BOREAS RSS-14 Level-1a GOES-8 Visible, IR, and Water-Vapor Images Summary The BOREAS RSS-14 team collected and processed several GOES-7 and GOES-8 image data sets that covered the BOREAS study region. The level-1a GOES-8 images were created by BORIS personnel from the level-1 images delivered by FSU personnel. The data cover 14-Jul-1995 to 21-Sep-1995 and 12-Feb-1996 to 03-Oct-1996. The data start out as three bands with 8-bit pixel values and end up as five bands with 10-bit pixel values. No major problems with the data have been identified. The differences between the level-1 and level-1a GOES-8 data are the formatting and packaging of the data. The images missing from the temporal series of level-1 GOES-8 images were zero-filled by BORIS staff to create files consistent in size and format. In addition, BORIS staff packaged all the images of a given type from a given day into a single file, removed the header information from the individual level-1 files, and placed it into a single descriptive ASCII header file. The data are contained in binary image format files. Table of Contents * 1 Data Set Overview * 2 Investigator(s) * 3 Theory of Measurements * 4 Equipment * 5 Data Acquisition Methods * 6 Observations * 7 Data Description * 8 Data Organization * 9 Data Manipulations * 10 Errors * 11 Notes * 12 Application of the Data Set * 13 Future Modifications and Plans * 14 Software * 15 Data Access * 16 Output Products and Availability * 17 References * 18 Glossary of Terms * 19 List of Acronyms * 20 Document Information 1. Data Set Overview 1.1 Data Set Identification BOREAS RSS-14 Level-1a GOES-8 Visible, IR, and Water-Vapor Images 1.2 Data Set Introduction For the BOReal Ecosystem-Atmosphere Study (BOREAS), the level-1a Geostationary Operational Environmental Satellite 8 (GOES-8) imagery, along with the other remotely sensed images, was collected in order to provide spatially extensive information over the primary study areas at varying spatial scales. The GOES-8 data set has a significant improvement in spatial resolution over the GOES-7 data from 1994 and early 1995. 1.3 Objective/Purpose The primary objective for the GOES-8 images in 1995 and 1996 was to collect visible, infrared (IR) and water-vapor channel data covering the BOREAS region at a sufficiently high temporal frequency for subsequent use in analyzing weather events and deriving temporal surface radiation parameters and patterns that existed during the Focused Field Campaign (FFC) and Intensive Field Campaign (IFC). 1.4 Summary of Parameters The level-1a GOES-8 data from 1995 and 1996 in the BOREAS Information System (BORIS) contains the following parameters: Image header and summary information; central geographic position; digital counts for half-hourly visible and IR images; latitude and longitude coordinate reference files. The level-1a GOES-8 data set starts out as a 3- band 8-bit data set in 1995 and changes to a 5- band 10-bit data set in 1996. The image sizes varied in 1995 as well but stabilized in 1996. 1.5 Discussion Dr. Eric Smith Florida State University (FSU) provided BORIS with the level-1 GOES-8 images that were used to create the level-1a products. BORIS staff processed the level-1a GOES-8 images by: 1) summarizing and extracting header information from the level-1 GOES-8 images and and placing it in an American Standard Code for Information Interchange (ASCII) file on disk, 2) reviewing the header file information for potential errors, 3) working with FSU personnel to remove erroneous files detected in step 2, 4) repackaging/reformatting the image data for a given day, 5) writing the reformatted data files to tape, and 6) loading the on line data base with needed information. 1.6 Related Data Sets BOREAS Level-1 GOES-7 Images from 1994 and 1995 BOREAS Level-1a GOES-7 Images from 1994 and 1995 BOREAS Level-2 GOES-7 1994 Surface Radiation Data BOREAS Level-1 GOES-8 Images from 1995 and 1996 2. Investigator(s) 2.1 Investigator(s) Name and Title 2.2 Title of Investigation GOES Imagery for the BOREAS Experimental Areas 2.3 Contact Information Contact 1 --------- Jeffrey A. Newcomer Raytheon STX Corporation NASA GSFC Greenbelt, MD (301) 286-7858 (301) 286-0239 (fax) Jeffrey.Newcomer@gsfc.nasa.gov Contact 2 --------- Dr. Eric A. Smith Florida State University Tallahassee, FL (904) 644-4253 (904) 644-9639 (fax) esmith@metsat.met.fsu.edu 3. Theory of Measurements The mission of the GOES satellite series is to provide the nearly continuous, repetitive observations that are needed to predict, detect, and track severe weather. GOES spacecraft are equipped to observe and measure cloud cover, surface conditions, snow and ice cover, surface temperatures, and the vertical distributions of atmospheric temperature and humidity. They are also instrumented to measure solar X-rays and other energetics, collect and relay environmental data from platforms, and broadcast instrument data and environmental information products to ground stations. The GOES system includes the satellite (with the GOES instrumentation and direct downlink data transmission capability); the National Environmental Satellite, Data and Information Service (NESDIS) facility at Wallops Island, VA; and the ground systems at NESDIS. 4. Equipment 4.1 Sensor/Instrument Description (GOES-8) The GOES I-M imager on the GOES-8 satellite is a five-channel (one visible, four IR) imaging radiometer for measuring radiant and reflected energy from Earth. Using a servo-driven, two-axis gimbaled mirror scan system with a Cassegrain telescope, the multispectral channels simultaneously sweep an 8-km (5-statute- mile) north-to-south swath along an east-to-west/west-to-east path, at a rate of 20 degrees east-west per second. The imager consists of electronics, power supply, and sensor modules. The sensor module containing the telescope, scan assembly, and detectors is mounted on a base plate external to the spacecraft, together with the shields and louvers for thermal control. The electronics module provides redundant circuitry and performs command, control, and signal processing functions; it also serves as a structure for mounting and interconnecting the electronic boards for proper heat dissipation. The power supply module contains the converters, fuses, and power control for interfacing with the spacecraft electrical power subsystem. The electronics and power supply modules are mounted on the spacecraft internal equipment panel. -------------------------------------------------------------------------------- Imager Instrument Characteristics Spectral Bands (micrometers) VIS IR2 IR3 IR4 IR5 -------------------------------------------------------------------------------- Wavelength (micrometers) 0.55 3.80 6.50 10.20 11.50 to to to to to 0.75 4.00 7.00 11.20 12.50 -------------------------------------------------------------------------------- Clouds X X X X X -------------------------------------------------------------------------------- Water Vapor X X X -------------------------------------------------------------------------------- Surface Temp 0 X 0 -------------------------------------------------------------------------------- Winds X X X -------------------------------------------------------------------------------- Albedo & IR Flux X 0 X 0 -------------------------------------------------------------------------------- Fires & Smoke X X 0 0 -------------------------------------------------------------------------------- X: Primary Spectral Channel 0: Secondary (supplementary) Spectral Channel -------------------------------------------------------------------------------- Field of View Defining Element: Detector -------------------------------------------------------------------------------- Optical Field of View: Square -------------------------------------------------------------------------------- 5-channel Imaging: Simultaneously -------------------------------------------------------------------------------- Scan Capability: Full Earth/Sector/Area -------------------------------------------------------------------------------- Channel/Detector Instantaneous Field of View (IFOV) Visible/Silicon : 1 km Short-wave/InSb : 4 km Moisture/HgCdTe : 8 km Long-wave 1/HgCdTe : 4 km Long-wave 2/HgCdTe : 4 km -------------------------------------------------------------------------------- Radiometric Calibration: Space and 290 Kelvin IR internal blackbody -------------------------------------------------------------------------------- Signal Quantizing (NE’delta’T) : 10 bits all channels S/N : Minimum 3X better than specifications -------------------------------------------------------------------------------- Frequency of Calibration Space : 2.2 sec for full disk; : 9.2 or 36.6 sec for sector/area Infrared : 30 minutes typical -------------------------------------------------------------------------------- System Absolute Accuracy : IR channel less than 0.1 K -------------------------------------------------------------------------------- Transmit Frequency : 1676.00 MHz -------------------------------------------------------------------------------- 4.1.1 Collection Environment The GOES-8 data were acquired using the FSU Direct Readout Ground System located in Tallahassee, FL, starting on 14-Jul-1995 and continuing through 23-Oct-1996. The GOES-8 satellite orbited Earth in a geostationary orbit at an altitude of 35,788 km (19,324 nautical miles). 4.1.2 Source/Platform Launch and data available dates for the GOES-8 satellite are: Satellite Launch Date Data Range --------- ----------- ----------------------- GOES-8 13-Apr-1994 0-1024 4.1.3 Source/Platform Mission Objectives The mission of the GOES satellite series is to provide the nearly continuous observations that are needed to predict, detect, and track severe weather. GOES spacecraft are equipped to observe and measure cloud cover, surface conditions, snow and ice cover, surface temperatures, and the vertical distributions of atmospheric temperature and humidity. They are also instrumented to measure solar X-rays and other energetics, collect and relay environmental data from platforms, and broadcast instrument data and environmental information products to ground stations. For BOREAS, the level-1 GOES-8 imagery, along with the other remotely sensed images, was collected in order to provide spatially extensive information over the primary study areas at varying spatial scales. The primary objective for the GOES-8 images in 1995 and 1996 was to collect visible, IR and water-vapor channel data covering the BOREAS region at a sufficiently high temporal frequency for subsequent use in analyzing weather events and deriving temporal surface radiation parameters and patterns that existed during the FFC and IFC. The GOES-8 data set has a significant improvement in spatial resolution over the GOES-7 data from 1994 and early 1995. 4.1.4 Key Variables Reflected radiation Emitted radiation Water vapor 4.1.5 Principles of Operation The GOES I-M program is a continuation of the previous National Oceanic and Atmospheric Administration (NOAA) National Aeronautics and Space Administration (NASA) collaboration to provide continuous monitoring of Earth's environment for weather forecasting and research. The objectives of the GOES I-M program are to maintain and expand the operational, environmental, and storm warning capabilities; to monitor Earth's atmosphere and surface and space environmental conditions; and to introduce improved atmospheric and oceanic observations and data dissemination capabilities. GOES I-M is a new series of five satellites that meet these objectives, providing significant improvements in weather imagery and atmospheric sounding information in accordance with current weather service requirements, particularly in regard to the forecasting of life-and property-threatening severe storms. A novel space-and ground-based computer and communication system provides users with calibrated and navigated (i.e., Earth-located) imagery and sounding data, in real time. The GOES I-M spacecraft meet the mission’s objectives by providing: • Independent imaging and sounding functions with instrument resolution, navigation, channelization, and signal-to-noise characteristics representing improvements over previous GOES missions • Full-time Weather Facsimile Transmission • Data Collection System Transponder Functions • Space Environment Monitor System • Search and Rescue Transponder Functions These functions have resulted directly from the overall system requirements developed by NOAA and NASA together with the National Weather Service (NWS), the system’s primary user. Many of the requirements are continued missions from the previous GOES D-H series of spacecraft. However, technological advancements have provided significant improvements in the reliability and availability of these systems for the new GOES I-M series. As in the previous GOES mission, the GOES I-M system provides the above services over a region covering the central and eastern Pacific Ocean, the 48 contiguous States, and the central and western Atlantic Ocean. This is accomplished by two satellites, GOES West located at 135 degrees west and GOES East (GOES-8 in this case) at 75 degrees west. A common ground station, the Command and Data Acquisition (CDA) station located at Wallops Island, VA, services both satellites. The GOES I-M Imaging and Sounding instruments provide significantly improved measurement capability over the previous GOES sensors. The GOES I-M five- channel Imager processes higher spatial resolution (i.e., 4 km for its IR channels) and higher radiometric sensitivity to improve the measurement of cloud and Earth's surface features. Sounding quality is improved by having more spectral channels (18 IR and 1 visible) with greatly improved radiometric sensitivity. The three-axis stabilized platform enables higher quality imagery and sounding data to be achieved through its dwell time advantage over a spinning satellite. The flexibility of scan control by both instruments combined with the three-axis stability enables rapid small-area coverage in addition to hemispheric or full-disk coverage. The new limited-area, higher frequency observation capability permits more continuous monitoring of severe weather development. The flexible nature of the Imager is used to provide a star-sensing capability. Time and location of a star is predicted very accurately and related to the spacecraft location and optical field. From a set of these data, the ground control system chooses a location and a time that is convenient within the imaging schedule. At the time for the scheduled starlook, the Imager is pointed to the predicted star location, which can be anywhere within its 21 degrees N-S by 23 degrees E-W view. (These viewing limits are for star sensing only. The maximum frame size during normal imaging operations is 19 degrees N-S by 19.2 degrees E-W). As the star passes through one or two of the eight elements of the visible array, it is sampled for Instrument Navigation & Registration (INR) purposes. The data is in the normal format and data stream for extraction and use at the ground station. The GOES I-M generation of spacecraft has been developed by the Space Systems/Loral, Inc. (SS/L). These satellites are three-axis body stabilized, meaning that the three axes of the satellite remain stationary relative to nadir. These satellites use internal momentum wheels to provide attitude control and require corrective action from the ground to compensate for the effects of thermal gradients and solar winds. Unlike the previous GOES D-H series, the GOES I-M spacecraft’s Imaging and Sounding instruments can be operated simultaneously and independently of one another. The spacecraft’s configuration is essentially a compact six-sided main body that carries the operational instruments, a continuous drive solar array attached to the south panel through a yoke assembly, a solar sail mounted off the north panel to offset solar pressure torque, a Telemetry and Command (T&C) antenna boom-mounted on the east end for full omni-directional coverage, and the Space Environmental Monitoring (SEM) magnetometer on a boom off the anti-Earth side of the satellite. The main body of the spacecraft accommodates the 5-channel visible and IR Imager and the 19-channel visible and IR Sounder, which sample radiance from Earth by identical two-axis scan systems and nearly identical telescopes in each unit. Scan control and data collection for the instruments are independent of each other and of most other activity on the spacecraft. 4.1.6 Sensor/Instrument Measurement Geometry During data acquisition for BOREAS, the GOES-8 satellite was stationed at approximately 0.0 degrees N, 75.0 degrees W. The Imager is a multi-channel instrument designed to sense radiant and solar- reflected energy from sampled areas of Earth's surface and atmosphere. The Imager’s multi-element spectral channels simultaneously sweep an 8-km north- south (N-S) (longitudinal) swath along an east-west (E-W) (latitudinal) path by means of a two-axis gimbaled mirror scan system. Position and size of an area scan are controlled by command. Beam splitters separate the spectral channels to the various IR detector sets, which are redundant. The 1-by-8 km visible detector array consisting of eight individual detectors is not redundant. Radiometric quality of the Imager is maintained by frequent views of space for Dark Current (DC) signal restoration. Less frequent views of the full aperture black-body (BB) establish a high temperature calibration point that determines the radiometric conversion factor for the IR channels. The frequency of radiometric calibration depends on the thermal and electrical stability of the system. In addition to radiometric calibration, the amplifiers and data stream are checked regularly from a 16-increment electronic calibration signal. This verifies the stability and linearity of the output data. Control of the Imager comes from a defined set of command inputs. The instrument is capable of full Earth imagery, sector imagery that contains the edges of Earth, and various sizes of area scans totally enclosed within the Earth scene. Area scan selection permits rapid, continuous viewing of local areas for monitoring of mesoscale phenomena and accurate wind determination. Area scan size and location are definable to less than one visible pixel to provide complete flexibility. Motion of the Imager and Sounder scan mirrors causes a small but well-defined disturbance of the spacecraft attitude. This effect is gradually reduced by spacecraft control but at a rate too slow for total compensation. Since all the physical factors of the scanners and spacecraft are known and the scan positions are continuously provided to the Imager and Sounder, the disturbances caused by each scan motion on the spacecraft and distributed to each instrument are calculated by the Attitude and Orbit Control System (AOCS). The Mirror Motion Compensation (MMC) signal is developed and used in the scan system server control loop to slightly modify the scan rate and position to offset the disturbance. This simple signal and control interface provides corrections that reduce any combination of effects. With this system in place, the Imager and Sounder are totally independent, maintaining image location accuracy regardless of the other unit’s operational status. If need be, this MMC scheme can be disabled by command. The AOCS also provides an Image Motion Compensation (IMC) signal that counteracts the spacecraft attitude, orbit effects, and predictable structural- thermal effects within the spacecraft-instrument combination. These effects are detected from ranging, star sensing, and landmark features. Corrective algorithms developed on the ground are fed through the AOCS to the instruments as a total IMC signal, which includes the MMC described above. Signal flow through the Imager maintains the maximum capability of each part of the optics, detection, and electronic subsystems to preserve the quality and accuracy of the sensed information. The signal flow starts with the radiation collected from the scene by the instrument’s optical system. This scene radiance is separated into appropriate spectral channels and imaged onto the respective detectors for each channel. Each detector converts the scene radiance into an electrical signal that is amplified, filtered, digitized, and put into a data stream for transmission to a ground station. The sensor assembly is mounted on a base plate external to the spacecraft, together with shields and louvers for radiation and heat control. The electronics module provides a structure for mounting and interconnecting the electronic boards with proper heat dissipation. The sensor assembly contains the telescope, scan assembly, and detectors. A passive radiant cooler with a thermostatically controlled heater maintains the IR detector temperature for efficient operation. The IR detectors operate at three patch temperatures: 94° K for 7 or 8 months that include the winter season, 101° K for the 4 or 5 months that include the summer season, and 104° K for radiative cooler contingencies. The visible detectors are at temperatures of 13 degrees C to 30 degrees C. Preamplifiers in the sensor assembly convert the low-level signals to higher level, low-impedance outputs for transmission by cable to the electronics module. A passive louver assembly and electrical heaters on the base aid thermal stability of the telescope and major components. The Imager instrument acquires radiometric data simultaneously for five distinct channels. These five radiometric channels are characterized by a central wavelength denoting primary spectral sensitivity within these channels. The five channels are split into two distinct classes, visible and IR, and comprise a total of 22 detectors as follows: • Visible-Channel 1 of the Imager contains eight visible detectors arranged in a linear fashion (v1-v8). Each detector provides an Instantaneous Geometric Field of View (IGFOV) of 28 microradians on a side. At the subsatellite point, this corresponds to a square pixel of 1 km per side. • IR-Channel 3 (6.75 µm) contains two square detectors, one primary and one redundant. Each provides an IGFOV of 224 microradians corresponding to an 8-km resolution at the subsatellite point. • IR Channels 2 (3.9 µm), 4 (10.7 µm), and 5 (12.0 µm) each contain four detectors: two of these are the primary detectors and the other two provide redundancy. Each of these detectors is square, providing an IGFOV of 112 microradians. At the subsatellite point, this corresponds to a square pixel having dimensions of 4 km per side. The Imager scans the selected image area in alternate directions on alternate lines. The imaging area is defined by a coordinate system related to the instrument’s orthogonal scan axis. During imaging operations a scan line is generated by rotating the scanning mirror in the E-W direction while concurrently sampling each of the active imaging detectors. At the end of the line, the Imager scan mirror performs a turnaround, which involves stepping the mirror to the next scan line and reversing the direction of the mirror. The next scan line is then acquired by rotating the scanning mirror in the opposite, west-east direction, again with concurrent detector sampling. Detector sampling occurs within the context of a repeating data block format. In general, all visible detectors are sampled four times for each data block (four times 1 km wide), while each of the active IR detectors is sampled once per data block (one times 4 km wide). 4.1.7 Manufacturer of Sensor/Instrument Aerospace/Communications Divisions of ITT McLean, VA 4.2 Calibration The calibration of the IR data and the normalization of the visible data are performed by the Operations Ground Equipment (OGE) on the raw data received from the spacecraft Imaging and Sounding sensors. The calibration/normalization function can be described in terms of those functions that occur during online processing and those that are performed during non-real-time operating modes. The real-time calibration and normalization of Imager and Sounder data can be divided into a continual application process and a periodic calibration coefficient generating process. In the real-time continual application process, factory-measured detector response characteristics together with in-flight measurements made while viewing space and BB targets are used by the Sensor Processing Subsystem (SPS) to convert raw Imager and Sounder sensor data to theoretical target radiance. All radiometric image data produced by the Imager and Sounder instruments must undergo calibration/normalization processing. This function is performed in the SPS and involves the conversion of instrument output from raw digital counts to its final physical units. For IR data calibration, this process uses the recalculated gain and bias factors to adjust for detector variations over time. This calibration process takes place in the SPS. The visible data normalization is performed so that all detectors of the same instrument produce the same readings when viewing an area of uniform brightness. The data produced by the eight Imager visible channels must be normalized to prevent striping. The normalization process is performed in the SPS with data provided by the Product Monitor (PM). These data are generated by an operator performing a histogram matching using data with the full range of intensities. The SPS maintains a current calibration data base for each satellite to be used in the real-time calibration of raw Imager and Sounder sensor data. The data base is maintained for both primary and redundant detectors. The SPS maintains the coefficients for the calibration equations that have been supplied to the data base prior to launch. This factory detector response information consists of Imager and Sounder IR nominal coefficients. The SPS data base has the equations for converting the BB thermistor output to temperature and BB temperature to equivalent target radiance. In addition, the data base contains the current calibration coefficients for the IR channels, which are based on the space and BB measurements. These calibration coefficients, computed by the SPS, are the gain and bias factors and coefficients of the quadratic terms. They must be recalculated periodically because it is expected that these factors will vary with the age and temperature of the instruments. This information is maintained, for both the Imager and Sounder, in a data base that resides in the SPS memory. Normalization for Imager visible data is performed in real time by the Sensor Data Interface (SDI) hardware, through use of look-up tables. For Imager and Sounder IR data, calibration is performed by the SPS software, using the calculated calibration coefficients. Imager data are ingested into the SPS via the SDI as a serial bit stream. Following ingest processing, including bit and frame synchronization, the SDI hardware under SPS software control performs normalization of the visible data. The SDI hardware uses the radiometric look-up tables to normalize Imager visible data by compensating for variation in the satellite detector response curves over time. The appropriate tables are loaded from the SPS memory into SDI hardware tables, under the control of the Imager Interface task. These look-up tables are used by the SDI to locate the correct value for a detector pixel. There are different tables for each of the eight Imager visible detectors. These radiometric tables are loaded with values that have been adjusted to incorporate the latest normalization information. Imager IR data are deposited in SPS memory without any linear transformations. Calibration is then performed by SPS software using the actual calibration equations for the IR detectors. Sounder data are deposited in the SPS memory by the SDI, following ingest processing. The Sounder Data Formatter task calibrates Sounder IR data and normalizes visible data, in SPS software, using the actual calibration equations for the IR and look-up tables for the visible. The periodic calibration coefficient generating process is performed to ensure that the best possible coefficient values are used when calibrating Imager and Sounder IR data. This process allows for periodic recalculation of the coefficients contained in the calibration data base of the SPS. Spacecraft instrument data are provided to the SPS in support of this operation. This section defines the data received from the Imager and Sounder and their use in the detector calibration process. The Imager and Sounder in the spacecraft perform periodic in-flight sequences that support IR calibration processing the OGE. The instrument information generated includes space look, BB calibration, and Electronic Calibration (ECAL). A space look sequence provides radiometric data for all detectors from a view of space located beyond the edge of Earth. The Imager can perform space looks in two ways. The instrument scan can be interrupted and the mirror pointed to space to take a reading; this is called a space clamp. The frame can also be made such that either the left side or the right side extends beyond Earth's edge, beyond Earthshine, into space, making the instrument scan space; this is called a scan clamp. In space clamp mode, the Imager performs a space clamp at the start of a frame with subsequent space clamps performed upon the execution of a timer. The timer is selected at 9.2 seconds or 36.6 seconds; the timer restarts upon expiration so that space looks are at fixed intervals. Upon timer expirations, the Imager completes the line it is scanning, slews horizontally to the preselected side to a point 10.2° from nadir to perform the space look, performs a turnaround sequence, and resumes scanning the frame. The turnaround sequence includes retracing the last line three times, so the direction of the interruption is proportional to the width of the frame. The total amount of time the scan is interrupted is proportional to the number of space looks performed, and therefore, to the total size of the frame. In scan clamp mode, the frame scan is never interrupted, but the frame actually scanned extends out into space, beyond the area of meteorological interest. The frame edge extends beyond Earth's edge, beyond Earthshine, a 0.5° wide ring around Earth, and out into space far enough to acquire a preset number of samples. Initially, the number of samples obtained by the GOES-8 Imager will be 400 IR samples, 64 microns on center, for a displacement of 1.47° beyond Earthshine. Scan clamp mode will typically be used for full disk frames, making the frame boundary lie 1.97° beyond Earth's edge at the Equator, 10.67° from nadir. In a full disk frame scanned thus, with the over scan into space applied only to one side of Earth, i.e., every other scan line, space calibration data are obtained every 2.2 seconds. The Sounder operates in a space clamp mode similar to that of the Imager. The Sounder space clamp mode is its only mode of operation. A space look is performed by the Sounder every 2 minutes. In normal and priority modes, after each location is sounded, a timer is examined to see if a space look is necessary. The Sounder then saves the current address and slews to the space location. Upon completion, the Sounder checks to determine if ECAL and BB calibration should be performed. If not, the instrument returns to the original location. A BB calibration consists of data samples from a view of the instrument’s internal BB. Both space look and ECAL data are transmitted as part of the BB calibration operation. The ECAL data generated are used by the PM to monitor linearity of the Imager and Sounder instrument’s electronic gain. In the Imager, the BB calibration is normally performed every 20 minutes, unless doing so would mean interrupting an image in process. At the most, the Imager may go 30 minutes between BB views. The BB calibration process does not interrupt any other operation and cannot occur during a frame. The only exception to this is if a frame is interrupted by a priority scan frame or a star sense operation. The BB calibration may then be performed after the “priority” operation is completed and before returning to the interrupted frame. When executing a BB calibration, the Imager outputs ECAL data, performs a space look operation, and then slews to the BB location to gather data. It then returns to space and executes another space look. In the Sounder, a BB calibration sequence is performed after every 20 minutes, unless inhibited by ground command. The Sounder BB calibration operation consists of outputting samples of ECAL, after the space look samples, and slewing to a BB. The scan mirror settles, then collects samples of BB target data prior to returning to the interrupted operation. During normal sounding, a space look or a BB calibration operation is performed when its respective timer has expired and only after the current location has been sounded before the scan system steps to the next location. For both the Imager and Sounder, the in-flight calibration data generated, together with the raw spacecraft sensor data, are transmitted to the SPS at the CDA for use in IR calibration processing. The calibration data, ECAL, space look, and BB measurements are extracted for the Imager and Sounder data and used for calculation of the IR calibration equations. After the data have been extracted, the SPS software verifies the data and performs out-of-limit checks prior to calculating calibration coefficients. This process is performed in the instrument calibration preprocessing tasks. Here the spacecraft calibration data from both the Imager and Sounder undergo statistical quality checks, limit checks, and verification. The good data are then averaged into a single value for each detector or detector/filter combination. The resulting space look and BB calibration data are received from the SPS memory and used by the calibration processing task to compute the needed coefficients. The SPS provides normal mode (Mode 1) calibration for both the Imager and Sounder, along with a number of “extended” modes. In the normal mode, the space look, BB, and BB temperature data resulting from the in-flight calibration sequences are used by the SPS to compute the gain and bias factors necessary for both Imager and Sounder IR calibration. The coefficients of the quadratic calibration terms are extracted from a table of factory-measured coefficients versus instrument temperature. These extended modes compensate for the effects of 1/f noise in the instruments as well as rapid changes in instrument and detector temperatures. Upon completion of the real-time data processing, the SPS formats the calibrated infrared and normalized visible Imager and Sounder data into GOES VARiable (GVAR) processed data streams. The newly computed IR calibration coefficients and the space look and BB statistics generated during their calculation are also included in GVAR for transmission to the OGE Performance Management System (PMS) and the user community. The GVAR processed data consist of data blocks numbered 0-11. The calibrated and normalized Imager data are buffered by the SPS until space look occurs. When a space look occurs, the SPS continues to buffer the Imager data until the end of the current scan line is reached. The SPS then formats these data to be included in the GVAR blocks 0-10. Following the transmissions of blocks 0-10 is the transmission of various block 11s. During Sounder operation, block 11s containing the Sounder sensor data that have been calibrated and normalized by the SPS as well as the raw Sounder data are queued in the SPS for output. Blocks containing the Imager and Sounder calibration statistics generated along with the IR coefficients and the visible Normalization Look-up Tables (NLUT) used in processing the instrument data are also queued for output, ordered according to their priority. The non-real-time calibration/normalization functions include the generation of visible normalization coefficients and the production of short- and long-term history files of data and archive. The visible NLUTs are used in the SPS for the normalization of Imager and Sounder visible data. These NLUTs are generated periodically through an analyst-interactive histogram matching technique performed with the PM. The PM provides the capability to analyze Imager and Sounder visible data statistically through normalization diagnostics. These are performed to monitor the quality of the normalization of Imager and Sounder visible data and to provide a means to determine whether new tables are needed. To perform the diagnostic, a full-resolution normalized image sector, transferred in GVAR, is received by the PM equipment. The operator selects a sector containing cloud cover, ocean, and land masses to maximize the dynamic range represented by the values. The analyst displays and examines the image to see if it has desirable characteristics. From the displayed image, the PM automatically generates histograms and cumulative (integrated) histograms of count levels for each channel and displays them along with associated statistics. The analyst then selects a reference channel, based on long-term stability, and uses it as a basis with which to compare the remaining channels. The data generated as a result of the normalization diagnostic are available for storage, display, and printing. The statistics are stored in a normalization history file maintained in the PM. Selected plots may be displayed simultaneously with the reference detector’s cumulative histogram for comparison. In addition, the delta values that result from the comparison of cumulative histograms can be displayed along with the maximum deviation from the reference detector. The analyst uses these statistics to determine the quality of the normalization performed on the visible image data and to determine whether a normalization update is warranted. To generate new NLUTs, a raw visible data transmission from the SPS is scheduled. This is done by disabling the normalization process in the SPS. The data are received by the PM, where a high-resolution visible image sector is recorded and displayed to verify desirable characteristics. From the displayed raw image data, cumulative histograms are automatically generated and displayed. A reference detector is selected for NLUT generation. New NLUTs are then calculated for the other detectors by matching the target and the reference detector’s cumulative histograms. For each target detector count value, the percentile becomes the new NLUT value for the detector. These new look-up tables are then applied to the raw image sector data, and the sector is displayed for review. Cumulative histogram statistics are generated directly from all the visible detector data stored in a high-resolution image sector. These statistics can then be plotted for further review. Once approved by the analyst, the newly generated NLUTs are sent to Orbit and Altitude Tracking Systme (OATS) and transmitted to the SPS, through GOES I-M Telemetry and Command System (GIMTACS), for use. The level-1 GOES-8 images have not had any calibration applied. Information on calibration procedures can be found at http://www.nnic.noaa.gov/SOCC/SOCC_Home.html. 4.2.1 Specifications The level-1 GOES-8 images have not had any calibration applied. Information on calibration procedures can be found at http://www.nnic.noaa.gov/SOCC/SOCC_Home.html. 5. Data Acquisition Methods The BOREAS level-1 GOES-8 images used in the level-1a product creation were obtained by Dr. Eric Smith at FSU and supplied to BORIS. The data were acquired using the FSU Direct Readout Ground System located in Tallahassee, FL, starting on 14-Jul-1995 and continuing through 23-Oct-1996. 6. Observations 6.1 Data Notes None. 6.2 Field Notes Not applicable. 7. Data Description 7.1 Spatial Characteristics The scanning system consists of a mirror that is stepped mechanically to provide north to south viewing, while the rotation of the GOES satellite provides west to east scanning. The mirror is stepped following each west to east scan. A sequence of 1,821 scans over 18.21 minutes is performed to provide a "full disk" view from just beyond the northern Earth horizon to just beyond the southern Earth horizon. Based on the level-1 GOES-8 images, the BOREAS level-1a GOES-8 images essentially cover the entire 1,000-km by 1,000-km BOREAS region. This contains the Northern Study Area (NSA), the Southern Study Area (SSA), the transect region between the SSA and NSA, and some surrounding area. 7.1.1 Spatial Coverage The North American Datum of 1983 (NAD83) corner coordinates of the BOREAS region are: Latitude Longitude -------- --------- Northwest 58.979 N 111.000 W Northeast 58.844 N 93.502 W Southwest 51.000 N 111.000 W Southeast 50.089 N 96.969 W 7.1.1.1 Spatial Coverage for GOES-8 Three-Band Data (14-Jul-1995 to 29-Jul-1995) The corner latitude and longitude coordinates for the GOES-8 three-band data from 14-Jul-1995 to 29-Jul-1995 are shown in the following tables, where the value of -999.000 indicates that the sensor was viewing past the Earth horizon into outer space. Visible Visible Latitude Longitude ---------- ----------- Northwest -999.000 -999.000 Northeast 73.054 -125.439 Southwest 45.774 -100.785 Southeast 45.323 -91.759 IR4 IR4 Latitude Longitude ---------- ----------- Northwest -999.000 -999.000 Northeast 73.117 -125.818 Southwest 45.823 -100.814 Southeast 45.372 -91.801 IR3 IR3 Latitude Longitude ---------- ----------- Northwest -999.999 -999.000 Northeast -999.000 -999.000 Southwest 38.350 -97.267 Southeast 38.067 -89.634 7.1.1.2 Spatial Coverage for GOES-8 Three-Band Data (01-Aug-1995 to 21-Sep-1995) The corner latitude and longitude coordinates for the GOES-8 three-band data from 01-Aug-1995 to 21-Sep-1995 are shown in the following tables, where the value of -999.000 indicates that the sensor was viewing past the Earth horizon into outer space. Visible Visible Latitude Longitude ---------- ----------- Northwest -999.000 -999.000 Northeast 65.462 -98.345 Southwest 47.809 -107.113 Southeast 46.773 -87.996 IR4 IR4 Latitude Longitude ---------- ----------- Northwest -999.000 -999.000 Northeast 70.047 -104.477 Southwest 46.147 -105.839 Southeast 45.222 -87.592 IR3 IR3 Latitude Longitude ---------- ----------- Northwest -999.999 -999.000 Northeast -999.000 -999.000 Southwest 38.546 -101.379 Southeast 37.971 -85.990 7.1.1.3 Spatial Coverage for GOES-8 Five-Band Data (12-Feb-1996 to 30-Apr-1996) The corner latitude and longitude coordinates for the GOES-8 five-band data from 01-Jan-1996 to 30-Apr-1996 are shown in the following tables, where the value of -999.000 indicates that the sensor was viewing past the Earth horizon into outer space. Visible Visible Latitude Longitude ---------- ----------- Northwest -999.000 -999.000 Northeast 65.478 -98.187 Southwest 47.808 -107.162 Southeast 46.793 -88.137 IR2 IR2 Latitude Longitude ---------- ----------- Northwest -999.000 -999.000 Northeast 70.040 -104.087 Southwest 46.150 -105.913 Southeast 45.243 -87.745 IR3 IR3 Latitude Longitude ---------- ----------- Northwest -999.000 -999.000 Northeast -999.000 -999.000 Southwest 38.555 -101.528 Southeast 37.991 -86.186 IR4 IR4 Latitude Longitude ---------- ----------- Northwest -999.000 -999.000 Northeast 70.040 -104.087 Southwest 46.150 -105.913 Southeast 45.243 -87.745 IR5 IR5 Latitude Longitude ---------- ----------- Northwest -999.000 -999.000 Northeast 70.040 -104.087 Southwest 46.150 -105.913 Southeast 45.243 -87.745 7.1.1.4 Spatial Coverage for GOES-8 Five-Band Data (01-May-1996 to 03-Oct-1996) The corner latitude and longitude coordinates for the GOES-8 five-band data from 01-May-1996 to 03-Oct-1996 are shown in the following tables, where the value of -999.000 indicates that the sensor was viewing past the Earth horizon into outer space. Visible Visible Latitude Longitude ---------- ----------- Northwest -999.000 -999.000 Northeast 65.354 -96.453 Southwest 47.885 -108.274 Southeast 46.777 -87.385 IR2 IR2 Latitude Longitude ---------- ----------- Northwest -999.000 -999.000 Northeast 69.565 -99.365 Southwest 46.293 -108.022 Southeast 45.208 -86.229 IR3 IR3 Latitude Longitude ---------- ----------- Northwest -999.000 -999.000 Northeast -999.000 -999.000 Southwest 38.644 -103.275 Southeast 37.971 -84.902 IR4 IR4 Latitude Longitude ---------- ----------- Northwest -999.000 -999.000 Northeast 69.565 -99.365 Southwest 46.293 -108.022 Southeast 45.208 -86.229 IR5 IR5 Latitude Longitude ---------- ----------- Northwest -999.000 -999.000 Northeast 69.565 -99.365 Southwest 46.293 -108.022 Southeast 45.208 -86.229 7.1.2 Spatial Coverage Map Not available at this time. 7.1.3 Spatial Resolution The spatial resolution of each pixel is dependent on the off-nadir scan angle of the sensor and increases from nadir to the scanning extremes. The satellite subpoint resolution of the various channels is: North/South East/West ----------- --------- Visible 1 km 1 km IR2 4 km 4 km IR3 8 km 4 km IR4 4 km 4 km IR5 4 km 4 km The spatial dimensions of each pixel can be calculated from the provided latitude and longitude coordinate information (see Section 8.2). 7.1.4 Projection The BOREAS level-1a GOES-8 images are stored in the same GOES ‘perfect’ projection as the level-1 images. The ‘perfect’ projection indicates that the satellite movement between temporal acquisitions has been removed so the images are aligned spatially. Detailed information about the projection is not currently available. 7.1.5 Grid Description Not available at this revision. 7.2 Temporal Characteristics 7.2.1 Temporal Coverage From 14-Jul-1995 through 21-Sep-1995, partial to complete data are available for 68 of the possible 70 days. From 12-Feb-1996 to 03-Oct-1996, partial to complete data are available for 201 of the possible 235 days. One large period of missing data is 19-Jun-1996 to 03-Jul-1996, which resulted from hardware problems at FSU. 7.2.2 Temporal Coverage Map Not available. 7.2.3 Temporal Resolution From 14-Jul-1995 to 21-Sep-1995, visible, IR4, and IR3 images were acquired at 15 and 45 minutes after the hour, 24 hours a day. During 1996, the visible, IR2, IR3, IR4, and IR5 images were acquired at 15 and 45 minutes after the hour, 24 hours a day. 7.3 Data Characteristics The image inventory data file characteristics are defined in the companion data definition file (goes81a.def). 7.3.1 Parameter/Variable The parameter contained in the image data files is: Digital Number (DN) 7.3.2 Variable Description/Definition The quantized digital number derived by the GOES-8 scanning system for the respective channel. 7.3.3 Unit of Measurement For the image data files: Digital Number (DN) - counts 7.3.4 Data Source The level-1a GOES-8 image bands were processed from the level-1 data. The raw data were received, processed and subset, and sent to BORIS by personnel within the Department of Meteorology at Florida State University (FSU). 7.3.5 Data Range In 1995, the maximum range of digital numbers in each GOES-8 image band is limited from 0 (zero) to 255 so that the values can be stored in a single 8-bit (byte) field. For the 1996 data, each digital number can range from 0 to 1023 and occupies 10-bits of the 16-bit field used to store the value. 7.4 Sample Data Record A sample data record for the level-1a GOES-8 images is not available here. 8. Data Organization 8.1 Data Granularity The smallest unit of data for level-1a GOES-8 image data is a single level-1a image. This includes extracted descriptive information in ASCII form, all visible and infrared images collected within a given day of 0015 to 2345 Greenwich Mean Time (GMT), and the reference latitude and longitude files for the images. All of the BOREAS RSS-14 Level-1a GOES-8 Visible, IR, and Water-Vapor Images data are contained in one dataset. 8.2 Data Format(s) The image inventory data file contains numerical and character fields of varying length separated by commas. The character fields are enclosed with a single apostrophe marks. There are no spaces between the fields. The image inventory data file format is defined in the companion data definition file (goes81a.def). One day of level-1a GOES-8 data from 1995 is contained in four physical tape files: 1) ASCII header file 2) Visible image data file 3) IR4 image data file 4) IR3 image data file One day of level-1a GOES-8 data from 1996 is contained in six physical tape files: 1) ASCII header file 2) Visible image data file 3) IR2 image data file 4) IR4 image data file 5) IR5 image data file 6) IR3 image data file The ASCII header file contains 80-byte ASCII records that describe the level-1a product, provide summary and detailed information about the ‘good’ and zero- filled images for the given day, and contain descriptive header information extracted from the visible, IR2, IR4, IR5, and IR3 images. Note that the data set consisted of three bands in 1995 and changed to five bands in 1996. In addition, the size of the images also changed during 1995 and 1996. The following table describes the changes, and the following paragraphs give details of the binary data formats. Number Number Number Start Date End Date Band of Bits of Lines of Pixels ----------- ---------- -------- ------- -------- --------- 14-Jul-1995 29-July-1995 visible 8 1024 1024 IR4 8 256 256 IR3 8 256 256 02-Aug-1995 21-Sep-1995 visible 8 824 2048 IR4 8 256 512 IR3 8 256 512 12-Feb-1996 30-Apr-1996 visible 16 824 2048 IR2 16 256 512 IR4 16 256 512 IR5 16 256 512 IR3 16 256 512 02-May-1996 03-Oct-1996 visible 16 824 2248 IR2 16 256 612 IR4 16 256 612 IR5 16 256 612 IR3 16 256 612 8.2.1 GOES-8 Three-Band Data Format (14-Jul-1995 to 29-Jul-1995) One GOES-8 visible image contains 1,024 unsigned 8-bit counts (i.e., values from 0 to 255) stored in unsigned 8-bit (1-byte) values in each of 1,024 lines. A full day of visible image data consists of 48 half-hourly images acquired every 30 minutes starting at 0015 GMT. The visible image data file on tape contains a full day of image data. Each visible image data file on tape consists of 3,072 records of 16,384 bytes. Each tape record contains 16 visible image lines of 1024 bytes (16,384 = 16 * 1024). Therefore, one visible image is contained in 64 tape records (64 = 1024 / 16). The sequence of image times in the file is 0015, 0045, 0115, 0145, ... 2315, 2345 GMT. If a given half-hourly visible image was not acquired, the level-1a product contains a set of 1,024 records containing 1,024 values of 0. The presence of these zero-filled images is shown in the ASCII header file. Placing these zero-filled images in the file for missing acquisitions keeps the images from the same time in the same physical location in each daily file. One GOES-8 IR4 image contains 256 unsigned 8-bit counts (i.e., values from 0 to 255) stored in unsigned 8-bit (1-byte) values in each of 256 lines. A full day of IR4 image data consists of 48 half-hourly images. The IR4 image data file on tape contains a full day of image data. Each IR4 image data file on tape consists of 192 records of 16,384 bytes. Each tape record contains 64 IR image lines of 256 bytes (16,384 = 64 * 256). Therefore, one IR4 image is contained in four tape records. The sequence of image times in the file is 0015, 0045, 0115, 0145, 2315, 2345 GMT. If a given half-hourly IR4 image was not acquired, the level-1a product contains a set of 256 records containing 256 values of 0. The presence of these zero-filled images is shown in the ASCII header file. Placing these zero-filled images in the file for missing acquisitions keeps the images from the same time in the same physical location in each daily file. One GOES-8 IR3 image contains 256 unsigned 8-bit counts (i.e., values from 0 to 255) stored in unsigned 8-bit (1-byte) values in each of 256 lines. A full day of IR3 image data consists of 48 half-hourly images. The IR3 image data file on tape contains a full day of image data. Each IR3 image data file on tape consists of 192 records of 16,384 bytes. Each tape record contains 64 water- vapor image lines of 256 bytes (16,384 = 64 * 256). Therefore, one IR3 image is contained in four tape records. The sequence of image times in the file is 0015, 0045, 0115, 0145, ... 2315, 2345 GMT. If a given half-hourly IR3 image was not acquired, the level-1a product contains a set of 256 records containing 256 values of 0. The presence of these zero-filled images is shown in the ASCII header file. Placing these zero-filled images in the file for missing acquisitions keeps the images from the same time in the same physical location in each daily file. The set of six reference latitude and longitude files for the GOES-8 images collected in 1995 is appended as the last six files of the delivered tape. The six files consist of a pair of latitude and longitude files for each of the visible, IR2, and IR3 image types. The reference latitude and longitude files for the visible images each consist of 1,024 records of 4,096 bytes. Each record of 4,096 bytes contains 1,024 signed 32-bit (4-byte) integer latitude or longitude values. The bytes of each 32-bit value are stored as low-order byte first. The unit of each latitude and longitude value is thousandths of a degree. To get the original decimal degree values, divide each value by 1000. The reference latitude and longitude files for the IR4 images each consist of 256 records of 1,024 bytes. Each record of 1,024 bytes contains 256 signed 32- bit (4-byte) integer latitude or longitude values. The bytes of each 32-bit value are stored as low-order byte first. The unit of each latitude and longitude value is thousandths of a degree. To get the original decimal degree values, divide each value by 1000. The reference latitude and longitude files for the IR3 images each consist of 256 records of 1,024 bytes. Each record of 1,024 bytes contains 256 signed 32- bit (4-byte) integer latitude or longitude values. The bytes of each 32-bit value are stored as low-order byte first. The unit of each latitude and longitude value is thousandths of a degree. To get the original decimal degree values, divide each value by 1000. 8.2.2 GOES-8 Three-Band Data Format (01-Aug-1995 to 21-Sep-1995) One GOES-8 visible image contains 2,048 unsigned 8-bit counts (i.e., values from 0 to 255) stored in unsigned 8-bit (1-byte) values in each of 824 lines. A full day of visible image data consists of 48 half-hourly images acquired every 30 minutes starting at 0015 GMT. The visible image data file on tape contains a full day of image data. Each visible image data file on tape consists of 4,944 records of 16,384 bytes. Each tape record contains eight visible image lines of 2,048 bytes (16,384 = 8 * 2048). Therefore, one visible image is contained in 103 tape records (103 = 824 / 8). The sequence of image times in the file is 0015, 0045, 0115, 0145, ... 2315, 2345 GMT. If a given half-hourly visible image was not acquired, the level-1a product contains a set of 824 records containing 2,048 values of 0. The presence of these zero-filled images is shown in the ASCII header file. Placing these zero-filled images in the file for missing acquisitions keeps the images from the same time in the same physical location in each daily file. One GOES-8 IR4 image contains 512 unsigned 8-bit counts (i.e., values from 0 to 255) stored in unsigned 8-bit (1-byte) values in each of 256 lines. A full day of IR4 image data consists of 48 half-hourly images. The IR4 image data file on tape contains a full day of image data. Each IR4 image data file on tape consists of 384 records of 16,384 bytes. Each tape record contains 32 IR4 image lines of 256 bytes (16,384 = 32 * 512). Therefore, one IR4 image is contained in eight tape records (8 = 256 / 32). The sequence of image times in the file is 0015, 0045, 0115, 0145, ... 2315, 2345 GMT. If a given half-hourly IR4 image was not acquired, the level-1a product contains a set of 256 records containing 512 values of 0. The presence of these zero-filled images is shown in the ASCII header file. Placing these zero-filled images in the file for missing acquisitions keeps the images from the same time in the same physical location in each daily file. One GOES-8 IR3 image contains 512 unsigned 8-bit counts (i.e., values from 0 to 255) stored in unsigned 8-bit (1-byte) values in each of 256 lines. A full day of IR3 image data consists of 48 half-hourly images. The IR3 image data file on tape contains a full day of image data. Each IR3 image data file on tape consists of 384 records of 16,384 bytes. Each tape record contains 32 IR3 image lines of 512 bytes (16,384 = 32 * 512). Therefore, one IR3 image is contained in eight tape records(8 = 256 / 32). The sequence of image times in the file is 0015, 0045, 0115, 0145, 2315, 2345 GMT. If a given half-hourly IR3 image was not acquired, the level-1a product contains a set of 256 records containing 512 values of 0. The presence of these zero-filled images is shown in the ASCII header file. Placing these zero-filled images in the file for missing acquisitions keeps the images from the same time in the same physical location in each daily file. The set of six reference latitude and longitude files for the GOES-8 images collected in 1995 is appended as the last six files of the delivered tape. The six files consist of a pair of latitude and longitude files for each of the visible, IR4, and IR3 image types. The reference latitude and longitude files for the visible images each consist of 824 records of 8,192 bytes. Each record of 8,192 bytes contains 2,048 signed 32-bit (4-byte) integer latitude or longitude values. The bytes of each 32-bit value are stored as low-order byte first. The unit of each latitude and longitude value is thousandths of a degree. To get the original decimal degree values, divide each value by 1000. The reference latitude and longitude files for the IR4 images each consist of 256 records of 2,048 bytes. Each record of 2,048 bytes contains 512 signed 32- bit (4-byte) integer latitude or longitude values. The bytes of each 32-bit value are stored as low-order byte first. The unit of each latitude and longitude value is thousandths of a degree. To get the original decimal degree values, divide each value by 1000. The reference latitude and longitude files for the IR3 images each consist of 256 records of 2,048 bytes. Each record of 2,048 bytes contains 512 signed 32- bit (4-byte) integer latitude or longitude values. The bytes of each 32-bit value are stored as low-order byte first. The unit of each latitude and longitude value is thousandths of a degree. To get the original decimal degree values, divide each value by 1000. 8.2.3 GOES-8 Five-Band Data Format (12-Feb-1996 to 30-Apr-1996) One GOES-8 visible image contains 2,048 10-bit counts (i.e., values from 0 to 1023) stored in 16-bit (2-byte) values (low-order byte first) in each of 824 lines. A full day of visible image data consists of 48 half-hourly images acquired every 30 minutes starting at 0015 GMT. The visible image data file on tape contains a full day of image data. Each visible image data file on tape consists of 9,888 records of 16,384 bytes. Each tape record contains four visible image lines of 4,096 bytes (16,384 = 4 * 4096). Therefore, one visible image is contained in 206 tape records (206 = 824 / 4). The sequence of image times in the file is 0015, 0045, 0115, 0145, ... 2315, 2345 GMT. If a given half-hourly visible image was not acquired, the level-1a product contains a set of 824 records containing 2,048 16-bit values of 0. The presence of these zero- filled images is shown in the ASCII header file. Placing these zero-filled images in the file for missing acquisitions keeps the images from the same time in the same physical location in each daily file. One GOES-8 IR2 image contains 512 10-bit counts (i.e., values from 0 to 1023) stored in 16-bit (2-byte) values (low order byte first) in each of 256 lines. A full day of IR2 image data consists of 48 half-hourly images. The IR2 image data file on tape contains a full day of image data. Each IR2 image data file on tape consists of 768 records of 16,384 bytes. Each tape record contains 16 IR2 image lines of 1,024 bytes (16,384 = 16 * 1024). Therefore, one IR2 image is contained in 16 tape records (16 = 256 / 16). The sequence of image times in the file is 0015, 0045, 0115, 0145, ... 2315, 2345 GMT. If a given half-hourly IR2 image was not acquired, the level-1a product contains a set of 256 records containing 512 16-bit values of 0. The presence of these zero-filled images is shown in the ASCII header file. Placing these zero-filled images in the file for missing acquisitions keeps the images from the same time in the same physical location in each daily file. One GOES-8 IR4 image contains 512 10-bit counts (i.e., values from 0 to 1023) stored in 16-bit (2-byte) values (low-order byte first) in each of 256 lines. A full day of IR4 image data consists of 48 half-hourly images. The IR4 image data file on tape contains a full day of image data. Each IR4 image data file on tape consists of 768 records of 16,384 bytes. Each tape record contains 16 IR4 image lines of 1,024 bytes (16,384 = 16 * 1024). Therefore, one IR4 image is contained in 16 tape records (16 = 256 / 16). The sequence of image times in the file is 0015, 0045, 0115, 0145, ... 2315, 2345 GMT. If a given half-hourly IR4 image was not acquired, the level-1a product contains a set of 256 records containing 512 16-bit values of 0. The presence of these zero-filled images is shown in the ASCII header file. Placing these zero-filled images in the file for missing acquisitions keeps the images from the same time in the same physical location in each daily file. One GOES-8 IR5 image contains 512 10-bit counts (i.e., values from 0 to 1023) stored in 16-bit (2-byte) values (low-order byte first) in each of 256 lines. A full day of IR5 image data consists of 48 half-hourly images. The IR5 image data file on tape contains a full day of image data. Each IR5 image data file on tape consists of 768 records of 16,384 bytes. Each tape record contains 16 IR5 image lines of 1,024 bytes (16,384 = 16 * 1024). Therefore, one IR5 image is contained in 16 tape records (16 = 256 / 16). The sequence of image times in the file is 0015, 0045, 0115, 0145, ... 2315, 2345 GMT. If a given half-hourly IR5 image was not acquired, the level-1a product contains a set of 256 records containing 512 16-bit values of 0. The presence of these zero-filled images is shown in the ASCII header file. Placing these zero-filled images in the file for missing acquisitions keeps the images from the same time in the same physical location in each daily file. One GOES-8 IR3 image contains 512 10-bit counts (i.e., values from 0 to 1023) stored in 16-bit (2-byte) values (low-order byte first) in each of 256 lines. A full day of IR3 image data consists of 48 half-hourly images. The IR3 image data file on tape contains a full day of image data. Each IR3 image data file on tape consists of 768 records of 16,384 bytes. Each tape record contains 16 IR3 image lines of 1,024 bytes (16,384 = 16 * 1024). Therefore, one IR3 image is contained in 16 tape records (16 = 256 / 16). The sequence of image times in the file is 0015, 0045, 0115, 0145, ... 2315, 2345 GMT. If a given half-hourly IR3 image was not acquired, the level-1a product contains a set of 256 records containing 512 16-bit values of 0. The presence of these zero-filled images is shown in the ASCII header file. Placing these zero-filled images in the file for missing acquisitions keeps the images from the same time in the same physical location in each daily file. The set of 10 reference latitude and longitude files for the GOES-8 images collected in 1995 is appended as the last 10 files of the delivered tape. The 10 files consist of a pair of latitude and longitude files for each of the visible, IR2, IR3, IR4, and IR5 image types. The reference latitude and longitude files for the visible images each consist of 824 records of 8,192 bytes. Each record of 8,192 bytes contains 2,048 signed 32-bit (4-byte) integer latitude or longitude values. The bytes of each 32-bit value are stored as low-order byte first. The unit of each latitude and longitude value is thousandths of a degree. To get the original decimal degree values, divide each value by 1000. The reference latitude and longitude files for the IR2 images each consist of 256 records of 2,048 bytes. Each record of 2048 bytes contains 512 signed 32- bit (4-byte) integer latitude or longitude values. The bytes of each 32-bit value are stored as low-order byte first. The unit of each latitude and longitude value is thousandths of a degree. To get the original decimal degree values, divide each value by 1000. The reference latitude and longitude files for the IR4 images each consist of 256 records of 2,048 bytes. Each record of 2,048 bytes contains 512 signed 32- bit (4-byte) integer latitude or longitude values. The bytes of each 32-bit value are stored as low-order byte first. The unit of each latitude and longitude value is thousandths of a degree. To get the original decimal degree values, divide each value by 1000. The reference latitude and longitude files for the IR5 images each consist of 256 records of 2,048 bytes. Each record of 2,048 bytes contains 512 signed 32- bit (4-byte) integer latitude or longitude values. The bytes of each 32-bit value are stored as low-order byte first. The unit of each latitude and longitude value is thousandths of a degree. To get the original decimal degree values, divide each value by 1000. The reference latitude and longitude files for the IR3 images each consist of 256 records of 2,048 bytes. Each record of 2,048 bytes contains 512 signed 32- bit (4-byte) integer latitude or longitude values. The bytes of each 32-bit value are stored as low-order byte first. The unit of each latitude and longitude value is thousandths of a degree. To get the original decimal degree values, divide each value by 1000. 8.2.4 GOES-8 Five-Band Data Format (02-May-1996 to 03-Oct-1996) One GOES-8 visible image contains 2,248 10-bit counts (i.e., values from 0 to 1023) stored in 16-bit (2-byte) values (low-order byte first) in each of 824 lines. A full day of visible image data consists of 48 half-hourly images acquired every 30 minutes starting at 0015 GMT. The visible image data file on tape contains a full day of image data. Each visible image data file on tape consists of 9,888 records of 17,984 bytes. Each tape record contains four visible image lines of 4,496 bytes (17,984 = 4 * 4496). Therefore, one visible image is contained in 206 tape records (206 = 824 / 4). The sequence of image times in the file is 0015, 0045, 0115, 0145, ... 2315, 2345 GMT. If a given half-hourly visible image was not acquired, the level-1a product contains a set of 824 records containing 2,248 16-bit values of 0. The presence of these zero- filled images is shown in the ASCII header file. Placing these zero-filled images in the file for missing acquisitions keeps the images from the same time in the same physical location in each daily file. One GOES-8 IR2 image contains 612 10-bit counts (i.e., values from 0 to 1023) stored in 16-bit (2-byte) values (low-order byte first) in each of 256 lines. A full day of IR2 image data consists of 48 half-hourly images. The IR2 image data file on tape contains a full day of image data. Each IR2 image data file on tape consists of 768 records of 19,584 bytes. Each tape record contains 16 IR2 image lines of 1,224 bytes (19,584 = 16 * 1224). Therefore, one IR2 image is contained in 16 tape records (16 = 256 / 16). The sequence of image times in the file is 0015, 0045, 0115, 0145, ... 2315, 2345 GMT. If a given half-hourly IR2 image was not acquired, the level-1a product contains a set of 256 records containing 612 16-bit values of 0. The presence of these zero-filled images is shown in the ASCII header file. Placing these zero-filled images in the file for missing acquisitions keeps the images from the same time in the same physical location in each daily file. One GOES-8 IR4 image contains 612 10-bit counts (i.e., values from 0 to 1023) stored in 16-bit (2-byte) values (low-order byte first) in each of 256 lines. A full day of IR4 image data consists of 48 half-hourly images. The IR4 image data file on tape contains a full day of image data. Each IR4 image data file on tape consists of 768 records of 19,584 bytes. Each tape record contains 16 IR4 image lines of 1,224 bytes (19,584 = 16 * 1224). Therefore, one IR4 image is contained in 16 tape records (16 = 256 / 16). The sequence of image times in the file is 0015, 0045, 0115, 0145, ... 2315, 2345 GMT. If a given half-hourly IR4 image was not acquired, the level-1a product contains a set of 256 records containing 612 16-bit values of 0. The presence of these zero-filled images is shown in the ASCII header file. Placing these zero-filled images in the file for missing acquisitions keeps the images from the same time in the same physical location in each daily file. One GOES-8 IR5 image contains 612 10-bit counts (i.e., values from 0 to 1023) stored in 16-bit (2-byte) values (low-order byte first) in each of 256 lines. A full day of IR5 image data consists of 48 half-hourly images. The IR5 image data file on tape contains a full day of image data. Each IR5 image data file on tape consists of 768 records of 19,584 bytes. Each tape record contains 16 IR5 image lines of 1,224 bytes (19,584 = 16 * 1224). Therefore, one IR5 image is contained in 16 tape records (16 = 256 / 16). The sequence of image times in the file is 0015, 0045, 0115, 0145, ... 2315, 2345 GMT. If a given half-hourly IR5 image was not acquired, the level-1a product contains a set of 256 records containing 612 16-bit values of 0. The presence of these zero-filled images is shown in the ASCII header file. Placing these zero-filled images in the file for missing acquisitions keeps the images from the same time in the same physical location in each daily file. One GOES-8 IR3 image contains 612 10-bit counts (i.e., values from 0 to 1023) stored in 16-bit (2-byte) values (low-order byte first) in each of 256 lines. A full day of IR3 image data consists of 48 half-hourly images. The IR3 image data file on tape contains a full day of image data. Each IR3 image data file on tape consists of 768 records of 19,584 bytes. Each tape record contains 16 IR3 image lines of 1,224 bytes (19,584 = 16 * 1224). Therefore, one IR3 image is contained in 16 tape records (16 = 256 / 16). The sequence of image times in the file is 0015, 0045, 0115, 0145, ... 2315, 2345 GMT. If a given half-hourly IR3 image was not acquired, the level-1a product contains a set of 256 records containing 612 16-bit values of 0. The presence of these zero-filled images is shown in the ASCII header file. Placing these zero-filled images in the file for missing acquisitions keeps the images from the same time in the same physical location in each daily file. The set of 10 reference latitude and longitude files for the GOES-8 images collected in 1995 are appended as the last 6 files of the delivered tape. The 10 files consist of a pair of latitude and longitude files for each of the visible, IR2, IR3, IR4, and IR5 image types. The reference latitude and longitude files for the visible images each consist of 824 records of 8,992 bytes. Each record of 8,992 bytes contains 2,248 signed 32-bit (4-byte) integer latitude or longitude values. The bytes of each 32-bit value are stored as low-order byte first. The unit of each latitude and longitude value is thousandths of a degree. To get the original decimal degree values, divide each value by 1000. The reference latitude and longitude files for the IR2 images each consist of 256 records of 2,448 bytes. Each record of 2,448 bytes contains 612 signed 32- bit (4-byte) integer latitude or longitude values. The bytes of each 32-bit value are stored as low-order byte first. The unit of each latitude and longitude value is thousandths of a degree. To get the original decimal degree values, divide each value by 1000. The reference latitude and longitude files for the IR4 images each consist of 256 records of 2,448 bytes. Each record of 2,448 bytes contains 612 signed 32- bit (4-byte) integer latitude or longitude values. The bytes of each 32-bit value are stored as low-order byte first. The unit of each latitude and longitude value is thousandths of a degree. To get the original decimal degree values, divide each value by 1000. The reference latitude and longitude files for the IR5 images each consist of 256 records of 2,448 bytes. Each record of 2,448 bytes contains 612 signed 32- bit (4-byte) integer latitude or longitude values. The bytes of each 32-bit value are stored as low-order byte first. The unit of each latitude and longitude value is thousandths of a degree. To get the original decimal degree values, divide each value by 1000. The reference latitude and longitude files for the IR3 images each consist of 256 records of 2,448 bytes. Each record of 2,448 bytes contains 612 signed 32- bit (4-byte) integer latitude or longitude values. The bytes of each 32-bit value are stored as low-order byte first. The unit of each latitude and longitude value is thousandths of a degree. To get the original decimal degree values, divide each value by 1000. 9. Data Manipulations 9.1 Formulae 9.1.1 Derivation Techniques and Algorithms See Section 9.2.1. 9.2 Data Processing Sequence 9.2.1 Processing Steps Using the BOREAS level-1 GOES-8 product as input, the data were processed to level-1a products. The processing included: 1) Separating the header record and image data records contained in level-1 GOES image files. 2) Unpacking the header record information for use in processing control and output to the level-1a ASCII header file. 3) Filling missing images with zero-filled records. 4) Writing the four files for each day's data to disk. 5) Copying a 4-week period of daily files to tape. 6) Appending the reference latitude and longitude files to the tapes. 9.2.2 Processing Changes None. 9.3 Calculations See Section 9.1.1. 9.3.1 Special Corrections/Adjustments See Section 9.1.1. 9.3.2 Calculated Variables See Section 9.1.1. 9.4 Graphs and Plots None. 10. Errors 10.1 Sources of Error The level-1a processing depended on the times provided in the level-1 GOES-8 image header records. If the time in the level-1 GOES-8 header record is incorrect, it would result in the image being placed in the incorrect sequence in the level-1a product. Any mixup between visible, IR2, IR3, IR4, and IR5, images could not occur because of the file type and size checking done during processing. 10.2 Quality Assessment 10.2.1 Data Validation by Source Whatever the processing level, the geometric quality of the image depends on the accuracy of the viewing geometry. Spectral errors could arise from image-wide signal-to-noise ratio, saturation, cross-talk, spikes, or response normalization caused by a change in gain. 10.2.2 Confidence Level/Accuracy Judgment Not available at this revision. 10.2.3 Measurement Error for Parameters Not available at this revision. 10.2.4 Additional Quality Assessments The level-1 GOES-8 images used to create the level-1a products were visually scanned for bad periods by FSU staff. 10.2.5 Data Verification by Data Center See Section 9.1.1. 11. Notes 11.1 Limitations of the Data Not available at this revision. 11.2 Known Problems with the Data To date, the following discrepancies/problems have been noted in the data: 1) Occasional reception problems, especially during NOAA rapid scan operations, may result in some images being truncated at the northern edge (top of image). When this occurs, visual review by FSU staff has ensured that the BOREAS areas of interest are still present. 2) The ASCII header files for the level-1a GOES-8 images up to day 97 of 1996 (06-Apr-19960) contain the incorrect relative file numbers for the IR2, IR4, IR5, and IR3 files. An incorrect header file shows the IR2, IR4, IR5, and IR3 data to be contained in files 03, 03, 03, and 04. The actual file numbers are 3, 4, 5, and 6. This was fixed for data after day 97. 11.3 Usage Guidance None. 11.4 Other Relevant Information None. 12. Application of the Data Set The GOES-8 images provide a high temporal resolution data set that would be useful for monitoring radiation loading and cloud development and movement over the BOREAS region. 13. Future Modifications and Plans None. 14. Software 14.1 Software Description BORIS staff developed software and command procedures for: 1) Extracting header information from level GOES-8 images on tape and writing it to ASCII files on disk for initial quality checking. 2) Processing the files of level-1 GOES-8 data for a given day to a level-1a product. 3) Writing the level-1a GOES-8 image files from tape to disk. 4) Creating binary files of scaled latitude and longitude coordinates from the original ASCII files. 5) Appending the latitude and longitude files to tape. The software mentioned above is written in the C language and is operational on VAX 6410, MicroVAX, and VAXstation systems at GSFC. The primary dependencies in the software are the tape input/output (I/O) library and the Oracle data base utility routines. 14.2 Software Access All of the described software is available upon request (See Section 15). BORIS staff would appreciate knowing of any problems discovered with the software, but cannot guarantee that they will be fixed. 15. Data Access 15.1 Contact Information Ms. Beth Nelson BOREAS Data Manager NASA GSFC Greenbelt, MD (301) 286-4005 (301) 286-0239 (fax) beth@ltpmail.gsfc.nasa.gov 15.2 Data Center Identification See 15.1. 15.3 Procedures for Obtaining Data Users may place requests by telephone, electronic mail, or fax. 15.4 Data Center Status/Plans The level-1a GOES-8 data are available from the Earth Observing System Data and Information System (EOSDIS), Oak Ridge National Laboratory (ORNL), Distributed Active Archive Center (DAAC). The BOREAS contact at ORNL is: ORNL DAAC User Services Oak Ridge National Laboratory Oak Ridge, TN (423) 241-3952 ornldaac@ornl.gov ornl@eos.nasa.gov 16. Output Products and Availability 16.1 Tape Products The level-1a GOES-8 data can be made available on 8-mm tapes or Digital Archive Tapes (DAT). 16.2 Film Products None. 16.3 Other Products None. 17. References 17.1 Platform/Sensor/Instrument/Data Processing Documentation Bobotek, A., A.S. Hechtman, R.J. Komajoa, and P.G. Woolner. July 1995. GOES I-M System description. MITRE Corporation. Rossow, W.B., C.L. Brest, and M. Roiter. 1996. International Satellite Cloud Climatology Project (ISCCP) New Radiance Calibrations. WMO/TD-No. 736. World Meteorological Organization. Rossow, W.B., C.L. Brest, and M.D. Roiter. 1995. International Satellite Cloud Climatology Project (ISCCP): Update of radiance calibration report. Technical Document, World Climate Research Programme (ICSU and WMO), Geneva, Switzerland, 76 pp. Rossow, W.B., Y. Desormeaux, C. L. Brest, and A. Walker. 1992. International Satellite Cloud Climatology Project (ISCCP): Radiance calibration report. WMO/Technical Document No. 520, World Climate Research Programme and World Meteorological Organization (ICSU and WMO), Geneva, Switzerland, 104 pp. 17.2 Journal Articles and Study Reports Sellers, P. and F. Hall. 1994. Boreal Ecosystem-Atmosphere Study: Experiment Plan. Version 1994-3.0, NASA BOREAS Report (EXPLAN 94). Sellers, P. and F. Hall. 1996. Boreal Ecosystem-Atmosphere Study: Experiment Plan. Version 1996-2.0, NASA BOREAS Report (EXPLAN 96). Sellers, P., F. Hall, and K.F. Huemmrich. 1996. Boreal Ecosystem-Atmosphere Study: 1994 Operations. NASA BOREAS Report (OPS DOC 94). Sellers, P., F. Hall, and K.F. Huemmrich. 1997. Boreal Ecosystem-Atmosphere Study: 1996 Operations. NASA BOREAS Report (OPS DOC 96). Sellers, P., F. Hall, H. Margolis, B. Kelly, D. Baldocchi, G. den Hartog, J. Cihlar, M.G. Ryan, B. Goodison, P. Crill, K.J. Ranson, D. Lettenmaier, and D.E. Wickland. 1995. The boreal ecosystem-atmosphere study (BOREAS): an overview and early results from the 1994 field year. Bulletin of the American Meteorological Society. 76(9):1549-1577. 17.3 Archive/DBMS Usage Documentation None. 18. Glossary of Terms None. 19. List of Acronyms AOCS - Attitude and Orbit Control System ASCII - American Standard Code for Information Interchange ATS - Application Technology Satellite BB - Blackbody BOREAS - BOReal Ecosystem-Atmosphere Study BORIS - BOREAS Information System BPI - Bytes Per Inch CCT - Computer Compatible Tape CDA - Command and Data Acquisition CD-ROM - Compact Disk-Read-Only Memory DAAC - Distributed Active Archive Center DAT - Digital Archive Tape DC - Dark Current DN - Digital Number ECAL - Electronic Calibration EOS - Earth Observing System EOSDIS - EOS Data and Information System ESD - Environmental Satellite Data, Inc. E-W - East-West FFC - Focused Field Campaign FSU - Florida State University GIMTACS - GOES I-M Telemtry and Command System GMT - Greenwich Mean Tim GOES - Geostationary Operational Environmental Satellite GSFC - Goddard Space Flight Center GVAR - GOES VARiable IFC - Intensive Field Campaign IFOV - Instantaneous Field of View IGFOV - Instantaneous Geometric Field of View IMC - Image Motion Compensation INR - Instrument Navigation & Registration I/O - Input/Output IR - Infrared ISLSCP - International Satellite Land Surface Climatology Project MMC - Mirror Motion Compensation NAD83 - North American Datum of 1983 NASA - National Aeronautics and Space Administration NESDIS - National Environmental Satellite, Data and Information Service NLUT - Normalization Look-Up Table NOAA - National Oceanic and Atmospheric Administration N-S - North-South NSA - Northern Study Area NWS - National Weather Service OATS - Orbit and Altitude Tracking System OGE - Operations Ground Equipment ORNL - Oak Ridge National Laboratory PANP - Prince Albert National Park PM - Product Monitor PMS - Performance Monitoring System RSS - Remote Sensing Science SDI - Sensor Data Interface SEM - Space Environmental Monitoring SMS - Synchronous Meteorological Satellites SPS - Sensor Processing Subsystem SSA - Southern Study Area SS/L - Space Systems/Loral, Inc. T&C - Telemetry and Command URL - Uniform Resource Locator VAS - VISSR Atmospheric Sounder VISSR - Visible and Infrared Spin Scan Radiometer 20. Document Information 20.1 Document Revision Dates Written: 25-Jul-1995 Last Updated: 07-May-1998 20.2 Document Review Dates BORIS Review: 24-Apr-1998 Science Review: 20.3 Document ID 20.4 Citation The level-1a GOES-8 images resulted from a joint effort between BOREAS staff at NASA GSFC and Dr. Eric Smith of FSU. The original data were acquired by FSU and processed to level-1 products. The present level-1a product was created by BORIS. The respective contributions of the above individuals and agencies to completing this data set are greatly appreciated. 20.5 Document Curator 20.6 Document URL Keywords GOES-8 EMITTED RADIATION REFLECTED RADIATION WATER VAPOR RSS14_GOES8_L1A.doc 05/26/98