Data Set Information
DATA_SET_NAME ANGLO-AUSTRALIAN OBSERVATORY DATA FROM SL9 IMPACTS
DATA_SET_ID EAR-J-AAT-3-EDR-SL9-V1.0
NSSDC_DATA_SET_ID
DATA_SET_TERSE_DESCRIPTION
DATA_SET_DESCRIPTION Data Set Overview: The data set contains imaging, spectroscopy, and spectral mapping ofJupiter from July 16, 1994 through July 23, 1994. This includes datafrom the C, D, G, K, N, R, V, and W impacts in addition to before andafter observations for baseline and comparison. The imaging for the V impact is composed of .455 second exposures of theplanet taken about 10 seconds apart with one image per file. This modehas a low duty cycle, but allows continuous monitoring of the eventwithout significant loss of large blocks of time which is encounteredwhen dumping a data cube to disk. The imaging for the N impact is composed of 100 .455-second exposurestaken one after another and then all dumped onto disk at once in a60x60x100 cube with the x and y dimensions corresponding to spatialdimensions (xy planes are IMAGES in this case, in comparison to thespectral mapping data, where the yz plane holds the image) and the zdimension corresponding to time. These large image cubes took asignificant amount of time to dump to disk, causing the N data to becomposed of 45 second periods of constant monitoring with excellenttemporal resolution interspersed with data gaps of approximately 40seconds. The majority of the impact sequences (C, D, G, K, R, W) were done usingspectral mapping. This is done by taking multiple spectra through aslit and then moving the slit back and forth across the planet to obtaina second dimension of spatial information. So: these cubes will (mostlikely) be 128x128x153 cubes with the x dimension being a spectraldimension and corresponding to a wavelength scale between 2.2 and 2.39microns for K-band data. The y dimension corresponds to spatialinformation, with high y values being farther to astronomical east. Thez dimension corresponds to both north-south spatial information andtemporal information as well. The slit was moved first from south tonorth, and then from north to south across the planet to produce eachdiscrete cube. Thus when a ytplane is cut through the cube at a desiredwavelength range, two images of Jupiter appear in the resulting image.The image on the left has south to the left and north to the right, andthe image on the right is flipped, having north to the left and south tothe right. It is important to also remember that the image on the rightis actually from a later point in time than the image on the left, thusfast moving objects (satellites) may change position slightly betweenthe two images and fast changing objects (impact sites) will change ineither morphology or intensity (or both) between the two images as well.For a full planet scan, the data cube takes about 4 minutes to completea double scan and 1-2 minutes to dump to disk. We therefore monitor theimpact region with a temporal resolution of 3 minutes in the full scancase, but this time interval can be decreased by sacrificing spatialcoverage. This compromise was made for parts of the C, K and Gsequences, resulting in temporal resolution as fast as 15 secondsbetween impact spectra. Therefore the spectral mapping achievescoarser temporal resolution compared to photometric imagingobservations, but in return provides near simultaneous acquisition ofspatially resolved spectra across the entire impact region. There are also several image cubes of nonstandard size, such as thosetaken with a small (60 pixel) window and/or those which scan across theimpact site only, but many times thus improving the temporal resolutionof the impact data. Lastly, there are many two-dimensional spectra such as sky frames andstandard stars which are similar to the spectral mapping cubes but withonly one frame. In these, like in the mapping cubes, the x dimensionscorrespond to spectral data and the y dimensions to spatial data. Parameters: Wavelengths are measured in microns and intensitiesin Watts per meter squared per micron.(W/(m^2*um) or sometimes W*m^-2*um^-1). In the spatial dimension, pixels are 0.6 arcseconds square (yz). The slit used was 1.2 arcseconds wide. Coordinate System: 1950AD equinox. Software: No software is provided with this data set. However, this data set waswritten using the Figaro software package which is publicly availablefrom the AAO. See the following URLs for access to this softwarepackage: http://www.aao.gov.au/ http://www.aao.gov.au/figaro.html Asample summary of how to view the images using Figaro follows: In thefollowing description, Figaro commands are shown lowercase in linecommands, and in uppercase in the general text. However, they shouldalways be typed all lowercase on Unix systems. Once Figaro is installed and running: Type FIGDISP to bring up the Figaro display window. On a Sun, press theF6 key to bring up the pixel position and value display on the top barof the Figaro window (pressing F5 will display help information on thefeatures of this image display program). To read in a given *.fit file, type RDFITS with a carriage return andanswer the prompts for input and output file names. On a Sun, take thedefaults (yes) for byte swapping and conversion to floating point. Notethat all the parameters can be entered on the command line, thus: rdfits filename.fit figaro_out \\ will read in filename.fit, take the defaults for byte swapping andconversion to floating point, and output a file figaro_out.sdf which isin Figaro format. To add the wavelength scale, use RDIPSO to convert the wavelengths.datfile into a Figaro file: rdipso wavelengths.dat 2 wavelength_scale \\ This will create a Figaro file wavelength_scale.sdf which should then beXCOPYd onto the Figaro data, so: xcopy figaro_out wavelength_scale figaro_out This will replace the x-axis in the file figaro_out with wavelengthvalue in microns. The above command will write the output file backover the input file. If you wish to keep both steps, change the second'figaro_out' to something else. The Figaro format cube can now be examined using the commands XYPLANE,YTPLANE, IMAGE and EXTRACT. YTPLANE will slice through the cube in the YT direction (correspondingto the YZ direction in the PDS data) to show image planes within thecube. The user must specify over which X range (wavelength) the imageis to be extracted. To extract an image in the wavelength range 2.32 to2.38um, use the command ytplane figaro_out 2.32 2.38 ytimage \\ If you have not tacked the wavelength scale on yet, simply specifying arange of pixel values will also work. Now, ytimage can be displayed: image ytimage reset opt:0 \\ will give an autoscaled version of the image in the figdisp window. Youcan also simply type in IMAGE, carriage return, and answer the promptsto optimize the display. The command XYPLANE works in much the same way, extracting and addingtogether a range of contiguous planes in the t (PDS z) direction. t inthis case corresponds to time, or scan step, and in the ytimage you havejust displayed, you can isolate which t pixels correspond to the desiredspatial region you wish to extract a spectrum from (e.g. the impacts,or perhaps the poles). The t pixels correspond to the x-axis directionin a ytplane. Once you have your chosen t range (say, t1 to t2, where t1and t2 are integers) from the ytplane image, use: xyplane figaro_out t1 t2 xyimage \\ image xyimage reset opt:0 \\ This will now show you the spatially resolved spectra for that scan timeand region, with the x axis corresponding to x in the cube, and beingthe wavelength axis, and y being a spatial dimension along the slit.Isolate which spectrum you wish to extract on the xyimage by choosing arange of y values (say y1 to y2) and then use: extract xyimage y1 y2 impact_spectrum \\ To display the spectrum, use the figaro SPLOT command: splot impact_spectrum wh au \\ will bring up a graphics window and autoscale the spectrum. Again,typing SPLOT and carriage return will allow you to step through theprompts and optimize the display. Further information on more detailed functions with Figaro can be foundat http://www.aao.gov.au/figaro.html The QUBE OBJECT does not have much software support in the PDS at thistime. However, because these files are in FITS format, they are easilyread in by a standard FITS reading routine such as in IDL.
DATA_SET_RELEASE_DATE 1997-10-01T00:00:00.000Z
START_TIME 1994-07-17T05:03:08.000Z
STOP_TIME 1994-07-22T12:25:59.000Z
MISSION_NAME SUPPORT ARCHIVES
MISSION_START_DATE 2004-03-22T12:00:00.000Z
MISSION_STOP_DATE N/A (ongoing)
TARGET_NAME SL9
JUPITER
TARGET_TYPE COMET
PLANET
INSTRUMENT_HOST_ID AAO
INSTRUMENT_NAME IRIS
INSTRUMENT_ID IRIS
INSTRUMENT_TYPE CCD/SPECTROGRAPH
NODE_NAME Planetary Atmospheres
Small Bodies
ARCHIVE_STATUS ARCHIVED
CONFIDENCE_LEVEL_NOTE Confidence Level Overview: The AAT data were taken in photometric conditions with seeing that wastypically 1 arcsec. Poorer seeing was experienced during the R impact(up to ~ 5''). However, although the data were taken in photometricconditions, the drift-scanning technique used to take the majority ofthe data contains inherent errors which compromise photometric accuracywhen used on point sources (such as the early stages of the impactphenomena). As the narrow slit is stepped across the object, variationsin seeing result in a time-varying point-spread function (PSF) whichchanges both FWHM and spatial position, making it difficult to correctlysample the object's PSF. This effect can be ameliorated by takingseveral scans back and forth across the object to average the effects ofthe seeing. However, for the rapidly changing phenomena associated withthe impact events, this was not possible. In an attempt to quantify theerror, we performed aperture photometry on the Galilean satellitespresent in the impact data (as approximations to point sources) andfound typical standard deviations of 10%, with errors as large as 23% on21 July (R impact). This error is correlated with the seeing, beinglarger on nights of poorer seeing. However, when the observations areaveraged, absolute fluxes derived for the same satellite observed ondifferent nights show a standard deviation of 12% for Europa, which wasobserved on 17, 19 and 21 July (if measurements taken only on 17 and 19July are considered, this discrepancy falls to 3%), and 1% for Io, whichwas observed on 19 and 21 July.
CITATION_DESCRIPTION Meadows,V., ANGLO-AUSTRALIAN OBSERVATORY DATA FROM SL9 IMPACTS, EAR-J-AAT-3-EDR-SL9-V1.0, NASA Planetary Data System, 1997.
ABSTRACT_TEXT This SL9 data set contains imaging, spectroscopy, & spectral mapping of Jupiter from 1994Jul16 through 1994Jul23. This includes data from the C,D,G,K,N,R,V, and W impacts in addition to before and after observations for baseline and comparison.
PRODUCER_FULL_NAME VIKKI MEADOWS
SEARCH/ACCESS DATA
  • Atmospheres Online Archives
    		•