Archive of Viking Lander 1 and 2 EDR Images

Edward A. Guinness
Department of Earth and Planetary Sciences
McDonnell Center for the Space Sciences
Washington University
St. Louis, Missouri 63130

Table of Contents

  1. Introduction
  2. Viking Mission
  3. Viking Lander Spacecraft
  4. Lander Camera System
  5. EDR Image Generation
  6. Archive Data Preparation
  7. Disk Directory Structure
  8. PDS Image and Browse Image File Formats
  9. Radiometric and Geometric Calibration
  10. Lander Imaging Science Plan
  11. References

1. Introduction

The two Viking Lander spacecraft were the first spacecraft to operate successfully for an extended period of time on the surface of Mars. Both spacecraft operated from 1976 through April 1980 and Viking Lander 1 (VL1) continued to operate until November 1982. In 1981, Viking Lander 1 was renamed the Thomas A. Mutch Memorial Station in honor of Tim Mutch, Lander Imaging Team Leader and later NASA Associate Administrator for Space Sciences. Tim Mutch was killed in a climbing accident in 1980. With respect, the Mutch Memorial Station will be referred to as VL1 throughout most of this document to be consistent with other documentation and labels in this archive. The primary scientific objective of the lander mission was to test for the presence of life on Mars. Secondary objectives were embodied in the seismology, meteorology, inorganic chemical analysis, imaging, magnetic, and physical properties investigations. Among the instruments aboard the landers was an imaging system that consisted of two identical cameras. These cameras operated throughout the mission and returned nearly 6600 images. This archive on two compact disks (CD-ROM) includes the Experiment Data Record (EDR) version of all available images acquired on Mars by the Viking Lander imaging systems.

This EDR dataset is the primary record of lander image data as it was received on Earth. EDR images were originally distributed to lander imaging team members and to the National Space Science Data Center (NSSDC) as photoproducts and digitally on 9-track magnetic tapes. This CD-ROM archive was produced by the Geosciences Discipline Node of NASA's Planetary Data System (PDS) to transfer the EDR image dataset to more stable media and to provide wide distribution of the dataset to interested scientific research organizations and universities. The images in this archive have not been processed in any form other than generating and attaching PDS labels and histograms to the original EDR data. This archive also includes documentation about the imaging system and radiometric and geometric calibration of the image data; browse versions of the EDR images; and metadata about the images in a format suitable for loading into spreadsheet or data base management programs. Derived Viking Lander image datasets will be included in future PDS archives.

The following material provides the primary documentation of the EDR image archive. Included in this document are sections on: A) the Viking Mission; B) the Viking Lander spacecraft; C) the lander imaging system; D) the processes for generating the original EDR images and this archive; E) the volume and PDS image file structure; F) image calibration procedures; and G) the major types of imaging sequences. Other documentation files are included elsewhere in this archive. Detailed explanations of file formats are found in the CDVOLSIS.TXT file. Full radiometric and geometric calibration information are found in the CALINFO.TXT and GEOMINFO.TXT files, respectively. The ERRATA.TXT file contains a list of comments and anomalies about the archive. Finally, the AAREADME.TXT file in the top-level directory has an introduction to the archive and a description of the archive structure.

2. Viking Mission

The Viking Mission consisted of four spacecraft: two identical orbiters and two identical landers [Soffen, 1977]. The landers were attached to the orbiters during cruise to Mars. After orbit insertion around Mars and landing site selection, the landers separated from the orbiters for descent to the surface and a soft landing. The primary scientific objective of the mission was to search for life on Mars. Other science questions addressed by the mission were the composition and physical properties of the atmosphere, the distribution of atmospheric water vapor, global and local meteorology, the composition and physical properties of the surface, nature of Mars seismicity, and gravity field of Mars. The combined orbiter and lander spacecraft supported thirteen science investigations. The orbiters had three mapping experiments (imaging, infrared thermal mapper, and water vapor mapper). An atmospheric investigation was conducted during lander descent to the surface. The landers had eight scientific experiments (see Section 3). In addition, a series of radio science investigations were done using both the lander and orbiter radio systems [Snyder, 1977; Snyder and Moroz, 1992].

The Viking 1 and 2 spacecraft were launched on August 20 and September 9, 1975, respectively, and the cruise to Mars phase lasted 10 to 11 months. Below is a table showing major events of the mission.

Orbiter 1 Lander 1 Orbiter 2 Lander 2
Launch 8/20/75 8/20/75 9/9/75 9/9/75
Orbit Insertion 6/19/76 6/19/76 8/7/76 8/7/76
Landing 7/20/76 9/3/76
Transmissions End 8/7/80 11/13/82 7/25/78 4/11/80

Viking Lander 1 (VL1) landed at 22.480 deg N latitude and 47.968 deg W longitude planetographic [Kieffer et al., 1992; Michael, 1979] in the western portion of Chryse Planitia, which is a region of smooth plains and impact craters. Viking Lander 2 (VL2) landed at 47.967 deg N latitude and 225.737 deg W longitude planetographic [Kieffer et al., 1992; Michael, 1979] about 200 km west of the Crater Mie in Utopia Planitia, which consists of plains with broad swales and bulges [Moore et al., 1987].

The four spacecraft operated independently after lander separation. The orbiters served as communication relay stations to transfer data from lander to orbiter, and then on to Earth, but data were also transferred directly from the lander to Earth. Each spacecraft operated for a number of years and long after their primary missions were completed. Viking Orbiter 2 (VO2) was the first of the four spacecraft to end its mission. It developed a leak in its propulsion system and lost its attitude control gas. VO2 was turned off after 706 orbits around Mars. Viking Orbiter 1 (VO1) consumed the last of its attitude control gas and was turned off after 1485 orbits around Mars. VL2 operated on the surface for 1281 Mars days and was turned off when its batteries failed. VL1 operated the longest of the four Viking spacecraft. It returned data for nearly 6.5 Earth years (2252 Mars days or over 3 Mars years). Its mission ended when communications with the spacecraft were lost [Arvidson et al., 1983].

3. Viking Lander Spacecraft

The Viking Landers were identical to each other and had the same instrument packages. The main lander structure was a hexagonal prism body that housed the spacecraft computers, tape recorder, batteries, several science instruments, and controls for the surface sampler, thermal system, and data handling system. The spacecraft body was supported above the surface by three legs, each with a saucer-shaped footpad. The legs were arranged in a triangle pattern with two at the front of the lander and one at the rear. Mounted to the sides of the spacecraft were three terminal descent engines, two propellant tanks, and the extendible surface sampler arm with a collector head and backhoe [Moore et al., 1987]. Also mounted on the spacecraft body were two cameras, two radioisotope thermoelectric generators (RTG) with covers, sample entry ports for the biology, organic chemistry and inorganic chemistry instruments, the seismometer, the meteorology boom, a magnifying mirror, a magnet, and three imaging reference test charts. The spacecraft had three antennas for communications; a high-gain S-band antenna (large dish antenna), a low-gain S-band antenna, and a UHF antenna. The two S-band antennas were used to communicate with Earth, whereas the UHF antenna communicated with the orbiters.

Each lander was about 1.5 m across. Nominal clearance between the spacecraft body and the surface was about 22 cm. After landing the spacecraft mass was about 610 kg. Power for the spacecraft was supplied by rechargeable batteries; the batteries were recharged by the two RTGs. Excess heat from the RTGs was used to heat the instruments and control systems in the spacecraft body [Soffen, 1977].

Lander science investigations employed seven instruments in addition to the cameras. The meteorology instrument was mounted on a mast and was capable of measuring atmospheric temperature, pressure, wind speed, and wind direction. Each lander had a three-axis, short-period seismometer to measure Mars seismic activity. The seismometer on VL1 failed to deploy and no data were returned, while the one on VL2 operated as planned. An X-ray fluorescence spectrometer (XRFS) measured the inorganic elemental composition of soils at the landing sites. A gas chromatograph mass spectrometer (GCMS) measured the composition of the atmosphere and searched for organic compounds in the soils. The biology investigation consisted of three experiments (and instruments) to search for biological metabolism, growth, or photosynthesis: carbon assimilation, labeled release of carbon-14, and gas exchange. The physical properties investigation used information from many lander operations, such as sampling activities, digging trenches, pushing rocks, forming soil piles, and footpad penetration during landing to characterize the properties of rocks and soils at the landing sites. The magnetic properties investigation employed two magnets on the backhoe of the sampler collector head and one magnet array on the center reference test chart on the lander body [Hargraves et al., 1979]. The reference test charts mounted on the lander also contained two patches coated with paint that degraded when exposed to ultraviolet radiation. Finally, lander communication systems were used for radio science experiments [Snyder and Moroz, 1992].

The lander mission can be divided into a number of periods (mission phases) in terms of activity level and the types of observations. The table below lists the lander mission phases and their dates. Snyder [1979] provides a narrative of the activities for most of these mission phases. The time frames for these mission phases apply to both Viking Landers.

Mission Phase Start End
Primary Mission 7/20/76 11/15/76
Extended Mission 11/15/76 5/31/78
Continuation Mission 5/25/78 2/26/79
Interim Period 2/26/79 7/19/79
Survey Mission 7/19/79 8/7/80
Completion Mission 8/7/80 11/19/82

The Primary Mission began with VL1 landing and continued until the solar conjunction in November, 1976. The Primary Mission had a high-level of activity and was highlighted by the collection and analysis of soil samples and characterization of the landing site and atmosphere. The Extended Mission began after the solar conjunction with a lower activity level due to a smaller staff. However, the Extended Mission provided an opportunity to monitor the surface of Mars through a complete Mars year and to perform a variety of experiments that were not possible during the Primary Mission. During the Extended Mission, additional soil samples were collected for biological and chemical analyses. Three deep holes (ranging from 8 to 23 cm in depth) were dug at both landing sites. The physical properties investigation executed many experiments with the surface sampler system [Moore et al., 1987]. The imaging experiment studied the surface and atmosphere through the cycle of Mars seasons.

During the first Mars winter of the Extended Mission, VL2 was programmed to operate in an automatic manner designed to allow the spacecraft to survive the cold winter temperatures and still return data (mainly imaging, meteorology, and seismology data). VL1 continued to operate normally during the winter due to its more equatorial location. VL2 returned to full operation after the winter passed. A major limitation was placed on VL2 when a portion of its communication system failed in the Extended Mission, leaving VL2 with no direct downlink to Earth.

At the end of the Extended Mission, all lander instruments were turned off except for imaging, meteorology, and XRFS. These three instruments continued to operate during the Continuation Mission in a fully automated manner. Observation sequences collected data on a 37 sol (Mars day) cycle that repeated throughout the Continuation Mission. There was an Interim Period after the Continuation Mission ended where communications with the Viking spacecraft were severely limited because of Voyager encounters with Jupiter. During the Interim Period, imaging and meteorology data were collected and returned when possible. A final VL2 surface sampler sequence was conducted during this period as an engineering test in the cold temperatures of mid winter.

The final phases of the lander mission were the Survey and Completion Missions. The plan was to collect image and meteorology data for as long as possible. For VL2 this meant for as long as VO1 provided a relay link because VL2 no longer had a direct downlink capability. Relay opportunities with VO1 occurred every seven weeks during the Survey Mission. VL2 was finally turned off after its batteries could no longer hold a charge. VL1 operated in a cyclic and automatic mode by returning data about once a week with the image sequences repeating every 37 Mars days. The VL1 high-gain antenna was programmed to track the Earth until December, 1994. However, communications with the Mutch Memorial Station (VL1) were lost in November, 1982 after a command sequence uplink, and thus, ended the Viking Lander mission [Arvidson et al., 1983].

4. Lander Camera System

The Viking Lander camera design was very different from vidicon framing or CCD array cameras. The lander camera was a facsimile camera with a single, stationary photosensor array (PSA), and azimuth and elevation scanning mechanisms. A lander image was generated by scanning the scene in two directions (elevation and azimuth) to focus light onto the photosensor array. The Viking Lander cameras were built by Itek Corp. A number of published papers described the characteristics and performance of the lander cameras. The scientific rationale and early design of the cameras were described in Mutch et al. [1972] and a detailed description of the flight cameras was given in Huck et al. [1975b]. Huck and Wall [1976] discussed image quality and Patterson et al. [1977] described camera performance during the Primary Mission. A summary of the information from these papers is given here as a high-level description of the camera and its operating modes.

Each camera was bolted to the top of the lander body. The camera had a lower stationary section containing electronics and an upper section that rotated in azimuth (i.e., around a vertical axis). The other prominent exterior component was a post assembly that protected the camera window from the Mars environment when the camera was not in use. Major internal camera components were two fused silica windows, a mirror, elevation and azimuth rotation assemblies, a lens, the photosensor array, electronics, and pinlights for calibration. The overall camera height was 55.6 cm. The lower assembly diameter was 25.6 cm and the upper assembly diameter was 14.4 cm. The mass of the camera was 7.26 kg. Below is a table that lists the serial numbers of the cameras and photosensor arrays as they were assigned to the flight cameras [Huck et al., 1975a].

Camera Serial Number PSA Serial Number
Viking 1, camera 1 FC-1B M017
Viking 1, camera 2 FC-2A M020
Viking 2, camera 1 FC-3A M015
Viking 2, camera 2 SPARE M019

The camera had several features to deal with possible dust abrasion or obscuration. When the camera was not in use, the upper assembly was rotated so that the window was behind the post assembly to prevent exposure to dust. Mounted in the post was a nozzle for releasing carbon dioxide gas to blow dust off the window. In addition, the outer window, known as the contamination cover, was designed to move out of the optical path if it became abraded or dust coated. The contamination cover window was hinged and spring loaded so that it could be moved aside by rotating the camera behind the post with enough force to release the spring lever. During the Extended Mission the contamination cover windows of camera 1 on lander 1 and camera 2 on lander 2 were opened. One drawback of the contamination cover window was that its frame caused a vignetting effect at elevation angles above 25 deg. [Patterson et al., 1977].

Light entered the camera through the windows, reflected off the mirror toward the lens, passed through the lens, and was sensed by one of the photodiodes. The light generated a voltage in the selected photodiode that was digitized by an analog-to-digital (A/D) converter. Below the lens was a black shutter that could be closed to sample photodiode dark current and to perform internal radiometric calibration. Between the lens and photosensor array was a light baffle to minimize internal reflections. The lens had a 0.95 cm aperture diameter and 5.37 cm focal length. For details on the transmittance and reflectivity of the optical components, see the documentation and data in the CALIB directory.

Mechanical scanning was done by rotating the mirror around a horizontal axis to scan each vertical line in an image. The entire upper camera assembly rotated around a vertical axis to scan successive vertical lines. Image data were collected while the mirror rotated upward through the elevation range in 512 steps (i.e., pixels were sampled from the bottom of the image to the top). The mirror rapidly rotated down to the start position while the camera turned in azimuth for the next scan. The mirror could rotate from 60 deg below a plane perpendicular to the azimuth rotation axis to 40 deg above it for a total range of 100 deg. The camera could see 342.5 deg in azimuth with the range limited by the post assembly. Image commands specified both a start and stop azimuth so that the widths of images varied from image to image. On receiving a command to acquire an image, the camera first rotated to a preset azimuth and then rotated to the start azimuth. The camera stepped in azimuth after each vertical scan until the stop azimuth was reached. The sequence ended by the camera moving back to the preset azimuth position and then to the park position with the window behind the post. The preset position at the beginning and end of the sequence prevented the camera from turning through mechanical stops. Internal radiometric calibration could be done by closing the shutter while the camera was in the preset position either before or after the image data were acquired. The camera had two scanning rates so that data could be collected at 250 or 16,000 bits/sec.

The photosensor array consisted of 12 silicon photodiodes (or diodes) sensitive to light between 0.4 and 1.1 micrometers. The diodes were arranged in a 2x6 array. There were four broad band, high resolution (0.04 deg angular resolution) diodes, known as BB1, BB2, BB3, and BB4. The distance between the lens and each high resolution diode was different to vary the in-focus distance of each diode. In-focus distances were 1.9, 3.7, 4.5, and 13.3 meters for BB1, BB2, BB3, an BB4, respectively. There was a low resolution (0.12 deg angular resolution), broad band diode known as the SURVEY diode. There were also six narrow band, low resolution diodes for color (BLUE, GREEN, and RED) and infrared (IR1, IR2, and IR3) images. The narrow bandwidth was generated by covering diodes with a set of interference filters, chosen to survive the spacecraft sterilization process. The interference filters had significant out-of-band leaks (see CALINFO.TXT). Also, the infrared filters were known to degrade due to neutron radiation from the lander RTGs [Patterson et al., 1977]. The twelfth diode was also a low resolution diode covered by a red filter for imaging the Sun. The in-focus distance for the low resolution diodes was about 3.7 meters.

Diodes in the camera generated a voltage proportional to incoming radiance. Each diode, except for the SUN diode, had an amplifier to enhance the voltage signal. The SUN diode did not need an amplifier because of the strong signal when directly viewing the Sun. Voltage from the amplifier was digitized as a 6-bit number in an A/D converter that also performed automatic dark current subtraction. Dark current was sampled after every line in slow scan mode and after every 64th line in rapid scan mode. The A/D converter had 6 gains and 32 offsets so that the full dynamic range of the diode output could be stored in 6-bits. Gain defined the voltage range sampled and its resolution. Low gain (high gain number) covered a wide range in voltage and had low voltage resolution. The offset could be varied from a slightly negative voltage to several volts in 32 small steps. Digital values from the A/D converter could be dumped to the spacecraft tape recorder, transmitted to Earth, or transmitted to an orbiter in real-time (usually done at beginning or end of a transmission link when the bit error rate was relatively high).

Each camera had two pinlights in the post assembly. Using a special command, the lights turned on and an image was acquired while viewing the pinlights. This command, known as a scan verification, monitored the mechanical scanning operation of the camera. A third pinlight with four different radiance levels was located between the dark current shutter and the photosensor array. This pinlight was used with the shutter closed for radiometric calibration of the camera. Scan verification and internal calibration sequences are further discussed in the CALINFO.TXT file.

Operation of Viking Lander cameras was versatile because many parameters for the image could be specified. Commands to acquire an image involved selecting the sampling mode, diode, start and stop azimuths, center elevation, gain and offset, scan rate, and specifying whether automatic dark current subtraction, internal calibration, rescan, or dusting were done. Image start time and whether the image was transmitted in real-time or sent to the tape recorder were also specified. The camera had seven sampling modes: low resolution (0.12 deg) stepping with three diodes (known as triplet mode); low resolution stepping with one diode; high resolution (0.04 deg) stepping with one diode; and four different calibration lamp intensity levels. In triplet mode, elevation scanning was repeated three times with three different diodes at every azimuth position. An unusual sampling mode effect was an elevation shift when the step size did not match the diode resolution. There was a -5.6 deg elevation shift when using a low resolution diode with high resolution stepping and a +5.6 deg shift when using a high resolution diode with low resolution stepping. These non-nominal modes were mostly used to generate high resolution color images or to better resolve the solar disk. For the triplet mode, the specified diode was the first of the three diodes, e.g., BLUE diode for a color image. Start and stop azimuths had to be multiples of 2.5 deg. Center elevation had to be a multiple of 10 deg. The elevation range was determined by the step size with a range of about 20 deg for high resolution and about 60 deg for low resolution because there were always 512 steps in an elevation scan. Rescan was done by inhibiting the azimuth rotation and repeatedly scanning in elevation. Rescan could be activated by commanding the camera to operate for a time longer than it took to scan the azimuth range of the image.

5. EDR Image Generation

The original EDR dataset was produced by the Viking Project and Lander Imaging Team. Levinthal et al. [1977a] explains the processing done to produce the original EDR images. Initial EDR processing included selecting and merging the best available data recorded at one or more DSN stations. After merging several playbacks, some imaging data were still not extracted from the telemetry stream because of noise in the data identification tags. Thus, special purpose software searched the telemetry stream to identify and extract imaging data with noisy tags. In some cases, this secondary processing greatly increased the amount of data recovered from the telemetry stream. Even with the secondary recovery, the EDR data contained random noise and missing lines. Random noise was usually due to transmission errors. Missing scan lines were filled with zero valued pixels. Missing data appear as black vertical lines in the image because the primary scan direction was vertical. As noted in the previous section, the camera voltage signal was digitized on the spacecraft to a 6-bit integer number (possible values between 0 and 63). However, the maximum brightness value output from the camera was 62 due to limitations of the system. As part of the standard EDR processing, the original 6-bit value was converted to an 8-bit integer number by multiplying the 6-bit values by 4. These 8-bit values are the data archived in this dataset.

6. Archive Data Preparation

Several steps were taken to insure that the best available image data and label information were included in this archive. The primary source of digital image data was the Viking Lander Imaging Team set of EDR 9-track magnetic tapes at Washington University. The data from these tapes were transferred to 8-mm tape and then to CD-WO before this archive was made. Because the original EDR files did not contain histogram or checksum values, automated data validation could not be done. Instead, each digital image was visually inspected and compared to a photoproduct version of the image. This check insured that the file contained the correct image data and that there were no obvious errors, such as missing or corrupted data. In a few cases errors were detected and a replacement image was obtained from the EDR set at the Jet Propulsion Laboratory (JPL).

The original EDR image files were stored in a VICAR format with metadata attached in a VICAR label. These metadata were used to populate the PDS label. Label parameters were checked for acceptable values and corrected if necessary. Remaining anomalies were documented in the ERRATA.TXT file. Catalogs [Tucker, 1981; Jones et al., 1981; Wall and Ashmore, 1985] and planning notes produced by the Lander Imaging Team were used to determine the purpose for acquiring a given image. This information was added to the PDS label as a NOTE field. PDS image files for this archive were generated by extracting label information from the VICAR header, computing the histogram and checksum for the data, and then writing the label, histogram, and image data in a PDS format. Note that the index table was also generated by extracting information from the original VICAR headers, so that the PDS labels and index table contain the same information.

7. Disk Directory Structure

The volume and directory structure of this CD-ROM conforms to the ISO-9660 standard, except that files do not contain any Extended Attribute Records (XAR). The directory content, file format, and supporting documentation conform to PDS standards version 3.2 [PDS, 1995]. The AAREADME.TXT file in the top-level directory has an outline of the archive directory structure. File names within this archive conform to the "8.3" convention.

EDR images, browse images, supplemental files, and documentation are located in separate directories. Directories with ancillary files and data are as follows: The CATALOG directory contains PDS high-level catalog information about the Viking Mission, Viking Lander spacecraft and cameras, and the EDR image dataset. The DOCUMENT directory contains documentation for the EDR dataset and the archive. The INDEX directory contains index tables about images from both landers. The CALIB directory contains documentation and data for radiometric calibration of Viking Lander images. The GEOM directory contains documentation and figures that describe the geometric aspects of the Viking landers and cameras. The BROWSE directory contains HTML files and browse images for use as a simple image browser.

EDR image files are subdivided into directories based on the image PRODUCT_ID. Image file names are also based on the PRODUCT_ID. For Viking Lander images, the PRODUCT_ID has the form LCXSSS-FFF. L is the lander number (1 or 2) and C is the camera number (1 or 2). The sequence XSSS is an image identifier where X is a letter and SSS is a number from 000 to 255. The initial value of this identifier was A001 for both landers. The number portion of the identifier was incremented by 1 for each image. However, the number could not exceed 255 because of spacecraft software. Thus, the letter portion was incremented and the number was reset to 000 after 255 was reached. For example, image identifier A255 was followed by B000. The sequence FFF in the PRODUCT_ID is an abbreviation for the filter (diode) name. EDR images are grouped in directories by the image identifier portion such that all images with the same first two characters of the identifier are in the same directory. For example, the directory B1XX has files with image identifiers between B100 and B199. This scheme was selected to have a reasonable number of files in each directory and to keep the images in chronological order. Image file names have the form LCXSSS.FFF where the components are the same as for the PRODUCT_ID.

Browse images are also subdivided into directories under the IMAGE subdirectory in the BROWSE directory. The scheme for grouping the browse images is the same as for the EDR image files. Names of browse image files have the pattern of LCXSSSFF.GIF, where the first six characters are the same as in the image PRODUCT_ID and FF is the first and last character of the filter name from of the PRODUCT_ID. This file naming scheme was adopted to: 1) maintain the "8.3" file name standard; 2) have the ".GIF" extension so that software will recognize the GIF format; and 3) have unique file names.

8. PDS Image and Browse Image File Formats

The structure of the PDS image file will be briefly described here. Additional details about the PDS image file can be found in the CDVOLSIS.TXT file. Each PDS image file in this archive has three components, the PDS label, the image histogram, and the image data. The file can be considered as having fixed length records even though there are no record delimiters per se in the file. The record size is determined by the number of samples in the image. If the PDS label or the histogram do not completely fill a record, they are padded with zero valued bytes to maintain the fixed length records. Thus, each file object starts on a record boundary.

The PDS label is the first object in the file. It has a "keyword=value" format [PDS, 1995] with sections that contain the file size and object locations, metadata about the images, and descriptions of the histogram and image objects. Each "keyword=value" pair in the label is delimited by a carriage control (ASCII 13) and line feed (ASCII 10) character pair. Keyword values are either numbers with optional units in brackets, standard values, or text strings. Double quotes are used to enclose text strings for keyword values. The CDVOLSIS.TXT file contains an example label and definitions for each of the label keywords.

The image histogram object is the second component in the file. It begins at the record specified by the ^IMAGE_HISTOGRAM keyword. The histogram consists of 256 numbers, each a 32-bit signed integer in binary format with most-significant byte first order. The value of the first item in the histogram is the number of image pixels with brightness value 0. The second item is the number of pixels with brightness value 1, and so on until the last item, which has the number of pixels with brightness value 255.

The third object in the file is the image data. The image object starts at the record specified by the ^IMAGE keyword. Each EDR image contains 512 lines. The number of samples in each image varies from image to image because the start and stop azimuth could be commanded independently. Each pixel value is an 8-bit unsigned integer. The value of the CHECKSUM keyword in the PDS label is the sum of all pixel values in the image.

Browse images were generated from the PDS images and were stored in GIF format. Images were reduced in size by a factor of four in lines and samples. Resampling was done by skipping pixels. Browse images were also contrast enhanced for better viewing of the data. An image browser is provided with this archive that uses HTML for easy and rapid image viewing. The file BROWINFO.TXT in the BROWSE directory has more information about the browser.

9. Radiometric and Geometric Calibration

Full quantitative use of Viking Lander image data requires an understanding of the radiometric and geometric properties of the Viking Lander cameras. EDR pixel values can be converted to radiance or reflectance at the sensor using the calibration data provided with this archive. Topographic data can be extracted by using the stereoscopic capabilities of the two cameras. Radiometric and geometric properties of Viking Lander cameras are described in detail in the documents and data found in the CALIB and GEOM directories. These calibration procedures will be briefly outlined here.

Radiometric calibration of a Viking Lander image is basically a linear scaling of pixel values that is dependent on the diode, gain and offset, and time (i.e., Mars-Sun distance). These scaling parameters can easily be computed using the equations given in the CALINFO.TXT file and data provided in the CALIB directory. Additional corrections can be made for atmospheric effects. The radiometric calibration data in the CALIB directory are based on pre-flight testing. In-flight calibration measurements (internal calibration images) are included in this archive because they are part of the EDR dataset. However, analysis of internal calibration images is not included in this archive. Internal calibration images are described in the CALINFO.TXT file.

The GEOM directory contains the GEOMINFO.TXT document that describes the location and orientation of the landers. The document also explains camera coordinate systems and how to extract the three-dimensional location of objects with stereo images. Three-dimensional measurements can be used to construct topographic maps [Liebes, 1982]. Such data can also be used to determine distances between points in the scene; the size, shape, and orientation of objects [Moore et al., 1987]; and emission and phase angles for objects. Given image coordinates of the same point seen in both camera 1 and 2 images (i.e., conjugate images), the equations in the GEOMINFO.TXT file can be used to compute the location of the point. These equations can be easily programmed in a spreadsheet or using one of many programming languages.

10. Lander Imaging Science Plan

This section outlines the scientific objectives of the imaging experiment and describes the imaging sequences used to investigate these objectives. The specific intent for acquiring each image is given by the NOTE keyword of the PDS label. The major scientific objectives of the Viking Lander imaging investigation were to analyze the geology, cartography, meteorology, and biology of the landing sites [Mutch et al., 1972]. Geologic studies included characterizing the morphology of rocks, soils, and other features from texture and color, determining the size distribution of rocks, and understanding sediment transport. Cartographic studies involved mapping features at the landing sites, measuring surface topography, and determining lander location by comparing features seen in both lander and orbiter images. Meteorological investigations with Viking Lander images determined atmospheric aerosol properties (abundance, size, composition, and distribution) and searched for evidence of dust and condensate clouds. Biological studies with Viking Lander image data consisted of searching for evidence of living things [Levinthal et al., 1977b]. Arvidson et al. [1989] review the major results of the lander imaging investigations.

Imaging sequences during the Primary Mission focused on: A) characterizing the surface and atmosphere at the landing sites; B) monitoring the sites for change; and C) supporting other experiments and sampling activities. Both landing sites were imaged with high resolution diodes during both the morning and afternoon. A noon-time high resolution set of images were also acquired at VL2. These high resolution images have been assembled into a set of mosaics [Levinthal and Jones, 1980]. The high resolution mosaics were the primary images used to generate systematic topographic maps [Liebes, 1982]. The scene at both sites was also imaged in color at noon-time, and selected areas were imaged with the three infrared diodes. Furthermore, a portion of the scene was imaged in the high resolution color mode (low resolution color diodes with high resolution stepping). Several types of image sequences were acquired for atmospheric studies. The sun diode was used to determine atmospheric optical depth [Colburn et al., 1988]. Aerosol distribution was analyzed from sky images in a color rescan mode at dawn and dusk (twilight rescan images). In another atmospheric sequence, known as a sky brightness sequence, images of the sky were taken at different azimuths from the Sun. Scan verification and internal calibration images were periodically acquired to monitor the health of the cameras. Several sequences were acquired to search for variable features using rescan images and by repeatedly imaging the same areas. Images were also obtained to study the photometry of the surface.

Image sequences supported several other lander investigations. Images were taken of the sample delivery ports of the Biology, GCMS, and XRFS experiments to check whether the ports opened after landing and to support sample collection. Sampling sites were imaged before and after sampling, including color and stereo images, to support the planning and collection of samples. Magnets on the lander and surface sampler backhoe were imaged periodically for the magnetic properties investigation. Images of spacecraft parts, trenches, and surface sampling activities were taken for the physical properties team. Problems with the surface sampling system were also analyzed with special images.

During the Extended Mission and beyond, the emphasis of imaging shifted to: A) monitoring the surface and atmosphere through the Mars seasons; B) supporting other lander investigations; and C) performing special experiments. Camera health was monitored throughout the Extended Mission. Surface monitoring included periodically looking at disturbed areas (trenches and soil piles) and undisturbed areas (e.g., drifts) for movement and change. Images were also acquired to search for evidence of sediment deposition. During the winter season, the spacecraft and surface at VL2 were monitored for frost formation. Images, known as "repro" images, were acquired to match the lighting conditions of Primary Mission images so that subtle changes could be detected without complications from lighting differences. Jones et al. [1981] and Wall and Ashmore [1985] have image lists sorted by sun position to identify such "repro" images. Atmospheric studies during the Extended Mission continued with optical depth, twilight rescan, and sky brightness sequences. Some of the special imaging sequences of the Extended Mission included images to search for fog and ozone, images for photometric studies, and rescan images to detect the passage of the shadow of Phobos over VL1.

11. References

Arvidson, R. E., E. A. Guinness, H. J. Moore, J. Tillman, and S. D. Wall, Three Mars Years: Viking Lander 1 Imaging Observations, Science, 222, 463-468, 1983.

Arvidson, R. E., J. L. Gooding, and H. J. Moore, The martian surface as imaged, sampled, and analyzed by the Viking Landers, Rev. Geophys., 27, 39-60, 1989.

Colburn, D. S., J. B. Pollack, and R. M. Haberle, Diurnal variations in optical depth at Mars: Observations and interpretations, NASA Tech. Memo. TM 100057, 1988.

Hargraves, R. B., D. W. Collinson, R. E. Arvidson, and P. M. Cates, Viking magnetic properties experiment: Extended mission results, J. Geophys. Res., 84, 8379-8384, 1979.

Huck, F. O., E. E. Burcher, E. J. Taylor, and S. D. Wall, Radiometric performance of the Viking Mars lander cameras, NASA Tech. Memo. TM X-72692, 1975a.

Huck, F. O., H. F. McCall, W. R. Patterson, and G. R. Taylor, The Viking Mars lander camera, Space, Sci. Instr., 1, 189-241, 1975b.

Huck, F. O., and S. D. Wall, Image quality prediction: An aid to the Viking Lander imaging investigation on Mars, Applied Optics, 15, 1748-1766, 1976.

Jones, K. L., M. Henshaw, C. McMenomy, A. Robles, P. C. Scribner, S. D. Wall, and J. W. Wilson, Viking Lander imaging investigation during extended and continuation automatic missions, NASA Reference Publication 1068, 2 vols., 1981.

Kieffer, H. H., B, M, Jakosky, and C. W. Snyder, The planet Mars: From antiquity to the present, In Mars, Kieffer et al., eds., Univ. of Arizona Press, Tucson, 1992.

Levinthal, E. C., W. Green, K. L. Jones, and R. Tucker, Processing the Viking Lander camera data, J. Geophys. Res., 82, 4412-4420, 1977a.

Levinthal, E. C., K. L. Jones, P. Fox, and C. Sagan, Lander imaging as a detector of life on Mars, J. Geophys. Res., 82, 4468-4478, 1977b.

Levinthal, E. C., and K. L. Jones, The mosaics of Mars as seen by the Viking Lander cameras, NASA Contractor Report 3326, 1980.

Liebes, S., Viking Lander Atlas of Mars, NASA Contractor Report 3568, 1982.

Michael, W. H., Viking Lander tracking contributions ot Mars mapping, Moons and Planets, 20, 149-152, 1979.

Moore, H. J., R. E. Hutton, G. D. Clow, and C. R. Spitzer, Physical properties of the surface materials at the Viking landings sites on Mars, USGS Professional Paper 1389, 1987.

Mutch, T. A., A. B. Binder, F. O. Huck, E. C. Levinthal, E. C. Morris, C. Sagan, and A. T. Young, Imaging experiment: The Viking Lander, Icarus, 16, 92-110, 1972.

Patterson, W. R., F. O. Huck, S. D. Wall, and M. R. Wolf, Calibration and performance of the Viking Lander cameras, J. Geophys. Res., 82, 4391-4400, 1977.

Planetary Data System Standards Reference, Version 3.2, JPL Doc. D-7669, 1995.

Snyder, C. W., The missions of the Viking Orbiters, J. Geophys. Res., 82, 3971-3983, 1977.

Snyder, C. W., The extended mission of Viking, J. Geophys. Res., 84, 7911-7933, 1979.

Snyder, C. W., and I. V. Moroz, Spacecraft Exploration of Mars, in Mars, Kieffer et al., eds., Univ. of Arizona Press, Tucson, 1992.

Soffen, G. A., The Viking Project, J. Geophys. Res., 82, 3959-3970, 1977.

Tucker, R. B., Viking Lander imaging investigation: Picture catalog of primary mission experiment data record, NASA Reference Publication 1007, 1978.

Wall, S. D., and T. C. Ashmore, Conclusion of the Viking Lander imaging investigation, NASA Reference Publication 1137, 1985.