THE CLEMENTINE NIR GLOBAL LUNAR MOSAIC Lisa Gaddis, Chris Isbell, Matt Staid*, Eric Eliason**, Ella Mae Lee, Lynn Weller, and Tracie Sucharski United States Geological Survey Astrogeology Team 2255 North Gemini Drive Flagstaff, AZ 86001 Paul Lucey, Dave Blewett***, John Hinrichs***, Donovan Steutel Hawaii Institute of Geophysics and Planetology University of Hawaii Honolulu, Hawaii 96822 May 1, 2007 Version 0.1 * Now at Planetary Sciences Institute 1700 East Fort Lowell, Suite 106 Tucson, AZ 85719 ** Now at Planetary Image Research Laboratory Lunar and Planetary Laboratory Department of Planetary Sciences The University of Arizona Tucson, AZ 85721 *** Now at NovaSol 733 Bishop Street Honolulu, HI 96813 TABLE OF CONTENTS Return to CD Home Page 1 - INTRODUCTION 2 - CLEMENTINE MISSION 3 - NEAR-INFRARED (NIR) CAMERA 4 - LUNAR ORBIT SUMMARY 5 - GEOMETRIC ACCURACY 6 - DATA PROCESSING 7 - RADIOMETRIC CALIBRATION 8 - PHOTOMETRIC FUNCTION NORMALIZATION 9 - EMPIRICAL FRAME-OFFSET CORRECTION 10 - CONVERSION OF 16-BIT PIXEL VALUES TO FLOATING POINT NUMBERS 11 - FILES, DIRECTORIES, AND DISK CONTENTS 12 - IMAGE FILE ORGANIZATION 13 - INDEX FILES 14 - ACKNOWLEDGMENTS 15 - REFERENCES APPENDIX A - KEYWORD ASSIGNMENTS APPENDIX B - GEOMETRIC DEFINITION OF A PIXEL Go to Table of Contents. 1 - INTRODUCTION The Clementine Near-Infrared (NIR) mosaic [Eliason et al., 2003; Staid et al., 2003] of Earth's Moon is a radiometrically and geometrically controlled, photometrically modeled global mosaicked Digital Image Model (DIM) [Batson, 1987; Batson, 1990]. The NIR DIM was compiled using more than 400,000 images from the Near-Infrared Camera onboard the Clementine orbiting spacecraft. The mosaic is produced in six wavelengths with band centers at 1100, 1250, 1500, 2000, 2600, and 2780 nanometers (nm). Cartographic processing of the NIR mosaic resembles that applied to the mosaic derived from data acquired by the five- band (415, 750, 900, 950, 1000 nm) Ultraviolet/Visible (UVVIS) Camera that also flew onboard the Clementine orbiter [e.g., Eliason et al., 1999]. The major lunar rock-forming minerals plagioclase, pyroxene, and olivine have diagnostic absorptions in this wavelength range, so the NIR data combined with UVVIS data can be used to characterize aspects of the mineralogic and lithologic diversity of the Moon [e.g., Le Mouelic et al., 1999; Gillis et al., 2004). These UVVIS and NIR multispectral data are complemented by data acquired by two additional Clementine science cameras, the High Resolution (HIRES) [e.g., Robinson et al., 2003] and Long-Wavelength Infrared (LWIR) instruments [e.g., Lawson et al., 2000]. The NIR mosaic is mapped in the Sinusoidal Equal-Area Projection [Snyder, 1987] at a resolution of 100 meters per pixel and requires approximately 68 gigabytes of digital storage. This database is partitioned into 996 quadrangles (quads) or "tiles" equivalent to those of the previously released Clementine 750 nanometer Basemap Mosaic (PDS volumes CL_3001 through CL_3015) and the UVVIS DIM (PDS volumes CL_4001 through CL_4078; Eliason et al., 1999). This 100 m/pixel version of the NIR DIM is presented on 13 DVD volumes (equivalent to 78 ‘virtual CD’ volumes). Tiles are stored as image files of approximately 2100 pixels on a side, covering approximately a 6-degree by 7-degree quadrangle (303 pixels/degree). Pixels are 16-bit signed integers. Reduced-resolution planet-wide NIR mosaics will be made available online at the PDS Map-a-Planet Web site (http://www.mapaplanet.org) along with the full-resolution NIR mosaic. Additional information in support of the NIR DIM, including a frame number (colorset_id) image, empirical frame offset values for each frame, and a source file index table ('srcindx.tab') will be made available online at http://astrogeology.usgs.gov/Projects/ClementineNIR/. The source file index table contains information about the Engineering Data Record (EDR) image collection used to assemble the NIR DIM and the images that make up 6-band color sets. This file contains an entry for each EDR image that was used in this database. A phase angle image database does not accompany the NIR DIM (recall that this information is available with the UVVIS DIM archive). Although the tiling scheme is identical to the original basemap, each 30-degree longitudinal section of the basemap now represents one DVD volume (equivalent to 6 virtual CD volumes). The full-resolution tiling and archive scheme is available in graphical form in the 'scheme.pdf' (Portable Document Format [PDF]) file found in the 'document' directory of each volume. PDS standards require inclusion of "catalog" files that describe aspects of the mission, spacecraft, instrument, and dataset. These files are located in the 'catalog' directory of each CD volume. To minimize the amount of redundant information we reference these files throughout this document. The catalog files, ending with the '.cat' extension, are organized as ASCII text files and contain PDS labels for automatically ingesting the documents into a catalog system. The DVD volume set contains ancillary data files that support the NIR mosaic. These files include browse images stored in a 'JPEG' format, 'HTML' documents that support a web browser interface to the volumes, image index files that tabulate the contents of the volume set, and documentation files that describe the archive collection. For more information on the contents and organization of the volume set, refer to the "Files, Directories, and Disk Contents" section of this document. Using a web browser application (e.g., Netscape or MS Internet Explorer), open the 'index.htm' file located in the 'root' directory of each CD. The html document will direct you to other informational documents and the image browser for rapidly viewing the image collection. The browse image consists of false color and color ratio images for each Clementine NIR DIM product. Three browse image renditions are available to the user: false color, color ratio, and black-and-white 2000 nm images. The false color images use the 1100 nm (blue), 1500 nm (green), and 2000 nm (red) spectral bands to portray the Moon in enhanced color. The color ratio browse images use the 1500/1100 (blue), 1500/2000 (green), and 1500 (red). The ratio rendition enhances color differences related to soil mineralogy as measured by reflectance at the given wavelengths. At NIR wavelengths, the ratios measure strength of absorption bands of common lunar minerals such as orthopyroxene, clinopyroxene, olivine, and plagioclase [e.g., Pieters, 1993]. In particular, the 1500/1100 nm ratio is sensitive to the abundance of clinopyroxene and olivine, as well as plagioclase, and the 1500/2000 nm ratio is related to the abundance of pyroxene in lunar soil. The NIR data are particularly useful for discriminating between olivine and pyroxene because pyroxene has two absorptions in the UVVIS and NIR (near 1 and 2 microns, or 1000 and 2000 nm) and olivine has only one (near 1 micron, or 1000 nm). Software tools for viewing and accessing the image collection are available through the Planetary Data System (PDS) Internet services. Refer to the 'aareadme.txt' located in the 'root' for more information on these tools. The program NASAView can be used to display the full resolution NIR image products. NASAView is a PDS product display tool that runs on multiple platforms with a common graphical user interface. The current version (version 2.13, 2006) supports the display of multi-band image cubes (up to three selected bands). See the PDS Software Download page (http://pds.jpl.nasa.gov/tools/software_download.cfm) for information on NASAView and its availability. Another tool, Map-a-Planet (MAP), provides online access to the NIR data via the MapMaker software. Map-a-Planet allows users to generate seamless image maps for any latitude-longitude region of the Moon at a variety of scales and map projections. For more information on Map-a-Planet and the MapMaker system, refer to the USGS Astrogeology Program Web page at http://www.mapaplanet.org/ . Additionally, each image file is associated with an Integrated Software for Imagers and Spectrometers (ISIS) [Eliason, 1997; Gaddis et al., 1997; Torson and Becker, 1997; Anderson et al., 2004] detached label. The detached ISIS label allows for images to be accessed and processed within the ISIS system (available through the USGS, Flagstaff, AZ at http://isis.astrogeology.usgs.gov/ ). Go to Table of Contents. 2 - CLEMENTINE MISSION The Clementine Mission [Nozette et al., 1994; McEwen and Robinson, 1997], officially known as the Deep Space Program Science Experiment (DSPSE), was the first in a planned series of technology demonstrations jointly sponsored by the Ballistic Missile Defense Organization (BMDO) of the Department of Defense and the National Aeronautics and Space Administration (NASA). The spacecraft was known as Clementine because, as in the song of the same name, it would be 'lost and gone forever' after completing its short mission. Clementine was launched on 1994- 01-25 onboard a Titan IIG rocket from Vandenburg Air Force Base in California. The mission included two months of systematic lunar mapping (1994-02-26 through 1994-04-21), which was to have been followed by a flyby of the near-Earth asteroid Geographos (1994-08-31). An onboard software error combined with improbable hardware conditions, on 1994-05-07 led to accidental spin-up of the spacecraft and loss of attitude control gas. These events precluded the flyby of asteroid Geographos. Clementine's primary objective was to qualify lightweight imaging sensors and component technologies (including a star tracker, inertial measurement unit, reaction wheel, nickel hydrogen battery, and solar panel) for the next generation of Department of Defense spacecraft. DSPSE represented a new class of small, low cost, and highly capable spacecraft that embraced emerging lightweight technologies to enable future long-duration deep space missions. A second objective of Clementine was the return of scientific data about the Moon and asteroid Geographos to the international civilian space community. The NIR mosaic was created using data from Clementine EDR Image Archive [e.g., Eliason et al., 1995] produced by the Clementine mission. The EDR (Engineering Data Record) data are raw images that contain the inherent properties of unprocessed and uncorrected data [Lewis, 1995]. The Clementine EDR Image Archive contains more than 1.9 million images acquired during active mission operations. For information on how to obtain this archive contact NSSDC (http://nssdc.gsfc.nasa.gov/database/MasterCatalog?ds=PSPG- 00309) or the PDS Imaging Node Clementine Mission pages (http://pdsimg.jpl.nasa.gov/Missions/Clementine_mission.html). Go to Table of Contents. 3 - NEAR-INFRARED (NIR) CAMERA The Clementine Near-Infrared (NIR) camera was a framing camera built jointly by Lawrence Livermore National Laboratory (LLNL) and Amber Engineering of Goleta, GA. The NIR camera obtained images of nearly the entire lunar surface at six wavelengths and its coverage was essentially identical to that of the UVVIS camera. The center wavelengths [and full-width half maximum (FWHM) bandpass widths] of the six filters were: 1100 nm (30 nm), 1250 nm (30 nm), 1500 nm (30 nm), 2000 nm (30 nm), 2600 nm (30 nm), and 2780 nm (60 nm) [e.g., Nozette et al., 1994]. [Note that the wavelength of NIR band 6 has been reported as 2690 nm in places, but that value is the cuton wavelength of the band 6 filter. 2780 nm is the more accurate wavelength of the center of NIR band 6.] Each image frame was 256 by 256 pixel elements, with a field of view of 5.6 degrees and an IFOV of 0.38 mrad. The NIR camera angular resolution (equivalent to spatial resolution) was lower than that of the UVVIS camera at 0.25 mrad, with a maximum spatial sampling of about 135 meters/pixel. The camera was based on an indium antimonide (InSb) CCD detector array produced by Amber which offers sensitivity from 1 to >5 microns. The NIR camera is described in detail by Priest et al. [1995] and online at the LLNL (http://www.llnl.gov/sensor_technology/STR42.html) and NSSDC (http://nssdc.gsfc.nasa.gov/database/MasterCatalog?sc=1994- 004A&ex=2) Web sites. For more information on the NIR camera refer to the 'nirinst.cat' file in the 'catalog' directory. Go to Table of Contents. 4 - LUNAR ORBIT SUMMARY The Clementine spacecraft maintained a polar orbit during the systematic mapping of the surface of the Moon [McEwen and Robinson, 1997]. Mapping of the lunar surface was done in two Earth months, or two lunar days. To obtain full coverage during these two months, the image overlap for both the UVVIS and NIR cameras was ~15% in the down-track and ~10% in the cross-track directions. This mapping scheme required an inclination of the orbit of 90 degrees plus-or-minus 1.0 degree with reference to the lunar equator; the periselene of the lunar orbit was maintained at an altitude of 425 plus-or-minus 25 km. To provide the necessary cross-track separation for the alternating imaging strips to cover the entire surface of the Moon, the orbital period was approximately 5 hours, during which the Moon rotated approximately 2.7 degrees beneath the spacecraft. Images were taken and recorded only in the region of periselene, leaving sufficient time to replay the data to Earth. The best data for lunar mineral mapping are obtained if both high and low solar phase angles are avoided, nominally at angles less than 30 and greater than 0 degrees. The solar phase angle is defined as the angle between the vector to the Sun and the vector to the spacecraft from a point on the Moon's surface. To maximize the time in which the solar phase angle is less than 30 degrees, the plane of the lunar orbit should contain the Moon- Sun line halfway through the two-month lunar mapping period. Therefore, insertion into the lunar orbit was selected so that, as the Moon-Sun line changes with Earth's motion about the Sun, the Moon-Sun line will initially close on the orbital plane, and then lie in the orbital plane halfway through the mapping mission. The angle between the Moon-Sun line and the orbital plane was near-zero (less than 5 degrees) for approximately five weeks before becoming zero. The table below lists Clementine's orbital parameters. For more information on the lunar orbit refer to the 'mission.cat' file located in the 'catalog' directory. Clementine Orbital Parameters ================================================================ Orbital Period: 4.970 hr < P < 5.003 hr Altitude of Periselene: 401 km < radius < 451 km Eccentricity: 0.35821 < e < 0.37567 Right Ascension: -3 deg < Omega < +3 deg (J2000) Inclination: 89 deg < i < 91 deg Argument of Periselene: -28.4 deg < w < -27.9 deg (1st month) 29.2 deg < w < 29.6 deg (2nd month) Go to Table of Contents. 5 - GEOMETRIC ACCURACY Both the Clementine NIR and UVVIS mosaics were geometrically controlled to the previously published 750-nm Clementine Basemap Mosaic [PDS volumes CL_3001 through CL_3015] by tying individual NIR images that make up a color set to the corresponding image used to produce the basemap mosaic. The 750-nm basemap mosaic was geometrically controlled using the methods described below. Although shortcomings have since been identified [e.g., Cook et al., 2002], the 750-nm basemap mosaic significantly improved the geometric control of the Moon from previous maps and ground control points. On the basis of best-effort measurements of the spacecraft orbit and pointing, UVVIS geometric distortions, and time tags for each observation, the Clementine Spacecraft, Planet Instrument, C-matrix, and Event kernels [SPICE; Acton, 1996] data alone provide positional accuracy better than 1 kilometer over most of the Moon. The geometric processing goal of the basemap was for 95% of the Moon (excluding the oblique observation gap fills) to have better than 0.5 km/pixel absolute positional accuracy and to adjust the camera angles so that all frames match neighboring frames to within an accuracy of 2 pixels [e.g., Lee et al., 1997; Eliason et al., 1998, Isbell et al., 1999]. To achieve these goals, camera alignment and pointing data were required to be accurate to a few hundredths of a degree. The absolute alignment of the UVVIS was determined with respect to spacecraft-fixed axes (A and B Star Tracker Camera quaternions) by analyzing a major subset of the over 17,000 images of Vega, over 6,000 images of the Southern Cross, and a few hundred images of the Pleiades, taken during the approach to the Moon and throughout the lunar mapping phase of the Clementine mission. Multiple star images within a single picture were used to determine the UVVIS focal length and optical distortion parameter values. Approximately 265,000 match points were collected at the USGS from ~43,000 UVVIS 750 nm images providing global coverage [e.g., Lee et al., 1997]. Approximately 80% of these points were collected via autonomous procedures, and the remaining 20% were collected manually. Streamlined procedures for the supervised collection of match points were developed and applied, and these procedures saved several person-years of effort. The automated success rate exceeded 90% along each spacecraft orbit track, where the overlap regions of successive images are highly correlated, but failed when the overlap region is narrow and/or nearly featureless. ('Failure' is defined as less than 3 points per image with correlation coefficients greater than 0.85; thus, many good match points were rejected because we could not be certain that the matches were valid without verification.) Across-track matching was more difficult due to changes in scale and illumination angle, but a fair success rate (~60%) was achieved via the use of 'window-shaping' (local geometric reprojections). The oblique gap-fill images were the most difficult to match and required substantial human intervention. Matching the polar regions was time-consuming because each frame overlaps many other frames. Most match point locations were found to a precision of 0.2 pixel. The USGS match points were provided to Tim Colvin and Mert Davies of the RAND Corporation for analytical triangulations. Using these match points, control points from the Apollo region, and the latest SPICE kernels from Navigation and Ancillary Information Facility (NAIF) at JPL [Acton, 1996], RAND determined improved camera orientation angles for the global set of UVVIS images. A constant lunar radius of 1737.4 kilometers was assumed, and this was later found to be a significant source of error near the oblique gap fills. The analytical triangulation is a least-squares formulation designed to adjust the latitude and longitude of the control points and the camera orientation angles to best fit the match points. The triangulation was first computed on 'packets' of match points (each covering about one-eighth of the Moon), then checked and rechecked at the USGS via plots and test mosaics to fix and add match points as needed. The final (global) analytical triangulation required the solution of ~660,000 normal equations. The mean error is less than 1 pixel. This effort was by far the largest analytical triangulation ever applied to a planetary body other than Earth. The results defined the planimetric geometry of the 750-nm basemap, to which all systematic Clementine mosaic products have been tied. Go to Table of Contents. 6 - DATA PROCESSING As was the case for the UVVIS mosaic, the U.S. Geological Survey Integrated Software for Imagers and Spectrometers (ISIS) processing system [Eliason, 1997; Gaddis et al., 1997; Torson and Becker, 1997; Anderson et al., 2004] was used to generate the NIR mosaic. Because the final steps of the NIR data processing followed the UVVIS processing by several years, the same version of ISIS (v.2.0, or 'Old ISIS') was frozen in place and used for both datasets. ISIS cartographic processing for the NIR mosaic includes radiometric correction, geometric control to the Clementine 750-nm basemap mosaic, spectral registration, photometric normalization, and image mosaicking to produce near-seamless, uniformly illuminated views of the surface of the Moon at 6 wavelengths. Radiometric correction applies 'flat fielding,' dark current subtraction, non-linearity correction, and conversion to radiometric units (usually radiance). Geometric transformations tie each raw image with the ground control network from the basemap mosaic and convert from raw image coordinates to the Sinusoidal Equal-Area projection. Photometric normalization is applied to balance brightness variations due to illumination differences among the images in a mosaic. The first four NIR bands (1100 to 2000 nm) have also been normalized to reflectance based on the approach previously applied to the calibrated UVVIS global mosaics [Pieters et al., 1999]. Images are then mosaicked together to form a global map of continuous image coverage for the entire planetary body. Basic cartographic processing of the NIR mosaic was performed in five stages or "levels". The processing sequence was designed to start with corrections with highest probability of accuracy to those with the lowest. The first level of processing, Level 0, ingests raw data and prepares it for processing by ISIS. The raw images are converted to ISIS format and ancillary data such as viewing geometry are added to the labels of the image file. Level 1 processing applies radiometric corrections and removes artifacts from the image. Level 2 performs geometric processing to remove optical distortions and convert the image geometry to a standard map projection. Level 3 performs photometric processing for normalizing the solar illumination geometry of an image scene. Level 4 performs mosaicking of individual images to create global or regional views of the planetary surface. More information on the general character of these cartographic processing steps for the NIR data is provided below. Further details on the calibration, photometric normalization, and empirical frame-offset correction required for the NIR mosaics have been described in several publications [e.g., Lucey et al., 1997; Lucey et al., 1998; Lucey et al., 2000; Eliason et al., 2003; Staid et al., 2003] and are summarized in the next sections. Level 0 The Level 0 processing step ingests the raw image data and associated metadata and prepares them for processing by the ISIS system. Level 0 processing consists of two program steps. The first step reads the format of the raw image, extracts the metadata from the input image labels, and creates an ISIS file. The metadata may contain information such as the instrument operating modes, temperature of the camera focal plane, UTC time of observation, and other information necessary to rectify an image. The second step extracts navigation and pointing data ("SPICE" kernel data) for inclusion in the ISIS file. Level 1 The next level of processing, Level 1, performs radiometric correction and cosmetic enhancement ("clean-up") on an image. Level 1 consists of a series of programs to correct or remove image artifacts such as 1) camera shading inherent in imaging systems, 2) brightness anomalies caused by dust specks in the optical path, 3) microphonic noise introduced by operation of other instruments on the spacecraft during image observations, and 4) data drop-outs and spikes due to missing or bad data from malfunctioning detectors or missing telemetry data. Level 1 processing results in an "ideal" image that would have been recorded by a camera system with perfect radiometric properties. In practice, residual artifacts and camera shading effects remain at a very low level. The density number (DN) values of a radiometrically corrected image are proportional to the brightness of the scene. The details for radiometric correction are described in section 7 below. Level 2 Production of both the Clementine UVVIS and NIR mosaics required geometric processing to be performed on the individual, single- band images. The images are geometrically transformed from spacecraft camera orientation to a common map coordinate system of a specific resolution. Before geometric transformation, images must first be geometrically matched to each other to establish relative geometric control among the multiband images in a single "color set" and then the image set must be tied to a ground control net to establish absolute ground truth. The process of matching images and tying the image set to ground truth minimizes the spatial misregistration along image boundaries. The 750-nm basemap provided the geometric control for the NIR multispectral DIM. This base map underwent rigorous cartographic processing to tie the imaging to a lunar geodetic network resulting in an absolute positional accuracy of better than 0.5 km/pixel for 95% of the surface. The six NIR spectral bands are coregistered to a precision of 0.2 pixel. Level 2 geometric processing also includes correcting camera distortions. This image transformation is based on the original viewing geometry of the observation (including the optical distortion model of the camera), relative position of the target, and the mathematical definition of the map projection. An additional resampling step is used to perform sub-pixel registration of the 6 bands in the color set. Using the ISIS system, the Sinusoidal Equal-Area projection NIR data can be transformed to other map projections as needed. Level 3 Photometric normalization is applied to images that make up the NIR DIM to balance brightness levels among images that were acquired under different lighting conditions. To illustrate, consider two images of the same area on the Moon where one image was acquired with the Sun directly overhead and the second with the Sun lower to the horizon. The image with the higher sun angle would be significantly brighter than the image with the low sun angle. Photometric normalization causes the two images to be adjusted to the same brightness level. Radiometrically calibrated spacecraft images display the brightness of a scene under specific angles of illumination, emission, and phase. Photometric normalization is effective only to the extent that all geometric parameters can be modeled. For an object (such as the Moon) without an optically significant atmosphere, this brightness is controlled by two major influences: 1) the intrinsic properties of the surface materials, including composition, grain size, roughness, and porosity; and 2) variations in brightness due to the local topography of the surface. For both Clementine UVVIS and NIR photometric processing [e.g., McEwen et al., 1996; Lucey et al., 1998, 2000; Eliason et al., 2003] the planetary surface is assumed to be a smooth sphere, so the effects of local topography are not included in the photometric correction. However, it is understood that illumination geometry at each pixel depends on local topography. This means that the topographic slope within a pixel is not accurately known and compensated in either the UVVIS or NIR mosaics, so the photometric correction cannot be perfect. Section 8 describes the photometric normalization. Level 4 Compilation of an accurate digital mosaic of the individual images is the final stage in the construction of the NIR DIM. The DIM is created by first generating a blank (or null) image that represents the regional or global image map of the Moon. The individual images are then mosaicked into the initially blank image map. The order in which individual images are placed into the mosaic is an important consideration. Because images are mosaicked one on top of the other, images that get laid down first are overwritten in the area of overlap by subsequent images that are added to the mosaic. Images that have the lowest data quality or resolution are laid down first, followed by images with highest quality. With this method the areas of image overlap contain the highest quality images. Go to Table of Contents. 7 - RADIOMETRIC CALIBRATION Radiometric calibration for the NIR data [Lucey et al., 1987; 1998] provides correction for camera gain and offset operating modes, pixel dependent nonuniformity in bias, dark current rate, and responsivity, dark-current dependence on temperature, removal of thermal background contamination, and conversion to radiometric units of radiance and/or reflectance. Details of these steps are provided below. Before radiometric processing could begin, two calibration issues specific to the NIR data were addressed [Eliason et al., 2003]: 1) characterization of the instrument operating modes, and 2) characterization of the instrument thermal background changes during an orbital observation pass over the Moon. Instrument Operating Modes The goal of this step was to determine an optimum set of calibration constants to minimize the difference between calibrated values for portions of the Moon imaged sequentially with different camera settings [Lucey et al., 2000]. Statistics were acquired for all images collected during systematic mapping that straddled camera-state boundaries (~19,800 cases). The calibration equation from Lucey et al. [1998] was applied to the entire set of data, an error function was refined iteratively, calibration constants were updated, and the process was repeated until little change occurred in the error function. In this manner the gain, exposure duration, digital offset, offset multiplier, and global bias were simultaneously optimized. Thermal Background Instrument thermal background problems were first observed when the initial pre-launch calibration algorithms were not satisfactorily modeling the NIR deep-space observations during cool-down tests [Lucey et al., 1998]. The goal of this step was to characterize the thermal background changes during an orbital pass and to define a set of corrections for each orbit. To characterize these effects, the NIR 1100 nm data were compared to the highly correlated UVVIS 1000 nm data held as "truth". The UVVIS and NIR images were radiometrically corrected (with the thermal background remaining in the NIR imaging) and geometrically coregistered. UVVIS radiance values were scaled to the NIR data, and the averages of the UVVIS and NIR images in areas of overlap were computed for all color sets and plotted as a function of the cryocooler duration. A simple difference of the NIR and UVVIS bands (NIR/1100 band subtracted by the scaled UVVIS/1000 band) characterizes the change in thermal background in an orbit, and a 3rd order least-squares fit to the data is a valid model of time-dependent thermal background for each orbit. Compensation for the two effects above was incorporated into the radiometric calibration steps, and residual effects were removed in later processing. The radiometric calibration process for the NIR data (performed by the ISIS program NIRCAL) converts raw density-number (pixel brightness) values to radiance, and these steps are described below. For all equations provided in the document the asterisk character (*) is used to denote a multiplication and the double asterisk (**) is used to denote exponentiation. EQUATIONS Term1 = (DR(x,y) - digital_offset ) / Gfact(gm) Term2 = Term1 - bias_global - BIAS(x,y) - (om * V) Term3 = Term2 / t Term4 = Term3 - global_dc - DC(x,y) Iterative for each background radiance coefficient (ncoef) which is orbit/filter dependent: background_radiance = background_radiance + background_radiance_coefficient(icoef) * ((cryocooler_duration/cryo_norm) ** (icoef-1)) Term5 = Term4 - background_radiance - THERMAL(x,y) R(x,y) = ( (Term5 / (FF(x,y) * OF(x,y)) ) - AF(x,y) ) * abscoef Let: x,y = line and sample position of pixel in an image R(x,y) = Result of correction in absolute radiance. DR(x,y) = Raw input density number FF(x,y) = Flat-field as a function of filter and line, sample position. This will be read from an ISIS cube file. OF(x,y) = Orbit dependent flat file correction. AF(x,y) = Additive component of flat field correction. DC(x,y) = Dark current as a function of compression type, line, sample position. BIAS(x,y) = Bias correction as a function of compression type, line, sample position. THERMAL(x,y) = Thermal shade correction based on thermal shade file at the north and south pole. digital_offset = 8.30690 bias_global = 2.15547 V = -0.954194 global_dc = 0.730 cryo_norm = 1.0 abscoef = 1.0 Gfact(gm)= Gain factor as a function of instrument gain_mode = 32.0235 (gm=0) = 28.2755 (gm=1) = 24.9144 (gm=2) = 10.9443 (gm=5) = 24.1835 (gm=8) = 21.3530 (gm=9) = 15.9844 (gm=11) = 7.77177 (gm=13) = 28.1618 (gm=16) = 24.8658 (gm=17) = 21.9100 (gm=18) = 18.6140 (gm=19) = 6.83130 (gm=22) = 3.48425 (gm=23) = 20.3218 (gm=24) = 17.9433 (gm=25) = 15.8104 (gm=26) = 13.4320 (gm=27) = 9.32361 (gm=28) = 6.95951 (gm=29) = 4.75472 (gm=30) = 2.43896 (gm=31) = 13.9238 (gm=33) = 12.2687 (gm=34) = 7.23501 (gm=36) = 7.04438 (gm=41) = 6.16495 (gm=42) = 3.57405 (gm=44) = 2.73995 (gm=45) = 1.88595 (gm=46) = 11.9078 (gm=48) = 9.26433 (gm=50) = 5.39513 (gm=52) = 4.08125 (gm=53) = 1.40899 (gm=61) = 0.964975 (gm=62) gm = Gain_mode of instrument(gm = 0,1,2,5,...,62) om = Offset_mode of instrument(om = 1,2,3,4,...,31) t = Optimal integration duration of observation in seconds (A function of EXPOSURE_DURATION on labels in milliseconds.) PROGRAM STRATEGY NIRCAL reads the keyword label area from the input file to obtain processing parameters to radiometrically correct the image. The following keywords are extracted from the keyword label area: TABLE OF IMAGE KEYWORDS USED BY NIRCAL -------------------------------------- INSTRUMENT_ID - Camera (Should be NIR) GAIN_MODE_ID - Instrument gain mode OFFSET_MODE_ID - Instrument offset mode CRYOCOOLER_DURATION - Cryocooler duration EXPOSURE_DURATION - Exposure time of camera BAND_BIN_FILTER_NAME - Instrument filter name (A,B,C,D,E,F) for (1100, 1250, 1500, 2000, 2600, 2780 nm, respectively) After obtaining the image keywords from the image label area and determining all processing parameters, NIRCAL opens the appropriate calibration files and processes the image. Go to Table of Contents. 8 - PHOTOMETRIC FUNCTION NORMALIZATION The Clementine data have large changes in brightness because they were acquired under a broad range of viewing conditions, with phase, emission, and incidence angles varying from 0 to 90 degrees. To create NIR mosaics with uniform scene brightness, a photometric normalization procedure was applied [McEwen, 1996; McEwen et al., 1998] to the individual images before compiling the global mosaic. The data are normalized to R30, the reflectance expected at an incidence angle (i) and phase angle (p) of 30.0 degrees and an emission angle (e) of 0.0 degrees matching the photometric geometry of lunar samples measured at the reflectance laboratory at Brown University [Pieters et al., 1991]. Differences in phase behavior, especially near 0.0 degrees, as a function of terrain type were observed in the resulting mosaics. A second-order correction to the phase function for near zero phase observations was developed and applied to the data. This procedure compared areas of overlap among low and high-phase imaging for determining a multiplicative correction to the low-phase imaging which forced overlapping areas to match. The resulting multiplicative corrections were then applied to the low-phase imaging in non- overlap areas. The modified Lunar-Lambert function used in the processing [McEwen, 1996; McEwen et al., 1998] is as follows: XL(i,e,p) = 2*L(p)*u0/(u+u0) + (1-L(p))*u0 where: i = incidence angle e = emission angle p = phase angle u = cos(e) u0 = cos(i) L = variation of the limb-darkening parameter expressed as a third-order polynomial: L(p) = 1.0 + A*p + B*(p**2) + C*(p**3) A = -0.019 B = 0.242E-3 C = -1.46E-6 The phase function correction based on work from Helfenstein is defined as follows: F(p) = B(p,h,b0)*[(1-f)*P(p,g1) + f*P(p,g2)] B(p,h,b0) = 1 + b0/(1+tan(p/2)/h) (Backscatter function) P(p,g) = (1-g**2)/(1+g**2+2*g*cos(p)**1.5) b0 = wave length dependent constant (see table below) The g1 parameter is expected to vary with albedo, modeled as: g1 = d*R30 + e (however, d=0 for Clementine processing, thus eliminating any correction for albedo). R30 = normalized albedo To describe the backscatter function at less than 2 degrees phase angle, the following linear functions are used based on results from Burratii [McEwen, 1996]: F(p) = 1.0 + xb xb = xb0 + xb1 * wave_length where: xb0 = -0.0817 xb1 = 0.0081 Final normalization to the RELAB geometry is given by: R30 = R(i,e,p)*[XL(30,0,30)/XL(i,e,p)]*[F(30)/F(p)] where: f,g2,b0,h are parameters determined by fits to a dataset. Values applied to the Clementine NIR data were derived for the UVVIS E filter (1000 nm) and these have been applied to all six NIR wavelengths. Note that d=0 means that we do not attempt to vary the photometric correction as a function of albedo. Filter b0 h d e f g2 -------------------------------------------------- NUA (1100 nm) 1.35 0.052 0 -0.226 0.5 0.36 Go to Table of Contents. 9 - EMPIRICAL FRAME OFFSET CORRECTIONS Processing of the NIR data based on the radiometric and photometric procedures described above resulted in global mosaics, which had achieved a "standard" level of processing and calibration. Examples of these data have been released on the USGS Clementine NIR data Web site: http://astrogeology.usgs.gov/Projects/ClementineNIR/ .In addition to this standard radiometric and photometric processing, empirically derived frame offset corrections were applied to produce a second version of the mosaics with reduced variability across camera modes and adjacent orbits [Staid et al., 2003]. This additional processing resulted in "frame offset corrected" mosaics extending from 70 degrees N to 70 degrees S and is described in more detail below. Note that polar regions above 70 degrees were not included in this additional processing (see also page 22). Frame Offset Corrections After radiometric and photometric calibrations were applied to the NIR data [e.g., Eliason et al., 2003], additional corrections were derived to adjust for remaining problems in the characterization of specific camera modes and the drift of radiometric properties over the two month Clementine observation period. Problems in the existing NIR calibrations (standard processing version) were apparent in both NIR albedo images and ratio images of NIR and NIR to UVVIS bands. Though such problem areas were typically associated with ten degree latitude strips where specific camera settings were poorly characterized, no systematic correction for these residual errors was apparent based on comparisons of image means with gain and offset camera mode values. As a result, a set of empirical corrections was developed to reduce within- and between-orbit variations for each NIR band as described below. To identify Clementine NIR frames with residual calibration errors, ratio images were created by dividing each radiometrically and photometrically calibrated NIR band to the USGS calibrated 750 nm UVVIS data. To make computational processing more efficient, the NIR data were divided by longitude into twelve 30-degree wide strips at a resolution of 1 km/pixel. Several thousand "truth" image frames were then selected for each NIR band of each strip based on orbits and camera mode states that demonstrated the least amount of radiometric variability across NIR/UVVIS ratio images. Image frames that were not included in the selected truth sets were then adjusted by computing an offset for each frame based on color ratio differences in overlapping regions of the frame being corrected and the surrounding truth-set frames. This procedure was applied to each longitude strip to derive a global set of offset frame corrections for each NIR band. Derived frame offset corrections for each NIR band were then applied to a global mosaic of Clementine NIR data at 2 km/pixel resolution. New ratio images of the NIR channels over the UVVIS 750 nm data were created from the frame offset-corrected NIR data to evaluate results. Uniformity across camera mode settings and orbits was improved in the resulting ratio images for most regions. Offset correction values for NIR frames that had not been improved by the automated offset correction algorithm due to within-image gradients or lack of adequate surrounding truth images were either eliminated from the offset corrections or adjusted manually based on visually matching values with surrounding frames in the ratio images. After applying the best set of frame offset corrections derived through the above steps, a final small offset correction was then computed for every image frame in the Clementine global mosaic to correct for residual between-orbit variations. This was carried out by differencing the global ratio of each NIR band over the 750 nm band by the same data after applying a longitudinal median filter. The width of the longitudinal median filter was varied until a filter was identified that captured albedo offsets across orbits while minimizing the influence of geologic albedo and color boundaries on the subsequently derived offset corrections. The mean difference of the pre- and post-filter ratio images was derived for all six bands of each Clementine frame, smoothed as a function of latitude within each orbit and added to the previous offset correction values. The final result of the offset deviation processing described above was a single text file containing an offset correction by NIR band for every frame in the Clementine mosaic (see nir_mosaic_frameoffset.txt). A file containing the corresponding area of Clementine frames is included along with the 0.5 km NIR mosaics available online at http://astrogeology.usgs.gov/Projects/ClementineNIR/ . These offset values were then subtracted from the radiometrically corrected NIR data to create the offset-corrected, full- resolution (100 m/pixel) mosaics. A final multiplicative correction equivalent to the month1-to- month2 correction applied to the calibrated global UVVIS data was also applied to the offset corrected NIR mosaic (frame-based NIR calibrations were made relative to the 750 nm UVVIS data before this final multiplicative correction had been applied.) Orbit-to-orbit multiplicative corrections as a function of Clementine frame color-set number are included in nir_mosaic_framemult.txt available at this Web site: http://astrogeology.usgs.gov/Projects/ClementineNIR/ . Apollo 16 Normalization Apollo 16 soil measurements previously used to normalize and calibrate the UVVIS data [Pieters et al., 1999] to units of reflectance were convolved through the first four NIR filter transmission curves. Values obtained from the soil spectra were then compared with DN values from the NIR data of the same Apollo 16 landing site location used to calibrate the UVVIS data. Multiplicative ‘reflectance normalization’ correction factors were derived for the first four bands of the NIR data (1100 - 2000 nm), and the data from NIR these bands were calibrated to reflectance. Multiplicative correction factors derived from the ‘standard’ data and the offset-corrected NIR data were consistent to ~2% or better for all four bands. Due to a decrease in orbit-to-orbit variability around the Apollo 16 landing site used for normalizations, correction values derived from the final offset-corrected mosaics were used for the final normalizations and are presented in the table below. The longest NIR wavelengths (2600 and 2780 nm bands) were not normalized to these soil measurements because reflectance information was not available at these wavelengths and their brightness’s may have been complicated by the additional presence of thermal emission signatures. Apollo 16 Coefficients for Calibration to Reflectance NIR Band Coefficient 1100 nm 0.127860 1250 nm 0.116715 1500 nm 0.158662 2000 nm 0.295333 Polar Regions The processing steps described above for the NIR mosaics were applied to data from 70 degrees N to 70 degrees S only. The polar regions were processed through 'standard' processing steps only, and the later empirical frame offset and multiplicative corrections were not applied. This was due to multiple overlapping frame coverage and the difficulty in selecting optimal image coverage and deriving a single representative offset correction value for a given frame to be corrected. Sources of Error These global Clementine NIR mosaics represent a large improvement in radiometric calibrations over the original mission calibrations [Lucey et al., 1998, 2000]. More recent work illustrates the utility of these NIR mosaics [Cahill et al., 2004; Staid et al., 2004; Gillis et al., 2005]. However, residual errors remain and these influence usage of the data. The presence and approximate magnitudes of such errors can be identified in ratio images of various bands of the global NIR data where color variations change across orbits and camera modes that cannot be attributed to surface materials. Though such color variations are less apparent in the frame corrected version of the NIR data, some residual patterns due to imperfect calibrations are known to exist in this latter version of the data as well. As a result, it may be useful to compare both sets of global data when conducting regional-scale analyses to identify local variations introduced by imperfect radiometric calibrations and empirical corrections. Due to the potential for such local variations in radiometric calibrations, it is advantageous to conduct spectral analyses of the data over relatively large regions that include several different orbits and camera modes. Possible sources of error in the Clementine NIR data include variations in photometric properties of materials within a scene, stray and/or scattered light and unknown and uncorrected phase and photometric effects for NIR wavelengths. Scattered light is possible anomalous brightness in which high-albedo units influence the measured values of low-albedo units (and vice versa) to varying degrees at different wavelengths. There is good evidence that the NIR camera suffered from stray light effects, but to a lesser and qualitatively different degree than the UVVIS camera [e.g., Lucey et al., 1998]. Other sources of error include the assumption that the Apollo 16 sample measurements are representative of (1) the larger lunar surface region used for normalization to reflectance, and (2) other lunar terrain types such as maria. Cahill et al. [2004] note that the noise levels in the corrected NIR data are near 1% (compared to ~0.5% for the UVVIS calibrated data) and that comparisons between Earth-based spectra and those derived from a dozen sites on the nearside generally correspond to within 1- sigma standard deviation. Go to Table of Contents. 10 - CONVERSION OF 16-BIT PIXEL VALUES TO FLOATING POINT NUMBERS To convert the 16-bit integer values found in the image arrays of the NIR multispectral mosaic to radiance and/or fractional reflectance an offset and scaling factor need to be applied as shown: FRACTIONAL_REFLECTANCE = (SCALING_FACTOR * DN) + OFFSET where: DN = 16-bit pixel value of NIR DIM image array. SCALING_FACTOR = 1.3500E-04 OFFSET = 0.00 The ISIS system automatically converts the 16-bit integer values to floating point values by applying the scaling factor and offset values. Go to Table of Contents. 11 - FILES, DIRECTORIES, AND DISK CONTENTS The files on each CD volume are organized starting at the root or 'parent' directory. Below the parent directory is a directory tree containing data, documentation, and index files. In the table below directory names are indicated by brackets (<...>), upper-case letters indicate an actual directory or file name, and lower-case letters indicate the general form of a set of directory or file names. DIRECTORY/FILE CONTENTS | |-AAREADME.TXT Introduction to the CD volume (ASCII Text) | |-INDEX.HTM Hypertext Markup Language (HTML) file for use | as a user-interface to files on this volume. | |-ERRATA.TXT Description of known anomalies and errors | present on the volume set (optional file). | |-VOLDESC.CAT A description of the contents of this | CD volume in a format readable by | both humans and computers. | |- Catalog Directory | | | |-CATINFO.TXT Describes Contents of the Catalog directory | | | |-DATASET.CAT Clementine NIR DIM Mosaic description | | | |-DSMAP.CAT Map Projection description | | | |-INSTHOST.CAT Clementine Spacecraft description | | | |-MISSION.CAT Clementine Mission description | | | |-PERSON.CAT Contributors to Clementine NIR Mosaic | | | |-REF.CAT References for Clementine NIR Mosaic | | | |-NIRINST.CAT NIR Camera description | |- Documentation Directory. The files in this | | directory provide detailed information | | regarding the Clementine NIR Mosaic. | | | |-DOCINFO.TXT Description of files in the DOCUMENT | | directory. | | | |-VOLINFO.TXT Documentation regarding the | | contents of this CD Volume. | | | |-VOLINFO.HTM HTML document for VOLINFO.TXT | | | |-VOLINFO.LBL PDS Label file describing the VOLINFO | | documents. | | | |-SCHEME.PDF Adobe Acrobat Portable Document Format file | showing NIR tiling and volume scheme. | |- Directory for the image index files. | | | |-INDXINFO.TXT Description of files in directory. | | | |-INDEX.TAB Image Index table specific to each volume. | | | |-INDEX.LBL PDS label for INDEX.TAB | | | |-CUMINDEX.TAB Image Index table for entire CD collection. | | | |-CUMINDEX.LBL PDS label for CUMINDEX.TAB. | |- Data directory containing NIR DIM tiles. | | | |- Data filenames where; | (For this NIR Volume Set) | t = N (Clementine NIR Mosaic) | | (For past or future Volumes) | = B (Clementine Basemap Mosaic) | = U (UVVIS Cube) | = N (NIR Cube) | = L (LWIR Image Data) | = H (Hi-res Image Data) | | s = (Resolution - km/pixel) | = A (0.004 km/pixel - future mapping) | = B-D (For future mapping as needed) | = E (0.02 km/pixel - future mapping) | = F-H (For future mapping as needed) | = I (0.1 km/pixel) | = J (0.15 km/pixel) | = K-L (For future mapping as needed) | = M (0.5 km/pixel) | = N-P (For future mapping as needed) | = Q (2.5 km/pixel) | = R-T (For future mapping as needed) | = U (12.5 km/pixel) | = V-Z (For future mapping as needed) | | pp = (00-90) Center latitude of Image File. | (Two digit truncated integer) | | y = N (Positive latitude) | = S (Negative latitude) | = (Not used for full latitude | coverage. i.e. -90 to 90) | | mmm = (000-360) Center longitude of Image. | (Three digit truncated integer) | xxx = IMG (PDS Labeled Image File) | = LAB (ISIS Detached Label File) | = JPG (JPEG small, medium, and | large Browse Images) | Directory Tree only | = HTM ( Directory Tree only) | | |- Directory tree containing enhanced color, | color ratio, and b/w (2000nm) “Browse” | (reduced resolution) JPEG images for each | image data product on the volume. | |-BROWINFO.TXT Description of content. | |-BRCOLOR.HTM Graphics (map)-based HTML interface to | enhanced color browse data | (Accessed by INDEX.HTM file). | |-BRRATIO.HTM Graphics (map)-based HTML interface to | color ratio browse data | (Accessed by INDEX.HTM file). | |-BRBW.HTM Graphics (map)-based HTML interface to | b/w (2000 nm) browse data | (Accessed by INDEX.HTM file). | |-LOCATOR.HTM Volume/Quad Locator Map. | |-CLEMLOGO.GIF Various icons, logos, and images |-USGSLOGO.GIF used by HTML documents on the volume. |-CLEMGRID.GIF |-GR17.GIF |-GR38.GIF | | |- | |- | |- Directory tree containing small, medium, | |- and large sized JPEG images (enhanced color, | color ratio, and b/w (2000nm)) for each |- DIM product. These JPEG images are primarily | |- used by the HTML documents on the volume. | |- small images are ~60x60 pixels | |- medium images are ~400x400 pixels | large images are ~1000x1000 pixels |-<2000nm> |- |- |- Go to Table of Contents. 12 - IMAGE FILE ORGANIZATION The image files are stored in a PDS compliant format. An image file contains a label area (header) at the beginning of the file followed by the image data. The number of bytes of the label area is a multiple of the number of bytes that make up an image line (number of samples * 2 bytes/pixel). The image label area contains ASCII text data that describing the image file (see Image Labels section below). The label area can be viewed with a simple ASCII editor on most computer systems. ISIS access to NIR DIM Each NIR image file (*.img filename extension) is associated with an Integrated Software for Imagers and Spectrometers (ISIS) 'qube object' detached label (.lab filename extension). This detached label allows for images to be accessed and processed within the ISIS system (available through the USGS, Flagstaff, AZ). Within ISIS, users should reference the detached label file when accessing image files. The ISIS detached label contains a pointer to the actual image file so both files (.img and .lab) should remain in the same directory if transferring data to other media. Pixel Storage Order The Clementine NIR mosaic is stored as image files with 16-bit signed integer pixels. The storage order of the pixels is "most significant byte order first." This is the storage order for Unix/Sun and Macintosh systems. For other systems such as IBM- compatible PC, Dec/Alpha, and VAX systems, the high and low order bytes of the pixels will need to be swapped before the data can be used. The ISIS program "convert" will automatically convert the data to the proper storage order. Special Pixel Values Pixels in an image array can contain special values to denote a special condition about a pixel. The following values designate a special pixel: -32768 - NULL - The pixel is "empty" and no data were acquired for this pixel location in the image array. This condition typically occurs when a gap exists in the image map such as when an image was not acquired for an area of the Moon. -32767 - Low Processing Saturation - Processing on the pixel caused its value to go outside the low-end dynamic range of the 16-bit signed integer. -32766 - Low Instrument Saturation - The pixel of the raw image was "bit clipped" and did not contain a valid value. For example, if the bias of the camera was set so that the low DN values could not be stored in the dynamic range of the raw image. -32765 - High Instrument Saturation - The pixel of the raw image was high-end saturated and could not be stored in the dynamic range of the raw image. For example, if part of an image scene was too bright to be recorded by the imaging instrument. -32764 - High Processing Saturation - Processing on the pixel caused its value to go outside the high-end dynamic range of the 16-bit signed integer. Image Labels The label area of an image file contains descriptive information about the image. The label consists of keyword statements that conform to version 3 of the Object Description Language (ODL) developed by NASA's PDS project. There are three types of ODL statements within a label: structural statements, keyword assignment statements, and pointer statements. Structural statements provide a shell around keyword assignment statements to delineate which data object the assignment statements are describing. The structural statements are: 1) OBJECT = object_name 2) END_OBJECT 3) END The OBJECT statement begins the description of a particular data object and the END_OBJECT statement signals the end of the object's description. All keyword assignment statements between an OBJECT and its corresponding END_OBJECT statement describe the particular object named in the OBJECT statement. The END statement terminates a label. A keyword assignment statement contains the name of an attribute and the value of that attribute. Keyword assignment statements are described in more detail in Appendix A of this document. These statements have the following format: name = value Values of keyword assignment statements can be numeric values, literals, and text strings. Pointer statements are a special class of keyword assignment statements. These pointers are expressed in the ODL using the following notation: ^object_name = location If the object is in the same file as the label, the location of the object is given as an integer representing the starting record number of the object, measured from the beginning of the file. The first label record in a file is record 1. Pointers are useful for describing the location of individual components of a data object. Pointer statements are also used for pointing to data or label information stored in separate files. An example of a detached label (i.e., label information stored in a separate file) is shown below: By convention, detached labels are found in the LABEL directory. ^STRUCTURE = 'logical_file_name' The value of 'logical_file_name' is the name of the detached label file containing the description. The keyword statements in the label are packed into the fixed- length records that make up the keyword label area. Each keyword statement is terminated by a carriage-return and line-feed character sequence. An example of a Clementine NIR DIM image label is shown below. Descriptions of the keywords used in the NIR label are found in Appendix A. Example PDS Label for Clementine NIR DIM Image files PDS_VERSION_ID = PDS3 /* FILE FORMAT AND LENGTH */ RECORD_TYPE = FIXED_LENGTH RECORD_BYTES = 3688 FILE_RECORDS = 10637 LABEL_RECORDS = 2 INTERCHANGE_FORMAT = BINARY /* POINTERS TO START RECORDS OF OBJECTS IN FILE */ ^IMAGE = 3 /* IMAGE DESCRIPTION */ DATA_SET_ID = "CLEM1-L-N-5-DIM-NIR-V1.0" PRODUCT_ID = "NI03N003" PRODUCER_INSTITUTION_NAME = "UNITED STATES GEOLOGICAL SURVEY" PRODUCT_TYPE = MDIM MISSION_NAME = "DEEP SPACE PROGRAM SCIENCE EXPERIMENT" SPACECRAFT_NAME = "CLEMENTINE 1" INSTRUMENT_NAME = "NEAR INFRARED CAMERA" INSTRUMENT_ID = "NIR" TARGET_NAME = "MOON" FILTER_NAME = ("A","B","C","D","E","F") CENTER_FILTER_WAVELENGTH = (1110.000,1250.000,1500.000, 2000.000,2600.000,2780.00) BANDWIDTH = (60.000,60.000,60.000, 60.000,60.000,120.000) START_TIME = "N/A" STOP_TIME = "N/A" SPACECRAFT_CLOCK_START_COUNT = "N/A" SPACECRAFT_CLOCK_STOP_COUNT = "N/A" PRODUCT_CREATION_TIME = 2006-05-25T13:30:03 NOTE = "NIR 6-BAND MOSAIC" /* DESCRIPTION OF OBJECTS CONTAINED IN FILE */ OBJECT = IMAGE BANDS = 6 BAND_STORAGE_TYPE = BAND_SEQUENTIAL BAND_NAME = "N/A" LINES = 2127 LINE_SAMPLES = 1844 SAMPLE_TYPE = MSB_INTEGER SAMPLE_BITS = 16 SAMPLE_BIT_MASK = 2#1111111111111111# OFFSET = 0.0 SCALING_FACTOR = 1.350000E-04 VALID_MINIMUM = -32752 NULL = -32768 LOW_REPR_SATURATION = -32767 LOW_INSTR_SATURATION = -32766 HIGH_INSTR_SATURATION = -32765 HIGH_REPR_SATURATION = -32764 MINIMUM = 878 MAXIMUM = 11120 CHECKSUM = 3245986915 END_OBJECT = IMAGE OBJECT = IMAGE_MAP_PROJECTION ^DATA_SET_MAP_PROJECTION = "DSMAP.CAT" COORDINATE_SYSTEM_TYPE = "BODY-FIXED ROTATING" COORDINATE_SYSTEM_NAME = "PLANETOGRAPHIC" MAP_PROJECTION_TYPE = "SINUSOIDAL" MAP_RESOLUTION = 303.2334900 MAP_SCALE = 0.1000000 MAXIMUM_LATITUDE = 7.0000000 MINIMUM_LATITUDE = -0.0132000 EASTERNMOST_LONGITUDE = 6.0131998 WESTERNMOST_LONGITUDE = 0.0000000 LINE_PROJECTION_OFFSET = 2123.6345297 SAMPLE_PROJECTION_OFFSET = 4549.5024429 A_AXIS_RADIUS = 1737.4000000 B_AXIS_RADIUS = 1737.4000000 C_AXIS_RADIUS = 1737.4000000 FIRST_STANDARD_PARALLEL = "N/A" SECOND_STANDARD_PARALLEL = "N/A" POSITIVE_LONGITUDE_DIRECTION = EAST CENTER_LATITUDE = 0.0 CENTER_LONGITUDE = 15.0000000 REFERENCE_LATITUDE = "N/A" REFERENCE_LONGITUDE = "N/A" LINE_FIRST_PIXEL = 1 SAMPLE_FIRST_PIXEL = 1 LINE_LAST_PIXEL = 2127 SAMPLE_LAST_PIXEL = 1844 MAP_PROJECTION_ROTATION = 0.0000000 VERTICAL_FRAMELET_OFFSET = "N/A" HORIZONTAL_FRAMELET_OFFSET = "N/A" END_OBJECT = IMAGE_MAP_PROJECTION END Go to Table of Contents. 13 - INDEX FILES Each CD volume in the Clementine NIR mosaic contains an image index table ('index.tab') with catalog information specific to each volume. A cumulative index table ('cumindex.tab') is also provided. It contains entries for the entire NIR DVD collection. The image index files and their associated PDS label files ('index.lbl' and 'cumindex.lbl') are located in the 'index' directory. The index table includes the file names, CD volumes, and mapping parameter information. Go to Table of Contents. 14 - ACKNOWLEDGMENTS The National Aeronautics and Space Administration is charged with the responsibility for coordination of a program of systematic exploration of the planets by U.S. spacecraft. To this end, it finances spaceflight missions and data analysis and research programs administered and performed by numerous institutions. The Astrogeology Program of the Geological Survey of the U.S. Department of Interior performs digital cartographic mapping in support of NASA's program of planetary exploration and scientific research. The Clementine NIR DIM was compiled for the National Aeronautics and Space Administration (NASA) by the United States Geological Survey (USGS) under the direction of Lisa Gaddis (USGS) and Eric M. Eliason (USGS, now at The University of Arizona). Paul Lucey, Donovan Steutel, Dave Blewett, and John Hinrichs (the latter two are now at NovaSol, Honolulu, HI) of the University of Hawaii characterized the behavior of the NIR camera and developed the radiometric calibration methods. Matt Staid (USGS, now at Planetary Sciences Institute, Tucson, AZ) provided expertise in the empirical correction of the NIR following standard processing of the NIR mosaic. Ella Mae Lee and Lynn Weller headed the USGS technical group responsible for processing the NIR image files through standard processing. Cartographic and archive design, data conversion for PDS compliance, and production of DVDs was performed by Chris Isbell. The Lunar Geometric Control network was derived by Mert Davies and Tim Colvin (both from the RAND Corporation). External validation and review were provided by Carle Pieters (Brown University) and Brad Jolliff (Washington University). Generation of the NIR DIM is a very complex process that involves and depends on an extended number of people. In Completion of this product is the result of contributions by the following individuals and groups in Astrogeology at USGS/Flagstaff: Data Processing: Ella Mae Lee, Lynn Weller, Bob Sucharski, Janet Richie, Ana Grecu, Alicyn Gitlin Data Review: Ella Mae Lee, Lynn Weller, Annie Bennett Scripting: Eric Eliason, Ella Mae Lee, Lynn Weller, Tammy Becker, Tracie Sucharski, John Shinaman, and Bob Sucharski Database: Janet Barrett, Ella Mae Lee, Bob Sucharski Data Management: Annie Bennett, Lynn Weller A variety of developers and programmers within the ISIS programming group of USGS/Flagstaff, AZ were also involved in the development of this archive. Go to Table of Contents. 15 - REFERENCES Anderson, J.A., S.C. Sides, D.L. Soltesz, T.L. Sucharski, K.J. Becker, Modernization of the Integrated Software for Imagers and Spectrometers, Lunar and Planetary Science Conference XXXV, abstract #2039, 2004. Acton, C.H., Ancillary Data Services of NASA's Navigation and Ancillary Information Facility: Planetary and Space Sciences, Vol. 44, No. 1, pp. 65-70, 1996. Batson, R.M., Cartography: in Greeley, Ronald, and Batson, eds. Planetary Mapping: New York, Cambridge University Press, pp. 60- 95, 1990. Batson, R.M., Digital Cartography of the Planets: New Methods, Its Status, and Its Future: Photogrammetric Engineering and Remote Sensing, Vol. 53, No. 9, p.1211-1281, 1987. Cahill, J.T., P.G. Lucey, J.J. Gillis, D. Steutel, Verification of quality and compatibility for the newly calibrated Clementine NIR Data Set, Lunar and Planetary Science Conference XXXV, abstract #1469, 2004. Cook, A.C., B. Semenov, M.S. Robinson, and T.R. Watters, Assessing the Absolute Positional Accuracy of the Clementine UVVIS Mosaic, Proceedings of the 36th Vernadsky-Brown Microsymposium, Moscow, Russia, CDROM abstract #ms107, 2002. Eliason, E.M., Production of Digital Image Models Using the ISIS System: Lunar and Planetary Science Conference 28th, pp. 331- 332, 1997. Eliason, E., C. Isbell, E. Lee, T. Becker, L. Gaddis, A. McEwen, M. Robinson, Mission to the Moon: The Clementine UVVIS Global Lunar Mosaic, PDS Volumes USA_NASA_PDS_CL_4001 through 4078, produced by the U.S. Geological Survey and distributed on CD media by the Planetary Data System, 1999. Eliason, E.M., E.M. Lee, T.L. Becker, L.A. Weller, C.E. Isbell, M.I. Staid, L.R. Gaddis, A.S. McEwen, M.S. Robinson, T. Duxbury, D. Steutel, D.T. Blewett, and P.G. Lucey, A Near-Infrared (NIR) global multispectral map of the Moon from Clementine, Lunar and Planetary Science Conference XXXIV, abstract #2093, 2003. Eliason, E.M., E.R. Malaret, and G. Woodward, Clementine Mission, The Archive of Image Data Products and Data Processing Capabilities: Lunar and Planetary Science Conference 26th, pp. 369-370, 1995. Eliason, E.M., A.S. McEwen, M.S. Robinson, E.M. Lee, T.L. Becker, L. Gaddis, L.A. Weller, C.E. Isbell, J.R. Shinaman, T. Duxbury, E. Malaret, Clementine: A Global Multi-Spectral Map of the Moon from the Clementine UVVIS Imaging Instrument: Lunar and Planetary Science Conference 30th, pp. 1933-1934, 1999. Eliason, E., A. McEwen, M. Robinson, P. Lucey, T. Duxbury, E. Malaret, C. Pieters, T. Becker, C. Isbell, E. Lee, Multispectral Mapping of the Moon by Clementine, New Views of the Moon 1998, abstract #6006, 1998. Gaddis, L.R., Anderson, J., Becker, K., Becker, T., Cook, D., Edwards, K., Eliason, E., Hare, T., Kieffer, H., Lee, E.M., Mathews, J., Soderblom, L., Sucharski, T., and Torson, J., An overview of the Integrated Software for Imaging Spectrometers (ISIS), Lunar and Planetary Science Conference XVIII, 387-388, 1997. Gillis, J.J., P.G. Lucey, B.A. Campbell, and B.R. Hawke, Clementine 2.7 micron data and 70-cm Earth-based radar data provide additional constraints for UVVIS-based estimates of TiO2 content for lunar mare basalts, Lunar and Planetary Science Conference XXVI, abstract #2254, 2005. Isbell, C.E., E.M. Eliason, K.C. Adams, T.L. Becker, A.L. Bennett, E.M. Lee, A.S. McEwen, M.S. Robinson, J.R. Shinaman, and L.A. Weller, Clementine: A Multi-Spectral Digital Image Model Archive of the Moon: Lunar and Planetary Science Conference 30th, pp. 1812-1813, 1999. Isbell, C.E., E.M. Eliason, E.M. Lee, K.C. Adams, T.L. Becker, A.L. Bennett, A.S. McEwen, Clementine: Reduced Resolution Digital Image Model of the Moon: Lunar and Planetary Science Conference 32nd, abstract #2076, 2001. Jet Propulsion Laboratory, Planetary Data System Standards Reference: JPL Document D-7669, Part 2, Jet Propulsion Laboratory, Pasadena, California, 2006. Lawson, S.L., B.M. Jakosky, H.-S. Park, and M.T. Mellon, Brightness temperatures of the lunar surface: Calibration and global analysis of the Clementine long-wave infrared camera data, J. Geophys. Res., 105, No. E2, 4273-4290, 2000. Lee, E.M., K.E. Edwards, T.L. Becker, D. Cook, E. Eliason, M.S. Robinson, A.S. McEwen, T. Colvin, M. Davies, T. Duxbury, T. Sorenson, 1997, Clementine UVVIS Multispectral Processing, Lunar and Planetary Science Conference XXVIII, abstract #1705, 1997. Le Mouelic, S., Y. Langevin, and S. Erard, A new data reduction approach for the Clementine NIR data set: Application to Aristillus, Aristarchus, and Kepler, Journal of Geophysical Research, v. 104, No. E2, pp. 3833-3843, 1999. Lewis, I., Clementine NIR and LWIR Cameras: Comments on Raw Data Sets, online document, http://astrogeology.usgs.gov/Projects/Clementine/nasaclem/sensor s/nir/nir_lwir_lewis_comments.html, 1995. Li, Lin, J.F. Mustard, and C.M. Pieters, The Effects of Scattered Light in the Clementine UVVIS Camera on Mixture Analysis, Lunar and Planetary Science Conference 30th, pp. 1356- 1357, 1999. Lucey, P.G., J.L. Hinrichs, and E. Malaret, Progress Toward Calibration of the Clementine Near Infrared Camera Data Set, Lunar and Planetary Science Conference XXXVIII, Abstract #1401, 1997. Lucey, P.G., J. Hinrichs, C. Budney, G. Smith, C. Frost, B.R. Hawke, E. Malaret, M.S. Robinson, B. Bussey, T. Duxbury, D. Cook, P. Coffin, E. Eliason, T. Sucharski, A. McEwen, C.M. Pieters, Calibration of the Clementine Near Infrared Camera: Ready for Prime Time, Lunar and Planetary Science Conference XXXI, Abstract #1576, 1998. Lucey, P.G., D.T. Blewett, E. Eliason, L.A. Weller, R. Sucharski, E. Malaret, J.L. Hinrichs, and P.D. Owensby, Optimized calibration constants for the Clementine NIR camera, Lunar and Planetary Science Conference XXIX, #1273, 2000. McEwen, A.S., E. Eliason, P. Lucey, E. Malaret, C. Pieters, M. Robinson, T. Sucharski, Summary of Radiometric Calibration and Photometric Normalization Steps for the Clementine UVVIS Images: Lunar and Planetary Science Conference 29th, pp. 1466-1467, 1998. McEwen, A.S., M. Robinson, Mapping of the Moon by Clementine: Adv. Space Research, Vol. 19, No. 10, pp. 1523-1533, 1997. McEwen, A.S., A Precise Lunar Photometric Function: Lunar and Planetary Science Conference 27th, pp. 841-842, 1996. Nozette, S., P. Rustan, L.P. Pleasance, D.M. Horan, P. Regeon, E.M. Shoemaker, P.D. Spudis, C.H. Acton, D.N. Baker, J.E. Blamont, B.J. Buratti, M.P. Corson, M.E. Davies, T.C. Duxbury, E.M. Eliason, B.M. Jakosky, J.F. Kordas, I.T. Lewis, C.L. Lichtenberg, P.G. Lucey, E. Malaret, M.A. Massie, J.H. Resnick, C.J. Rollins, H.S. Park, A.S. McEwen, R.E. Priest, C.M. Pieters, R.A. Reisse, M.S. Robinson, D.E. Smith, T.C. Sorenson, R.W. Vorder Breugge, and M.T. Zuber, The Clementine Mission to the Moon: Scientific Overview: Science, Vol. 266, pp. 1835-1839, 1994. Pieters, C., Compositional diversity and stratigraphy of the lunar crust derived from reflectance spectroscopy, Chap. 14 in Remote Geochemical Analysis: Elemental and Mineralogical Composition, C. Pieters and P. Englert (ed.), University of Cambridge, Cambridge, England, 167-198, 1993. Pieters, C.M., The Moon as a Spectral Calibration Standard Enabled by Lunar Samples: The Clementine Example, in New Views of the Moon II: Understanding the Moon Through the Integration of Diverse Datasets, September 22-24, 1999, Flagstaff, Arizona, abstract 8025 (http://www.planetary.brown.edu/pds/Pieters_NV99_8025.pdf), 1999. Pieters, C.M., S. Pratt, H. Hoffmann, P. Helfenstein, and J. Mustard, Bi-directional Spectroscopy of Returned Lunar Soils: Detailed "Ground Truth" for Planetary Remote Sensors: Lunar and Planetary Science Conference 22nd, pp. 1069-1070, 1991. Priest, R.E., I.T. Lewis, N.R. Sewall, H. Park, M.J. Shannon, A.G. Ledebuhr, L.D. Pleasance, M.A. Massie, and K. Metschuleit, Near-infrared Camera for the Clementine Mission, Proceedings of the Society of Photo-optical Instrumentation Engineers (SPIE), 2475, pp. 393-404, 1995. Robinson, M.S., A.S. McEwen, E.M. Eliason, E.M. Lee, E. Malaret, P. Lucey, Clementine UVVIS Global Mosaic: A New Tool for Understanding the Lunar Crust: Lunar and Planetary Science Conference 30th, pp. 1931-1932, 1999. Robinson, M.S., E. Malaret, and T. White, A calibration for the Clementine HIRES Camera, Jour. Geophys. Res., 108, No. E4, 5028, 2003. Snyder, J.P, Map Projections - A Working Manual: U.S. Geological Survey Professional Paper 1395, United States Government Printing Office, 1987. Staid, M.I., C. Isbell, L. Gaddis, E. Eliason, E.M. Lee, T. Becker, L. Weller, J. Shinaman, D. Steutel, P. Lucey, J. Gillis, A. McEwen, M. Robinson, T. Duxbury, E. Malaret, B. Jolliff, C. Pieters, Clementine Near-Infrared Global Multispectral Map, First Release, Online document and images (http://astrogeology.usgs.gov/Projects/ClementineNIR/), 2003. Staid, M.I., L.R. Gaddis, C.E. Isbell, Global comparisons of mare volcanism from Clementine NIR data, Lunar and Planetary Science Conference XXXV, abstract #1925, 2004. Torson, J.M. and K.J. Becker, ISIS - A Software Architecture for Processing Planetary Images: Lunar and Planetary Science Conference 28th, pp. 1443-1444, 1997. Go to Table of Contents. APPENDIX A - KEYWORD ASSIGNMENTS This section defines the keywords used in the image label area of the Clementine NIR mosaic. PDS_VERSION_ID = PDS3 This dataset conforms to version 3 of the PDS standards. RECORD_TYPE = FIXED_LENGTH This keyword defines the record structure of the file as fixed- length record files. RECORD_BYTES = xxxx Record length in bytes for fixed-length records (number of samples *2). FILE_RECORDS = xxxx Total number of fixed-length records contained in the file. LABEL_RECORDS = x Number of fixed-length label records in the file. INTERCHANGE_FORMAT = BINARY Data are organized as BINARY values. ^IMAGE = x Pointer to the first record that contains image data. (The first record in the file is designated as record 1.). DATA_SET_ID = "CLEM1-L-N-5-DIM-NIR-V1.0" The PDS defined dataset identifier for the Clementine NIR mosaic. PRODUCT_ID = "NI03N003" Unique product identifier for this image file. This value is the same as the file name. (Format described in the "FILES, DIRECTORIES, AND DISK CONTENTS" section above.) PRODUCER_INSTITUTION_NAME = "UNITED STATES GEOLOGICAL SURVEY" Identifies the producer organization of this data product. PRODUCT_TYPE = MDIM This keyword identifies the image product as a Mosaicked Digital Image Model (MDIM). MISSION_NAME = "DEEP SPACE PROGRAM SCIENCE EXPERIMENT" The keyword identifies the product name of the mission. (This is the official name of the Clementine Mission.) SPACECRAFT_NAME = "CLEMENTINE 1" Name of the spacecraft that acquired the data. INSTRUMENT_NAME = "NEAR INFRARED CAMERA" Name of the instrument that acquired the image data. INSTRUMENT_ID = "NIR" Abbreviated name of the instrument that acquired the image data. TARGET_NAME = "MOON" Target of the data product. FILTER_NAME = ("A","B","C","D","E","F") Images acquired from each filter of the NIR camera were used to complete the Clementine NIR DIM mosaic. CENTER_FILTER_WAVELENGTH = (1100.000,1250.000,1500.000, 2000.000,2600.000,2780.000) The center filter wavelength (nanometers) of each NIR filter. BANDWIDTH = (60.000,60.000,60.000, 60.000,60.000,120.00) The bandwidth (nanometers) of each NIR filter. START_TIME = "N/A" STOP_TIME = "N/A" SPACECRAFT_CLOCK_START_COUNT = "N/A" SPACECRAFT_CLOCK_STOP_COUNT = "N/A" Start_Time, Stop_Time, and clock counts are not applicable (N/A) for this data product but are required keywords. PRODUCT_CREATION_TIME = 2006-05-25T13:30:03 Time at which the image product was produced. NOTE = "NIR 6-BAND MOSAIC" Note field always says NIR 6-BAND MOSAIC. OBJECT = IMAGE BANDS = 6 There are five spectral bands in the NIR DIM mosaic. BAND_STORAGE_TYPE = BAND_SEQUENTIAL Storage order is band sequential. BAND_NAME = "N/A" Band name keyword is not applicable. LINES = xxxx Number of lines (rows) in image array. LINE_SAMPLES = xxxx Number of samples (columns) in image array. SAMPLE_TYPE = MSB_INTEGER Data are stored in "Most Significant Byte" order first format. This is the storage order of Sun workstations and Macintosh computers. Other systems, such as IBM/PC compatible computers and DEC/VAX systems will need to reverse the byte order of the 16-bit pixels before the data can be used. SAMPLE_BITS = 16 There are 16 bits per sample (2 bytes). SAMPLE_BIT_MASK = 2#1111111111111111# This keyword indicates all bits within a 16-bit word are used in the expression of the value. OFFSET = 0.0 SCALING_FACTOR = 1.350000E-04 The OFFSET and SCALING_FACTOR keywords contain values used to convert the 16-bit integer pixel value to radiometric units. FRACTIONAL_REFLECTANCE = (PIXEL* SCALING_FACTOR) + OFFSET VALID_MINIMUM = -32752 Lowest valid value that can be stored in pixel (always -32752). NULL = -32768 Value of empty pixels or missing data (always -32768). LOW_REPR_SATURATION = -32767 Value of pixel if processing caused a low-end value pixel to go outside dynamic range of a 16-bit signed integer (always -32767). LOW_INSTR_SATURATION = -32766 Value if pixel was low-end saturated (always -32766). For example, if the bias of the camera was set so that low DN values could not be stored in the pixel. HIGH_INSTR_SATURATION = -32765 Value of pixel if processing caused a high-end value pixel to go outside dynamic range of a 16-bit signed integer (always -32765). HIGH_REPR_SATURATION = -32764 Value if pixel was high-end saturated (always -32764). For example, if the scene was too bright for the image to record at the pixel value became saturated. MINIMUM = xxxx Minimum value in image array. MAXIMUM = xxxx Maximum value in image array. CHECKSUM = xxxxxxxx Sum of all bytes in the image object. Used to validate that an image file was properly stored on the media. END_OBJECT = IMAGE OBJECT = IMAGE_MAP_PROJECTION ^DATA_SET_MAP_PROJECTION = "DSMAP.CAT" Name of file containing additional information about the map projection. DSMAP.CAT is located in the 'catalog' directory. COORDINATE_SYSTEM_TYPE = "BODY-FIXED ROTATING" COORDINATE_SYSTEM_NAME = "PLANETOGRAPHIC" Coordinate system used in the map projection. MAP_PROJECTION_TYPE = "SINUSOIDAL" Name of map projection. MAP_RESOLUTION = xxx.xxxxx Map resolution (pixels per degree) at the reference point of the projection. MAP_SCALE = x.xxxxxx Map scale (kilometers per pixel) at the reference point of the projection. MAXIMUM_LATITUDE = xx.xxxxxxx Maximum latitude of the image file MINIMUM_LATITUDE = xx.xxxxxxx Minimum latitude of the image file. EASTERNMOST_LONGITUDE = xxx.xxxxxxx Easternmost longitude of the image file. WESTERNMOST_LONGITUDE = xxx.xxxxxxx Westernmost longitude of the image file LINE_PROJECTION_OFFSET = xxxxx.xxxxxxx SAMPLE_PROJECTION_OFFSET = xxxxx.xxxxxxx Projection offsets are used to define the relationship between line and sample of the image array and the latitude and longitude coordinate on the surface of the planet. See 'dsmap.cat' file located in the 'catalog' directory for information on these keywords. A_AXIS_RADIUS = 1737.4000000 B_AXIS_RADIUS = 1737.4000000 C_AXIS_RADIUS = 1737.4000000 Three-axis radius of the Moon used in the derivation of the map products that make up the NIR 6-band mosaic. FIRST_STANDARD_PARALLEL = "N/A" SECOND_STANDARD_PARALLEL = "N/A" Standard parallels of map, not used in this Sinusoidal Equal- Area projection. POSITIVE_LONGITUDE_DIRECTION = EAST The Moon coordinate system uses a positive longitude direction of east. Longitude values increase in the eastern direction. CENTER_LATITUDE = 0.0 Center latitude of the map projection. CENTER_LONGITUDE = xxxx.xxxx Center longitude of the map projection. REFERENCE_LATITUDE = "N/A" REFERENCE_LONGITUDE = "N/A" Reference latitude and longitudes are not used in the Sinusoidal Equal-Area projection. LINE_FIRST_PIXEL = 1 SAMPLE_FIRST_PIXEL = 1 The first pixel (upper left) in the image array is defined as line 1, sample 1. LINE_LAST_PIXEL = xxxx SAMPLE_LAST_PIXEL = xxxx The last pixel (lower right) in the image arrays is defined by these keywords. MAP_PROJECTION_ROTATION = 0.0000000 Map projection rotation always 0 for the Clementine NIR DIM. VERTICAL_FRAMELET_OFFSET = "N/A" HORIZONTAL_FRAMELET_OFFSET = "N/A" These keywords are not applicable for the Sinusoidal Equal-Area projection. END_OBJECT = IMAGE_MAP_PROJECTION END Go to Table of Contents. APPENDIX B - GEOMETRIC DEFINITION OF A PIXEL The purpose here is to describe the spatial or geometric definition of a pixel used in the generation and utilization of the digital image products. A broad range of factors enters into this question. For example, is a pixel to be conceived of as a point or as an area? The point definition would be most convenient, for instance, when dealing with coordinate grid overlays. This results in an odd number of pixels across a map that has an even number of spatial increments. For changing scales (for instance by even powers of 2) this definition becomes a problem. In this case it makes more sense to treat a pixel as a finite area. Then an even number of pixels covers an even number of spatial increments and decreasing/increasing scales by a power of 2 becomes trivial. However, grids now fall between pixels, at least in a mathematical sense. Their treatment in the generation of hardcopy therefore becomes an issue. It was decided that the area concept of a pixel was the better choice; we would have to live with the asymmetries introduced in things like cartographic grids. There are various solutions: (1) use two pixels for the width of a grid line, (2) stagger grid pixels back-and-forth across the mathematical position, (3) use a convention whereby grid lines are systematically drawn offset from their mathematical position. The next issue is the conversion between integer coordinates and real coordinates of the pixel mesh. We adopt the convention that pixels are numbered (or named if you like) beginning in the upper left corner with line 1, sample 1 (pixel 1,1); lines increase downward; samples increase to the right. (Even this is not a universal standard; some astronomical systems begin, perhaps more logically, in the lower left corner.) There are three reasonable possibilities for aligning a real, or floating point, coordinate system with the pixel mesh: the coordinate 1.0, 1.0 could be the upper left, the center, or the lower right of pixel 1,1. The convention historically used for geometric calibration files (Reseau positions) and also used in the Multimission Image Processing Laboratory at the Jet Propulsion Laboratory, is that the center of the pixel is defined as its location in real coordinates. In other words, the real coordinates of the center of pixel 1,1 are 1.0, 1.0. The top left corner of the pixel is 0.5, 0.5 and the bottom right corner is 1.49999...,1.499999. The bottom and right edge of a pixel is the mathematically open boundary. This is the standard adopted in the image products. Cartographic conventions must also be defined. The map projection representation of a pixel is mathematically open at the increasing (right and lower) boundaries, and mathematically closed at its left and upper boundaries. An exception occurs at the physical limits of the projection; the lower boundary of the lowest pixel is closed to include the limit of the projection (e.g. the south pole). The figure below shows the coordinates of Pixel 1,1. Coordinates of Pixel 1,1 longitude 180.0 179.00001 | | latitude | | line 90.0 -- ----------------- -- 0.5 | | | | | | | | | + | | (1.0,1.0) | | | | | | | 89.00001 -- ----------------- -- 1.49999 | | | | sample 0.5 1.49999 We must also select a convention for drawing grid lines for various cartographic coordinates on planetary images and maps. The convention used for image products is that a grid line is drawn in the pixels that contain its floating point value until the open boundary is reached and then an exception is made so the outer range of latitude and longitude will always appear on the image. This means, in the example above, a 10 degree grid would start on pixel 1 and be drawn on every tenth pixel (11, 21, 31,...) until the open boundary is reached. Then the line would be drawn on the pixel previous to the open boundary (line 180 instead of line 181, or sample 360 instead of 361). To summarize, the conventions are: 1) Pixels are treated as areas, not as points. 2) The integer coordinates begin with 1,1 (read, "line 1, sample 1") for the upper-left-most pixel; lines increase downward; samples increase to the right. 3) Integer and floating point image coordinates are the same at the center of a pixel. 4) Grids will be drawn in the pixels that contain the floating point location of the grid lines except for open boundaries, which will be drawn to the left or above the open boundary. Go to Table of Contents. 04/25/07 final?? 2