orbit. This has proven difficult since there is no convenient target. It has been proposed to use the edge of the moon as a high contrast target, offering the possibility of performing a 'knife-edge' MTF test. The following treatment produces a visible, east-west spread function (ESF) from imager scans over the lunar limb. The ESF is used both to produce a visible knife-edge MTF estimate as well as to produce, through differentiation, a visible line spread function, which also is used to produce an MTF. While the results do not rigorously conform to ground tests, they track low and medium spatial frequency test values reasonable well and offer a means to test the instrument's imaging fidelity on-orbit.

The GOES-8 Imager and Sounder Radiometric Performance Models are flexible tools intended to provide a intuitive user interface to instrument sensitivity analysis. The models were developed at Lockheed Martin IR Imaging Systems and transferred to the GOES modeling community as the standard for projecting the instruments' IR sensitivity in units of noise-equivalent delta radiance, under arbitrary user- imposed conditions. In addition, the models provide an efficient means for selecting flight IR detectors with the highest possible instrument performance. THe radiometric models have shown the way to several instrument-level performance improvements, and have also been a valuable tool in troubleshooting anomalous measurements at various stages of the GOES instrument build. The operation of the GOES-8 SOunder Radiometric Performance Model is described herein in terms of input parameters, calculational flow, and output results.

During the Fall of 1994 eclipse season, GOES-8 Sounder telescope secondary mirror temperature exceeded qualification limits, while operating normally in the vicinity of the Sun. Based on a simple conservative thermal analysis, which was performed in January of 1995, it was determined that there was a health and safety issue with operating Imager or Sounder in the vicinity of the Sun. As a result, a set of operating constraints were levied on both instrument normal operations, limiting them from scanning in the vicinity of the Sun. Since the simple thermal analysis was believed to be conservative, ITT was tasked to develop a detailed model of the secondary assembly and to determine the effect of focused energy on the secondary assembly critical components. The detailed model of the secondary assembly was developed and several full disk frame cases were analyzed to determine if the keep out zone may be eliminated or reduced. In addition to the full disk frame cases several CEI initialization cases were also analyzed for the Imager post eclipse instrument urn on scenario. The results of the detailed analysis indicates that the simple thermal analysis done in January of 1995 was in fact not conservative. It was determined that the Imager secondary temperature will far exceed the mission allowable temperature of 47C, while performing full disk frames in the vicinity of the Sun. The maximum inner baffle temperature was predicted to get as high as 130C for the worst full disk frame case analyzed. The thermal modeling effort as well as the results of the analysis and recommendations for future instrument qualification test and design changes will be documented in this paper.

Starting with SN05 Imager the optical MTF of the visible optics assembly has been measured at subsystem and at system level testing. For the visible detector MTF at 18,000 cycles/radian relative differences of 16 percent have been observed between system and subsystem data. The test setups are quite different, however, the MTF values should be dominated by the detector's field of view MTF and thus the differences should be small. Descriptions of the two configurations are given along with a discussion of some of the important differences. The investigation into the differences led to the testing of the imager and test collimator with a parabolic mirror as collimator and an MTF analyzer. The MTF values using the parabola with the instrument were about 12 percent better than with the test collimator. Observations of the image of a pinhole at the Cassegrain focus led to the analysis of spider vane diffraction which accounts for 2.7 percent MTF reduction. In addition analysis has also been performed on the collimator interferogram that indicates a reduction in MTF due to a number of steep zones within the collimator wavefront. The analysis predicts a 9.7 percent reduction in MTF, at 18,000 cycles/radian due to the test collimator of which 1.4 percent is a result of the collimator secondary spider vanes. The verification of the analysis awaits a new collimator with slope errors reduced to less than 0.12 waves/inch.

The mission of the NOAA Space Environment Center (SEC) is to serve the nation's need to reduce adverse effects of solar- terrestrial disturbances on humankind's activities. To meet this need, SEC: 1) acquires, interprets, and disseminates space weather information; 2) prepares and disseminates forecasts and alerts of conditions of the space environment; 3) conducts research and development in solar-terrestrial physics and in techniques to improve monitoring and forecasting; 4) prepares high quality data for national archives; 5) uses its expertise to advise and educate those affected by variations in the space environment. Users are provided information in the form of forecasts, nowcasts, data, advice, support, expertise, and publications about conditions in the solar-terrestrial environment. The space environment monitors on GOES spacecraft provide space weather observations from the Sun to Earth and form the basis of the SEC real-time operation. The X-ray sensor (XRS) monitors solar flare activity and serves as the international standard for rating the intensity of flares. Flares are observed and classified according to the intensity of their emission on the XRS sensor. Forecasts of the occurrence of solar flares are expressed in terms of the measurements from the XRS. SEC also issues a daily index of the background solar radiation based on the XRS measurements. SOlar particle events (SPE) and energetic electron event s are detected using the energetic particle sensor (EPS) on GOES. Alerts and forecasts of the occurrence of SPE are made in terms of the fluxes of charged ions (mostly protons) measured by the EPS. Alerts are issued for energetic electron events based on the EPS measurements.

A central mission of the GOES space environment monitor program is to provide continuous, real-time monitoring of the near-Earth space environment. The GOES energetic particle sensor (EPS) and high energy proton and alpha detector (HEPAD) obtain measurements of the energetic electron, proton, and alpha particle fluxes at geosynchronous orbit. These measurements serve as the basis for a variety of services that are provided to numerous government and private organizations, including the forecasting of adverse conditions, providing a characterization of the current conditions in the Earth's environment, and providing the necessary measurements to allow comprehensive post-event analyses. Numerous products are derived from the GEOS EPS measurements and made available to industry, government agencies, and research institutions throughout the world. The particle fluxes in the Earth's environment are comprised of three main components: 1) particles trapped within the geomagnetic field, 2) particles of a more direct solar origin, and 3) a cosmic ray background. These particle fluxes are quite dynamic, varying on times-scales ranging from seconds to decades. In this paper, we describe the GOES-8 and GOES-9 EPS and HEPAD instruments and the operational uses of the energetic particle measurements made at geosynchronous orbit.

The X-ray sensor (XRS) on the GOES provides a standard reference for essentially continuous monitoring solar activity and characterizing solar flares. Disk-integrated x- ray fluxes observed by XRS are used by forecasters and researchers around the world as a measure of the strength and duration of solar flares. The peak 0.1-0.8 nm x-ray flux during flares is used to distinguish between C, M, and X flares; flares that differ by an order of magnitude in the peak flux. Forecasters use this peak flux to predict the magnitude of proton events, and the x-ray duration is used to estimate whether coronal mass ejection may have occurred that could cause a geomagnetic disturbance if it hits the Earth. Recipients of the data use the peak flux and the duration of the flare to estimate the disturbances expected on radio communication systems. The magnitudes of XRS- observed flares are also used to determine when to issue alerts of changed communication systems. The magnitudes of XRS-observed flares are also used to determine when to issue alerts of changed ionospheric conditions that can disrupt communications and GPS signals. XRS fluxes are also used to augment solar radio observations to alert users of radio frequencies of times when the solar signal may interfere with their operations. The non-flaring x-ray flux, otherwise known as the x-ray background flux, is used as a proxy for he solar EUV emissions that are used to predict the atmospheric density as satellite orbits; variations in the daily averaged solar x-ray flux are used to estimate changes in the atmospheric drag on spacecraft orbits.

Magnetic field measurements have been made form geosynchronous orbit for more than 20 years. These measurements are important for monitoring 'space weather' and for providing a unique data base that can be used for improving our knowledge of the Earth's magnetosphere and solar-terrestrial interactions. This paper will focus on the variety of products and services provided by these measurements--those currently available, and those under consideration for the future. The magnetic field assist forecasters in qualitatively assessing the level of geomagnetic disturbance, to interpret changes in energetic particle measurements, to provide data to the National Geophysical Data Center, to support in real-time scientific activities such as rocket launches, and to conduct research for a better understanding of the space environment. One important use of magnetometer data in the Space Environment Center is to alert customers when shocks occur in the solar wind. These shocks have the potential for energizing particles to multi-MeV levels, causing Single Event Upsets in spacecraft electronics, and at lower energy ranges causing deep-dielectric charging that produces spacecraft anomalies. Data from the new GOES-8 and GOES-9 spacecraft will be discussed along with prospects for future products and services.

The first solar x-ray imager (SXI) will provide a major advance in real-time, continuous monitoring of solar- terrestrial conditions. This instrument, which will fly on a Geostationary Operational Environmental Satellite (GOES), will provide full-disk images of the Sun once a minute in the 0.6-6 nm range with 5 arcsec pixels. SXI's images will complement x-ray fluxes from the disk-integrating GOES x-ray sensor and optical images from ground-based observatories. THe automated sequence of SXI images will make it easy for forecasters, researchers, and image processing algorithms to interpret the images. SXI is being built to meet five operational goals for real-time prediction of solar- terrestrial disturbances: 1) SXI will provide clear evidence of x-ray coronal holes that are associate with recurrent geomagnetic storms. 2) SXI will provide flare locations that are used to estimate the magnitude and timing of energetic particle events, including flares from regions behind the solar limb that are not visible at other wavelengths. 3) SXI could provide a significant improvement in forecasting geomagnetic disturbances if CME-associated brightenings can be readily observed. 4) SXI images will show the complexity of the active regions, which will be used to estimate each region's flare potential.

Conditions in the near-Earth space environment are of every increasing importance to our human activities on Earth and in space. The provision of the space environment services required in future depends on improving our understanding of solar activity and the coupling of this activity to our local region of space, as well as improving our remote sensing and in-situ monitoring of conditions and events in the solar system. Our present service is largely analogous to the state of terrestrial weather forecasting rom a local weather office before the advent of numerical modeling and remote atmospheric sensing technology. Numerical models of the local space environment and of interplanetary space are being developed. However, these models are limited in performance by our understanding of the underlying physical processes, and their practical applications is restricted by the paucity of observational data. Instruments on the GOES provide a critical resource to NOAA's space environment services. GOES is our most effective operational platform for real-time remote sensing of the Sun, the near-Earth environment, and processes in interplanetary space. It also makes important in-situ measurements in a critical region of space that is now of huge commercial importance. This paper will discuss the planned and potential extensions of the GOES space environment monitor in the overall context of the data required to meet the future needs for space environment services.

This paper discusses the issues associated with correcting GOES spacecraft magnetometer data for the effects of the spacecraft platform. The effects discussed include: varying fields due to moving permanent magnets, stray spacecraft fields resulting form the flow of electric currents, the effects of spacecraft noise, and magnetometer shift due to temperature. Issues associated with the interpretation of on-orbit data are discussed, and approaches to overcoming these issues are offered. The use of ground processing to compensate for spacecraft magnetic effects is analyzed based on the results from on-orbit data and pre-launch test data. The results form analyzing GOES-8 and GOES-9 on-orbit data indicate that the combined effect of the errors associated with torque current ambiguity, torquer corrections, other uncorrected stray fields, results in a root of the sum of the squares error of 0.4 nT. The most significant concern in achieving high accuracy in magnetic measurements of the Earth's magnetic field with these GOES spacecraft was found to be the control of magnetic contamination on the magnetometer boom.

The GOES-8 solar x-ray sensor (XRS) detects solar x-rays in two wavelength bands of approximately 0.5 to 3 angstrom and 1 to 8 angstrom. The XRS uses a dual ion chamber design with beryllium windows and Xenon or Argon gas fills to provide the x-ray detection, and which determine the wavelength response functions. GOES spacecraft before GOES-8 were spinning, and the previous XRS design used this property to measure the solar x-rays as a 'pulse' above the ambient particle background during the time when the XRS FOV scanned across the sun. GOES-8 and later spacecraft are three-axis stabilized, and the XRS now views the sun constantly from a mount on the solar panel yoke. This puts a severe requirement on the XRS for shielding of the ion chamber, since the background current form ambient particles must be will below the current from the design threshold x-ray fluxes. The design and calibration of the XRS is described, as are results from electron irradiation which were used to verify the immunity to ambient particle fluxes.

The energetic particle sensor (EPS) and the high energy proton and alpha detector (HEPAD) measure protons, alpha particles and electrons in several energy ranges. The latest version of the EPS for GOES-8 and succeeding GOES spacecraft detects protons in 7 energy ranges from 0.8 to 500 MeV, alpha particles in 6 energy ranges from 350 to greater than 700 MeV, and alpha particles in 2 energy ranges from 2560 to greater than 3400 MeV. Most of the particle channel responses of the EPS and HEPAD have been calibrated directly at available accelerators. The GOES particle data are used to monitor the ambient, trapped, particle radiation, which is predominately MeV electrons, and the proton and alpha particle fluxes from solar particle events. The data are used for various types of warnings and are archived by the NOAA space environment center.

The GOES-8 magnetometer is one of the most sophisticated and highest performance fluxgate magnetometers to ge flown in space. Its 16-bit analog-to-digital converter (ADC) can resolve magnetic field strengths as low as 0.03 nT. It can measure fields to an accuracy of 1 nT over a temperature range of -80 degrees C to 72 degrees C. This paper describes the GOES-8 magnetometer design and performance. The magnetometer consists of a very low noise three-axis analog magnetometer, a high resolution dual slope integrating ADC, and a redundant digital serial spacecraft interface. Another feature of the magnetometer is its ability to generate a calibration sequence upon command. The 16-bit ADC design was the most challenging. There were no 16-bit grade 1 ADCs on the market at the time (1986), so it was necessary to design the ADC using the limited number of grade 1 parts that were available.

An engineering design of an imager upgrade is under consideration for the GOES-N/Q series. The upgrade consist of adding up to three new IR channels at 4 km resolution and doubling the earth coverage rate. The design approach introduces advanced technology as needed to minimize impacts to the sensor optical and mechanical configuration, electronics box layout, data communication links and ground systems. By operating presently redundant portions of detector arrays and developing double rate signal processors, the scanning servo system and electronics modules are largely retained. Bicolor detectors and optical coatings are used to add channels while retaining the present relay optics and radiant cooler layouts. Lossy data compression of visible imagery and lossless data compression of IR imagery are used to preserve the sensor data and precessed data relay communications links. System NEDT and SNR performance and implementation issues for new technologies are addressed.

Replacement of the sounder filter wheel in one of the clones of the GOES I-M satellite series with a Fourier transform infrared (FTIR) interferometer would improve retrieval performance and could be a pathfinder for technology employed in the next generation GOES series. Based on examination of an earlier proposal, NOAA decided to pursue the feasibility of such a FTIR replacement. The FTIR interferometer is to be as nearly a 'plug-in' replacement for the sounder filter wheel as possible. THis imposes some interesting and challenging optical, mechanical, and electronic constraints on the design and fabrication of the interferometer. This overview describes a brassboard design for a replacement FTIR interferometer and provides the background and the status of GHIS brassboard work-in- progress at MIT Lincoln Laboratory.

Interim results of a current study on upgrading the GOES infrared Sounder are presented. Considered are a 15 cm diameter telescope to reduce instrument size and weight, use of a Fourier transform infrared (FTIR) interferometer for high spectral wavelength resolution, a small detector focal plane array operating at 65K, and combining the instrument with a microwave sounder. Retrieval performance improvement from the FTIR sounder is estimated.

The results of a conceptual design study of an advanced Imager for GOES-R is presented. Tentative performance requirements are established from NOAA operational and science communities. A performance scaling law is developed which provides a quantitative measure of sensor capabilities and insight into the key design parameter tradeoffs. The basic design concept is described and estimates for sensitivity and resolution are given.

An evolutionary improvement to the existing GOES I-M Sounder has been proposed as the GOES high-resolution interferometer Sounder (GHIS). This instrument will be based upon technology developed during successful demonstrations of an airborne Michelson interferometer instrument for the temperature and moisture sounding of the atmosphere from passive infrared measurements. The spectral resolution afforded by the interferometer technology will substantially improve retrieval accuracy and vertical resolution of the sounding information. Simulations of the spectral radiance measurements have been used to study the expected retrieval operator in an iterative, non-linear scheme. The effects of instrument noise and spectral operating mode have been considered in the retrieval simulation. A measure of the vertical resolution of the retrieved profiles has been developed. This measure considers the correlation in the retrieval error between atmospheric levels of the retrieved parameters. The effects of instrument noise and spectral operating mode have been considered in the retrieval simulation. A measure of the vertical resolution of the retrieved profiles has been developed. This measure considers the correlation in the retrieval error between atmospheric levels of the retrieved parameters. The effects of both the instrument characteristics and those of the retrieval algorithm are thus considered in this measure. The results indicate that the GHIS is expected to provide temperature retrieval accuracy on the order of 1 degree C throughout the troposphere which will be a significant improvement over the existing GOES-8 sounder. The vertical resolution simulations also show improvement over GOES-8 capabilities.

Future GOES sounders will likely require increased spectral resolution and number of channels in order to meet the sounding vertical resolution and accuracy specifications set forth by the science community. The sensor technology and systems group at MIT Lincoln Laboratory has pursued the use of a high resolution interferometer (GHIS) is order to meet these spectral needs. This paper will focus on the electronics and signal processing necessary for achieving the noise fidelity, dynamic range and data rate constraints for the GHIS instrument. This will include the absolute fringe counting interferometer, analog filtering, A/D conversion, and digital processing. The paper will highlight techniques for reducing the interferometer dynamic range and data rate in order to simplify a GHIS flight design.

The GOES Sounder is a 19-channel discrete filter radiometer. Studies by the University of Wisconsin and MIT Lincoln Laboratory indicate that retrieval performance could be improved y replacing the filter wheel with a Michelson interferometer. The interferometer system is to replace the relay optics and filter wheel and is required to use the existing GOES telescope and detector subsystems. Numerous configurations were considered in an effort to define a design which meets the optical specifications and fits within the limited available space with sufficient clearance around the optical components for the mechanical housings. The selected interferometer subsystem design fits within the available volume without rotation of the star sensor. Performance, diffraction and sensitivity analysis results are included.

The GHIS FTIR spectrometer is designed to be a plug-in replacement for the low-resolution filter wheel spectrometer employed in current GOES sounders. State-of-the-art retrieval algorithms will be able to produce atmospheric temperature and humidity profiles with improved vertical resolution and smaller uncertainties as a results of the instrument's higher spectral resolution. MIT Lincoln Laboratory has designed, assembled, and performed visible- wavelength tests of a GHIS brassboard spectrometer. Details of the assembly procedure will be presented, showing how the optical components were individually tested, then pre- aligned and integrated into the instrument at room temperature for eventual operation at a nominal 220K.

GOES long wave sounder (LWS) detector requirements have always pushed the state-of-the-art for longwave detectors operating in the vicinity of 102 K. Performance and yield of acceptable detectors have always been problems and continue to be important issues affecting the performance of instruments of both present and future design. GSFC has been examining new device and operational concepts aimed at producing significant improvements in performance and yield. Our approach has been directed towards mitigating the deleterious effects of operating small geometry HgCdTe PC devices under heavy bias, that is, under minority carrier sweepout, as is typical in conventional LWS detector operation. Specifically, theory indicates that detectors of the new design operating under optimal bias conditions have significantly higher responsivity, lower power dissipation,and lower 1/f noise knees than conventional LWS detectors. In this paper we will describe the new LWS detectors fabricated at GSFC, present detector data, and review the theory of operation of these devices.

There has been significant progress made during the past several years in PV HgCdTe technology for advanced long wavelength remote sensing applications. Useful cutoffs wavelength shave been extended to beyond 17.0 micrometers . Junction quality has been improved to the point that D* > 3 X 10¹¹ cm-(root) Hz/W can be achieved at temperatures of 60-65 K. The Atmospheric Infrared Sounder (AIRS) instrument, scheduled for launch in the year 2000 as part of the NASA EOS program, uses long linear multiplexed arrays of PV HgCdTe detectors with cutoff wavelengths extending as far as 15.0 micrometers at 60 K. PV HgCdTe offers many advantages over PC HgCdTe for multiplexers are possible, backside-illuminated 2D arrays of closely spaced elements, 10X-100X better linearity, dc coupling for measuring the total incident photon flux, and a (root) 2 higher BLIP D* limit. In this paper we compare the relative merits of PV and PC HgCdTe for advanced remote sensing instruments, we review recent data for linear arrays of PV HgCdTe with cutoff wavelengths as long as 17.5 micrometers at 70 K, and we project that PV HgCdTe should be able to meet or exceed the present demanding GOES LW Sounder D* requirements at T equals 100 K, with the additional benefits of negligible 1/f noise and better linearity.

The GOES Imager and Sounder instruments each utilize several HgCdTe photoconductive (PC) detectors and detector arrays for detection over the 6.5 to 14.7 micrometers region. These high performance detectors are integrated with germanium aplanat lenses and mounted in miniature hermetically sealed housings. There are demanding requirements on the radiometric performance of these detector assemblies. For LW Sounder detectors, the highest possible sensitivity achievable by a practical HgCdTe photoconductor at the operating temperatures of 100 to 105 K was required. Lockheed Martin designed, fabricated, tested, packaged, qualified, and delivered 7 of the 11 HgCdTe PC detector assemblies for GOES-8, and 9 of the 11 assemblies for GOES- 9. All the n-type HgCdTe starting material was grown at Lockheed Martin.

Advanced detector fabrication technology and high reliability packaging processes have been developed for the manufacturing of high performance HgCdTe photoconductors. These infrared array assemblies for use in GOES and other weather satellites operate at radiative cooler temperatures ranging from 95 to 115 K. A large quality of flight detectors sensitive to spectral wavelengths ranging from 7 to 16 micrometers have been fully characterized. Detector performance data as a function of bias, temperature, frequency and cutoff wavelength are presented. A computerized model has been developed and reasonable agreements between computer projections and measured performance are obtained. This model has been used successfully to identify an optimum set of materials and device parameters for a given set of system requirements. In addition, advanced assembly and packaging techniques have been developed to ensure tight alignment tolerances, long- life hermeticity, low-outgassing and low internal reflection. Detector array assemblies have been demonstrated to withstand extensive qualification and environmental tests and the results are summarized.

The low frequency noise characteristics of HgCdTe photoconductors are very important to the weather satellite user community because of the extremely long integration time used in these sensor applications. There are several detector technologies critical to the reduction of 1/f noise including surface passivation, bulk material selection, defect-free wafer thinning, contact metallurgy, off-HgCdTe bonding,and planar/low pressure substrate mounting. Each of these technical building blocks is discussed. When integrated to form a combined process, these critical technologies lead to a significant improvement in 1/f noise. Tow widely accepted empirical models of 1/f noise are reviewed. Experimental results validate Kruse's model but repudiate Broudy's model, namely, 1/f noise is inversely proportional to the square root of the detector volume but does not depend on the gr noise of the detector. Furthermore, we find no correlations between 1/f noise and total detector surface area or the detector contact effects. A novel test structure is presented which suggests that by using innovative detector geometries, the detector designers may be able to increase D* at low frequencies compared to the conventional square or rectangular detector structures.

NOAA has commissioned a solar x-ray imager to be built for use on the GOES spacecraft. The mission of the SXI is to provide soft x-ray imagery of the Sun. The current instrument design employs a microchannel plate detector stack to convert the incident x-rays to an electrically detectable signal. In this paper, we discuss the SXI performance improvements possible by replacing the detector with a back-illuminated, x-ray sensitive CCD fabricated using technology developed at MIT/LL. In addition to a description of the x-ray sensitive CCD, we discuss possible improvements in data quality, reduction in instrument mass and power requirements, and simplified instrument handling.

The modulation transfer function is one of the GOES Imager and Sounder system's key optical performance indicators. Though the visible images from GOES-8 and GOES-9 are excellent, the MTF values measured during ground tests are below expectations. The 4490 cycle per radian frequency MTF is particularly interesting because it is not significantly affected by small changes in focus or aberrations of the optical system. The primary contributors to the ow MTF values which are within the instrument are the ghost reflections from the transmissive components and the detector and its package. The ghost reflections that exist are certainly a factor in the stray light. An estimate of each ghost's intensity produces a worst case effect on MTF of approximately 2.5 percent. The detector itself appears to be the largest contributor to the stray light. There are several ways that the detector and its package contribute to the stray light. The most prominent among them is the reflection from the aluminum coated substrate combined with the closeness of the detector window. The stray light form the detector reduces the MTF by approximately 4 percent. Instantaneous geometric field of view (IGFOV) curves verify the existence of the stray light and calculations show approximately the magnitude of difference between measured and expected MTF data. A new detector design appropriate for the Imager visible channel or the Sounder star sense channel incorporates the changes recommended to reduce stray light within the detector THe first of the new detectors qualifies for the Sounder star sense detector. IGFOV curves show the reduction in stray light levels by about a factor of 2 to 4. Sounder star sense optical MTF data shows approximately 8 percent increase in the MTF at 4490 cycles per radian compared to SN05 Sounder.

The calibration processing algorithms currently in effect for the GOES-8 and -9 imagers and sounders are described. The algorithms have been changed since the launch of GOES-8. Smoothing of the calibration slopes has been instituted to suppress the effects of noise in the blackbody sequences. In the visible channel, the signal, originally transmitted to users in absolution counts, is now provided relative to the signal from space. The operational calibration equations were generalized to eliminate errors caused by the variation of the reflectances of the instruments' scan mirrors with east-west scan position, which was discovered after the launch of GOES-8. We are in the process of developing processing changes to mitigate the effects of two other anomalies--the appearance of east-west stripes in the infrared images from the imagers; and erroneous calibrations around midnight during six months of the year, caused by the presence of unwanted radiation from hot components of the imagers' and sounders' telescopes during the calibration sequences. Instructions are given on how to convert the counts in the GVAR data stream to radiances, temperatures, and mode-A counts.

East-west stripes are observed in images in channels 4 (10.7 micrometers ) and 5 (12 micrometers ) of the GOES-8 imager. These channels use two detectors arranged in a north-south array to sweep out alternate lines of an image. The stripes are caused by differences in the outputs of the two detectors. There is a clear correlation between scene temperature and the magnitude of the striping. Measured in temperature units, the striping is more severe at the cold end of the spectrum, and therefore affects meteorological products which depend on observations of cold targets. The striping can be divided into three components: 1) within-frame random, 2) systematic within-frame but random from frame-to-frame, and 3) systematic over many frames. The first two components are caused by 1/f noise during data-taking and calibrations. The cause of the third component of the striping is not known. Examples of each striping element are presented and discussed. A brief comparison of GOES-8 and GOES-9 imager striping is made.

Since the GOES-8 launch in 1994, unexpected values of imager calibration slopes have ben observed near satellite midnight, especially in the 'shortwave' channels 2 and 3. Concurrent imager measurements of apparently decreased ocean temperatures suggest that the slope changes do not reflect actual changes in the instrument, but rather errors in the onboard calibration process. One type of error is believed to be caused by heating of the instrument surfaces around midnight when the sun is above the equator. Reflected by the onboard blackbody, excessive radiation from these hot surfaces reaches the detectors when the blackbody look is performed during the calibration process. This decreases the magnitude of the slope, causing the earth scene to appear cooler. Another type of error is introduced by scattered solar radiation, which contaminates the instrument space- look signal around midnight during the two annual eclipse seasons. When the calibration uses the contaminated space signal, it increases the magnitude of the slope, causing the earth scene to appear warmer. Both midnight effects alter the slope by as much as 4 percent in the shortwave channels, thus leading to temperature errors of up to 1 K for a 300 K target. This paper describes a study of these effects and an algorithm to correct them. The algorithm estimates the correct slope at midnight form correlations between the slope and the temperatures of optic components within the instrument. In particular, it is found that the slope can be predicted from the telescope primary-mirror temperature with an rms error of less than 1 percent. Our study also indicates that the effective temperature of the hot surfaces may reach 360 K in the middle of summer, and the shortwave channel space look may be contaminated with radiation from an apparent 'target' with temperature as high as 250 K around the eclipse seasons.

The present GOES imager exhibits East-West stripes in IR images due to low frequency errors in the calibration of the adjacent North-South detectors. Striping makes delineating boundaries of structures in images difficult, especially in the case of cold scenes. A computer program has been developed that generates simulated IR images using detector noise parameters as inputs. The simulation includes errors due to background drift between space clamps, drift during a space clamp, and errors determining the first order gain during the internal blackbody calibration. The results of the simulation agree well with on orbit measurements of GOES 8 and 9 striping in channels 4 and 5. These simulations can also predict striping performance of future GOES imagers from detector noise parameters allowing for improved detector selection constraints.

Techniques which use empirical distribution functions have been used successfully to intercalibrate detectors on the same instrument or channels on different instruments. In this paper the procedure is developed as a statistical technique for estimating a function. Since it is a statistical procedure, it is possible to use classical statistical techniques to evaluate the accuracy of the results. Confidence regions can be constructed for the estimated parameters. Given a required degree of accuracy it is possible to estimate the required sample sizes. Using hypotheses testing techniques questions about possible changes in the inter-calibration functions can be answered. The theory, which uses quantiles from the distributions of random variable, is outlined. The large-sample properties of the estimators of these quantiles are presented. The theory requires that the data be continuous. However, it is discrete. Preliminary simulation results indicate that the theory for continuous data can be applied to the discrete data with little loss of accuracy.

Future GOES missions, augmented by the addition of very stable visible calibration, could measure long-term changes in the earth's albedo. Direct sunlight can be viewed through a plate perforated by small holes, attenuating the radiance at the center of the solar image by both blockage and diffraction. Attenuation by a factor of 50,000 produces a full-aperture, end-to-end calibration source with radiance comparable to a high-albedo target. Since its radiance depends only upon geometry, the perforated plate technique avoids the materials degradation problems inherent to calibration techniques based on reflection or refraction. Our analysis indicates that a plate with 50 micrometers holes spaced 4.9 mm apart in a hexagonal array has the potential to satisfy the requirement for a source stable to < 1 percent over a 5-year GOES mission. Some randomization of the hole positions may be desirable to suppress interference effects. We have made preliminary measurements of the large- scale angular structure of a simulated solar image viewed through a perforated screen, and have demonstrated their agreement with out theoretical predictions.

As reported by Weinreb et. al., both the imager and sounder exhibit an east-west radiance gradient due to reflection off the coating of the scan mirrors. While this problem was basically eliminated with ground processing in the imagers as of the spring and summer of 1995, the sounder instruments were corrected on March 26, 1996 and June 19, 1996 for the GOES-9 and GOES-8, respectively. This paper will investigate the process of minimizing the sounder east-west radiance variation by a radiance bias adjustment prior to the use of radiances in product generation. The sounder ground- corrected data show that the radiance bias adjustment does not change as much as a function of longitude, and that the biases between those observed from the sounder and those calculated from the 'truth' profiles are reduced. The artificial east-west gradient across the earth is difficult to observe due to several affects, including clouds, land/sea differences, topography and natural variations. Reducing/eliminating the artificial east-west gradient helps the retrieving of atmospheric parameters, like moisture. The second part of this paper presents the verification of satellite derived moisture profiles. Even before the ground correction of the east-west gradient, atmospheric moisture retrievals have been shown to improve upon the 'first guess' profile.

The SXI detector is designed to cover the 6 angstrom-60 angstrom soft x-ray band and operate in a single photon or integrating camera mode. The detector is based upon a MCP intensifier/converter coupled to a CCD via phosphor coated fiber optic. This paper discusses the detector characterization and calibration to date.

This paper presents the mathematical formulation and implementation techniques used in the GOES orbit and attitude determination (OAD) software. Also presented are some recent results obtained by analyzing data from the operational GOES. The OAD software, a key component of the orbit and attitude tracking subsystems, has been in use for operating the GOES spacecraft successfully since the beginning. Three kinds of observables are ingested by OAD: range, star line-of-sight direction, and landmark line-of- sight direction. Raw observations are preprocessed to rid of known systematic errors. Among the solve-for parameters are six orbital elements, dynamic model constants, range bias, and a number of instrument attitude coefficient. The attitude of either the imager or the sounder instrument is characterized by five angles: roll, pitch, yaw plus two rotation angles specifying the misalignment of the optical axis of the instrument. Each attitude angle is modeled as the sum of Fourier terms and optional polynomial and exponential terms. The original version of the software was implemented as a least squares batch filter, ingesting observations in batches of one to two days at a time. A recent enhancement to the software to include a priori knowledge of the orbit and attitude solution and their statistical properties has been implemented. By reprocessing operational data using this new software, it is shown that the sequential batch solution obtained exhibits better daily stability as well as providing a better model for next day predictions compared to the batch solution.

This paper presents jitter performance of the GOES spacecraft and instruments. Of primary interest is the validation of design and ground test procedures with on- orbit data. System requirements, spacecraft and instrument disturbances, ground testing techniques, and on-orbit testing are discussed. Dynamic spacecraft signatures during appendage deployment are illustrated and used to determine spacecraft modal response frequency and damping. Specific analysis results form angular displacement sensor data, used to measure jitter dynamics from 0.2 to 160 Hertz, and digital integrating rate assemblies, used to measure low frequency, are detailed with regard to solar array, momentum wheel, and other spacecraft disturbances including those generated by instruments during scanning and slewing. Emphasis is also given to feed-forward pointing compensation and optimization of the solar array stepping. This optimization reduces pointing errors, improves image registration, and reduces pixel smear. Performance is illustrated with sensor statistics, fast Fourier transforms, and Earth images. Both long term as well as short term evaluations are presented.

Following the successful transition of GOES-8 to the on- orbit mode, small disturbances were observed at approximately 14_30 SLT. Further analysis has shown that the disturbance is caused by shadowing of the solar sail by the magnetometer boom. The disturbance is most prominent when the sun is at lower declinations causing the shadow to rise higher along the sail boom. Data for the solar sail snap caused by the magnetometer boom shadow is analyzed, showing the time dependence of the attitude motion in the wheel angular momentum data and earth sensor data. Additional data is presented showing the profile of the shadow as viewed through the imager and sounder cooler radiator patch heater control voltage data. The presentation will demonstrate the following: 1) The attitude variation is caused by the magnetometer boom shadow as it is correlated with the true position of the sun; 2) The magnitude of the spacecraft motion for GOES-8 and GOES-9 is measured and compared over the seasonal duration of the effect; 3) The amount of deflection of the sail boom required to cause the observed attitude motion is estimated; 4) The dependence of the magnitude of the attitude motion is compared with sun angle subtended by the spacecraft north panel at the shadow point on the sail boom; and 5) Use of the onboard computer's reprogrammability function has provided a mechanism of compensating for the disturbance.

The Earth Sensors on GOES-8 exhibit seasonal pointing errors in both the dual and single chord operating modes of the sensors. The errors were largely compensated for by a software patch uploaded to the satellite. The combination of detailed analyses and laboratory test results established that the observed error signatures are the result of stray solar radiation. The nature of the stray radiation paths giving rise to the seasonal errors is such that only the most significant stray path was eliminated in the GOES-9 hardware. The Earth Sensors on GOES-9 show significantly improved pointing performance over GOES-8, validating the origin of the most significant error source in the GOES-8 Earth Sensors. Low level seasonal pointing errors are observed, as expected, in the GOES-9 Earth Sensors. These errors again are effectively compensated for by an additional software patch developed to permit satisfactory single chord Earth Sensor operation. The operating principles of the Earth Sensor are described. On-orbit data of the seasonal anomalies are presented for both the uncompensated and compensated hardware and operating modes.

The dynamics of GOES-8 is simulated using an articulated, flexible, multi-body dynamics model. The solar array on this spacecraft is flexible, and is articulated with a stepper motor. Three stepping configurations are available: 1 0.007 degree single step every 1.69 seconds, and two double stepping options that were provided in order to deadbeat the solar array oscillations. In this paper, models that are commonly employed for simulating the dynamics of spacecraft with flexible appendages are reviewed and the advantages of the present model over other methods are emphasized. Simulation results show good agreement with the flight data when orbital environmental effects are accounted for. This study has significant implications for both the design and dynamic modeling of articulating, flexible solar arrays attached to the spacecraft, especially when payloads such as the Solar X-ray Imager with tight pointing requirements are to ge located on the solar array. The implications are i) appropriate dynamic models must be employed, ii) the solar array actuating device must be such that vibrational modes of the array are not excited, and iii) conservative damping estimates must be used when empirical values are not available.

An independent audit of the in-orbit behavior of the GOES-8 and GOES-9 satellites has been conducted for the NASA/GSFC. This audit utilized star and landmark observations from the GOES imager to determine long-term histories for spacecraft attitude, orbital position, and instrument internal misalignments. The paper presents results from this audit. Long-term drifts are found in the attitude histories, whereas the misalignment histories are shown to be diurnally stable. The GOES image navigation and registration system is designed to compensate for instrument internal misalignments, and both the diurnally repeatable and drift components of the attitude. Correlations between GOES-8 and GOES-9 long-term roll and pitch drifts implicate the Earth sensor as the origin of these observed drifts. This results clearly demonstrates the enhanced registration stability to be obtained with stellar inertial attitude determination replacing or supplementing Earth sensor control on future GOES missions.

Fourteen days of sounder observations were scheduled during the GOES-9 post launch test to perform an inflight measurement of the effects of IR radiance variations in the stratosphere on altitude control using the GOES Type II IR Earth Sensor. Two 16.33 degree by 1.65 degree scenes corresponding to the Earth sensor north and south chords were scheduled every four hours for the 14 days from July 19, 1995 until August 2, 1995. The data from IR channels 1 through 4 are integrated over a 1.65 degree square FOV and weighted to synthesize the GOES ES medium passband RI filter and a narrow IR passband filter. The integrated scenes were converted to time dependent Earth sensor radiance input pulses for each passband, north and south chord, for six times per day over the 14-day interval. The qualitative features of these synthesized Earth sensor IR input pulses are compared graphically and correlated with image navigation registration (INR) star residuals for the operational GOES-8 spacecraft. Comparisons of the altitude of the IR horizon on the raw sounder scans in channel 1 with channel 7 are proposed to detect variations the 14.7 micron horizon altitude. Schemes for removing the effects of attitude motion from the sounder scenes are discussed and the Earth sensor pulses are processed through a linear model of the Earth sensor electronics to evaluate the relative error response to the synthesized IR input pulse shapes. These results appear to be influenced by thermal mechanical distortion of the spacecraft. Correlation analysis shows that the GOES Earth sensors are responding to nonrepeatable and therefore unpredictable variations in stratospheric radiance and that the time evolution of these variations is consistent with the INR star residual data from GOES-8. Geometric and time adjustments are made for the stratospheric rotation rate and the 15 degree eastward displacement of GOES-8 relative to GOES-89. The performance of the narrow and wide passband filters shows no difference for the pitch and roll errors computed at this level of accuracy. Finally, based on this work methods of implementing sounder data from the Earth IR limbs scenes into the daily INR process to improve image registration accuracy may be developed.

The magnetic field data from the GOES satellite must be corrected on the ground for the contribution from the roll and yaw torquer coils which comprise important elements of the spacecraft attitude and orbit control system (AOCS). The correction may be calculated from measurements of the torquer current which are telemetered at the same rate as the magnetic field data or from torquer input command data. The baseline plan was to allow torque current changes only at infrequent intervals of once per 20 minutes since it was suspected that torquer switching noise might be a problem. However, for improved navigation and registration performance in the radiometric instruments, the actual mode of operation has been to calculate the torquer current drive more frequently and to rate limit the current changes to one commanded increment per step. The magnetic signature of the 3.2 ma step is an much as 1.2 nanoTeslas. Both the current level and the telemetered value change essentially instantaneous when compared to the frequency response of the magnetometer, which incorporates a 0.5 Hz bandwidth five- pole Butterworth anti-aliasing filter in each data channel. Straightforward correction of the data using the current telemetry therefore results in objectionable noise spikes in the derived bandwidth-limited output data. The problem is complicated by the fact that the clock which governs the AOCS current commands to the torquers is not synchronized with the telemetry unit clock which governs magnetometer data sampling. Fortunately, it has been possible to incorporate data in the spacecraft telemetry which provides improved information for determining the time of a current step in the telemetry unit time base as well as the torquer input command data mentioned as one method of compensation. This paper describes the development of a numerical filter used in ground processing of the telemetered current to simulate the five-pole Butterworth response of the magnetometer output filter. Since the timing information is determined to an accuracy of 1 part in 8, one of eight sets of filter weights is selected based on the telemetered timing information. Examples are given of the improvement in output data quality resulting from the technique.

The GOES I/M series of geostationary weather satellites are 3-axis stabilized and incorporate an optical trim surface, or 'trim tab', to balance solar radiation pressure torque between the south moment arm, formed by a single solar array, and the north moment arm, formed by a solar sail on an extended boom. This design minimizes the need for active roll/yaw momentum management from thrusters or magnetic dipoles and provides margin for uncertainties and changes such as center of mass shift and degradation of surface optical properties. The trim tab has one rotational degree of freedom and is actuated vie open loop ground command to a motor on the rotation axis. Normal operational procedure is to command a small incremental trim tab slew daily to track the change in north/south solar radiation pressure torque with Sun declination. Proper management of the trim tab is critical to spacecraft roll/yaw pointing performance and the open loop design of the system requires sophisticated ground based algorithms to determine trim tab position error and developed a daily commanding profile. Flight experience with the GOES-8 and GOES-9 spacecraft has proven the feasibility of the trim tab design and has led to refinement of the ground based trim tab analysis algorithm. THis paper summarizes the performance of the trim tab roll/yaw momentum control system and describes analysis methods developed by the GOES mission operations team to support trim tab operations.

The GOES-8 and -9 Sun analog sensor (SAS) flight data is analyzed to evaluate the attitude motion environment of payloads mounted on the solar array. The work was performed in part to extend analysis in progress to support the solar x-ray imager to be flown on the GOES-M. The SAS is a two axis sensor mounted on the x-ray sensor pointing (XRP) module to measure the east/west error angle between the SUn and the solar array normal and to provide a north south error angle for automatic solar pointing of the x-ray sensor by the XRP. The goal was to search for evidence of solar array vibrational modes in the 2 Hz and 0.5 Hz range and to test the predicted amplitudes. The results show that the solar array rotates at the rate of the mean Sun with unexpected oscillation periods of 5.6 minutes, 90 minutes, and 1440 minutes originating from the two 16.1 gear drive train stages between the solar array drive stepper motor and the solar array yoke. The higher frequency oscillations are detected as random noise at the 1/16 Hz telemetry sampling rate of the SAS. This supports the preflight predictions for the high frequency modes but provide s no detailed measurement of the frequency as expected for this data period. In addition to this the data indicates that the solar array is responding unexpectedly to GOES imager instrument blackbody calibration events.

The GOES I-M image navigation and registration (INR) system was developed and patented by Space Systems/Loral in response to NASA/NOAA pointing requirements of Imager and Sounder payloads. This INR system is divided into space segment and ground segment. The space segment consist of image motion compensation (IMC) system that compensates for deterministic errors caused by orbit and attitude motions and mirror motion compensations system that compensates for spacecraft attitude motion caused by Imager and Sounder scanner mirror motion. The ground segment consists mainly of Orbit and Attitude Tracking System (OATS) that determines the IMC orbit and attitude coefficients from star, landmark, and range measurements.

GOES 8/9 presents a new challenge to the National Environmental Satellite, Data, and Information Service's (NESDIS) Office of Satellite Operations. In the previous series of spin stabilized GOES satellites, the image navigation was performed off-line with little operator intervention or oversight. This system provided accuracy to 10 KM. However, the more stringent 4 KM navigation requirements of the new generation of three axis stabilized GOES satellites has resulted in a far more demanding image navigation and registration (INR) operation. The new system requires continuous observation gathering and monitoring of performance along with commanding real time corrections to the on-board image motion compensation system to maintain accuracy. All this must be accomplished while providing images as frequently as every 7.5 minutes in rapid scan mode and operating in the independent sounder. Once the GOES 8/9 navigation system was operational, it became apparent that there were several sources of pointing error which were not anticipated prior to launch. Most of these problems are related to special operational modes of the Earth sensor. It became necessary to devise new operational work-arounds for these problems to minimize the impact on the user community and continue to provide within specification INR performance. This paper will describe the GOES 8/9 INR operation and how it is integrate into the instrument and spacecraft command schedules. It will also describe the various operational solutions which NESDIS devised to overcome INR errors.

This paper describes methods, algorithms, and software developed to select and represent landmarks from cartographic databases and to automatically determine the landmark position in GOES imagery. A simulation study was conducted to demonstrate that the landmark position can be measured with subpixel accuracy.

This paper presents the results of an analysis of on-orbit image navigation and registration (INR) performance using GOES-8 and GOES-9 imager operational data. Two time intervals were analyzed: a 35-day span in March/April 1996 during eclipse season and a 35-day span in May/June 1996. These data were used to derive the performance of each specified INR parameter and then compared to its performance requirement. Significant sources of INR error discovered on- orbit and the methods adopted to reduce their effect are discussed. The study results confirm that the operational use of short span attitude adjusts is critical to achieving acceptable INR performance. Also shown is that landmark measurement errors currently limit the ability to adequately assess INR performance.

This paper presents a summary of the basic simulation parameters and results of a study for the GOES. The study involved the simulation of minor modifications to the current spacecraft, so that the relative performance of these modifications can be analyzed. The first modification studied requires the placement of a baseline inertial reference unit, such as the Dry Rotor Inertial Reference Unit or the Space Inertial Reference Unit, onto the spacecraft. The second modification involves using the imager/sounder assembly for real-time on-board attitude determination information. The third modification studied is the addition of star trackers to provide precise attitude knowledge. Simulation results are presented for each modification.

In this paper, the improvement in the image navigation and registration (INR) obtained by using two star trackers onboard a GOES satellite is investigated. The star trackers would augment the current INR system to eliminate the nonrepeatable errors caused mainly by the Earth sensor. The focus is on decreasing the spacecraft bus attitude error component in the total INR error budget with minimal changes to the current system.

An autonomous star tracker (AST) is a "starfield in, 3-axis attitude out" device capable of determining its attitude rapidly without requiring any a-priori attitude knowledge. Following attitude acquisition, the AST switches to a track mode where it outputs its attitude and rate at typically 5 Hz. The Lockheed Martin Advanced Technology Center has developed the first version of a line of reliable, highly accurate, low cost, ASTs. This "AST-201" is to be test flown on a sounding rocket and will function as a key attitude sensor aboard NASA's Small Spacecraft Technology Initiative (SSTI) "Clark" spacecraft scheduled for launch in 1996. The AST is of modular construction and comprises athermalized, radiation hard refractive optics, a frame-transfer CCD with a sensitive area of 512 by 512 pixels, camera electronics, a single board computer, an all-sky guide star database and a highly effective sunshade that allows operation to within 25 degrees from the Sun. Initial star identification is performed by a memory-efficient algorithm that provides an attitude acquisition success rate of better than 99.85% anywhere in the sky. The AST achieves its high accuracy through use of a moderate 8.8 by 8.8 degree field of view and by tracking 27 stars on average. On geostationary satellites a single AST can provide 3-axis attitude with greater precision, at lower cost, and with higher reliability than is possible with a combination of Earth sensors, fine Sun sensors, and a high performance gyro system. In addition, ASTs are devoid of angular range limitations, avoid Sun and Moon interference by proper orientation on the spacecraft, and enable rapid fault recovery, a capability especially important to geostationary missions where outages are usually costly. Two ASTs can provide a geostationary spacecraft with an attitude accuracy of 5 microradians (1 a). The paper contains a description of the AST, a summary of the functions enabled or improved by the device, real-sky AST test results, and accuracy statistics for an AST on a geostationary spacecraft, as obtained through realistic simulations. Keywords: star tracker, autonomy, attitude determination, pattern recognition, geostationary satellites, spacecraft attitude control.

The performance and functionality of star trackers has greatly improved over the past decade due to technological advances in charge-coupled devices, microprocessors, and application-specific integrated circuits. These technological advances have also lowered the cost of star trackers to the range of competing attitude sensors such as horizon sensors. This study evaluates the attitude reference requirements of a three-axis geosynchronous satellite (GOES- NEXT) and addresses the use of star trackers to replace a traditional attitude determination sensor suite based on horizon sensors and digital fine sun sensors. Although the GOES-NEXT satellite is used as a specific example in this study, the results are applicable to other geosynchronous remote sensing or communications satellites.

Advances in receiver technology permit the use of microwave spectrometers at 100-500 GHz in combination with economically affordable antennas on geostationary meteorological satellites. Microwave systems operating at 118 GHz, 183 GHz, and 425 GHz have been compared. With an antenna of just over 2 meters in diameter, a system operating at 380 GHz and 425 GHz is capable of sounding temperature and water vapor profiles with 20-km horizontal resolution, and a system operating at 118 GHz and 183 GHz is capable of 70-km resolution. Analyses show that integration of such microwave spectrometers with a high performance infrared sounder would provide a nearly all-weather continuous-observation geostationary sounding system with accuracies approaching those of low-earth-orbit observatories.

The GHIS optics assembly was designed as a replacement for the filter wheel assembly in an upgraded GOES Sounder sensor of the year 2000 +. An overview of the entire effort is presented elsewhere in these proceedings. The assembly includes miniature optics consisting of a Michelson interferometer with compatible input and output optics. It is to be aligned at room temperature and operated at -55 degrees C without realignment; therefore, an athermal design approach was chosen. Several aspects of the opto-mechanical design will be discussed herein, namely: the support, attachment, and adjustment of the miniature optics; investigation and fabrication of various types of aluminum mirrors; a remotely adjusted mirror for on-orbit alignment using PZT actuators; a moving-mirror supported on a leaf- spring, flexure mechanism moved by a voice-coil actuator; a laser, reference-beam, optical assembly, with fiber optic input and output, to control the velocity and position of the moving-mirror; the support structure; and the thermal control design.