> 2006 Goldstoneradarimagery. I understand is located here:
>
http://pds-geosciences.wustl.edu/geo/arcb_nrao-l-rtls_gbt-4_5-70cm-v1...
>
> - Kurt
Kurt,
Sorry about any confusion regarding radar maps of the Moon's poles
that I may have caused.
At this point I can't recall (or find) what I may have posted, but I
asuume it had something to do with radar-derived digital elevation
maps (DEM's), and my wish to simulate views from Earth with them.
I did indeed look at the large (file size) radar maps you mention
being available on NASA's Planetary Data System (PDS) at:
http://pds-geosciences.wustl.edu/missions/lunar_radar/
This site contains radar backscatter maps (not the same as a DEM) of
most of the visible disk of the Moon produced over a three year period
by transmitting radar pulses from the 305-m diameter Arecibo dish in
Puerto Rico and interpreting the signals received ~2.5 seconds later
by a smaller dish in Green Bank, West Virginia. According to the data
labels, the south polar data (spole_pol_... and spole_dep_...) was
acquired on October 19, 2003 at 11:04-12:17 UT when the librations
(from West Virginia) would have been -6.82 (lon)/-6.39 (lat). The
north polar data (byrd_pol_... and byrd_dep_...) was acquired on
October 19, 2006 at 14:11-15:01 UT when the librations would have been
+1.28 (lon)/+0.80 (lat). These maps are said to have a resolution as
fine as 320 m x 450 m at the limb (rather better than optical images
from Earth). Note, however, that the libration at the time of the
north polar map (if the label is correct) was not particularly strong,
so aside from the good resolution it isn't expected to show anything
that couldn't be seen by visual observations (using sunlight) from
Earth.
There is a good write-up explaining how the Arecibo-Greenbank data
were acquired in the "LUNAR_RADAR.PDF" document at the above URL.
The preceding maps were acquired using a 70-cm transmitting
wavelength. Other Arecibo-Greenbank maps using a 12.6 cm wavelength
and reportedly achieving a resolution of 20 m, have were announced at
aobut the same time on an undated webpage at:
http://www.nasm.si.edu/ceps/research/moon/radar_south_images.cfm
That page says the higher resolution data would also be released via
the PDS, but I don't know if it has been, or not.
Based on the Arecibo-Greenbank documentation, my vague understanding
of the way "delay-Doppler" radar maps are made is that a large part of
the Moon is blasted by a single, largely unfocused radio-frequency
pulse. The Moon fragments this into a collection of reflected pulses
that (as received) vary in delay, frequency and intensity. The delay
and frequency shift separately restrict the possible points of origin
of the reflection to specific areas on the Moon (annular rings of
constant distance and velocity), and combining the two assigns each
returned pulse to a unique two-dimensional position in the sky (on the
assumption that the Moon is a perfect constant radius sphere rotating
in accordance with its known librations). To improve signal to noise,
the results from many pulses over several hours are combined. Although
confined to the pieces of surface that are directly visible in the
projection seen from the radar station on Earth (and the "lighting" is
much like that experienced at Full Moon), the reduced data are
typically presented in the form of a standard map projection, such as
a stereographic projection for the poles. What look like shadows in
such re-projected radar maps are areas not visible from Earth (much as
the shadows in optical wavelength images are areas not visible from
the Sun).
To supply the missing third dimension, the delay-Doppler information
received simultaneously at two spatially-separated receiving stations
can be combined interferometrically, and the phase differences
interpreted in terms of height variations at different latitudes and
longitudes on the surface. In this way a height is assigned to each
point in the backscatter map, producing a Digital Elevation Map
(DEM). The only instrument that seems to have this capability is the
Goldstone Solar System Radar (GSSR) which uses three antennas of
NASA's Deep Space Network, one configured as a transmitter, and two as
receivers. Lunar data were first acquired with this system on October
6, 1997 (south pole: 20:30 to 22:10 UT ?), and again on October 19,
1997 (north pole and Tycho: 11:00 to 13:30 UT ?). The data reportedly
has spatial resolution of about 150x150 m, although the polar data
were lumped into 600x600 m bins. It is this data which is unofficially
(?) available at:
http://www.astro.cornell.edu/~jlm/out/pds/
This is the data I was hoping to use to create simulated topographic
views of the Moon's poles as viewed from Earth, and it would also
appear to be this data (or perhaps the 150 m version of it) that is
the ultimate source of the cold trap maps used in the Bart (2008)
presentation to illustrate the locations of the LCROSS impact targets;
although the cold trap maps are illustrated on the backscatter map
(showing reflectances) rather than the DEM (showing elevations).
There is a good write-up about the 1997 GSSR data at (subscription
required):
http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=841991
or as a free reprint on the GSSR website (
http://gssr.jpl.nasa.gov/)
at:
http://gssr.jpl.nasa.gov/IEEEtransRemoteSens.pdf
Although there are some ambiguities and inconsistencies in the times
quoted for these maps, it looks like the 1997 south polar data was
acquired when the librations (from the GSSR site) were about -5.95
(lon)/-6.09 (lat); and the north polar data with +3.97 (lon)/+5.67
(lat).
As you have pointed out, a reprint of the /Science/ magazine article
showing the backscatter maps for both poles with and without the cold
traps colored can also be found for free on the GSSR website:
http://gssr.jpl.nasa.gov/margotetl1658.pdf
The GSSR has since been upgraded, and ~90-minute data runs on the
Moon's south pole were repeated on September 13, 2006 (O9:00 - 10:15
UT) and at least two subsequent dates. It is this data which possibly
has resolution to 40 m which was announced with such fanfare at:
http://www.nasa.gov/mission_pages/exploration/mmb/022708.html
but it has never been publicly released, and may never be. The
librations on Sept 13th would have been +8.05 (lon)/-6.16 (lat). A
version of that data set degraded to 600-m spatial and 256 height
levels vertical resolution can be found in an article on polar
illumination by James Fincannon at:
http://gltrs.grc.nasa.gov/Citations.aspx?ID=4166
It is quite possible that still more recent data have been acquired
with both the Arecibo-Greenbank and the GSSR instruments. To the best
of my knowledge both produce backscatter maps; but only the GSSR
produces DEM's (since two receiving antennas are required to do that).
-- Jim