Data Set Overview
  =================
    The gravitational signature of the Moon was determined from
    velocity perturbations of the Lunar Prospector (LP) spacecraft as
    measured from the Doppler shift of the S-band radio tracking
    signal.  LP was tracked by NASA's Deep Space Network (DSN) at
    Goldstone, California, Canberra, Australia, and Madrid, Spain.
    The tracking data were used to determine the LP orbit about the
    Moon, as well as the lunar gravity field [KONOPLIVETAL1998].
 
    The LP data were combined with S-band tracking observations from
    Lunar Orbiters 1, 2, 3, 4, and 5 and from the Apollo 15 and 16
    subsatellites [KONOPLIVETAL1993B] and from Clementine
    [LEMOINEETAL1997].
 
    In advance of the LRO mission, the Lunar Prospector and other
    historical tracking data were re-analyzed and the GLGM-3 model was
    developed. The model is documented in a JGR paper
    [MAZARICOETAL2010]. In addition, the GLGM-3 model was extended to
    degree 200 using a Line-of-Sight (LOS) analysis technique
    [HANETAL2011] to produce the LPE200 model.
 
  Data
  ====
    There are 2 data types for the gravity products found on this
    volume: tabular and array data.  The files containing the
    spherical harmonic coefficients of the Moon's gravity field
    (GLGM3150, LPE200) are in tabular (ASCII) and array (BINARY)
    format, with each row in the table containing the degree index m,
    the order index n, the coefficients Cmn and Smn, and the
    uncertainties in Cmn and Smn. The binary array file is a
    little-endian, row ordered upper triangular matrix.
 
  Coordinate System
  =================
    The coordinate system for the gravity data, and the coefficients
    in the GLGM3150 and LPE200 gravity fields is selenocentric, center
    of mass, longitude positive east.  The location of the pole and
    the prime meridian are defined by the principal axes as given by
    the integrated lunar librations of DE421 [WILLIAMSETAL2008].
 
  Processing
  ==========
    The GLGM3 gravity solution consists of 4,324,171 observations, of
    which 3,570,901 were contributed by LP.  The data were divided
    into 1225 independent arcs based on considerations of data
    coverage and timing of maneuvers.  The table below summarizes the
    number of observations and arcs from each spacecraft:
 
    Satellite            Number        Total  Periapsis     Apoapsis
                        of Arcs      Observ.       (km)         (km)
    Lunar Orbiter-1          70       48,575         50         1830
    Lunar Orbiter-2          90       77,726         50         1870
    Lunar Orbiter-3          73       62,264         50     1820/320
    Lunar Orbiter-4          32       48,688    2700/75    6000/4000
    Lunar Orbiter-5          70       42,916    100/170    1750/2000
    Apollo-15 subsatellite   93       52,500         75          160
    Apollo-16 subsatellite   46       42,579         30          190
    Clementine               40      378,022        370         2960
    LP (nominal)            184    2,198,751         90          110
    LP (extended)           127    1,372,150          25          45
        Total              1225    3,570,901
 
    For each arc certain parameters were determined: for example, the
    spacecraft state (position and velocity), solar radiation pressure
    coefficients, Doppler biases for each station over the arc to
    account for frequency biases, and increments in velocity to
    account for spacecraft manuevers.  The DE421 set of planetary and
    lunar ephemerides was used in the analyses.
 
    Average data fits and the final effective data weight for each set
    of the data in the solution are as follows:
 
    Satellite                Avg. RMS of fit        Eff. Data Wt.
                                      (cm/s)               (cm/s)
     Lunar Orbiter 1                    0.24                3.16
     Lunar Orbiter 2                    0.11                3.16
     Lunar Orbiter 3                    0.07                3.16
     Lunar Orbiter 4                    0.05                0.55
     Lunar Orbiter 5                    0.21                0.49
     Apollo 15 subsatellite             0.12                0.95
     Apollo 16 subsatelite              0.15                0.32
     Clementine                         0.31                1.34
     Lunar Prospector (nom)             0.02                1.41
     Lunar Prospector (ext)             0.25                3.78
 
 
  Media/Format
  ============
    This data set is stored online at the Planetary Data System
    (http://pds.nasa.gov/) and may be downloaded using a web browser
    or FTP software. A copy may be requested on physical media if
    downloading is not possible. The Planetary Data System maintains
    backup copies of this data set on various media.

Confidence Level Overview
  =========================
    The data noise on the historic Doppler data ranged from 0.3 to
    several mm/s, depending on the arc, and most of the data were at a
    count interval of 60 seconds.  The Clementine Doppler data from
    the DSN stations had a data noise of 0.25 mm/s with a count
    interval of 10 seconds.  Clementine was also tracked by a 30 meter
    antenna of the Naval Research Lab in southern Maryland.
    [KONOPLIVETAL2001] provides a detailed description of the Lunar
    Prospector tracking data. The overall data accuracy was about
    0.3 mm/s but there was some variability due to antenna type
    (34 m vs. 26 m). In addition during the extended mission,
    [KONOPLIVETAL2001] reports that multipath signals make the LP
    Doppler data closer to the poles more noisy. We note that
    [KONOPLIVETAL2001] applied a deweighting scheme (cf pp.5 of
    [KONOPLIVETAL2001]) due to convergence of the ground tracks. No
    such deweighting by latitude was performed with GLGM-3.
 
    Errors in the Free-air gravity anomalies computed from the full
    error covariance to 150x150 range from a minimum of 2.4 mGals on
    the nearside equator to a maximum of 78.1 mGals on the mid
    latitude regions of the lunar farside. The lunar nearside is
    characterized by errors of 2-30 mGals. The lunar farside is
    characterized by errors of 30 to 78 mGals. Similarly errors in the
    geoid, computed from the full error covariance to 150x150 range
    from a minumum of 0.1 meters on the nearside to a maximum of 24.8
    meters on the mid-latitude regions of the farside. On the
    nearside, the geoid errors are under 8 meters. Geoid errors on the
    lunar farside range from 8 to 25 meters. These errors are computed
    as commission error to degree 150, and do not include omission
    error (the contribution of error from degrees higher than 150).
 
  Review
  ======
    The volume containing the LP gravity dataset was formally
    reviewed by the PDS in October 2011.
 
 
  Data Coverage/Quality
  =====================
    Lunar Prospector is the first circular polar orbiter for the Moon
    in a low altitude orbit of 100 km. The only previous low altitude
    data (about 100 km) was from the Lunar Orbiters and the Apollo
    subsatellites at low inclination (which provide coverage at < 30
    degrees of latitude). Lunar Orbiters 4 and 5 provided some near
    equatorial data at low altitude from their polar eccentric orbit,
    but no low altitude data for the high latitudes.  Clementine, with
    its eccentric near-five hour orbit, had a periapse altitude of 415
    km and so provided long wavelength information on the gravity
    field.
 
    Because of the Moon's synchronous rotation, spacecraft cannot be
    directly tracked from Earth over a large part of the lunar
    farside, so there is no tracking data from 120 to 240 degrees
    longitude in the +/- 70 degree latitude band.
 
 
  Limitations
  ===========
    See Data Coverage/Quality above.