From CedarWiki
Jump to: navigation, search

Kitt Peak H-alpha Fabry-Perot Interferometer


Contact Persons

Web Sites


The Wisconsin H-alpha Mapper (WHAM) Fabry-Perot Interferometer is operated at the Kitt Peak National Observatory by the Wisconsin Galactic Astronomy Group with support from the National Science Foundation.


Data Description

The Wisconsin H-alpha Mapper (WHAM) Fabry-Perot is located at the Kitt Peak Observatory (31.98N, 111.60W; alt 2120.0 m), and has been operated remotely by the University of Wisconsin Galactic Astronomy Group since 1997 [Reynolds, 1997; Haffner et al., 2003]. The geocoronal atomic hydrogen is studied using the H-alpha (656.274 nm) and H-beta (486.1 nm) emissions. The WHAM instrument is a double-etalon Fabry-Perot coupled to a Charge Coupled Device (CCD) camera where the optics are designed to image the Fabry-Perot annular interference pattern onto the CCD. Hence, the CCD images are spectral and not spatial. The geocoronal atomic hydrogen can be observed both at H-alpha (6563 A, ~1-10 Rayleighs), and H-beta (4861 A, ~0.25-1 R).

The terrestrial H-alpha emission is primarily excited by the line center portion of the solar Lyman-beta emission and thus depends on both the solar excitation flux and the density distribution of upper atmospheric hydrogen. To first order, the column of H-alpha emissions (code 2502) observed by the Fabry-Perot comes from the sunlit portion of the geocorona above the Earth's shadow. Since density falls off with height, the shadow height (code 186) is roughly just below the peak emission height. However, a portion of the signal (~1-2 R) is also due to multiple scattering of Lyman-beta radiation below the Earth's shadow height in darkness. This contribution becomes increasingly significant for observations at higher shadow altitudes. The solar zenith angle (code 180) for all observations is below the horizon (e.g. nighttime conditions).

The semi-raw spectral images are saved in FITS (.fts) format where only a portion of the chip was used and the read noise was reduced using 4 x 4 on-chip binning [Haffner et al., 2003]. The CEDAR Database FITS files include additional header information on the observing geometry, wavelength (H-alpha set at 6564 A in the original .fits headers starting in 2001), kinst (5190) and kindat (7001 for H-alpha or 7101 for H-beta).

The kindat 17000 series files contain information from the H-alpha spectral profile (.spe) where the annular interference pattern in the .fts file is summed. Equal area annuli correspond to equal spectral intervals (measured in wavenumber, so the arbitrary pixel intensities are summed in equal area annuli to produce a spectral profile of relative emission as a function of wavenumber expressed as spectral displacement in velocity units km/s with an arbitrary zero (code 2416). The relative emission is divided by the exposure time (code 60) from the .coors file to create a relative emission rate (codes 4145 for NAN calibration files and 4146 for other files). Hot pixels due to cosmic rays, dark counts, reflections and a constant average bias have been removed. The standard deviation of the relative pixel intensity in each annuli divided by the exposure time is reported as the 'error bar' for the data point (codes -4145 and -4146). Codes 2507 and 2509 are the relative annular emission after the data have been normalized with a white light flat field (from the .dat files) divided by the exposure time to make an emission rate. The initial 31 spectral displacement points (below the arbitrary value of -50 km/s) are outside the aperture and are removed in the original .dat file and also in kindats 17000-3,17100-3.

Additional information in nightly catalog files come from the headers from the FITS files and from shadow altitude calculations. This information is listed in the nightly .shad and .coors files. The UT times in the names (codes 4142 and 4143) of the original FITS files (.fts) are from .log file UTs that are contained in companion nightly catalog files. The actual UT start times listed in the FITS headers are a few seconds later than the UT start times in the .log file and reflect when the commands were started by the WHAM instrument.

Astronomical locations are given in two coordinate systems: (1) galactic coordinates with galactic longitude b in degrees (code 193) and galactic latitude l in degrees (code 195), and (2) equatorial coordinates with right ascension (RA) in hours (code 192) and declination in degrees (code 194).

The hour angle (code 191) is defined as the difference between the Local Siderial Time and the Right Ascension of the object. The hour angle is zero when the target transits the Local Meridian, which passes through the zenith direction. Hour angles are negative when the object is rising, and are positive when the object is setting. Atmospheric extinction is minimized when zenith angles are restricted to 50 degrees or less (ZA<50), which are equivalent to elevation angles (code 142) of 40 degrees or more (el>40).

Typically, WHAM locks onto a given astronomical location for integration times (code 60) of 30 to 600 seconds, so the azimuth and elevation angles (codes 132 and 142) are initial values, while RA and DEC (codes 192 and 194) remain constant.

All observations of H-alpha column emissions from the thermosphere plus exosphere are made at night during moonless conditions. All observations reported to the CEDAR Database are also made during clear sky conditions, since even high cirrus clouds can affect the H-alpha signal. Acceptable values of the quality code 4144 are:

  1. =A Excellent conditions
  2. =A- Clouds sighted much later or earlier in the night
  3. =B Clouds sighted within a few hours of the observation OR spectrum close to sunset or sunrise at the observatory OR ZA>50 (el<40)
-32767    Missing (consider ZA<50 or el>40 as best data) 

If the wing of a spectrum is truncated, then its grade is reduced. Sometimes a grade is given for each spectrum, sometimes for a single night, and sometimes grades are missing (-32767). Single nightly grades show as text in the nightly catalog record, and are marked as missing for each spectrum without an individual grade. Spectra with ZA>50 (el<40) are of lower quality. For ZA>80 (el<10), the data can be so compromised (e.g. very low emissions rates) that they were not included in this database.

The regions of low galactic emission are useful for geocoronal observations because uncertainty in the retrieval of the geocoronal emission due to the presence of galactic emission is minimized. The fitting does not require removing the galactic emission in the low galactic region of observations. Geocoronal observations are also available from zenith and other directions, especially in survey runs. The analysis for survey data may include one or two Gaussians in the fit, while the cleaner low galactic emission regions are usually analyzed using two Gaussians to account for fine structure. The absolute intensities (code 2502) are calibrated using the bright H-alpha emission at (RA 20.97, DEC 44.6 or l,b 85.60 -0.72) of 800 R +/-10% from the North American Nebula (NAN). There is an additional ~5% uncertainty in the relative calibration due to night-to-night variability in the transmittance of the atmosphere. These uncertainties are larger than the standard deviation (codes -4145 and -4146) of pixel intensity in the annuli divided by the exposure time.

Locations (code 4141) used for geocoronal observations or calibrations combined with the H-alpha kindats are:

  • -1/17001 = NAN (North Amercian Nebula calibration source)
  • 0/17000 = zenith (geoz, useful to compare with some models)
  • 1/17003 = surveys or other at all different (non-zenith) directions
  • 7/17002 = Lockman (low galactic emission region at l,b 148.5 53)
  • 8/17002 = Newoff (low galactic emission region at l,b 163.5 53.5)
  • 9/17002 = HD93521 (low galactic emission region at l,b 183.14 62.15)
  • 10/17002 = Off01 (low galactic emission region at l,b 166.7 26.31)
  • 11/17002 = Off02 (low galactic emission region at l,b 140.97 50.07)

Further details of the procedure are in the references, in the generic kindat catalog records, and are available from the contact persons.

References for the instrument and data processing procedures

Coakley, M.M., F.L. Roesler, R.J. Reynolds, and S. Nossal (1996), Fabry-Perot/CCD annular summing spectroscopy: Study and implementation for aeronomy applications, Appl. Opt. 35, 6479-6493.
Gaustad, J.E., McCullough, P.R., Rosing, W., VanBuren, D., 2001. A robotic wide-angle H-alpha survey of the southern sky. Publications of the Astronautical Society of the Pacific, 113, 1326.
Haffner, L.M., R.J. Reynolds, S.L. Tufte, G.J. Madsen, K.P. Jaehnig, and J.W. Percival (2003), The Wisconsin H-alpha mapper northern sky survey, Astrophys. J., 149, 405-422.
Hausen, N.R., Reynolds, R.J., Haffner, L.M., Tufte, S.L., 2002. Interstellar H-alpha line profiles toward HD 93521 and the Lockman Window. Astrophysical Journal 565, 1060-1068.
Leen, T. (1979), Application of radiative transfer theory to photometric studies of astronomical objects, M.S. thesis, Univ. of Wis., Madison.
Meier, R.R., 1995. Solar Lyman-series line profiles and atomic hydrogen excitation rates, Astrophysical Journal 452, 462.
Mierkiewicz, E.J., 2002. Fabry-Perot observations of the hydrogen geocorona, Ph.D. Thesis, University of Wisconsin, Madison, WI.
Mierkiewicz, E.J., F.L. Roesler, S.M. Nossal, and R.J. Reynolds, (2006), Geocoronal hydrogen studies using Fabry-Perot interferometers, Part 1: Instrumentation, observations, and analysis, J. Atmos. Solar-Terr. Phys., 68, 1520-1552, doi:10.1016/j.jastp.2005.08.024. (Link to JASTP .pdf at: and to author's .pdf at: TBD)
Nossal, S., 1994. Fabry-Perot observations of geocoronal hydrogen Balmer-alpha emissions. Ph.D. Thesis, University of Wisconsin, Madison, WI.
Nossal, S.M., E.J. Mierkiewicz, F.L. Roesler, R.J. Reynolds, and J. Bishop, Geocoronal hydrogen studies using Fabry-Perot interferometers, Part 2: Long-term observations, (2006), J. Atmos. Solar-Terr. Physics, 68, 1553-1575, doi:10.1016/j.jastp.2005.08.025. (Link to JASTP .pdf at: and to author's .pdf at: TBD)
Nossal, S. Roesler, F.L., Coakley, M.M., Reynolds, R.J., 1997. Geocoronal hydrogen Balmer alpha line profiles obtained using Fabry-Perot annular summing spectroscopy: effective temperature results. Journal of Geophysical Research, 102, A7, 14541-14553.
Nossal, S.M., F.L. Roesler, E.J. Mierkiewicz, and R.J. Reynolds, Observations of solar cyclical variations in geocoronal H-alpha column emission intensities, GRL, 31, L06110, doi:10.1029/2003GLO18729, 2004.
Nossal, S., F.L. Roesler, J. Bishop, R.J. Reynolds, M. Haffner, S. Tufte, J. Percival, and E.J. Mierkiewicz, Geocoronal H-alpha intensity measurements using the Wisconsin H-alpha Mapper Fabry-Perot facility, J. Geophys. Res., 106, A4, 5605, 2001.
Nossal, S., F.L. Roesler, and M.M. Coakley (1998), Cascade excitation in the geocoronal hydrogen Balmer alpha line, J. Geophys. Res., 103, 381-390, 1998.
Nossal, S. R.J. Reynolds, F.L. Roesler, and F. Scherb, Solar cycle variations of geocoronal Balmer-alpha emission, J. Geophys. Res., 98, 3669-3676, 1993.
Reynolds, R.J. (1997), Ionizing the galaxy, Science, 277, 1446-1447.
Scherb, F., 1981. Hydrogen production rates from ground-based Fabry-Perot observations of comet Kohoutek. Astrophysical Journal 243, 644.
Shih, P., F.L. Roesler, and F. Scherb, Intensity variations of geocoronal Balmer alpha emission: 1. Observational results, J. Geophys. Res., 90, 477-490, 1985.
Tufte, S.L., 1997, The WHAM Spectrometer: design performance characteristics and first results. Ph.D. Thesis, University of Wisconsin, Madison.

NAN Calibration Plots for Kitt Peak H-alpha Fabry-Perot Interferometer

The North American Nebula (NAN) is used to calibrate various H-alpha instruments. The calibration for the WHAM instrument at Kitt Peak is 800 R. 'N' is shown on the plot for those nights with this NAN calibration. The dashed line is the position of local midnight.

Summary Plots for Kitt Peak H-alpha Fabry-Perot Interferometer

Summary plots of the absolute column emission in R from thermosphere-exosphere atomic hydrogen. Most of this is from above the plotted shadow height, and so is a strong function of shadow height. The dashed line is the position of local midnight.

-Revised 18 Jan 2008 by Barbara Emery