MINI-TES Calibration Report

DATE	SECTIONS CHANGED	REASON FOR CHANGE	REVISION
06/15/04	All	First draft	v 1.0
08/15/04	Cover pages, Section 6	Cover pages and contents added;	v1.1

1. Instrument Overview

The Miniature Thermal Emission Spectrometer (Mini-TES) (Figure 1) is a Fourier transform interferometer/spectrometer intended to provide remote determination of the mineral composition of the rocks and soils, surface temperature, and atmospheric properties in the scene surrounding the Mars Exploration Rovers. The Mini-TES collects high resolution infrared spectra over the wavelengths where distinctive vibrational spectral bands are best observed, providing a direct means of identifying crystal structure, and hence mineralogy, of all geologic materials including silicates, carbonates, sulfates, phosphates, oxides, and hydroxides. Mini-TES will also measure the temperature of the lower atmospheric boundary layer, and provide information on suspended dust, water-ice, and water vapor opacity. The specific scientific objectives of the Mini-TES investigation are to:

1) determine the mineralogy of rocks and soils

2) determine the thermophysical properties of selected soil patches

3) determine the temperature profile, dust opacity, water-ice opacity and water vapor abundance in the lower boundary layer of atmosphere

The Mini-TES covers the spectral range from 5 to 29.5 µm (1997.06 to 339.50 cm^-1) with a spectral sampling of 9.99 cm^-1. A 6.35-cm diameter Cassegrain telescope feeds a flat-plate Michelson moving mirror mounted on a voice-coil motor assembly that provides the optical path difference necessary for the Michelson interferometer while achiving excellent tilt performance. A single deuterated triglycine sulfate (DTGS) uncooled pyroelectric detector gives a spatial resolution of 20 mrad; a actuated field stop reduces the field of view to 8 mrad. For a complete description of the Mini-TES optical, mehanical, and electronic design specifics, see Christensen et al., [2003].

Mini-TES is mounted within the Rover's Warm Electronics Box and views the terrain using its internal telescope looking up the hollow shaft of the Pancam Mast Assembly (PMA) to the fixed fold mirror and rotating elevation scan mirror in the PMA head located ~1.5 m above the ground. The PMA provides a full 360° of azimuth travel and views from 30° above the nominal horizon to 50° below. The overall envelope size of Mini-TES is 23.5 x 16.3 x 15.5 cm and the mass is 2.40 kg. The average power consumption is 5.6 W. Mini-TES I was delivered to JPL on Aug. 16, 1999, initially for integration into the 2001 Mars Lander; Mini-TES II was delivered on June 7, 2002. A summary of the instrument characteristics is given in Table 1.

Table 1. Mini-TES Design and Performance Parameters

Parameter	Description
Spectral Range	339.50 to 1997.06 cm^-1 (5.01 – 29.45 µm)
Spectral Sampling Interval	9.99 cm^-1
Field of View	8 and 20 mrad
Telescope Aperture	6.35 cm diameter Cassegrain
Detectors	Uncooled Alanine doped Deuterated Triglycine Sulphate (ADTGS) Pyroelectric detector
Michelson Mirror Travel	-0.25 – 0.25 mm
Mirror Velocity (physical travel)	0.0325 cm/sec
Laser Fringe Reference Wavelength	978 ± 2 nanometers
Interferometer Sample Rate	645 samples/sec
Cycle Time per Measurement	2 seconds (1.8 seconds Michelson mirror forward scan; 0.2 second retrace)
Number of Scans to Achieve 400 SNR at 1000 cm^-1; 270 K Scene Temperature; 0° C Instrument temperature	2 (20 mrad); 80 (8 mrad)
Number Samples per Interferogram	1024
Number Bits per Sample - Interferogram	16
Number Samples per Spectrum	167
Number Bits per Sample – Spectrum	12
Dimensions	23.5 x 16.3 x 15.5 cm
Mass	2.40 kg
Power	5.6 Watts (operating), 0.3 Watt (daily average)
Operational Temperature Range (Instrument Temperature)	Survival and Operability -45, +50C; Performance within Spec -10, +30C

The Mini-TES spectrometer provides data to the rover computer at a fixed rate of one interferogram every two seconds (1.80 second Michelson scan, 0.20 second retrace) whenever it is powered. The rover flight software performs the Fast Fourier Transform (FFT) processing, spectral co-adding, and lossless compression. The FFT algorithm transforms the raw interferograms, containing up to 1120 16-bit interferogram samples, into 16-bit spectra during the 200 ms retrace period. The Mini-TES Zero-Path-Difference (ZPD) algorithm locates ZPD using a selectable choice of the positive-, negative-, or mid-point of the interferogram peak-to-peak amplitude, and selects the central 1024 points of the interferogram. The FFT generates a spectrum of 512 16-bit samples, from which the 167 with useable response are selected. Co-addition of two spectra reduces the data rate by an additional factor of two, resulting in an average data rate of approximately 400 bits per second.

2. Mission Measurement Requirements

2.1 Science Requirements

The Mini-TES science objectives can be translated into a specific set of measurement requirements. Mineralogic mapping has three measurement requirements: (1) radiometric accuracy and precision necessary to uniquely determine the mineral abundances in mixtures to within 5% absolute abundance; (2) spectral resolution sufficient to uniquely determine the mineral abundances in mixtures to within 5% absolute abundance; and (3) spatial resolution of ≤25 cm at 10 m distance (25 mrad) necessary to resolve and identify individual rocks 0.5 m in size or larger in the rover near field. The thermophysical objective requires determining the thermal inertia to within ±10 J-m^-2-K^-1-s^-1/2. The determination of atmospheric temperature profiles, aerosols, water vapor, and condensates has two measurement requirements: (1) radiometric accuracy and precision necessary to determine the opacities of atmospheric dust and ice to ±0.05 and temperature to ±2 K; and (2) spectral resolution sufficient to uniquely identify dust, water-ice, water-vapor, and sound the atmosphere, and monitor their physical and compositional properties.

2.2 Radiometric Requirements

The radiometric requirements for determining mineral abundances to 5% accuracy in mixtures depend critically on the specific minerals in question. However, in general it is necessary to resolve the relative depths of mineral absorption bands to ~2% and the absolute mineral band depths to ~10% of their typical band depth (0.15 emissivity). This produces a relative (precision) emissivity requirement, stated as the Noise Equivalent Delta Emissivity (NEDe) of 0.003 and an absolute requirement of 0.015. At typical daytime temperatures of 270 K and a reference wavenumber of 1000 cm^-1 (10 µm), these requirements correspond to an absolute spectral radiance accuracy of 9 x 10^-8 W cm^-2 sr^-1 /cm^-1 and a 1-sigma (s) radiometric precision, or noise equivalent spectral radiance (NESR), of 2 x 10^-8 W cm^-2 sr^-1 /cm^-1. The accuracy and precision necessary to sound the atmospheric temperature profile with a <1 K temperature error and to determine the opacity of dust and ice aerosols are comparable to the mineral requirements. Determination of thermal inertia to ±10 J-m^-2-K^-1-s^-1/2 requires an absolute accuracy of the surface temperature determination to within ±2 K for typical night (170 K) and day (270 K) temperatures.

For calibration Mini-TES views two beam-filling blackbody calibration targets whose temperatures are well known. The primary calibration target is placed within the Pancam Mast Assembly (PMA), where it can be viewed by the instrument and where it is protected from martian dust. The secondary target is on the rover deck. Because the outer surface of the PMA is painted white and the target on the rover deck is black, the temperature difference between the targets is expec ted to be substantial (>20° C). The absolute accuracy of 1.5% can be achieved using internal and external calibration targets with an emissivity of >0.98, known to within ±0.005 over the Mini-TES spectra range, and absolute knowledge of the target temperature to within ±0.2° C [Ruff et al., 1997].

2.3 Spectral Requirements

The spectral requirements are determined by the width and position of the key spectral features in the materials and mixtures of materials. Determining mineral abundances within mixtures to 5% requires the capability to sample mineral spectral band to ~10% of their width. For geologic materials, the full width at half maximum (FWHM) width of typical spectral bands is ~100 cm^-1, and the minimum in the Si-O stretching band undergoes a shift of over 150 cm^-1 for differing crystal structures from low-to high-silica content, and an offset of up to 500 cm^-1 from the fundamental C-O, S-O, and P-O stretching bands [Farmer, 1974; Salisbury et al., 1992; Christensen et al., 2000a]. Therefore, a spectral sampling of 10 cm^-1 is sufficient for identifying key minerals and deconvolving mineral mixtures [Ramsey and Christensen, 1998; Feely and Christensen, 1999; Hamilton and Christensen, 2000].

2.4 Field of View and Pointing Requirements

Mini-TES views the terrain around the rover using its internal telescope looking up the hollow shaft of the PMA to a fixed fold mirror and a rotating elevation scan mirror located in the PMA head. The PMA provides a full 360° of azimuth travel, but limitations on the size of opening in the mirror assembly atop the PMA restrict the total elevation range to 80°. The mirrors were oriented to provide a 30° elevation view above the nominal horizon, allowing observation of the sky, and 50° below the nominal horizon, allowing the terrain with ~2 m of the rover to be observed. The PMA's mirror assembly is located ~1.5 m above the ground. The offsets between the Mini-TES and the Pancam were measured pre-flight to within 2 mrad, allowing data from these two instruments to be co-registered to within a single Mini-TES pixel.

The Mini-TES integration time required to achieve a given signal-to-noise ratio (SNR) is inversely proportional to the fourth power of the angular resolution. This strong dependence drove the selection of two spatial resolution modes: a 20-mrad (full width half maximum (FWHM)) mode for rapid surveys of large areas, and an 8-mrad mode for detailed study of limited high-priority regions. The Mini-TES telescope aperture size results from a trade between the integration time required to obtain an adequate SNR (large telescope), and the mass of the PMA and the shadowing of the rover's solar array (small telescope). A telescope aperture of 6.35 cm (2.42 inches) was selected, resulting in a PMA that is manageable and achieves the required SNR under all daytime conditions in the 20-mrad mode by co-adding two spectra. The required SNR can be obtained with just one spectrum for the warmest mid-day conditions.

The key Mini-TES requirements, along with the values achieved in the flight instruments, are summarized in Table 2.

Table 2. Mini-TES Performance Requirements and Actual Performance

Parameter	Requirement	Actual
Mineral Determination
Precision (NEDe)	±0.003	±0.003
Absolute Accuracy (NEDe)	0.015	0.004 to 0.008
Radiometric Precision (NESR)	±2x10^-8 W cm^-2 sr^-1 /cm^-1	±1.8x10^-8 W cm^-2 sr^-1 /cm^-1 between 450 and 1500 cm^-1; instrument temperature 10 to 30 °C
Radiometric Accuracy	9x10^-8 W cm^-2 sr^-1 /cm^-1	5x10^-8 W cm^-2 sr^-1 /cm^-1
Spectral Sampling	10 cm^-1	9.99 cm^-1
Field of View	Azimuth: 0-360°	0-360°
	Elevation: -50 to +5° from horizon	-50 to +30°
Spatial Resolution	25 mrad	17.5 and 6.9 mrad FWHM

Atmosphere Studies
Radiometric Precision (NESR)	±2x10^-8 W cm^-2 sr^-1 /cm^-1	±1.8x10^-8 W cm^-2 sr^-1 /cm^-1 between 450 and 1500 cm^-1; instrument temperature 10 to 30 °C
Radiometric Accuracy	9x10^-8 W cm^-2 sr^-1 /cm^-1	5x10^-8 W cm^-2 sr^-1 /cm^-1
Spectral Sampling	10 cm^-1	9.99 cm^-1
Field of View	Azimuth: 0-360°	0-360°
	Elevation: -5 to +30° from horizon	-50 to +30°
Spatial Resolution	40 mrad	17.5 and 6.9 mrad FWHM

Thermophysical Properties
Temperature	2 K absolute	0.4 (day) to 1.5 K (night)

3. Pre-Launch Calibration Tests

3.1 Overview

The initial Mini-TES calibration and test was performed at SBRS prior to delivery to JPL, and a subset of these tests was performed on the integrated Mini-TES/PMA assembly. The objectives of these tests were to determine: (1) the field-of-view definition and alignment; (2) the out-of-field response; (3) the spectrometer spectral line shape and spectral sample position; and (4) the spectrometer radiometric calibration.

Bench-level testing of the Mini-TES instrument was performed at SBRS in two phases. The first phase consisted of piece-part and system-level testing of the spectral performance of each sub-section under ambient conditions. The second phase consisted of field of view and out-of-field tests conducted before and after vibration and thermal-vacuum testing to determine and confirm the instrument field-of-view and alignment. Mini-TES I was operated for a total of 166 hours and Mini-TES II was operated for 594 hours at SBRS prior to initial delivery to JPL.

The Mini-TES spectrometer, without the PMA, was tested and calibrated in vacuum at SBRS at instrument temperatures of -30, -10, 10, and 30 °C. A matrix of calibration tests were performed viewing two precision calibration reference blackbody standards, one set at 223 K, 243 K, 263 K, and 283 K. while the second was varied at temperatures of 145 K, 190 K, 235 K, 280 K, and 325 K °C. The Mini-TES/PMA systems were radiometrically calibrated in 6 mbar of nitrogen at instrument temperatures of -30, 0, and 30 °C over a range of calibration blackbody temperatures (Table 3). These tests determined: (1) the emissivity and effective temperature of the internal reference surface; (2) the instrument response function and its variation with instrument temperature; (3) the absolute radiometric accuracy; (4) the spectrometer noise characteristics; and (5) the spectrometer gain values.

Table 3. Mini-TES Calibration Temperatures

	Instrument Temperature
	-30 °C		0 °C		30 °C
Mini-TES I	BCU-1	BCU-2	BCU-1	BCU-2	BCU-1	BCU-2
	223K	145 K	253 K	145 K	190 K	325 K
		190 K		235 K
		235 K
		280 K
		325 K

Mini-TES II	BCU-1	BCU-2	BCU-1	BCU-2	BCU-1	BCU-2
	223 K	145 K	253 K	145 K	283 K	145 K
		235 K		235 K		235 K
		325 K		325 K		325 K

3.2 Calibration Targets

The two precision calibration reference blackbody standards (BCU-1 and BCU-2) used in the thermal vacuum calibration were identical, 7.25" diameter, 15° half-angle cones machined at ASU and assembled and painted at SBRS. These targets are shown in Figure 2 in the thermal vacuum chamber at SBRS before the installation of the thermal blankets. Each blackbody was instrumented with two pairs of platinum thermistors, with one pair placed near the apex of the cone and the second pair approximately half-way between the apex and the opening of the cone. These thermistors were calibrated prior to shipment from the manufacturer to an absolute accuracy of 0.1 °C (±0.5% resistance). The digitization of the reference blackbody telemetry points varies from ~0.01 °C at -190°C to 0.02 at 35 °C [Christensen, 1999; Christensen et al., 2001a]. The temperature stability was within these digitization levels over >1 minute time periods. The front and back temperatures agree to within 0.7 °C for cold temperatures, increasing to 1.5 °C for hot temperatures.

The Mini-TES internal and external flight calibration targets are formed with parallel groves machined with 15° inclined surfaces with a depth of 1.73 mm and a spacing of 1.2 mm. The reference surface was machined from aluminum alloy and painted with Aeroglaze Z302 blank paint to a thickness of ~6 mil.

The external target will be exposed to the martian atmosphere and will collect airfall dust during the mission. Dust build up can affect the emissivity of this target, although the geometric (e.g. V-groove) character provides a high emissivity regardless of the nature of the target surface. Immediately following landing both targets will be observed and used to determine the instrument response function. It is expected based on TES data [Christensen et al., 2001a] that the Mini-TES instrument response function will not vary significantly from its pre-launch value. In this case the pre-launch or initial landed instrument response function can be used throughout the mission, adjusted for instrument temperature, and adequate calibration can be obtained from a single point calibration using only the internal target.

Each calibration target is instrumented with two platinum thermistors that are bonded to the underside. These thermistors were delivered from the manufacturer with a measured absolute accuracy of better than 0.1° C for temperatures from -130° to 110°. Resistances from the flight calibration target thermistors are digitized through the rover telemetry and converted to temperature. The digitization of the resultant temperatures is ~0.02° C at 0° C. The temperature differences between the two independent temperature readings for each calibration target were determined during JPL thermal vacuum tests to be <0.5° C over the full range of operating temperatures.

Immediately prior to launch one the internal reference thermistors in the MER-A PMA head failed. It was determined that the thermistors had been bonded incorrectly prior to delivery of the PMA to JPL in a manner that enhanced fracturing of the thermistor element. In response, both PMAs were disassembled and one of the two thermistors on each assembly was replaced. This action preserved the pre-launch calibration for the remaining thermistor, while providing improved reliability.

3.3 Field of View

Field of view characterization data were acquired at SBRS for the Mini-TES spectrometer using a precision collimator. Thermal and visual sources were projected through a 1 mrad wide, 40 mrad long slit into the Mini-TES aperture [Christensen, 1999]. The Mini-TES was manually rotated to move the slit at 1-mrad spacing across the focal plane; 31 points were measured from -15 to +15 mrad in elevation, 31 points were measured from -15 to +15 mrad in azimuth. The results from the final pre-shipment bench alignment tests for Mini-TES I in both 20- and 8-mrad mode in elevation and azimuth are shown in Figure 3. The FWHM of the Mini-TES 20 mrad field of view mode is 17.5 mrad in azimuth and elevation. The FWHM of the 8 mrad field of view mode is 6.6 mrad in azimuth and 6.9 mrad in elevation. After the Mini-TES instruments were delivered to JPL, it was aligned to the PMA and tested to verify that the view is not vignetted by any elements of the PMA over the expected range of rover tilt angles.

The near-field out-of-field response was measured at the bench level using a 2.4 mrad square slit that was stepped in 2.45 mrad increments from -14.7 to +14.7 mrad in azimuth and -19.6 to +19.6 mrad in elevation. The far-out-of-field response was determining over an extended area using a 40 x 32 mrad slit that was stepped in increments of 32.7 mrad in azimuth and 40 mrad in elevation from -196 to +196 mrad in azimuth and from -200 mrad to +200 mrad in elevation. No measurable out-of-field energy was detected in either test.

3.4 Sample Position and Spectral Line Shape

In an ideal interferometer with an on-axis point detector, the spectral samples are uniformly distributed in wavenumber and the full-width, half-maximum (FWHM) of each sample is determined by the optical displacement of the Michelson mirror. The Mini-TES uses a laser diode with a line at 0.978 µm in the visible interferometer to sample the IR interferometer. The ideal sample spacing of the interferometer is given by:

(9)

where Npts is the number of points in the FFT and is equal to 1024 for the Mini-TES.

4. Radiometric Calibration

4.1 Calibration Method

For each observation the Mini-TES acquires an interferogram signal, measured in voltage, that is transformed to a signal as a function of frequency. This signal is given at each wavenumber (n; subscripts omitted) by:

V_measured = {(R_emitted + R_reflected) - R_i} * f (1)

where:

V_measured is the fourier-transformed voltage signal generated by the Mini-TES looking at the scene

R_emitted is the emitted radiance of the surface (W cm^-2 sr^-1 /cm^-1)

R_reflected is the background radiance reflected off of the surface (W cm^-2 sr^-1 /cm^-1)

R_i is the radiance of the instrument (W cm^-2 sr^-1 /cm^-1)

f is the instrument response function (V / W cm^-2 sr^-1 /cm^-1)

High emissivity targets (>0.995) were used for all of the calibration tests and the R_reflected term can be ignored. Solving for the emitted radiance of the scene gives:

(2)

During the system-level Mini-TES/PMA thermal-vacuum tests, observations were acquired of the two precision reference blackbody standards, one cold and one hot, and the flight external and internal calibration targets. The emitted radiance (R_emitted) of each target is given by eB, where e is the emissivity of each target and B is the Planck function radiance at the target temperature. The relationships between the measured instrument signal (V) and the instrument and target radiance for the cold, hot, internal and external reference views are given by:

V_cold = (e_coldB_cold - R_i) * f (3)

V_hot = (e_hotB_hot - R_i) * f (4)

V_{internal ref} = (e_{internal ref}B_{internal ref} - R_i) * f (5)

V_{external ref} = (e_{external ref}B_{external ref} - R_i) * f (6)

The instrument response function has been shown through testing on the TES [Christensen et al., 2001a] and Mini-TES to be independent of signal magnitude, but is a function of instrument temperature. The spectra from each target were acquired over a relatively short period of time (<3 min) under highly stable conditions with a constant instrument temperature, so R_i and f are assumed to be constant in Equations 3 thru 6. The temperature of the external targets and reference surface was determined to be constant to within 0.1 °C using the thermistors located in or on each target surface, and the average temperature of each target over the time interval of data collection was used in the calibration.

Equations 3 and 4 give two equations and four unknowns (e_cold, e_hot, R_i, and f). The calibration blackbodies have equal emissivities (e_cold = e_hot) that are >0.995 [Christensen et al., 2001a; Christensen, 1999], and are assumed to be unity. With this substitution Equations 3 and 4 give:

(7)

and

(8)

4.2 Flight Calibration Target Emissivity and Temperature

The emissivity and temperature of the internal and external flight calibration targets were determined during thermal vacuum testing by computing the calibrated radiance from each target using Equation 2, together with the R_i and f values determined by viewing the calibration reference standards. A Planck function was fit to the target calibrated radiance and the temperature of this best-fit function was assumed to be the kinetic temperature of the target. The target emissivity was determined using the ratio of the measured calibrated radiance to a blackbody at this kinetic temperature. The emissivity of the external target was also measured in the ASU Thermal Spectroscopy Lab. The temperature sensors on both targets were calibrated by comparing the derived kinetic temperature with the telemetry reading for each target. Analysis of these data is on going.

4.3 Radiometric Performance

The instrument response varies with instrument temperature due to changes in detector response and interferometer alignment with temperature. The variation of instrument response with temperature for is shown in Figure 4 for the system-level calibration tests done at JPL for Mini-TES I. The actual instrument response function will be determined in flight using observations of the internal and external calibration targets and Equations 7 and 8.

The noise equivalent spectral radiance (NESR) of the Mini-TES was determined in thermal vacuum testing by converting the standard deviation in signal to radiance using the instrument response function. This approach produces an upper limit to the noise levels (a lower limit to the SNR) because other sources of variance may be present that are not related to the instrument itself. The most likely of these are variations in the target radiance due to minor variations in target temperature.

Figure 5 gives a representative NESR for Mini-TES I at instrument temperatures of -30, 0, and 30 °C observing the internal reference surface. The 1-s radiance noise level of an individual spectral sample in a single Mini-TES spectrum varies from 1.5 to 3.5 x10^-8 W cm^-2 sr^-1 /cm^-1 for the central wavenumbers from ~450 to 1500 cm^-1, increasing to ~6 x10^-8 W cm^-2 sr^-1 /cm^-1 at shorter (300 cm^-1) and longer (1800 cm^-1) wavenumbers (Figure 5). For the planned observing scenario in which two spectra are summed, these values reduce to 2.1 and 4.2 x10^-8 W cm^-2 sr^-1 /cm^-1 respectively. As shown in Figure 4 the instrument response function decreases with decreasing temperature. However, the noise also decreases with temperature, so that there is only an approximately 35% increase in the noise level between the highest (30° C) and lowest (-30 °C) instrument temperatures.

The absolute calibration of the Mini-TES spectrometer was determined during thermal vacuum testing by computing the instrument response and instrument radiance using observations of the two precision calibration reference blackbodies. These values were used to convert observations of a separate V-groove target to calibrated spectral radiance using Eq. 2. An example of the comparison between the V-groove target calibrated radiance and the blackbody radiance computed using the measured V-groove temperature is given in Figure 6 for Mini-TES II at an instrument temperature of -10°C. Figure 6b shows the difference between the calibrated radiance and Planck function radiance at the measured V-groove temperature. At low scene temperatures (<220 K) there are errors of up to 2 x10^-7 Watt cm^-2 sr^-1 /cm^-1 when comparing the derived calibrated radiance to the radiance calculated using the measured V-groove target temperature. However, the errors are less than 5 x10^-8 Watt cm^-2 sr^-1 /cm^-1 when compared to the best-fit derived V-groove target temperature. Based on the quality of the best-fit Planck curves, we estimate that the primary error is in the calibration of the V-groove thermistors at low temperatures [Christensen, 1999; Christensen et al., 2001a], and conclude that the absolute radiance error of the Mini-TES instrument is <5 x10^-8 Watt cm^-2 sr^-1 /cm^-1.

Systematic errors in radiance can occur in the calibration process where noise in the internal and external calibration target observations is mapped into the calibrated scene spectra as a function of the instrument temperature and the temperature difference between the scene and the instrument. This noise will be reduced by acquiring and averaging ~5 consecutive observations of both calibration targets.

The Mini-TES temperature is expected to vary from -10 to +30° C over the course a day, with most surface composition data collected at instrument temperatures >10° C and scene temperatures >270 K. For this combination of temperatures, and assuming the nominal operational mode of the Mini-TES in which two spectra are collected for each observation, the radiometric precision is ±1.8 x10^-8 W cm^-2 sr^-1 /cm^-1 between 450 and 1500 cm^-1. The absolute error, averaged over the wavenumber range where the scene temperature will be determined (1200-1600 cm^-1), will be ~1 x10^-8 Watt cm^-2 sr^-1 /cm^-1 (Figure 6b). The worst-case sum of these random and systematic radiance errors correspond to a best-fit absolute temperature error of ~0.4 K for a true surface temperature of 270 K, and ~1.5 K for a surface at 180 K. This temperature error is mapped into a smoothly varying offset in the emissivity spectrum that varies from 0.001 at 400 cm^-1, to a maximum of 0.005 at ~1000 cm^-1, to essentially 0 at 1400 cm^-1 for a scene temperature of 270 K. This subtle curvature has a negligible effect on the derived surface composition.

5. Rock Calibration Target Observations

A rock calibration target containing 14 samples was constructed to provide mineral detection calibration and validation data for each of the MER science instruments [Morris and Graff, 2002; Squyres et al., 2003]. The rocks ranged in size from 6.5 x 7.2 cm to 14.8 x 14.8 cm and were cut and polished. Both of the Mini-TES instruments viewed this target during system-level thermal vacuum testing at JPL. The target was placed 2.5 m from the Mini-TES telescope aperture, resulting in a projected, out-of-focus Mini-TES field of view of ~11.3 cm in size. A 41 x 33 raster image was obtained by Mini-TES I of roughly 3/4 of the target; test anomalies and time limitations only permitted the last column of rock targets to be observed with a single Mini-TES elevation scan.

The calibrated spectral radiance from the rock targets was determined using the instrument response and instrument radiance obtained viewing the two calibration reference blackbodies immediately prior to the rock target observations. This spectral radiance consists of the emitted radiance from the rocks and the radiance emitted by the environment and reflected off of the rocks (Eq. 1). The reflected component was removed by approximating the emitted radiance from the environment (B_env) assuming Planck function emission at the temperature of the vacuum chamber walls (~0° C), and simultaneously solving for the emissivity (e_rock) and reflectivity (1-e_rock) of each rock. With these substitutions Eq. 1 becomes:

V_rock = {(e_rockB_rock + (1-e_rock)B_env) - R_i} * f (10)

Solving for the emissivity of the rock gives:

(11)

A similar procedure will be performed at Mars using the measured or modeled atmospheric downwelling radiance as e_envB_env and solving for the emissivity of the surface rocks and soils.

Examples of the rock emissivity spectra with the environmental radiance removed are given in Figure 7. These spectra represent averages of ~5 Mini-TES spectra acquired at an instrument temperature of 0° C viewing rock targets heated to ~35° C (310 K). Figure 7 also shows the spectra of the same rock targets measured in the ASU Thermal Spectroscopy Laboratory. This figure shows the excellent agreement between the laboratory and flight instrument spectra, verifying the calibration and background radiance algorithms used for the Mini-TES instrument, and demonstrating the laboratory quality of the spectra that will be returned from Mars.

Each of the rock targets was analyzed using the linear deconvolution method developed and used extensively for TES data analysis [Adams et al., 1986; Ramsey and Christensen, 1998; Christensen et al., 2000b]. A subset of 57 mineral spectra from the ASU thermal emission spectral library [Christensen et al., 2000a] were selected for deconvolution of the rock target spectra. In addition, a high-silica glass spectrum was obtained along with additional goethite, magnesite, hematite, and chert samples. Two other endmembers used in the subset were: (1) a linear slope, to account for the variable brightness temperatures of the rock and rock mounting surface contained within the Mini-TES field of view; and (2) blackbody (emissivity =1.0) to account for spectral contrast differences between the particulate endmembers in the ASU library (700-1000 mm) and polished rock slabs. Finally, the average emissivity spectrum from the first rock target was used as an endmember to deconvolve the mixed spectra from nearby samples due to a lack of mineral spectra that modeled this target well (corundum and blackbody were the best fit to this target, with RMS errors of ~12%). The extreme ends of each spectrum (339-399 cm^-1 and 1727-1997 cm^-1) were excluded from the deconvolution, due to lower SNR in these spectral regions in both the measured and endmember emissivity. The band range used for each target was varied, based on model fits. Although different subsets of the endmember library were found to provide better fits for each target, the results reported here used the same 64-endmember set for each rock.

The derived mineral abundances for each rock target are given in Table 4, normalized to remove the blackbody component. The individual model results have been grouped into the sum of each major mineral group. The slope endmember was used at the 1-2% level for all targets except Rock 4 (8%) and Rock 5 (3%). At the distance to the rock target the Mini-TES field of view was too large to completely resolve any individual target, and the emissivity includes components from surrounding targets and the background black paint. Examples of the measured and best-fit mineral mixture are shown in Figure 8. Future work will compare these analyses with mineral abundances measured using laboratory techniques as well as the composition of these rock targets determined from each of the other MER instruments.

Table 4. Mini-TES Rock Target Results

	Mineral Abundance (%)*
Rock Number	1	2	3	4	5	6	7	8	9	10	11	12	13
Feldspar		44	76	2		81	9	69	51	10	39		55
Pyroxene		2	5		8	6		2	3	1		2	17
Amphibole		2						16	3	11	39
Olivine		12	6				10		6		9		9
Silica				9	61				4	4		36
Oxide	100	12	9	61	16	9	13	6	7	54	1	58	8
Carbonate		1	1	1	2	1	67	1	1	16	2	4	1
Sheet-silicates		12	2		10	1		2	2	1	5	1	7
Apatite		1		12					5
Al2SiO5				5				1
Pyroxenoid											2
High-Si Glass									16
Garnet		2	1			1					2		3
Epidote		6				1		4	3	2
Sum	100	94	100	90	97	100	99	101	101	99	99	101	100
Blackbody	0	-97	-45	-15	28	-136	-55	-94	19	-15	-85	-169	-76
RMS Error	0.000	0.714	0.470	0.549	1.290	0.691	3.225	0.494	0.355	0.520	0.748	1.694	0.720
*Abundances are normalized for blackbody and rounded to the nearest whole percentage.

6. In-Flight Calibrated RAdiance Algorithm

The calibrated radiance algorithm will be included in the next revision.

Initially the response function for each scene observation will be determined using a linear interpolation between the response functions for bounding IE-pairs. However, the noise in the response function from a single IE-pair can be reduced by combining multiple determinations over a period of time, taking into account the changes due to variations in instrument temperature. During the mission, the data will be recalibrated using a low-noise response function determined by fitting a function to the complete set of instrument response data over long (~10 day) periods. Once the instrument radiance is determined for each internal calibration target observation, it will be interpolated over time for all of the intervening scene observations and used with the response function to determine the calibrated radiance for each scene spectrum (Eq. 2). Initially a linear interpolation between bounding values will be used; with time a more complex function will be determined to account for repetitive, periodic variations in instrument temperature.

7. References

Adams, J.B., M.O. Smith, and P.E. Johnson, Spectral mixture modeling: A new analysis of rock and soil types at the Viking Lander 1 site, J. Geophys. Res., 91, 8098-8112, 1986.

Christensen, P.R., Calibration Report for the Thermal Emission Spectrometer (TES) for the Mars Global Surveyor Mission, pp. 228, Mars Global Surveyor Project, Jet Propulsion Laboratory, Pasadena, CA, 1999.

Christensen, P.R., J.L. Bandfield, V.E. Hamilton, D.A. Howard, M.D. Lane, J.L. Piatek, S.W. Ruff, and W.L. Stefanov, A thermal emission spectral library of rock forming minerals, J. Geophys. Res., 105, 9735-9738, 2000a.

Christensen, P.R., J.L. Bandfield, V.E. Hamilton, S.W. Ruff, H.H. Kieffer, T. Titus, M.C. Malin, R.V. Morris, M.D. Lane, R.N. Clark, B.M. Jakosky, M.T. Mellon, J.C. Pearl, B.J. Conrath, M.D. Smith, R.T. Clancy, R.O. Kuzmin, T. Roush, G.L. Mehall, N. Gorelick, K. Bender, K. Murray, S. Dason, E. Greene, S.H. Silverman, and M. Greenfield, The Mars Global Surveyor Thermal Emission Spectrometer experiment: Investigation description and surface science results, J. Geophys. Res., 106, 23,823-23,871, 2001a.

Christensen, P.R., J.L. Bandfield, M.D. Smith, V.E. Hamilton, and R.N. Clark, Identification of a basaltic component on the Martian surface from Thermal Emission Spectrometer data, J. Geophys. Res., 105, 9609-9622, 2000b.

Christensen, P.R., G.L. Mehall, S.H. Silverman, S. Anwar, G. Cannon, N. Gorelick, R. Kheen, T. Tourville, D. Bates, S. Ferry, T. Fortuna, J. Jeffryes, W. O'Donnell, R. Peralta, T. Wolverton, D. Blaney, R. Denise, J. Rademacher, Miniature Thermal Emission Spectrometer for the Mars Exploration Rovers, J. Geophys. Res., 108 (E12), 8064, doi:10.1029/2003JE002117, 2003.

Farmer, V.C., The Infrared Spectra of Minerals, 539 pp., Mineralogical Society, London, 1974.

Feely, K.C., and P.R. Christensen, Quantitative compositional analysis using thermal emission spectroscopy: Application to igneous and metamorphic rocks, J. Geophys. Res., 104, 24,195-24,210, 1999.

Hamilton, V.E., and P.R. Christensen, Determining of modal mineralogy of mafic and ultramafic igneous rocks using thermal emission spectroscopy, J. Geophys. Res., 105, 9717-9734, 2000.

Morris, R.V., and T.G. Graff, Athena instrument validation and data library development for the Mars Exploration Rover (MER) Mission, Eos Trans. AGU, 83, Fall Meet. Suppl., Abstract P21C-03, 2002.

Ramsey, M.S., and P.R. Christensen, Mineral abundance determination: Quantitative deconvolution of thermal emission spectra, J. Geophys. Res., 103, 577-596, 1998.

Ruff, S.W., P.R. Christensen, P.W. Barbera, and D.L. Anderson, Quantitative thermal emission spectroscopy of minerals: A technique for measurement and calibration, J. Geophys. Res., 102, 14,899-14,913, 1997.

Salisbury, J.W., and A. Wald, The role of volume scattering in reducing spectral contrast of reststrahlen bands in spectra of powdered minerals, Icarus, 96, 121-128, 1992.

Squyres, S.W., R.Arvidson, E. Baumgartner, J.F. Bell, III, P.R. Christensen, S.Gorevan, K.Herkenhoff, G.Klingelhöfer, M.Madsen, R.V. Morris, R.Rieder, and R.Romero, The Athena Mars Rover Science Investigation, J. Geophys. Res., 108 (E12), 8062, doi:10.1029/2003JE002054, 2003.

8. Figures

Figure 1 The Mini-TES I spectrometer following completion of testing at SBRS. (a) The instrument with its cover on and the optical aperture facing upward. (b) The Mini-TES with its protective cover off, showing the integrated packaging of the electronics and optics required to meet the size and weight constraints of the rover mission.

Figure 2. Mini-TES thermal vacuum test setup at Santa Barbara Remote Sensing. The Mini-TES instruments were tested at vacuum using precision calibration reference blackbodies. The cylindrical aluminum exterior of the two reference blackbodies can be seen in this image, along with Mini-TES (black box) that is pointing downward toward a pointing mirror that permitted the two reference blackbodies, a V-groove target (not visible) and a quartz crystal to be viewed.

Figure 3. Mini-TES field of view. The response was mapped viewing a thermal source through a collimator and a 1x40 mrad slit that was stepped at 1-mrad increments across the aperture. (a) 20 mrad mode; (b) 8 mrad mode.

Figure 4. Variation in instrument response with temperature. The instrument response determined for the integrated Mini-TES I/PMA is shown for the three instrument temperatures (-30° C, 0° C, and +30° C) measured in 6 mbar of nitrogen at JPL. Also shown are the instrument response functions determined for the Mini-TES I spectrometer separately at SBRS

Figure 5. The noise equivalent spectral radiance (NESR) for the integrated Mini-TES I/PMA measured in 6 mbar nitrogen at JPL. Values are computed for a single spectrum. During operations at Mars two spectra will typically be averaged together, reducing the NESR by ö2.

Figure 6. Absolute spectral radiance error. A representative example of the absolute radiance error for the Mini-TES I spectrometer is shown for an instrument temperature of 0° C measured in vacuum at SBRS. (a) Comparison of the derived calibrated spectral radiance viewing a V-groove target in thermal vacuum at SBRS and a Planck function fit to the calibrated radiance. (b) The difference between the V-groove calibrated spectral radiance and the best-fit Planck function.

Figure 7. Mini-TES I spectra of selected rock targets. In all cases the Mini-TES field of view was larger than the rock target. The spectra shown here are averages of ~5 individual spectra. The reflected background radiance has been removed. These spectra illustrate the quality of spectra that will be obtained of the rocks and soils at the MER landing sites. Spectra from these rock targets obtained in the ASU Spectrometer Laboratory are shown for comparison.

Figure 8. Mineral deconvolution results for selected rock targets. The modeled spectra were determined using a linear deconvolution technique and a library of 62 spectra (see text for details). The model spectra fit the measured spectra to within the noise of the Mini-TES instrument. The derived mineral abundances are given in Table 4. (a) Rock Sample 11. (b) Rock Sample 12. (c) Rock Sample 13.