NJPL1I00PDS100000000 = SFDU_LABEL RECORD_TYPE = STREAM OBJECT = TEXT NOTE = "Introduction to this IRTM Experiment" END_OBJECT END IRTM.TXT: The Viking IRTM Experiment H. Kieffer, 22aug89 [This information was extracted from the PDS catalog templates in early August, 1989. Two general references for the IRTM investigation are: Kieffer, H.H., T.Z.Martin, A.R.Peterfreund, B.M.Jakosky, E.D. Miner & F.D.Palluconi, Thermal and ALbedo Mapping of MArs during the Viking Primary Mission, Journal of Geophysical Research, vol.82, p. 4249-4291, (1977); referred to below as KIEFFER_ETAL_1977 Chase, S.C., J.L.Engel, H.W.Eyerly, H.H.Kieffer, F.D.Palluconi & D. Schofield, Viking Infrared Thermal Mapper, Applied Optics, vol. 17, p. 1243-1251 (1978); referred to below as CHASE_ETAL_1978.] SCIENTIFIC OBJECTIVES SUMMARY The objective of the Viking Orbiter infrared thermal mapper (IRTM) is to measure the thermal emission of the Martian surface and atmosphere and total reflected sunlight with high spatial and flux resolution. INSTRUMENT DESCRIPTION The IRTM contains four small Cassegrainian telescopes which each image the same, seven circular areas. There is a total of twenty- eight channels in four surface and one atmospheric thermal band from 6 micrometer to 30 micrometer and a broad solar reflectance band. All channels are sampled simultaneously, using the spacecraft scanning capability to map the radiance over small and large areas of the planet. All channels use thermopile detectors; spectral passbands are determined by a combination of interference filters, detector lens materials, antireflection coatings, and reststrahlen optics. The scan modes are described in INSTRUMENT MODE DESCRIPTION below. IRTM Passbands IRTM BAND D C1 C2 B C3 A DESIGNATION SOLAR T7 T9 T11 T15 T20 MINIMUM_WAVELENGTH 0.3 6.1 8.3 9.8 14.56 17.7 CENTER_FILTER_WAVELENGTH 1.6 7.2 9.0 11.2 15.0 21.0 MAXIMUM_WAVELENGTH 3.0 8.3 9.8 12.5 15.41 30.0 OPTICS DESCRIPTION The A telescope (17.7-24 micrometer) is shown schematically in Fig. 3 of CHASE_ETAL_1978. It is an f/3.5, 20.3-cm focal length Cassegrainian design with an aperture diameter of 5.8-cm., spherical surfaces, and, except for mirror materials, is identical to the B and C telescopes. By using relatively slow fore optics, degradation of filter sharpness normally caused by operating an interference filter in a low f-number beam is negligible. The focal plane contains a field-defining aperture plate with seven 0.107-cm diameter holes arranged in a chevron pattern. The fields of view thus defined are nested with those of the MAWD and imaging systems (Fig. 4 of CHASE_ETAL_1978). Behind each hole in the field stop plate is a lens which produces a 0.0254-cm diam image of the telescope aperture on the detector, which itself is about the same size. The final optical speed at the detector is f/1. Optical materials used in the four telescopes are shown in Table II, and the resulting spectral response is shown in Fig. 5 of CHASE_ETAL_1978. Mirrors are made of hot-pressed uncoated zinc oxide for both primary and secondary mirrors. The reststrahlen reflection properties of ZnO are the major factors in the A telescope spectral response. Minimizing extrafield sensitivity (EFS) was an important aspect of the optical design since on previous Mariner radiometers EFS contribution seriously compromised observations of scenes near large temperature contrasts (points near the planetary limb and polar caps). During instrumentation development, IRTM image quality was determined in two angular regions. In the near-field region, a laboratory collimator and ir source were used to measure the 2-D spatial response out to 16-mrad diam (three fields of view). Point source field of view measurements in this region are shown in Fig. 6 of CHASE_ETAL_1978. For far field measurements, sensitivity constraints dictated an approach in which the fraction of energy within a given angular annulus is measured. A 30.5-cm diam, concentric grooved, blackbody plate with a series of restricting apertures was used at several distances (30.5-cm, 140-cm, and 610-cm) to define angular response regions from about one field of view out to 1-rad diam. That is, with the telescope focused at 610-cm, a disk 3.17-cm in diameter at that distance defines one half-response field of view (5.2 mrad). The source was held at 95 degrees C. by a heater/regulator and integral water jacket. To prevent difficulties with atmospheric transmission, the entire apparatus was contained in a polyethylene bag flushed with dry N2. The EFS problem was more severe for the longer wavelength A telescope than the others, possibly owing to the higher reflectance at longer wavelengths of the black paint used inside the telescope. Tests using this apparatus led to several telescope modifications designed to reduce EFS (see Fig. 3 of CHASE_ETAL_1978): (1) A postfocal baffle was placed between the field lens and the detectors to confine energy to the sensitive area of the detectors. (2) A spider baffle, added to the outer edges of the secondary mirror support spider, was designed to reduce reflection off the sides of the spider legs. (3) A cone baffle coated with CTL 15 black paint was placed on the central dead spot of the secondary mirror. This was designed to prevent focal plane reflections from falling on the detectors. Of these three modifications, only the cone baffle gave significant improvement, although all three were incorporated in the design. These results of the final EFS measurements are shown in Fig. 6 of CHASE_ETAL_1978. The calculated response due to diffraction and the measured values are shown. The integrated EFS response between 12 mrad and 1-rad diam was about 4%. Of this, about 1/2 is due to diffraction effects. The effect of response outside of the nominal field of view can be estimated directly from data obtained on scans across the hot (subsolar) planetary limb. Assuming that the response is circularly symmetric, and all evidence indicates this to be closely followed, the signature of a half space would also be symmetric. A plot of fractional energy derived from a Viking 1 IRTM scan across the sunlit limb of Mars is shown in Fig. 6 of CHASE_ETAL_1978. The alignment was determined using a 20.3-cm (8-in.) collimator to illuminate all four telescopes with a small source of high temperature blackbody radiation. Measurements were taken simultaneously in twenty-eight channels over a 1.5-mrad square grid pattern. For each channel, a parabolic ellipsoid was fit to data where the measured intensity was more than 10% of the peak intensity in that channel. The alignment of each telescope was ascertained by combining the center of response so determined for the seven channels in the telescope. This procedure allowed for the alignment of the four telescopes to be determined with an estimated precision of 0.1 mrad. The back of the secondary mirror of the B telescope was aluminized and used as the alignment reference for this procedure and for instrument alignment on the spacecraft. The instrument pointing direction was verified in the same manner just prior to planetary encounter using Mars as a 5-mrad diameter source and using the science platform motion to generate a 5-mrad spaced grid. The in-flight alignment is shown in Fig. 4 of CHASE_ETAL_1978. The offset angles listed below are from that figure. Instrument Pointing Relative to Scan Platform Axis (L-vector) V0-1 V0-2 CONE OFFSET ANGLE 0.07 deg 0.04 deg CROSS CONE OFFSET ANGLE 0.03 deg -0.06 deg DETECTOR DESCRIPTION The seven-element thin-film antimony-bismuth thermopile array used in the IRTM is shown in Fig. 7 of CHASE_ETAL_1978. The chevron arrangement was based on the need for uniform coverage irrespective of scan platform orientation; it also allowed the detectors to all be approximately the same distance from the telescope optic axis. In this application thermopiles were found to be better than other thermal detectors because they operate to dc and exhibit no 1/f noise. Thus, no optical chopper is needed. Also, no bias supply, another potential source of 1/f noise, is needed. Cooled quantum detectors were not practical, considering the duration and weight constraints of the Viking Mission. The array was made by evaporating the various components onto a sapphire film using photoetched masks for dimensional control. The film, about 200 nm thick, is supported by a sapphire disk. The film was made by anodizing aluminum foil and etching away the aluminum. The black circular dots in the figure are the sensitive areas overlaid with bismuth oxide smoke which has good ir absorptivity but low thermal mass. Characteristics of the detectors are Active area 7.E-4 cm^2 Number of junctions 6 Resistance 13.E3 ohm Time constant 80-100 msec Responsivity 130 V/Watt Detectivity (D*) 2.E8 cm_Hz^0.5_W^-1 To obtain full sensitivity the detectors must be evacuated. Therefore, during ground testing the detector packages were pumped down through a permanently attached manifold. At other times the detector packages were backfilled with xenon to protect the detectors while still allowing gross sensitivity checks. To avoid exposure to moisture during the long period prior to launch when the IRTM was mounted on the spacecraft and could not be sealed, the manifold was kept at a slight positive pressure by a continuous flow of high purity nitrogen. The manifold was opened to space by launch vehicle separation. ELECTRONICS DESCRIPTION The signal channels use a synchronous demodulation scheme to provide good stability and to avoid 1/f noise in the preamp. The input FET chopper is a full-wave type operating at 471 Hz. This and the center-tapped thermopile allow voltage doubling of the detector signal and noise and thus reduce the preamp noise contribution which otherwise would be significant. The differential input connection, while suffering a square root (2) noise disadvantage compared to single-ended input, provides excellent common mode rejection of chopper spikes and other input noise. Temperature dependence of the thermopile, about - 0.5%/degree C., is compensated by a thermister network external to the hybrid package. Preamp gain is adjustable with an external resistor. Following the half-wave synchronous demodulator is an integrate, hold, and reset circuit with an integrate time of 981 msec. The integrator serves as a low pass filter while the hold feature ensures spatial simultaneity of corresponding detectors in each telescope. After completion of sampling by the multiplexer, all channel hold circuits are reset to ensure independence of data samples. The IRTM analog signals, which have a range of +/- 6V, are digitized by the analog-to-pulse width converter and flight data subsystem (FDS) counter into +/- 2**9 levels, yielding 1023 data numbers (DN) which are nearly linear with radiance in each channel. The IRTM multiplexer consists of sixty-eight FET switches and a buffer signal amplifier. In addition to thirty-two data channels (twenty-eight active and four spare), thirty-two channels of engineering data are also sample. These include eight temperature measurements from thermisters located at four locations on the reference plate, the electronics module, and each of the three ir detector packages (telescopes A, B, C). Three power supply voltages and the pre-dc restore voltage of twenty-one channels (telescopes A, B, and C) are monitored. The pre-dc restore monitors are diagnostic to determine the presence of large thermal or detector offsets. The scan mirror is driven by a four-position stepper motor through a 50/1 gear reduction. A motor drive pulse duration of 40 msec allows a 90 degree mirror rotation in 2 sec. The mirror position is sensed by a two-bit encoder on the motor shaft; the contacts at the three desired positions are about half of the width of 1.8 degree mirror step. The motor stepping is controlled by the FDS using a comparison of the encoder readout with the desired position originating either from the FDS normal mode clock or direct ground command; the motor cannot be directly commanded. In addition to the restore which occurs automatically in the normal model when the mirror reaches the space position, restores can be ground commanded when the IRTM is in the fixed planet or fixed space mode; in either case housekeeping data are multiplexed into the data stream during the 1-sec restore period. Whenever the mirror reaches the reference position, the calibration lamp is turned on for the next two integration periods. The lamp is at full radiance throughout the second integration period, which is used for gain determination of the D telescope channels. In the fixed reference mode, science and housekeeping data are sampled alternatively. Instrument Operational Modes INSTRUMENT MODE DESCRIPTION Other than power on or power off, the only command options for the IRTM controlled the motion of the scan mirror. This mirror could be commanded to 3 positions, separated by 90 degrees; PLANET, SPACE, and REFERENCE. In the PLANET position, data were acquired in all 28 channels. In SPACE, the zero-radiance level was reset to near the bottom of the dynamic range. In the REFERENCE position, gain calibration data was obtained using both a grooved blackbody at ambient instrument temperature and a small lamp and hemispherical diffuser for the solar telescope. The REFERENCE position was also the 'safe' position, used when the instrument was turned off. The NORMAL mode sequenced the pointing to PLANET most of the time, with pointing to SPACE at intervals of 64 ICKs, and to REFERENCE at intervals of 256 ICKs. ICK is a one syllable acronym for 'incremental counter keeper' and represents 1.12 second duration, the basic time interval of IRTM operation. It required 2 ICKs for the mirror to move 90 degrees and settle; whenever the mirror was in motion the downlink data contained housekeeping information about instrument status and detector voltage levels. In NORMAL mode, ICKs 1 and 2 were housekeeping data while the mirror moved to space. ICK 3 was the space level before reset. During ICK 4, the electronics for all channels were reset so that the sensed radiance (meant to be the cosmic background level of essentially zero radiance) yielded a data number (DN) of a few. There was a filter on the reset so that the voltage change was only about 2/3 of the way to the space radiance level; this smoothed out the zero setting, but also meant that several cycles were required to recover from a serious drift. ICK 5 was still in the space position and yielded the DN response to space. Ick 6 and 7 were housekeeping data while the mirror moved to the reference position. ICK 8 and 9 were the DN response to the reference surface; only the second reading was used in calibration of the solar channel to allow the lamp filament to warm up completely. ICKs 10 through 13 were housekeeping data while the mirror moved to the planet position. The 57 ICKs 14-64 were planet data. The 7 ICK cycle to space, with reset of the zero-radiance DN level in all channels, was repeated each 64 ICKs, beginning on ICKs 65, 129, and 193, with motion directly back to the planet position for another 57 ICKs of planet data. A MODIFIED normal mode was also available, in which the only space view was the first of the cycle, followed by 243 ICKs of planet data. In NORMAL mode, ICKs 1 and 2 were housekeeping data while the mirror moved to space. ICK 3 was the space level before reset. During ICK 4, the electronics for all channels were reset so that the sensed radiance (meant to be the cosmic background level of essentially zero radiance) yielded a data number (DN) of a few. There was a filter on the reset so that the voltage change was only about 2/3 of the way to the space radiance level; this smoothed out the zero setting, but also meant that several cycles were required to recover from a serious drift. ICK 5 was still in the space position and yielded the DN response to space. Ick 6 and 7 were housekeeping data while the mirror moved to the reference position. ICK 8 and 9 were the DN response to the reference surface; only the second reading was used in calibration of the solar channel to allow the lamp filament to warm up completely. ICKs 10 through 13 were housekeeping data while the mirror moved to the planet position. The 57 ICKs 14-64 were planet data. The 7 ICK cycle to space, with reset of the zero-radiance DN level in all channels, was repeated each 64 ICKs, beginning on ICKs 65, 129, and 193, with motion directly back to the planet position for another 57 ICKs of planet data. In the FIXED SPACE mode, the mirror was commanded to the space position from wherever it had been, with 2 ICKs of housekeeping data if the mirror was not already in the space position. Thereafter the instrument reported space data until another command was received. In the FIXED REFERENCE mode, the mirror was commanded to the reference position; 0 to 5 ICKS of housekeeping data were possible while the mirror was in motion. Thereafter, the instrument reported radiance readings or housekeeping data on alternate ICKs untill the next command was received. In the FIXED PLANET mode, the first 0,2, or 4 ICKs could be housekeeping data while the mirror was in motion to the planet position. Thereafter the instrument reported planet data untill the next command was received. INSTRUMENT CALIBRATION DESCRIPTION. Relative spectral response of all channels was measured end to end using a Perkin-Elmer 16 U monochrometer with appropriate gratings and order filters. A globar at 1400 K was used in the 2- 25 micrometer range; shortward of 2.0 micrometers a tungsten source at 2700 K was used. The reference detector was a thermocouple for all but the 0.4-1.1 micrometer range, where a calibrated silicon photodiode was used. Out-of-band measurements were made by replacing the spectrometer grating with a plane mirror and ir materials having known cutoff and cuton wavelengths. Flux calibration of the IRTM was performed under a simulated space environment using a vacuum chamber operated typically at a pressure of 1.E-6 Torr. The IRTM was operated by means of a console which simu-lated the interfaces and functions of the spacecraft FDS. A minicomputer was used to provide all operational sequences and modes. Data were recorded on magnetic tape for subsequent computer processing. The calibration fixture consisted of two identical blackbodies, one located in front of the space port and maintained at liquid nitrogen temperature and the other in front of the planet port and adjustable in temperature from 77 K to 350 K; eleven settings from 140 K to 330 K were used. Blackbody temperatures were measured with platinum resistance thermometers having an absolute accuracy of +/- 0.1 degrees C. traceable to the National Bureau of Standards. The digitizer used in the test console provided ten times the resolution of the FDS digitizer, thus making the digitizing uncertainty during calibration insignificant compared to the noise. The calibration data thus produced are IRTM output in digitization level (DN) as a function of blackbody temperature. Radiometrically measured internal reference surface temperatures showed close agreement (+/- 0.5 degrees C.) with those measured independently with a thermister. The IRTM temperature was controlled by regulating the temperature of a mounting base plate and the thermal shield inside the vacuum chamber. Calibration was performed at 10 degrees C. spacing across the range of operating temperatures expected during flight. Typical IRTM channel response to scene brightness temperatures is shown in Fig. 8 of CHASE_ETAL_1978. The one-sample noise on the thermal channels is less than 1 DN except for the 15 micrometer channel where it is about 2.5 DN. The dynamic ranges of the surface thermal bands are based on temperatures expected for the Martian surface. The 300 K maximum chosen for the A telescope might be exceeded by midday summer temperatures, but temperatures above the 310-K limit of the B telescope should not be exceeded unless active volcanic areas were found; temperatures to 320 K and 330 K could be measured by the 9 micrometer and 7 micrometer bands. The 15 micrometer band dynamic range was set quite large as its resolution is noise limited rather than digitization limited. Telescope D channels were calibrated using a different method. The radiance source was a mercury-xenon lamp and narrowband filter centered at 0.896 micrometer with a bandwidth of 425 nm. The in-band radiance of the lamp was known by direct comparison with a standard lamp acquired from the National Bureau of Standards, using a silicon photodiode as a transfer standard. The relative spectral response measurements then allowed extension of the one-point absolute calibration to the entire passband. Gains for the D channels were set to give full scale for 75% of the diffuse reflection of solar irradiance at Mars average distance from the sun. Using integrals of the Planck function and the measured spectral response, the flux response of the IRTM is found to be close to linear in the thermal channels. The best fit quadratic functions, normalized to full scale, typically had constant and quadratic coefficients of 0.002 and 0.02, respectively. The solar band channels, which had much higher absolute flux levels at full scale, showed a decrease in response at high signal levels corresponding to a quadratic coefficient of 0.07. With the IRTM in the vacuum chamber, the instrument response was measured at four lamp currents. An additional series of wide band measurements utilizing a NBS standard lamp and a barium sulfate diffusing screen, in which only the lamp-screen distance was changed, was used to determine in detail the solar band nonlinearity. During spacecraft thermal-vacuum testing and in flight, a small drift of about 1-min duration was found to be induced when the scan mirror moved to the reference position in normal mode. This appears to be caused by the decrease in radiative heat loss from the instrument when the telescopes do not view space. The shape of this postreference drift was accurately determined during normal mode sequences when the spacecraft was well away from Mars, and this effect is removed in the data reduction. The change of the thermal state of the IRTM caused by large scan platform slew or planetary radiation near periapsis can introduce significant drifts of the zero-flux level. These shifts have a time constant of 1-2 min or longer, and their magnitude increases with inband wavelength and preamplifier gain. It is probably due primarily to very small temperature gradients induced in the detector packages as the general instrument temperature changes. A significant design feature of the IRTM is that the space DN level of each channel is measured immediately prior to and after the restore which occurs each minute in normal mode. A linear interpolation between these zero-flux DN levels is used in data decalibration. The remaining quadratic and higher order drift is generally negligible. IRTM Dynamic range and noise levels (Radiance in WATT_METER^-2_MICROMETER^-1) Channel A B C1 C2 C3 D Maximum Radiance .00203 .00312 .00319 .00198 .00634 135.84 Noise Level 1.3E-6 .9E-6 1.3E-6 .6E-6 3.E-6 .03 OPERATIONAL CONSIDERATION DESCRIPTION Most low and moderate resolution IRTM data were acquired through using 'box scans'. These were commonly acquired between 1-6 hours from periapsis, and utilized the scan platform to slew back and forth in cone angle (in the direction the IRTM chevron points) with small offsets in the same direction between these oscillating slews. Ignoring spacecraft motion, this pattern would generate bi-directional evenly spaced scans with the seven IRTM detectors. Spacecraft motion during the scan sequence, typically of 10-40 minutes duration, created some distortion in this otherwise uniform pattern. Typical resulting scans across the planetary surface are shown in Figure 3 of Kieffer et al., 1976. These scans were usually designed to extend slightly off the limb of the planet on at least one side. These 'planet port' off- planet data provided the best estimates of the zero radiance response of the instrument. When the spacecraft was near periapsis, the apparent motion of the planetary surface relative to the spacecraft was too rapid to allow oscillating slews. At these times, the instrument would simply 'stare' in one direction and use the spacecraft motion to sweep the detector pattern across the surface. These observations were usually acquired in Normal Mode, but occasionally Fixed Planet was used. At irregular times through the mission, 'phase function' observations were made. These involved using the two axis scan platform to follow one point on the ground as the spacecraft went from horizon to horizon relative to this surface point. In actuality, this sequence was acquired using a small number of discrete scan platform moves, allowing the instrument to 'stare' across a short stripe centered on the target point between slews. Such 'phase function' observations typically yielded about 10 different viewing geometries within a single sequence. These observations were particularly useful in determining the influence of the atmosphere. In preparation of the IRTM data set, all observations which were more than 1 1/2 degrees apparent angle above the nearest limb of the planet were deleted. DATA SET DESCRIPTION This data set contains the Infrared Thermal Mapping (IRTM) data of Mars acquired by the Viking orbiters. The database contains the time, geometry, and radiative parameters obtained by the IRTM instrument. Included in the database for each measurement are model temperatures of Mars; this model attempts to represent the average Martian response to the temporal variation of insolation. The reference 'Thermal and Albedo Mapping of Mars during the Viking Primary Mission', Journal of Geophysical Research, Vol. 82, No. 28, 1977, describes how the thermal emission and reflected energy measured by the various channels on the Infrared Thermal Mapper are used to reveal wide variations in the Martian surface properties. There is a description of how the surface thermal inertias are mapped and correlated with the surficial geologic units. The derivation of model and surface temperatures are described. The data set includes latitude and longitude for each measured point, as well as emission angle, solar incidence angle, the position of the sun, and other parameters that help to locate the position of the spacecraft relative to Mars. (For further information see the CONFIDENCE LEVEL NOTE). The parameters contained in the IRTM dataset as briefly described in the following table. The first 6 items occur once per ICK (sampling time), and the geometry is for the center of the IRTM chevron array (the instrument L-vector). Items 7 to 17 are present for each of the seven detector spots at each observation time, making a total of 83 items. Abbreviation col v scale Name unit Description print_format 1 TI 1 TIme FDS/4 1.12 seconds from sequence start F6.0 2 QU 1 QUAL - Quality, bit encoded, left=worst F6.0 3 PH 80 Phase degree Phase angle (at L-vector) . F6.2 4 DI 1 DIstan Kmeter Distance from S/C to target . F7.0 5 A_ 80 dT/dA Kelvin Derivative of model wrt albedo . F6.2 6 I_ 80 dT/dI Kelvin Derivative of model wrt inertia. F6.2 7 IN 80 INcidA degree Solar Incidence angle . F6.2 8 EM 80 EMissA degree Viewing angle (from vertical) . F6.2 9 LA 80 Lat. degree Latitude . F6.2 10 LO 80 Long. degree Longitude . F7.2 11 LI 80 Limb degree Limb angle off the planet . F6.2 12 HO 800 HOur day/24 solar hour angle; 12 = noon . F6.2 13 T2 80 T_20 Kelvin 20 micron brightness temper. . F6.2 14 T1 80 T_11 Kelvin 11 micron brightness temper. . F6.2 15 TC 80 T_C Kelvin 7,9 or 15 micron bright. temp. . F6.2 16 VI 10000 VisBri - Solar Brightness . F6.4 17 TM 80 T_mod Kelvin MOdel temperature . F6.2 Brightness temperature is the kinetic temperature of an ideal blackbody which, if observed with the spectral response of the instrument spectral channel in question, would yield the observed radiance. Visual brightness is the radiant energy measured in the IRTM solar band (0.3-3.0 microns), normalized to that expected for an ideal Lambert reflector observed under normal incidence at the current heliocentric range of the target body. Model temperature is the surface kinetic temperature calculated using a nominal set of parametric values for a smooth planet of uniform physical properties and for the latitude, time of day, and season of the observation. The model used for the Viking IRTM is described in the appendix of KIEFFER_ETAL_1977. Derivative of model with albedo is the differential dependence of model temperature upon the assumed surface bolometric bond albedo, at the nominal value of model thermal inertia. Derivative of model with inertia is the differential dependence of model temperature upon the assumed surface thermal inertia, at the nominal value of model bolometric bond albedo. Lambert albedo is the reflectance of a grey Lambert (ideal diffuse) smooth surface parallel to the target body geoid which would yield the observed reflected radiation in the absence of any atmosphere. The following parameters are computed by the XG software when requested: Minnaert albedo is the albedo computed presuming a Minnaert behavior, with the Minnaert exponent k equal to 0.95. A_m = Visual_Brightness * cos(incidence angle)**-k * cos(emission angle)**(1.-k). Phase corrected albedo is the albedo computed using a phase dependence derived from IRTM observations (See KIEFFER_ETAL_1977). A_g = A_l * (1.+0.000146*exp(0.075*phase_angle) where A_l = Visual_brightness / cos(incidence angle) Single-point thermal inertia is the thermal inertia which would yield the observed brightness temperature, based on the observed Lambert albedo, the model temperature and partial derivatives of model temperature with albedo and inertia, and an approximation of the non-linear dependence of model temperatures with respect to thermal inertia. If the observation was at night, so that no measure of albedo was made, the albedo is assumed to be the same as in the nominal thermal model. CONFIDENCE LEVEL NOTE Initially in the mission, orbit updates were done daily; later, they were done only once a week. Therefore at some times, the precise position of the spacecraft was not known accurately enough to calculate the position of the spacecraft around Mars. These timing offsets in the position of the spacecraft in turn introduced errors in latitude and longitude of the thermal observations. The user should be aware of these discrepancies in latitude and longitude, which affect primarily near-periapsis data late in the mission; see "Geometry Errors" below. The following is a description of event and error flags that are captured in the quality word definition for each 1.12 second interval (the time between successive measurements, known as ICK). QUALITY WORD DEFINITION Bit flags pertaining to the data at the corresponding time (ICK). Bits are numbered from least significant to most significant, 0 to 15. Bits 0 through 9 represent conditions at the time of observation, and do not necessarily imply poor quality. Bits 11- 14 form a sequence of data quality indicators, from good (all 4 bits off) to terrible (bit 14 on). Thus the word itself may be used in an algebraic quality test. (QUAL < 2048 = 2**11 if data must be of best quality) VAL BIT 1 0 DC restore has just occurred in the IRTM instrument 2 1 Scan platform is in motion 4 2 Telemetry was used for pitch, roll and yaw computations 8 3 Telemetry was used for cone and clock computations 16 4 Image motion compensation geometry correction required 32 5 Geometry corrected unless bit 14 is on 64 6 Cone slew rate less than or = 1/8 deg/ICK (.1 deg/sec) 128 7 Clock slew rate GE 1/8 deg/ICK (.1 deg/sec) 256 8 Cone slew direction negative 512 9 Clock slew direction negative 1024 10 VO-2 anomaly correction applied (orbits 55-84) (bit 13 also set) or quadratic fit used for space values (VO-2 orbits greater than 84) 2048 11 Some channel(s) despiked 4096 12 Uncertainty < 8 DN for some channel(s) but missing space value or high drift rate 8192 13 VO-2 encoder dc restore on reference plate (visual band data probably OK) 16384 14 Uncertainty 7 DN or greater for some channel(s) due to missing space value or high drift rate or imaging sequence with only 1 space look 32768 15 Totally bad data at this ICK - do not use Geometry Errors Due to Uncertain Timing: Early and late during the Viking mission, orbital solutions based on the tracking telemetry were determined every few days. During VO-1 revolution 175-603 and VO-2 revolutions 118-521, orbital solutions were often separated by a week or more. Because there is significant irregularity in the Martian gravitional field, these irregularities could slowly influence the orbit of the Viking spacecraft in unpredictable ways. The primary influence was in the period of the orbit, resulting in uncertainty as to exactly where the spacecraft was along its orbit at any specific time. These uncertainties were as large as 75 seconds in the worst case. Far from periapsis, these timing uncertainties were not of major significance because the spacecraft velocities were low and the projected fields of views on the planet were large. However, near periapsis, the IRTM field of view could move across the surface equivalent to its full width in as little as one second. Thus, when there was a large timing error, the computed ground intercept locations could be in error by many fields of view. In the worst case, these positions may be in error by up to 200 km. When the magnitude of this problem was discovered, the SEDR (geometry calculations) for the imaging instrument was rerun with revised orbit solutions. However, it was impractical to regenerate the IRTM SEDR and these errors have not been corrected. There was an attempt by the navigational team to estimate the magnitude of the timing error for both Viking spacecraft for those revolutions through the affected part of the Viking mission. This is described in the 1980 April 14 memo by Frank Palluconi, which contains estimates of the magnitude of the error for each revolution. Hugh Kieffer has a copy (the sole surviving copy?) of this memo. A direct determination of the timing offset can be made from the IRTM data alone in those instances when thermal patterns can be unambiguously identified with surface features. Since the dominant geometric error is in time, maps of thermal patterns (typically as contours of observed temperature minus the calculated standard model temperature) can be slid across the cartographic map parallel to the subspacecraft track (if the instrument was in fixed planet mode, this is simply sliding the IRTM trace along its own path) until the thermal and cartographic features are aligned. Because there are small gaps in the IRTM coverage every 64 ICKs, the amount the IRTM pattern must be shifted to agree with the surface morphology can be scaled directly into a timing offset in seconds. This has been done for a variety of high resolution scans across Arsia Mons (by Jim Zimbelman) and for many scans across Valles Marineris (by David Paige and Hugh Kieffer). A set of known offsets is slowly accumulating.