The Earthshine Spectrum in the Near Infrared M. Turnbull1, W. Traub2, K. Jucks3, N. Woolf4, M. Meyer4, N. Gorlova4, M. Skrutskie5, J. Wilson5 1 Carnegie Institute, Washington, DC, USA 2 JPL, Pasadena, CA, USA 3 Harvard-Smithsonian Center for

1
The Earthshine Spectrum in the Near InfraredM.
Turnbull1, W. Traub2, K. Jucks3, N. Woolf4, M.
Meyer4, N. Gorlova4, M. Skrutskie5, J. Wilson51
Carnegie Institute, Washington, DC, USA2 JPL,
Pasadena, CA, USA3 Harvard-Smithsonian Center
for Astrophysics, Cambridge, USA4 University of
Arizona, Tucson, AZ, USA5 University of
Virginia, Charlottesville, VA, USA

• What is the Earthshine spectrum?
• The spectrum of the light reflected from the
  Earth
• to dark side of the moon and back to Earth.
• The ratio of the spectrum of the dark side to the
  
• bright side gives a spectrum of light that has
  passed
• through the Earths atmosphere, reflected off a
• surface, and back through the atmosphere (see
  figure 1).
• It represents the illuminated disk average
  spectrum
• of the Earth.
• This is exactly the type of observation that
  would be
• observed from an extra-solar planet around its
  star.
• The resulting spectrum contains a combination of
  many conditions, like reflection off of the
  ocean, plants,
• soil, and clouds of various heights and types.
  All these
• paths must be considered in analyzing these data.

Figure 2 Ephemeris images of the Earth at the
time Of the Earthshine observations. The right
image shows the superimposed GOES-12 cloud
image. The illuminated portion of the Earth
contains a combination of ocean, clouds vegetated
and non-vegetated land.
Figure 1 Schematic of the observational geometry
for the Earthshine observations.

• How were the observations made?
• Used the near infrared CorrMass Spectrograph
  telescope on Mt. Graham in Arizona.
• Spectrometer disperses the grating orders onto a
  NICMOS3 CCD detector (see figure 3).
• The orders are set for the edges to fall within
  the atmospheric water vapor bands (see figure 5).
• Spectra were taken for the Moon dark side, the
  Moon bright side, and background at a distance
  from the bright side that is similar to the
  distance of the dark side to account for
  scattering.
• Raw data were reduced using IRAF.
• Earthshine is calculated with
• ES (MoonD-skys)/MoonB x RB/RD

Figure 4 Sample overlapped raw extracted
spectra from the different orders for the light
scattered off the bright side of the moon. The
long Wavelength end shows some emission From the
telescope dome.
Figure 5 Example of the background spectra
obtained which must be subtracted off the dark
moon spectra. The spectra show many significant
Meinel band spectral features.
Figure 3 Sample NICMOS image of the dispersed
orders for the spectrum of the light scattered
off the bright side of the moon.

• What we learned
• We could reasonably model the observed data using
  
• a rather simple model that includes a
  combination of
• 3 atmospheric transmission/reflection schemes
• Reflection off the surface
• Reflection off 10 km ice clouds
• Reflection off 4 km water clouds
• In the near infrared, one can observe spectral
  features
• from H2O, CO2, O2, O3, and CH4, as well as
  clouds.
• The reflectance signatures from the surface (
  such
• as the red edge at .7 microns) are difficult to
  observe.
• The subtraction of the background light must be
• carefully handled to properly interpret these
  types of
• data.
• These types of studies are good test cases for
  what
• might be expected for the future observed
  spectra of
• extra-solar planets like what could come from
  future
• missions like TPF-C and Kepler.

Figure 7 The merged Earthshine spectra
combining this near infrared data with the
visible data from Woolf et al. 2003. The
spectral features comprising the primary
chromophores are denoted.
Figure 6 The punchline! The final averaged
near infrared Earthshine spectra from this study.
The spectral fit included the 3 scenarios shown
in the plot using a linear least square
reduction.