The Advanced Geosynchronous Studies Imager (AGSI) system design combines the latest available technologies into an instrument design concept which could deliver the improved performance sought by the National Weather Service at NOAA and meet NASA earth system science goals in a joint program. The instrument could cover the Earth disk every 15 minutes with subsatellite point resolution form 1/2 kilometer in the visible to 2 kilometers in the long wave IR. Simultaneously, it could provide coverage of a 3000 by 5000 kilometer region in 5 minute intervals and 30 second updates of a 1000 kilometer square region containing a weather system of interest. We found that performance margins could be improved even as we drove the design interactions with emphasis on reducing the mass. Scan speed was chosen by maximizing performance while trading off the acceptable impact on the total systems. The resulting 18-channel design could deliver vastly improved performance over the present GOES without great increases in mass or volume, while still paying close attention to control of development cost sand impact on the host spacecraft. The design could be adapted to changed requirements or descoped to have lower data rates and fewer channels.

An unobscured, f/4, Earth scanning telescope was designed by optimizing an off-center, off-axis Cassegrain telescope with a unit magnification spherical camera mirror. A flat scan mirror in front of the 30-cm diameter telescope directs the 1-degree field-of-view. Observation form geosynchronous orbit results in sunlight illumination of the scanning, primary, and secondary mirrors near satellite local midnight. The 'all-reflective' design has a primary- secondary spacing of 87-cm. A field stop helps control stay sunlight illumination of the camera mirror and beam division optics. Exit pupil stops further control stray illumination. Sub-satellite pixel image scale is 1/2-, 1-, and 2-km over the spectral range 0.4- to 13.5-microns. The general requirements are for the first diffraction dark ring to be imaged on the pixels. The flat image plane is separated into four bands using dichroic beamsplitters and flat mirrors to accommodate linear array detector technology. Sub-band channels are defined with filters in near contact with the detectors. All but the visible detectors are enclosed behind transmission windows so that all bands can be operated without thermal vacuum facilities during ground testing. All transmission elements are 'wedged' to correct for astigmatism introduced by using them in converging light. The powered mirrors and scan flat are to be silicon carbide. The mechanical structure will be a near-zero expansion graphite fiber reinforced plastic. Tolerance analysis and thermal modeling show optical element displacements well below requirements. Instrument volume corresponds approximately to that of the current GOES Imager.

Accurate and automatic image navigation and registration (INR) of remotely sensed data will be an essential element of future NASA satellite observation systems. INR describes the process by which geographic locations of the image pixels are computed and successive images from the same sensor are aligned to each other over time. For sensors such as the Advanced Geosynchronous Studies Imager (AGSI), a number of distortions prevent successive images from being perfectly registered to each other or to a fixed coordinate system. Most distortions in such images are the combined effects of sensor operation, satellite orbit and attitude, and atmospheric and terrain effects. These distortions are usually corrected by two methods; systematic correction, which relies on image acquisition models taking into account satellite orbit and attitude, sensor characteristics, platform/sensor relationship, and terrain models, and precision correction, which is feature-based, starting from the result of the systematic correction, and refining the geolocation or relative registration to subpixel precision. This paper describes the AGSI INR requirements and concepts, the image navigation model, a description of some potential precision correction methods utilizing edge and wavelet features, and a study of all the different error sources. The issues of swath-to-swath correlation and channel-to- channel coregistration are also described.

The AGSI is a visible and IR instrument being proposed to satisfy both NOAA's operational weather and NASA's geostationary science requirements. It scans the full earth disk with a plane scan mirror in object space, mounted on a two-axis gimbal system. Image rotation is an intrinsic problem: scanning about one gimbal axis rotates the projection of the focal plane array (FPA) onto the earth's surface. The AGSI's needs for both higher angular resolution and higher radiometric resolution are satisfied by time delay and integration (TDI) in several FPA's. The electronic and opto-mechanical scan vectors must match to maintain image quality: the projection of the TDI axis of each FPA onto the earth's surface must always coincide with the scan direction and the scan rates must be equal. A new gimbal geometry, focal plane layout, and associated scanning techniques have been developed to scan the earth's surface in a series of conical arcs that satisfy these conditions. This technique has the additional advantages that the outer gimbal axis remains stationary during the data-taking portion of the scan pattern and that the magnitude of the angle of reflection remains relatively constant during a single scanning arc.

The AGSI design permits scan rates slow enough to detect stars as dim as visual magnitude eight in the coarse of normal imaging. This gives many times the number of stars seen with the current Geosynchronous Operational Environmental Satellite (GOES) Imager and can eliminate the need to schedule special star looks. Besides improving image navigation and registration accuracy, the frequency observations enable the Imager to fly aboard a spacecraft with loose attitude control. The slow scan rate is thanks to the long CCD detector arrays and to the time delay integration made possible by the unique windshield wiper scan pattern. The Bremer star detection algorithm describe can be implemented onboard to reduce downlink requirements and so permit star detection across a dedicated full silicon passband. The wide passband increases the number of detectable stars, and cross checking with narrower science passbands eliminates false alarms from high energy particles while preserving low detection thresholds and sensitivity.

A focal plane system to generate images in 18 spectral channels ranging from visible to long-wave IR has been designed for the AGSI. The system is a line scan imager, comprised for four focal planes populated with linear detector arrays optimized for operating in the various AGSI spectral regions. Light from the AGSI telescope is diverted to the focal planes by spectrally tuned dichroic beam splitters. At the focal planes narrowband interference filters placed in close proximity to the detector arrays further filter the light. Multispectral silicon CCDs are used for visible and near IR channels. Key to system performance is the ability to use time delay and integration (TDI) in some of the narrower or less photon rich spectral channels. Detector arrays are supported by highly modular and therefore flexible and low risk signal processing and control circuits. Performance predictions have been generated for all of the spectral channels and show that the focal planes will meet or exceed NASA's requirements for an advanced imaging observatory to observe weather and climate processes and NOAA's requirements for an advanced GOES imager.

The AGSI instrument contains four focal plane arrays (FPAs) including the long wave IR (LWIR), medium wave IR (MWIR), short wave IR (SWIR), and the visible (VIS) FPA. The LWIR and MWIR FPAs are housed in a dewar. The SWIR FPA is housed in a separate dewar. The visible FPA will be located outside of the dewar. The AGSI instrument thermal subsystem is required to maintain the LWIR and MWIR FPAs at 65K, the SWIR FPA at 210 K, and the visible FPA between 210 K and 300 K. The stability of all four FPAs must be +/- 0.1 degrees C. The remainder of the optical system, telescope structure and instrument electronics must be maintained between 0 degrees C and 40 degrees C. Telescope alignment must also be maintained through the cooldown of the instrument optical bench, and during the diurnal and seasonal variations in orbital environment. The thermal design will accommodate such directly in the field-of-view for short periods of time without damage to the instrument, and with the ability to recover science operation quickly. This paper will discuss the thermal design needed to achieve these goals as well as detail the cryocooler and solar intrusion issues.

The ALI, which will be flown on the NASA New Millennium Program's EO-1 mission, has been completed and is being integrated with the spacecraft. The motivation for the EO-1 mission is to flight-validate advanced technologies that are relevant to next generation satellites. The ALI telescope is a reflective triplet design having a 15-degree cross-track field-of-view that employs silicon carbide mirrors. It incorporates a multispectral detector and filter array with 10 spectral bands that cover a wavelength range from the visible to the short-wave IR. The paper will describe the instrument and its operation, review test result, and suggest application to a future Landsat instrument.

The pre-launch measurements and test required for calibration and characterization of the advanced land imager (ALI), which will be flown on NASA's EO-1 mission, have been completed. The instrument level performance testing was conducted at MIT Lincoln Laboratory with the ALI in an operational environment. The overall calibration strategy, which includes both pre-launch and post-launch components, will be described in this paper. The fundamental sensor calibration data comprise five measurement categories: angular position in object space for each pixel; normalized spectral response functions; response coefficients; zero signal offsets; and modulation transfer functions. Performance and characterization test include measurements of noise, SNR, linearity, repeatability, image artifacts, stray light rejection, and cross-talk. An overview of the facilities, equipment, tests and results is presented here.

Spatial calibrations have been performed on the Advanced Land Imager (ALI) of the EO-1 satellite. Topics discussed in this paper include end-to-end imaging test, measurements of system modulation transfer function (MTF), and pixel lines of sight. The MTF measurements were made by recording scans of a knife-edge past the pixels. The techniques used to place the focal plane at the correct focal position are described, since they make use of MTF measurements. Line-of- sight measurements combine theodolite measurements of the telescope distortions and the photolithographic patterns of the detector arrays with images of a stationary Ronchi ruling recorded with the instrument at its normal operating conditions in a thermal vacuum chamber.

The spectral response of the Earth Observing 1 Advanced Land Imager (ALI) has been characterized at Lincoln Laboratory as a fully assembled instrument in a thermal vacuum chamber at operational temperatures. The focal pane for this instrument is partially populated over a 3 degree cross-track segment with 9 multispectral bands having a 30-meter ground sample distance (GSD) and a single panchromatic band having a 10- meter GSD. These bands were selected to mimic the six Landsat-7 VNIR/SWIR bands with three additional bands covering 0.433-0.453, 0.845-0.890, 1.20-1.30 micrometers . The instrument system level response was characterized spectrally form 400-900 nm in 2 nm increments and from 900- 2500 nm in 4 nm increments using a collimated monochromatic beam. Spectral artifacts introduced by the monochromatic and the collimator were accounted for using spectrally calibrated silicon and lead-sulfide detector which sampled the beam at each measurement interval. In this paper we describe the techniques employed during spectral calibration, present the measured in band and out of band spectral response for all VNIR bands, and compare the results to those obtained at the component level.

The radiometric calibration of the Earth Observation 1 Advanced Land Imager (EO-1 ALI) was completed in the Spring of 1999 at Lincoln Laboratory. This calibration was conducted with the ALI as a fully assembled instrument in a thermal vacuum chamber at operation temperatures. The ALI was calibrated radiometrically at the system level from 0 to > 100 percent Earth-equivalent albedo using a combination of internal and external halogen and Xenon lamps attached to a large integrating sphere. Absolute radiometric calibration was achieved by measuring the output of the integrating sphere at each radiance level prior to ALI illumination using a NIST-traceable spectroradiometer. Additional radiometric characterization of this instrument was obtained from data collected using a collimator designed for the spectral calibration of the ALI. In this paper we review the techniques employed during radiometric calibration and present the measured gain, linearity, offset, signal-to- noise ratio and polarization sensitivity of each pixel. The testing result of a novel, in-flight solar calibration technique are also discussed. Finally, the results from a Lincoln Laboratory/Goddard Space Flight Center Landsat transfer radiometric study are presented.

The calibration and performance assessment of the Earth Observing-1 (EO-1) Advanced Land Imager (ALI) required a ground data system for acquiring and processing ALI data. In order to meet tight schedule and budget requirements, an automated system was developed that could be run by a single operator. This paper describes the overall system and the individual Electrical Ground Support Equipment (EGSE) and computer components used. The ALI Calibration Control Node (ACCN) serves as a test executive with a single graphical user interface to the system, controlling calibration equipment and issuing data acquisition and processing requests to the other EGSE and computers. EGSE1, a custom data acquisition syste, collects ALI science data and also passes ALI commanding and housekeeping telemetry collection requests to EGSE2 and EGSE3 which are implemented on an ASIST workstation. The performance assessment machine, stores and processes collected ALI data, automatically displaying quick-look processing results. The custom communications protocol developed to interface these various machines and to automate their interactions is described, including the various modes of operation needed to support spatial, radiometric, spectral, and functional calibration and performance assessment of the ALI.

The performance assessment of the EO-1 Advanced Land Imager (ALI) requires software for processing data collected during sensor integration and test, ground calibration, and on- orbit operations. This paper describes the software developed for performance assessment processing and analysis of data collected on the ground during ALI and ground calibration. This involves various characterizations and calibrations of the ALI, including functional test, focus, MTF, radiometric response, spectral response, functional image reconstruction, and internal calibration lamp data processing. Processing examples are given, including results. Also, each section describes the use of this software to support the analysis of data collected on-orbit during mission operations, as well as the types of data to be collected and processed. The interface between MIT Lincoln Laboratory and the EO-1 Mission Operations Center at NASA Goddard Space Flight Center is also described.

This paper covers the design and initial performance analysis of the radiometric calibration pipeline software for the EO-1 Advanced Land Imager (EO-1). The design and implementation of the software and of the radiometric calibration database are discussed. Radiometric calibration data were gathered in the laboratory during integration and test and were used for initializing the database. This database will be updated in the future with data collected on orbit. Initial performance results have been obtained after applying the radiometric calibration to imaging data collected during ALI integration and test and ground calibration. During the initial test it was noted a few detectors leaked energy into other detectors, thereby requiring special calibration processing for the affected detectors. The handling of this anomaly is discussed, and the initial performance results of the calibration pipeline with the anomaly corrections will be shown.

The QuikScat satellite carrying the SeaWinds Scatterometer was developed as a replacement mission for the aborted Japanese Advanced Earth Observation System-I (ADEOS-I) mission carrying the NASA Scatterometer. Like NSCAT, SeaWinds is an active microwave remote sensor designed to measure winds over the ocean from space. SeaWinds can measure vector winds over 95 percent of the Earth's ice-free oceans every day, a significant improvement over previous scatterometers. Such data is expected to have a significant impact on weather forecasting and will support air-sea interaction studies. SeaWinds will also fly aboard ADEOS-I sensor scheduled for launch in Nov. 2000. QuikScat was successfully launched on June 19, 1999, though as of this writing the instrument has not ben turned on. This paper provides a brief overview of the SeaWinds instrument and discussed new applications of scatterometer data for the study of land and ice.

NASA has developed a new scatterometer, SeaWinds, which is scheduled to be launched on two missions, one in 1999 and the other in 2000. SeaWinds will measure the speed and direction of near-surface winds over the ocean. A fundamental part of the processing of SeaWinds data is the computation of (sigma) degree which includes solving the radar equation for every pulse sent and received by the scatterometer. However, for SeaWinds, this calculation is too computationally intensive to perform in real tie. Instead, a method of tabularizing most of the calculation has been developed. This method includes corrections for attitude and orbit perturbations and accounts for the elevation of the local topography. It also includes a table which will be used to assist ground data processing in determining the locations of the measurements. Tests comparing the tabulated result with the actual numerical calculations have shown that using this table is accurate to within +/- 0.1 dB and gives locations that are accurate to within 160 meters. We provide a description of this algorithm. The innovations developed as part of this algorithm may be of interest in processing data from other remote sensing systems.

In the next few years, there will be a substantial increase in the number of commercial space-based and airborne SAR systems and 3D SAR systems. This will result in affordable, new types of data that can be used to complement other sensor systems, e.g. LandSat, SPOT, and in some cases, solve serious data collection deficiencies. This is of particular importance to a number of agricultural applications. The promise of low cost, high resolution radar data available at a high revisit rate over the earth will lead to many new uses for the data. These include government inventory and monitoring programs and commercial precision farming programs. The availability of this data has resulted in high interest in products derived from this data. This paper describes a methodology for developing such products and present result from applying the methodology to a potential agriculture application.

The post-launch performance of the NOAA-15 Advanced Microwave Sounding Units-A was evaluated by analysis of the AMSU-A Level 1B data. We examined the main instrument characteristics, including the noise-equivalent uncertainty of temperature (NE(Delta) T), channel gains, instrument temperatures, and the cold space and blackbody radiometric counts, which are fused for on-board calibration. Short- and long-term trends of these instrument characteristics are presented. Effect of the instrument temperature on the radiometric calibration counts and channel gains are also demonstrated. It shows that the radiometric counts and channel gains correlate linearly with the instrument temperature. The NE(Delta) T values calculated from the AMSU- A 1B data compare favorably to those obtained from the pre- launch test data. All channels on the NOAA-15 AMSU-A have functioned well since its powering up after launch, except Channel 14 which has shown an anomalous behavior since early January 1999. THis study summarizes our evaluation of the AMSU-A performance during a four-month period of the NOAA-15 post-launch operation, and introduces the new AMSU-A data to data users.

A dual channel Airborne Microwave Radiometric Imaging system (AMRI) was designed and constructed for regional environment mapping. The system operates at 35GHz, which collects radiation at horizontal and vertical polarized channels. It runs at mechanical conical scanning with 45 degrees incidence angle. Two Cassegrain antennas with 1.5 degrees beamwidth scan the scene alternately and two pseudo- color images of two channels are displayed on the screen of PC in real time. Simultaneously, all parameters of flight and radiometric data are sorted in hard disk for post- processing. The sensitivity of the radiometer (Delta) T equals 0.16K. A new displaying method, unequal size element arc displaying method, is used in image displaying. Several experiments on mobile tower were carried out and the images demonstrate that the AMRI is available to work steadily and accurately.

With the launch of the SPOT5 satellite, planned in November 2001, the SPOT system will significantly improve its performances and capacities. Thanks to a specific detector, SPOT5 will provide panchromatic images at a resolution of 5 meters and up to 3 meters using sophisticated ground processing, although multispectral images will have a resolution of 10 meters and 20 meters.

The Ozone Monitoring Instrument (OMI) is a Dutch-Finnish contribution to NASA's EOS-Chemistry satellite, which is due to be launched in December 2002. The aim of OMI is to contribute to climate monitoring and atmospheric chemistry research by providing daily global measurements of the total ozone column, ozone profile, NO₂ column, other trace gases like SO₂ and BrO₂, aerosols, cloud fraction, cloud to pressure, and surface UV irradiance.

This paper reviews the work performed in Russia and the Former Soviet Union in the last twenty year in the development and application of acousto-optic spectrometer technology for Remote Sensing and other applications. A family of spectrometers designed for remote sensing of the Earth surface from a satellite or from an aircraft is described. There are presented the collection of visible range spectra obtained in test experiments and space mission. The collection includes spectra of the oceans, Azov, Barents, Black, Japan, Caspian, Aral seas, Dnieper and some other rivers, and different landscapes.

The Defence Evaluation Research Agency (DERA) have coordinated a UK feasibility study in to a low-cost, optical remote sensing satellite invovling DERA Space, Surrey Satellite Technology LTD. (SSTL), Rutherford Appleton Laboratory (RAL) and Matra Marconi Space, sponsored jointly by UK MoD and British National Space Centre. The main aim is to demonstrate the ability to provide useful imagery, to the point of use, quickly, if possible in near real time. To achieve this a design for a low cost satellite, which could obtain 2.5 m resolution imagery form 600 km altitude, is posed which will demonstrate a low cost progression to a 1 m capability. Time delay integration by satellite maneuver is used to increase dwell time and thereby SNR; satellite rotation in all 3 axis is required. An optical design for the telescope is outlined which uses 3 mirrors in an off- axis configuration. A modified, but pre-existing, microsatellite bus/structure can support the instrument. Target budget for build of a demonstration satellite is 16 million dollars. Low cost would allow a constellation of such satellites to provide high resolution imagery with high timeliness. The design of a communications system is an integral part of the sty. The satellite should provide an indication of the capability and cost of providing 1 m resolution imagery using a similar approach.

NAST-1 is a Fourier transform interferometric sounder that provides very high spectral and spatial resolution measurements of the Earth's atmosphere. The interferometer provides two dimensional, low noise data from the NASA ER-2 aircraft suitable for synthesizing data products of future satellite-borne sounding instrument candidates. It is the first such high altitude aircraft or satellite borne instrument. The instrument provides a 2.6 km nadir footprint and a cross-track field of regard of +/- 48.2 degrees. The instrument has a continuous spectral range of 3.6-16.1 micrometers , spectral resolution of 0.25 cm^-1, and radiometric noise on the order of 0.25 K. NAST-1 has proven to be an extremely reliable instrument generating over 100 hours of high-quality flight data, and was delivered to the sponsor on a very tight schedule. Using a first principles model, the noise performance of the instrument was modeled and found to be in close agreement with noise measured in- flight. Alignment jitter has been identified as the major contributor to the system NEdN. This paper describes the mode used to predict the instrument noise performance and discusses the comparison to actual flight data.

The NASA has built two airborne sensors to simulate space- borne instruments to be flown on the EOS AM-1 Terra satellite, scheduled for launch in 1999. The intent of these airborne systems is to provide initial, or 'precursor' data set to EOS investigators for algorithm development, as well as to conduct calibration and validation under-flights after the AM-1 platform has been launched. The MODIS Airborne Simulator (MAS), and the MODIS/ASTER Simulator, were designed to support investigations by the Moderate Resolution Imaging Spectro-radiometer, and the Advanced Spaceborne Thermal Emission and Reflection Radiometer science teams, respectively.

A collimator was required to qualify the Advanced Land Imager (ALI) instrument of the EO-1 new millennium satellite for focus, MTF measurements, and distortion. It was used during assembly of the instrument and during thermal cycling while the instrument was in a vacuum tank. It had to be diffraction-limited over a 3 X 3 degree field of view and over a waveband of 400 to 2500 nm, have no obstruction, and have a virtual exit pupil that could be imaged onto the entrance pupil of the telescope. To satisfy these requirements the collimator, external to the vacuum tank, was built comprising a spherical mirror with exit pupil at its center of curvature, a full-aperture beamsplitter in collimated space, and a single lens to flatten the field. The optical layout and method of verifying collimation will be presented as well as optical performance, of interest since no corrector plate could be used as in the usual Schmidt camera configuration.

The Advanced Land Imager (ALI) of the EO-1 satellite has been calibrated under its normal operating conditions in a thermal vacuum chamber. The optical equipment for the calibrations had to be placed outside the chamber. The effects on the calibrating beam of the vacuum chamber window are described here. Since the instrument senses the reflected solar spectrum, the existing fused-silica window was found to have adequate spectral transmission properties. It also was measured to cause little wavefront error by itself. When the chamber is under vacuum, and the interior cold shroud is cooled to approximately 100 K, the window develops significant optical power. This is a result of radiational cooling of the window, coupled with change of its index of refraction with temperature. Such an effect seriously compromises the spatial calibrations of the instrument, particularly MTF measurements. This effect was overcome in two ways: A heating plate placed outside the window was used to alter the temperature distribution until its effect on the wavefront was negligible. The more practical solution was to measure the window power with an interferometer, and compensate for it by shifting the target reticle of the collimator.

Mirrors in remote sensing instruments require durable dielectric coatings, both to prevent oxidation of the reflective surface and to protect it during cleaning. IR absorption bands within widely-used SiO_x coatings produce scene radiance and instrument background variations as a function of scan mirror angle which motivate the search for possible substitute materials. In this work several candidate coatings are evaluated including CeF₃, HfO₂, MgF₂ SrF2, and Y₂O₃. This evaluation consists of reflectance, adhesion, and durability measurements of mirrors with an aluminum reflective surface over-coated with these materials. S-polarized and P- polarized reflectance measurements are presented between 2 and 20 micrometers for incidence angles between 40 and 50 degrees. This angular range is sufficient to scan the earth disk from geostationary orbit. Additional measurements at 45 degrees incidence are presented between 2 and 55 micrometers , covering the IR wavelength range of interest for earth radiation budget sensors. Comparisons are drawn with measurements of scan- mirror witness samples from the imaging and sounding instruments used in the Geostationary Operational Environmental Satellite (GOES). These witness samples exhibit reflectance variations arising from IR absorption bands in the SiO_x protective coatings used in these mirrors. The spectral characteristics of several of the alternate materials are found to be quite attractive, however durable coatings of some of these materials require elevated deposition temperature which are incompatible with the nickel-coated beryllium scan mirror substrate construction used in GOES. This work present the achievable reflectance and durability of these alternate dielectric protective coatings at the deposition temperature constraints imposed by the scan mirror substrate. The prospects for substituting one of these coatings for SiO_x are evaluated, and contrasted with the capability of radiometric calibration techniques to deal with the reflectance variations produced by SiO_x coatings.

There are some special requirements for meteorology and oceanography remote sensor. For example: (1) Wide cross- track swath. The swath width of the modern remote sensor can be +/- 55 degrees against the nadir of the satellite. (2) Low Linear Polarization Sensitivity (LPS). The LPS should be LSEQ 2 degrees within 0.43 micrometers <EQ (lambda) <EQ 2.2 micrometers . The scan subsystem that can perform wide swath scanning and reduce LPS must be searched as a support to the requirements of these sensor. The two-mirror compensating/active configuration is the preferred candidate for the preliminary baseline.

In the design of the multispectral remote sensor, the primary design content is to obtain multispectral radiation from the earth scene. Several approaches have been taken to solve the problem of the spectral discrimination. The MMFA is the preferred candidate for the preliminary baseline.

This paper describes several methods for assigning tasks to Earth Observing Systems Satellites (EOS). We present empirical results for three heuristics, called: Priority Dispatch (PD), Look Ahead (LA), and Genetic Algorithm (GA). These heuristics progress from simple to complex, from less accurate to more accurate, and from fast to slow. We present empirical results as applied to the Window-Constrained Packing problem (WCP). The WCP is a simplified version of the EOS scheduling problem. We discuss the problem of having more than one optimization criteria. We will also discuss the relationship between the WCP and the more traditional Knapsack and Weighted Early/Tardy problems.

The large size is a typical feature of earth observation images. The increasing number of bands simultaneously available on satellites and the launch of system able to achieve a ground resolution as low as one meter is making this problem more a nd more hot. Indeed, the large size of typical images dramatically reduces the possibility to distribute them to a wide number of interest parties. In the wide majority of cases, this problem is currently tackled either by using lossy compression schemes such as basic versions of JPEG or by narrowing the ground extension of the pictures. Of course, both of them are unsatisfactory. Indeed, the partial loss of data may be acceptable only for non-quantitative analysis while narrower pictures may not carry all the needed informant. An alternate possibility is the use of an efficient lossless algorithm. Among others, JBIG has been preferred for this purpose because it achieves very high compression rate, overperforming the well known ZIP algorithm in the wide majority of cases. Further advantages are due to its progressive nature and to its availability as an ITU international standard. In order to have a very performant syste, this algorithm has been implemented by the development of an application specific integrated circuit designed to compress/decompress large volumes of data with throughput rate greater than 1 Gb/min. The performances achieved by the system when dealing with typical visual and radar satellite images and the perspective applications are described.

Boundary layer cumulus clouds are hard to detect in satellite imagery, especially for GOES imagery due to the coarse resolution of the IR channels. Two different approaches for the detection cumulus clouds in GOES satellite imagery are discussed and intercompared. The first step, structural thresholding, uses the morphology of cumulus cloud fields for detection. The second type, uses 1) classifiers based on texture and spectral, 2) edge detection and spectral, and 3) purely spectral features. For five selected scenes, cumulus cloud masks are created using these various methods and are compared against the expert-labeled masks. The structural thresholding method has the highest percentage of correct classification, followed by classifier based on Laplacian edge detection features. The classification time is lowest for the structural thresholding method, followed by classifiers based on spectral, edge detection, textural features. The structural thresholding method also is capable of detecting individual cumulus clouds within cloud fields. For the five scenes investigated, the average percentage of correct labeling of cumulus clouds by the structural thresholding method is 86 percent.

Many scientific data archives today are highly inflexible collections of static data objects that can be retrieved and distributed, but not easily manipulated to suit the users' requirements. Scientists, as well as their analysis and visualization tools, are significantly restricted by the limitations of data archive centers. These limitations include inability of the data centers to search archives based on the data content, inaccurate and ineffective geographic searches on swath data, inability to subset data products, lack of custom data products, lack of choices for gridding schemes and limitation of format choices. The Next Generation Information System being developed under the auspices of NASA's Passive Microwave Earth Science Information Partner (PM-ESIP) will address all the above limitations. This innovative system is being built around the flexible and extensible Algorithm Development and Mining System (ADaM) developed by the Information Technology and SYstems Center at the University of Alabama in Huntsville. ADaM provides an extensible framework for data mining and other types of processing including subsetting, gridding and mapping. The main emphasis for the design of the new PM-ESIP information system is to allow the user flexibility and ease in accessing and utilizing data. The PM-ESIP's next generation information system will be a scaleable distributed processing system that can grow as the need of the user community for on-demand processing increases. This paper will discus the design of this system along with some of the technology being used to build it.

MTF testing of optical systems incorporating high-speed time-delay-integrate device (CCD) arrays has typically been a challenging task, more so than area or linear arrays. TDI imaging depends upon the synchronized motion of an image with the clocking of TDI lines. Simulating this motion often requires complicated and very precise mechanical equipment such as a moving belt or rotating drum. An alternative to this mechanical approach involves the use of a flashlamp to freeze the motion of an object during one of the TDI integration stages. This flashlamp method is straightforward and eliminates MTF errors due to velocity mismatch and scan misalignment. This paper describes the approach used to obtain sine wave and knife-edge images for MTF analysis using the flashlamp method. Test data and results for an instrument with a large aperture that is used in R and D for high-resolution space-based imaging is presented.

A noise characterization of the Landsat-7 Enhanced Thematic Mapper Plus (ETM+) instrument was performed as part of a near-real time performance assessment and health monitoring program. Performance data for the integrated Landsat-7 spacecraft and ETM+ were collected before, during, and after the spacecraft thermal vacuum testing program at the Lockheed Martin Missiles and Space (LMMS) facilities in Valley Forge, PA. The Landsat-7 spacecraft and ETM+ instrument were successfully launched on April 15, 1999. The spacecraft and ETM+ are now nearing the end of the on orbit engineering checkout phase, and Landsat-7 is expected to be declared operational on or about July 15, 1999. A preliminary post-launch noise characterization was performed and compared with the pre-launch characterization. In general the overall noise levels in the ETM+ are at or below the specification levels. Coherent noise is seen in most bands, but is only operationally significant when imaging in the panchromatic band (band 8). This coherent noise has an amplitude as high as -.3 DN (peak-to-peak, high gain) at the Nyquist rate of 104 kHz, and causes the noise levels in panchromatic band images at times to exceed the total noise specification by up to —10%. However, this 104 kHz noise is now much weaker than it was prior to the successful repair of the ETM÷ power supplies that was completed in May 1998. Weak and stable coherent noise at —5 kHz is seen in all bands in the prime focal plane (bands 1-4 and 8) with the prime(side A) electronics. Very strong coherent noise at —20 kHz is seen in a few detectors of bands 1 and 8, but this noise is almost entirely in the turn-around region between scans when the ETM+ is not imaging the Earth. Strong coherent noise was seen in 2 detectors of band 5 during some of the pre-launch testing; however, this noise seems to be temperature dependent, and has not been seen in the current on orbit environment. Strong 91 kHz coherent noise was observed in the redundant (side B) panchromatic band data after the completion of spacecraft thermal vacuum testing. The cause of this coherent noise was identified as a failed capacitor that was replaced prior to launch, and this noise was not seen in the final pre-launch test.

Keywords: Landsat-7, ETM÷, Noise, Radiometry

The launch-related spectral band radiance change of the Spectro-Radiometric Calibration Assembly (SRCA) of MODurate Resolution Imaging Spectrometer (MODIS) is mainly attributable to its lamp temperature variations. Comparison of the spectral profiles, measured by the SRCA at different times, provides a way of tracking change in the band radiance while the integrated SRCA radiance remains constant. Prelaunch the SRCA was calibrated against a ground Spherical Integration Source (SIS100). Meanwhile the SRCA calibrations were run in spectral and radiometric modes. Comparison of the on-orbit data from the SRCA spectral and radiometric modes to prelaunch data will transfer SIS100 ground calibration to orbit. For validation, data form TV at nominal temperature plateau will define the radiometric transfer; this transfer will be applied to SRCA data measured at other temperature plateaus and compared with the measured SIS100 radiance values.

This paper describes the point source functions (PSF's) of the clouds and the Earth's Radiant Energy System (CERES), Earth Observing System (EOS), afternoon platform (PM), Flight Model 3 (FM3), and Flight Model 4 (FM4) scanning instruments. The PSF is vital to the accurate geo-location of the remotely sensed radiance measurements acquired by the instrument. This paper compares the characteristics of the FM3 and FM4 instruments with the earlier Proto Flight Model on the tropical rainfall measuring mission platform, and the FM1 and FM2 models on the EOS morning orbiting (AM) platform, which has recently been renamed 'Terra'. All of the PSF's were found to be quite comparable, and the previously noted 'spreading' characteristic of the window channel PSF is analyzed.

This document describes the calibration and performance of the SCIAMACHY instrument, to be launched on ESA-ENVISAT in 2000, after the main on-ground calibration and performance verification phases. A number of calibration and performance parameters will be discussed and results will be shown.

The Meteosat Second Generation (MSG) program consists of a series of 3 geostationary satellites, the objectives of which were defined by the European meteorological community led by the European Space Agency (ESA) and the European Organization for the Exploration of Meteorological Satellites. The development and procurement of MSG is under the responsibility of ESA. Alcatel France leads an industrial consortium of more than 50 European companies which develop, manufacture, integrate and test and spacecraft. The main payload, the imagin radiometer called spinning enhanced visible and IR imager, is developed by Matra Marconi Space France. The objective of the MSG program is to provide a continuous and reliable collection of environmental data in support of weather forecasting and relating services. A major element of this objective is fulfilled by the imaging mission, which corresponds to a continuous image taking of the earth using 12 channels with a baseline repeat cycle of 15 minutes, including the on- board calibration, the retrace and the overall satellite stabilization process. The imaging tasks are performed by the SEVIRI. Provision is also made on-board to carry as a passenger the so-called GERB instrument. This paper will focus on the calibration, characterization and sensitivity analysis of the SEVIRI radiometer/Imager.

ScaRaB is a radiometer built to observe the Earth Radiation budget at the top of the atmosphere. This instrument has been produced by means of a joint program that gathers France, Russia and Germany. SCARAB has been planned to fill the gap between the flights of NASAs instruments ERB and CERS. Thus SCARAB mission is similar to the one fulfilled by ERB and CERES. SCARAB FM1 has been launched in January 1994. At the very beginning of the flight, a critical problem occurred on-board: three calibration lamps over 6 failed and were unusable for the rest of the mission. An alternative way of calibration was found that saved the mission. This method, explained in a previous paper, and its accuracy are briefly recalled in the article. FM2 has been launched in July 1998. The first data show a perfect functioning of the on-board calibration devices. It is an opportunity to test the 3 in-flight calibration methods developed: a method using on-board lamps and black-bodies as previously planned on FM1, another one using the dependence of gains to temperature as used on FM1, and a last one using only black body-simulators. After describing more in detail these different methods, a comparison of their performances is established. These results are compared with FM1 performances. Future prospects of the calibration are given in conclusion.

The Advanced Solid-state Array Spectroradiometer (ASAS), an airborne hyperspectral off-nadir pointable pushbroom imager, has participated in a variety of remote sensing field campaigns and has historically been calibrated using a radiative transfer technique. Recent campaigns have provided several candidate calibration target image sites that can and have been used to validate instrument radiometric performance. Limitations in the sensor's CCD sensitivity and integrating sphere source existence provide a less than satisfactory calibration in the blue and near-IR wavelength bands and a vicarious calibration of these bands is under consideration. This paper will compare 'at sensor radiance' retrieved from a target of known spectral radiance using detector gains derived from laboratory calibration to radiances retrieved using detector gains derived from a well characterized ground calibration site. Physical reasons for the discrepancies and problems will be offered and discussed along with arguments for a combined vicarious and laboratory radiative transfer calibration.

The imagers and sounders aboard NOAA's Geostationary Operational Environmental Satellites (GOES) provide quantitative data for weather forecasting and studies of the Earth's atmosphere and surface. This paper describes the post-launch radiometric testing of the imagers and sounders and present result from the instruments aboard the GOES-8, - 9, and -10 satellites. In these tests, we measure such quantities as nose and signal-to-noise ratios, radiometric responsivities and their variability, and detector 1/f noise. Performance anomalies specific to GOES, such as variation of scan-mirror reflectance with east-west scan angle, are characterized. The on-orbit results are compared with the performance specification, with pre-launch test result, and with results, and with results of post-launch tests of the imagers and sounders on other GOES satellites.

The Modular Optoelectronic Scanner MOS of the German Aerospace Center (DLR) has now been working about three years on board the INdian Remote Sensing Satellite IRS-P3 very successfully. It consists of three instruments: the spectrometer MOS-A, the spectrometer MOS-B and the camera MOS-C. To meet the sophisticated radiometric and spectral requirements especially for ocean purposes, a 16 bit dynamic range and a qualified in-flight calibration concept including sun calibration and internal lamp calibration have been established. All three instruments have shown a remarkably high data stability and quality during the three years mission time. The highest changes of radiometric sensitivity of +7 percent were found in the SWIR- channels of MOS-B and the lowest changes of -1 percent in the MOS-A channels. Spectral shifts of center wavelengths could not be found. Small differences between the result of the two calibration methods are due to the fact that they do not cover the same optimal components exactly. But this enables us to allocate the sensitivity changes to those components of the instruments which cause them. The measured in-orbit calibration values were used to update the calibration coefficients.

A single data set of spatially extensive hyperspectral imagery is used to carry out vicarious calibrations for multiple Earth observation sensors. Results are presented based on a data acquisition campaign at the newell County rangeland test site in Alberta in October 1998, which included ground-based measurements, satellite imagery, and airborne casi hyperspectral data. This paper present new calibration monitoring obtained for NOAA-14 AVHRR, OrbView-2 SeaWiFS, SPOT-4 VGT, Landsat-5 TM, and SPOT-2 HRV.

The Clouds and Earth's Radiant Energy System (CERES) missions were designed to measure broadband earth-reflected shortwave solar and earth-emitted longwave radiances as well as earth-emitted narrow-band radiances in the water vapor window region between 8 and 12 micrometers . However, the CERES scanning thermistor bolometer sensor zero-radiance offsets were found to vary as much as 1.0 Wm^-2sr^-1 with the scan angle measurement geometry due to gravitational forces and systematic electronic noise. To minimize the gravitational effects, the Tropical Rainfall Measuring Mission (TRMM) Spacecraft CERES sensors; offsets were derived on-orbit as functions of scan elevation and azimuth angles from the January 7-8, 1998 radiometric observations of deep cold space, representative of a 3K blackbody.

In August, 1998, a Clouds and the Earth's Radiant Energy System (CERES) instrument telemetry housekeeping parameter generated a yellow warning message that indicated an on- board +15V Data Acquisition Assembly (DAA) power converter deregulation anomaly. An exhaustive investigation was undertaken to understand this anomaly and the long-term consequnces which have severely reduced CERES operations on the Tropical Rainfall Measuring Mission (TRMM) spacecraft. Among investigations performed were groudn tests that approximated the on-board electronic circuitry using a small quantity of flight identical components exposed to maximum spacecraft bus over-voltage conditions. These components include monolithic integrated microcircuits that perform analog signal conditions on instrument sensor signals and an analog-to-digital converter (ADC) for the entire DAA. All microcircuit packages have either a bipolar silicon design with internal current limiting protections or have a complementary metal oxide semiconductor (CMOS) design with bias protections. Ground test that have been running for approximately 8 months have indicated that these components are capable of withstanding as much as twice their input supply voltage ratings without noticeable performance degradation. This data provided CERES operators with assured confidence to be able to continue instrument science operations over the remaining life of the TRMM. This paper will discuss this anomaly and some possible cause, a simulator of affected electronics, test results, prognosis for future CERES operations, and conclusions.

The National Aeronautics and Space Administration (NASA) will be launching complex satellite remote-sensing platforms for monitoring the earth's radiation budget, land use, and atmospheric physics for periods exceeding 10 years. These Earth Observing Satellite (EOS) platforms will strive to detect man-made and natural variations in the Earth's climate. Form 1993 to the present (1999), the National Renewable Energy Laboratory and the King Abdulaziz City for Science and Technology (KACST) in Riyadh, Saudi Arabia, conducted a joint solar radiation resource assessment project to upgrade the solar resources assessment capability of the Kingdom of Saudi Arabia. KACST has deployed a high quality 12-station network in Saudi Arabia for monitoring solar total horizontal, direct beam, and diffuse radiation. One- and five-minute network data is collected and assessed for quality. 80 percent or more of the network data fall within quality limits of +/- 5 percent for correct partitioning between the three radiation components. This network will provide measured data for validating the NASA remote sensing systems. We describe the network, quality assessment procedures, and the result of estimating aerosol optical depth and precipitable water vapor. These are important for validating satellite estimates of radiation fluxes in and at the top of the Earth's atmosphere.

The present study is part of an investigation aimed at optimizing the use of desertic sites for absolute or relative calibration of satellite visible sensors. This effort includes characterization of the surface, gathering of climatology or atmospheric data sets, ground- and air- based measurements as well as result of calibration of various sensors over these sites. All these measurements and estimates are stored in a repository and made available to various methods for calibration. Post-launch degradation and relative sensitivity of various sensor have been estimated using north african desertic sites as radiometrically stable targets. The selected area have first been characterize in terms of bidirectional and spectral reflectances by making use of POLDER capabilities, then to cross-calibrate SeaWifs, VEGETATION on-board SPOT4 and AVHRR on-board NOAA-14 by reference to POLDER. Results are compared with absolute and relative calibration issued from other sources. Extensive period of time are spanned to assess the ability of this method to monitor long term trends in sensor evolutions. Results of this cross calibration will be presented. The method developed for this study will be presented as well, in order to make it applicable to other sensor. A sensitivity study has also been realized, considering synthetic data, allowing to evalute the main contributions to the error budget. The need for aerosol optical thickness is then evidenced, and will lead to the set up of a sun photometer on one of the selected sites in 1999.

During 1997-98, the NASA-NOAA Advanced Geosynchronous Studies (AGS) program sponsored work to explore the possibility of designing a high-quality imaging radiometer for the future GOES platforms. The AGS Imager (AGSI) design calls for the acquisition of 12-bit digital images every few minutes in 18 spectral bands with horizontal resolutions ranging from 0.3 to 1.5 km on a full-earth disk. The resulting raw sensor data stream is approximately 300 Mbits/sec, uncompressed. The AGSI ground system must convert the sensor data into calibrated, earth-located 'Level 1b' images and deliver them to NOAA and to the science community within minutes of reception. To accomplish this high-speed digital-image delivery, the AGSI downlink design calls for lossless compression to 150 Mbits/sec, packetized using CCSDS standards with error-correction, and broadcast in Ka- band to a low-precipitation site like White Sands, NM.

Scatterometers are active microwave instruments designed to measure ocean wind speed and direction. Recently they have been tested for the mapping of sea ice. In preparation for the launch of QuikSCAT, the goal of this research was to develop an ice delineation algorithm for the Bering Sea using reconstructed NASA scatterometer data. The algorithm was designed to utilize simple backscatter, image texture, and vertical/horizontal polarization ratios to track the advance of the Bearing Sea ice edge during late winter of 1997. The results of the image processing were compared to the published National Ice Center maps for the same period. There was good agreement between the two products. The greatest challenge in developing the algorithm was detecting the sea ice in areas of very low concentration.

Scatterometers were originally designed and employed based on their proven ability to measure near-surface winds over the ocean. However, they are also providing useful in global and regional studies of vegetation and soil moisture. In this paper we examine the C-band and Ku-band radar signatures of vegetation over South America using data for the ERS-1/2 and NSCAT scatterometers. We compare the seasonal response of various types of vegetation based on both Matthew's classification and the University of Maryland AVHRR-based classification and use the seasonal response in a simple vegetation classification experiment. A time-series of enhanced resolution (sigma) degree images are generated for both sensors. The classifier is used to classify the vegetation coverage of each pixel in the image frame into broad classes of vegetation type. Considering the accuracy and resolution limitations of the reference vegetation maps, the classifications result exhibit a high degree of accuracy and consistency with the primary confusion observed between related vegetation classes of similar vegetation canopy density.

This paper describes the x-ray camera for the Atmospheric X- ray Observatory (AXO) proposed for the Danish Small Satellite Program, which is under evaluation for the next mission in 2003. AXO is aimed at localizing the origin of the Terrestrial Gamma Flashes (TGF) that have been observed with BATSE. An additional objective is a detailed mapping of the auroral x-ray and optical emission. The x-ray camera to be used must be capable of detecting quite weak and pointlike, short-duration emission from TGF, and also to handle with the rather intense and extended radiation from auroral activity. The x-ray energy range is 5-200 keV and the angular resolution about 2 degrees. The requested satellite orbit is polar with an altitude of 500 km so that the phenomena can be seen from a close range. The design of a coded mask camera matching these requirements is discussed in terms of energy and angular resolution, sensitivity, count rates, and time resolution. Detailed simulations of the camera imaging capabilities are presented.

Construction of compact 20-channel airborne spectrometer 8- 13 mkm range is presented. Spectrometer optical part is based on the only metal optical elements as well as a carrier constructive elements. The result of laboratory testing of the spectrometer sensitivity and accuracy are presented. A possible way for unit sensitivity and accuracy improvement are shown. It has been carried out the analysis of possible ways for development of non-imaging spectrometer for 8-13 mkm spectral range. The selected scheme of spectrometer design is optimal from a point of view of ensuring the minimal error of temperature measurement with desired temperature sensitivity. The basic requirements to each of three spectrometer modules and some important parts were determined. Calibration module - the radiative temperature reproducibility relative error has to be less than 0.2 percent. Optical module - the IR energy dissipation in optical system has to be less than 55 percent, the collective objective area has to be approximately 200 cm², the aperture ratio of collective objective should be approximately 1, the error of spectral channel location has to be less than 120 nm. Electronic module - the magnitude of equivalent input noise should be less than 8 nV Hz^-1/2. The 20-element IR radiation detector on the basis of MCT with detectivity D* approximately 1.2 10¹⁰ cm Hz^1/2 Wt^-1 and photosensitive elements sizes of 250 X 250 mkm has been used.

This paper present the basic design of an information system developed for resources assessment and environmental monitoring in the Yonezawa region of Japan. The system combines the module of satellite data and image processing with the database modulating an integrated manner. The system delivers several products, of primary and secondary nature, as well as provides for various user-defined reports for in depth analysis based on the existing database system. Landcover, soil-cover and vegetation index maps, elevation model and other topographical parameters like slope and aspect are generated by the system. Temporal monitoring of the landcover and vegetation changes is also provided by the system, as are several reports highlighting the various geographical the various geographical features int eh study region. The paper presents an interesting work since it demonstrates the use of satellite data for operational environmental monitoring while using a relatively simpler design with minimal system resources.

In an effort to provide accuracy assessment of the visible and shortwave IR (SWIR) channels of MODIS Airborne Sensor (MAS), the result of three in-flight calibration/validation experiments from 1996-1998 are presented. Both MAS and Airborne Visible IR Imaging Spectrometer (AVIRIS) were flown together on NASA's ER-2 over two bright targets. Additionally MAS acquired another bright target, Ivanpah Playa, in 1998. AVIRIS data were spectrally and spatially convolved to MAS specifications for cross comparisons. Predicted at sensor radiances using MODTRAN 3.5 constrained with in situ data are then compared to MAS and convolved AVIRIS radiances for absolute radiometric calibration.

A new modular Active Cavity Radiometer Irradiance Monitor (ACRIM) has been developed for the Earth Observation System (EOS) ACRIM III experiment. ACRIM III will be launched on a dedicated small satellite (ACRIMSAT) in late 1999 for a five year EOS Phase I total solar irradiance (TSI) monitoring mission. The new ACRIm instrument is compact, reduced in mass and size to less than 1/2 that of previous ACRIM's. The enhanced structure, electronics and shutter system, designed collaboratively by the Columbia-based Principal Investigator and Jet Propulsion Laboratory instrumentation teams, provide a versatile, precise and accurate, state-of-the-art TSI observational capability for the EOS and other potential applications.

SCIAMACHY is a UV-vis-near IR spectrograph for atmospheric trace gas monitoring, due to be launched on the ESA ENVISAT in 2000. The SCIAMACHY Instrument Simulation Software was used to aid in the design of the instrument, support on- ground performance and calibration measurements, develop the level 1 algorithms and establish in-flight operational scenarios. The instrument simulator is based on a traveling spectrum approach from entrance aperture to detector electronics. Relevant optical properties of the individual components are used to model the spectrum and polarization state of the light at each component in the instrument. Subsequently, noise properties and electronic effects are added. The resulting simulated signal is directly comparable with recently performed measurements on the assembled instrument under flight representative conditions. In this paper the measured SCIAMACHY instrument performance is discussed on the basis of these simulations, in aid of better instrument knowledge required during space operation.