Data Set Overview
  =================
  Near-infrared spectral observations of 27 asteroids located at ~2.5 AU were
  obtained using the NASA IRTF SpeX instrument covering the 0.7 to 2.5 micron
  spectral interval. The reflectance spectra were obtained using the
  low-medium resolution spectrograph SpeX (RAYNER et al. 2003).  SpeX was used
  in the low- resolution spectrographic mode (asteroid mode) for two reasons:
  1) its ability to obtain spectra with a fairly high signal to noise ratio
  (SNR) even for weak signal received from asteroids and 2) its ability to
  resolve broad absorption features produced by mafic silicate minerals.
 
  The data in each asteroid spectral file contains three columns: wavelength
  (in microns), relative reflectance, and uncertainty in relative reflectance.
  The index.tab includes a listing of the product name (spectrum asteroid
  number and name), data file name (asteroid number.tab), target name(asteroid
  name and number), target type (asteroid), UT start and end dates of the
  first and last asteroid observations used in creating the final reduced,
  calibrated average spectrum, number of exposures used in creating the final
  reduced, average calibrated spectrum, net integration time, apparent V-mag,
  phase angle, geocentric distance, heliocentric distance of the asteroid at
  the time of the respective asteroid observations.  The solar analog star
  used for each asteroid's spectral calibration is listed. The listed airmass
  is the airmass at the start of the first observation of the respective
  asteroid.
 
  Observations
  ============
  During an observing run the telescope nodded between the A beam and the B
  beam(For SpeX, the telescope nod distance is 7.5 arcsec, positioning the
  spectrum image at 1/4 and 3/4 distance along the 15 arcsec slit), so images
  were taken in spectral image pairs of the target asteroid, local standard
  star, solar-analog stars, and calibration flat-field and argon arc-lamp
  images.  To produce high quality NIR asteroid reflectance spectra, we
  empirically derived the atmospheric extinction coefficients at each
  wavelength for each night or portion of the night. To model the atmospheric
  extinction over Mauna Kea, the slopes/intercepts of the relationship between
  the log of the flux (apparent magnitude) vs. airmass were calculated for
  each local standard star observation series.  The observational procedure to
  produce the slopes and intercepts required pairing each asteroid with a
  nearby solar type star that experienced the same atmospheric conditions
  (temporally and spatially). The local standard star and the asteroid
  observations were temporally and spatially related as well and were
  interspersed within the same air mass range (typically observed at airmasses
  less than 1.5). The IRTF SpeX instrument uses an argon lamp as its
  wavelength reference. Each night several argon arc spectra were obtained for
  wavelength calibration.
 
  Data Reduction
  ==============
  The data were reduced and analyzed at the University of North Dakota.  The
  obtained raw spectra were in the form of Flexible Image Transport System
  (FITS) images.  Two software packages were used to process the data: 1) the
  Unix-based Image Reduction and Analysis Facility (IRAF) from the National
  Optical Astronomy Observatories (NOAO), and 2) SpecPR a Windows-based
  program for reduction and analysis of near-infrared spectra stored in
  one-dimensional arrays (CLARK 1980; GAFFEY 2003). The extraction of spectra,
  background sky subtraction, summing the image rows encompassing the object
  flux, conversion to text files, and determination of wavelength calibration
  were done using IRAF.
 
  Spectral processing using SpecPR involved many important phases to achieve a
  final, reduced spectrum.  Important operations included: 1) calculation of
  starpacks from standard stars, 2) channel shifting to account for
  instrumental flexure, 3) averaging routines, and 4) data analysis (division
  of individual asteroid spectra by the relevant starpack and solar
  analog/standard star, polynomial fits, and determination of band
  positions/centers/band area ratios).  Detailed descriptions of this method
  can be found in (FIEBER-BEYER 2010; REDDY 2009; ABELL 2003; HARDERSEN 2003;
  GAFFEY 2003; GAFFEY ET AL. 2002). A brief overview of how SpecPR creates a
  final, reduced averaged nightly spectrum is as follows: each asteroid
  observation was divided at each wavelength by the star flux calculated from
  the selected starpack, which not only encompassed the asteroid in airmass,
  but also most effectively removed the 1.4 micron and 1.9 micron telluric
  water vapor absorption features from the spectrum.  Specifically,
  atmospheric extinction coefficients (starpacks) are computed from two or
  more sets of standard star observations. Extinction coefficients are
  determined for all sets of standard star observations (whole night
  starpacks), for portions of each night, and for individual sequential sets
  of standard star observations.
 
  Starpacks are used to calculate the standard star flux as a function of
  wavelength at the same airmass as each asteroid observation. The individual
  asteroid flux spectra were divided by the computed standard star flux ratio.
  The asteroid/star spectrum which most accurately canceled the atmospheric
  water vapor absorptions is selected as the best reduction. Since the stars
  and the asteroids were measured with the same instrument, the
  wavelength-dependent instrumental response was cancelled out when the
  asteroid flux measurements were ratioed to the extinction-corrected standard
  star flux measurements. The solar analog star data were reduced by the same
  method and used to correct for any non-solar behavior of the local standard
  stars producing a reflectance spectrum. Individual best spectra were reduced
  by this technique then after inspection for spurious sets - averaged
  together to produce a nightly average spectrum for each asteroid.  The best
  reductions were normalized across a spectral interval to avoid the noise
  that may be associated with any individual point in the spectrum. The
  normalization value was the average of the values in SpeX channels 380 to
  410. (1.5 to 1.7 microns). This interval was chosen because it is
  essentially unaffected by either the 1.4 or 1.9 micron atmospheric water
  vapor absorptions, or by the mafic silicate absorption feature centered in
  the 1.8 to 2.5 micron interval, common in asteroid spectra.
 
  A total of 27 asteroids were included in the survey.  The data have been
  published in FIEBER-BEYER 2010; FIEBER-BEYER AND GAFFEY 2011; FIEBER-BEYER
  ET AL. 2011a; and FIEBER-BEYER ET AL. 2011b.  This is an ongoing survey and
  data will be added to this archive yearly as they are published.
 
  References
  ==========
  Abell, P.A., Near-IR reflectance spectroscopy of main belt and near-Earth
  objects: A study of their composition, meteorite affinities and source
  regions. Ph.D. dissertation. Rensselaer Polytechnic Institute, Troy,NY, USA,
  2003.
 
  Clark, R.N., A large-scale interactive one-dimensional array processing
  system. Publ. Astron. Soc. Pac. 92, 221-224, 1980.
 
  Fieber-Beyer, S.K., Mineralogical characterization of asteroids in/near the
  3:1 Kirkwood Gap, Ph.D. Dissertation, University of North Dakota, Grand
  Forks, 203 pp, 2010.
 
  Fieber-Beyer, S.K., and M.J. Gaffey, Near-infrared Spectroscopy of 3:1
  Kirkwood Gap Asteroids (3760) Poutanen and (974) Lioba. Icarus, 214,
  645-651, doi:10.1016/j.icarus.2011.06.014, 2011.
 
  Fieber-Beyer,  S.K., M.J. Gaffey, and P.A. Abell, Mineralogical
  characterization of Near Earth Asteroid (1036) Ganymed, Icarus, 212,
  149-157, doi:10.1016/j.icarus.2010.12.013, 2011a.
 
  Fieber-Beyer, S.K., M.J. Gaffey, M.S. Kelley, V. Reddy, C.M. Reynolds, and
  T. Hicks, The Maria Asteroid Family: Genetic Relationship and a Plausible
  Source of Mesosiderites near the 3:1 Kirkwood Gap. Icarus, 213,
  doi:10.1016/j.icarus.2011.03.009, 524-537, 2011b.
 
  Gaffey, M.J., Observational and Data Reduction Techniques to Optimize
  Mineralogical Characterizations of Asteroid Surface Materials. Lunar.
  Planet. Sci. XXXIV, [abstract 1602], 2003.
 
  Gaffey, M.J., E.A. Cloutis, M.S. Kelley and K.L. Reed, Mineralogy of
  asteroids. In Asteroids III (W. F. Bottke, A. Cellino, P. Paolicchi and R.
  P. Binzel, Eds.), Univ. of Arizona Press, pp. 183-204, 2002.
 
  Hardersen, P.S., Near-IR Reflectance Spectroscopy of Asteroids and Study the
  Thermal History of the Main Asteroid Belt. Ph.D. Dissertation. Rensselaer
  Polytechnic Institute, Troy, New York, USA, 2003.
 
  Rayner, J.T., P.M. Onaka, M.C. Cushing and W.D. Vacca, Four Years of Good
  SpeX, in Ground-based Instrumentation for Astronomy, A. F. M. Moorwood and
  M. Iye, Eds., Proceedings of the SPIE, vol. 5492, pp. 1498-1509, 2004.
 
  Reddy, V., Mineralogical Survey Of Near-Earth Asteroid Population:
  Implications For Impact Hazard Assessment And Sustainability Of Life On
  Earth, Ph.D. dissertation. University of North Dakota, Grand Forks, 2009.

Confidence Level Overview
  =========================
    The spectra were obtained over several years (2000-2009); thus, nightly
    sky conditions varied.  Uncertainties and point to point scatter differ in
    each asteroid spectrum.  The uncertainty in each spectrum is affected by
    the level of the signal to noise achieved, changing atmospheric
    conditions, and instrument stability. The atmospheric water absorption
    bands at 1.4 and 1.9 microns may not be completely removed from the
    averaged spectrum.  The uncertainties associated with each channel in each
    asteroid spectrum are standard errors of the mean of the values averaged
    to get the reflectance in that channel.  These errors are not based on
    Poisson statistics, but are instead a measure of systematic variations
    among the individual data making up the combined spectrum.
    The May 2009 spectra start and end times of observations are listed as
    hh:mm as recorded from the observing logs; the fits headers for each
    spectrum were not able to be extracted due to a fire shortly after the
    IRAF extracted spectrum was transferred to a windows based machine.