LATERAL VARIATIONS IN BULK DENSITY AND POROSITY OF THE UPPER LUNAR
CRUST FROM HIGH-RESOLUTION GRAVITY AND TOPOGRAPHY DATA:
COMPARISON OF DIFFERENT ANALYSIS TECHNIQUES
D. Wahl1, J. Oberst1,2
1 Technische Universit¨at Berlin, Chair of Planetary Geodesy, 10623 Berlin, Germany - daniel.w[email protected]
2 German Aerospace Center (DLR), Institute of Planetary Research, 12489 Berlin, Germany - juergen.oberst@dlr.de
Commission III, ICWG III/II
KEY WORDS: Gravity Field, Bouguer Anomalies, Correlation Analysis, GRAIL, LOLA
ABSTRACT:
We map lateral variations in bulk density of the upper lunar highland crust using the most recent GRAIL gravity field solution of degree
and order 1500 in combination with LOLA topography data, both truncated to an upper limit of degree and order 700. Our maps have
a spatial resolution of 0.75◦, where each grid point was calculated using circular analysis regions of 3◦radius. We apply two methods,
which yield similar results for most parts of the study area. The first method minimizes the correlation between topography and Bouguer
anomalies, the second maximizes the smoothness of the Bouguer anomalies. Both approaches suffer in the case that terrain is flat and
lacks topographic features; consequently, this is where results from the two methods differ. We also mapped porosity of the crust using
grain densities derived from Lunar Prospector spectrometry and sample analysis. It appears that variations in bulk density are mostly
related to differences in crustal porosity. We find that high porosity is often associated with areas of impact basins. This confirms earlier
studies, that impacts changed the geophysical characteristics of the lithosphere sustainably and that the high porosity of the upper lunar
crust is most likely impact induced.
1. INTRODUCTION
While the interior of the Moon is comparably homogeneous
on global scale, the near-surface crustal structure is complex.
In-situ seismic data as well as gravity measurements reveal that
compaction of the crustal rocks increases quickly with depth.
Also, there is evidence for significant regional variety in upper
crust composition and physical properties.
The lunar near-surface structure is of great interest, as it reveals
details of early crustal formation from the lunar magma ocean as
well as subsequent evolution during the late heavy bombardment.
The density may constrain the composition of the upper crust
and availability of lunar resources. Besides, bulk density of the
upper crust is needed to model mass distribution of the lower
crust, independently from topographic features (calculation of
Bouguer gravity).
In the past years, techniques have been demonstrated to obtain
the density of the lunar crust using gravity data from Gravity
Recovery and Interior Laboratory (GRAIL) mission (Zuber et
al., 2013) in combination with topography data. Besserer et al.
(2014) performed a localized, multitaper spectral analysis on
gravity and topography data, to study variations in the vertical
density structure. While for the lunar highlands an average
increase of density with depth of around 35 kg m−3km−1was
found, mare regions reveal dense basaltic material to overly low-
density anorthositic rock. Wieczorek et al. (2013) mapped the
lateral variations in crustal density by analyzing the correlation
between the gravity field and topography. A gravity field model
of degree and order 420 was used to produce a crustal density
map with a grid of 60 km spacing. Bulk density was found
to vary between 2300 and 2900 kg m−3globally. An average
bulk density of the Moon’s highland crust of 2550 kg m−3was
determined, substantially lower than estimated in earlier studies.
From bulk density and grain density, the porosity can be deter-
mined. Studies by Wieczorek et al. (2013) reveal that the lu-
nar upper crust is highly fractured. An average porosity of 12%
was determined, varying between 4 and 21%. A correlation was
found between high crustal porosity and the location of large im-
pact basins, which confirms the idea that crustal fracturing is most
likely impact induced.
The derived gravity field models from GRAIL are available in
an unrivaled resolution and accuracy compared to other planetary
bodies. With data from GRAIL’s extended mission (Lemoine et
al., 2014) and improved processing strategies, gravity field mod-
els of the Moon are now available in higher accuracy and reso-
lution. In this study we map bulk density and the porosity of the
lunar crust in high spatial resolution, using most recent gravity
and topography data. We assume a homogeneous density of the
crust in the vertical, as Wieczorek et al. (2013), and seek for lat-
eral variations. In our approach, we benefit from the fact that the
measured gravity signal at short wavelengths strongly correlates
with the topography (Turcotte and Schubert, 2014). In this pa-
per, we applied two separate methods, not contemplating alone
on the correlation between gravity anomalies and the overlying
topography, but also on the roughness of the anomalies.
2. DATA
We use the gravity field solution GL1500E from data obtained
during the primary and the extended GRAIL mission (Park et al.,
2015), which is typically given in a series of spherical harmonic
coefficients. The relation between the coefficients (frequency
domain) and gridded surface data (spatial domain) may be
expressed as
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume IV-2/W5, 2019
ISPRS Geospatial Week 2019, 10–14 June 2019, Enschede, The Netherlands
This contribution has been peer-reviewed. The double-blind peer-review was conducted on the basis of the full paper.
https://doi.org/10.5194/isprs-annals-IV-2-W5-527-2019 | © Authors 2019. CC BY 4.0 License.
527
f(θ, φ) =
∞
X
l=0
l
X
m=0
¯
Plm(cos θ)(¯clm cos mφ+ ¯slm sin mφ)(1)
where ¯clm and ¯slm are the given 4πnormalized spherical har-
monic coefficients of degree land order m,¯
Plm are the fully
normalized associated Legendre functions, and θand φcolatitude
and longitude, respectively (Hofmann-Wellenhof and Moritz,
2005).
GL1500E has a resolution of spherical harmonic degree and order
1500, corresponding to a spatial grid size of around 7 arcminutes
or 3.6 km at the equator. To assess the noise level of the dataset,
we compute the root mean square (RMS) power spectra of the
gravity signal following Kaula (1966) and compare these with
the uncertainties, provided for each gravity coefficient (Fig. 1).
For degree higher than 910, errors become larger than the power
of the signal.
Likewise, we computed the power spectrum of Bouguer anoma-
lies (see explanation later in the text) using a global crustal den-
sity of 2550 kg m−3(Wieczorek et al., 2013). The RMS power
approaches the GL1500E error near degree 700. Consequently,
we decided to truncate the coefficient series at degree and order
700. Also, we removed coefficients smaller than degree 150, to
avoid contributions of the crust mantle interface gravity signal
(Wieczorek et al., 2013). The remaining set of coefficients is as-
sumed to represent attraction from topography only.
Figure 1. Power spectra of the GRAIL gravity field model
GL1500E, the associated uncertainties of the coefficients, and
the Bouguer anomalies, calculated with a crustal density of
2550 kg m−3
In addition to GRAIL gravity we use the topography from Lunar
Orbiter Laser Altimeter (LOLA) (Smith et al., 2017). To match
the resolution of the gravity field, we reduced the topography to
the same spherical harmonic extension of degree and order 700.
Both, topography and gravity field, are given in the principal axis
(PA) reference system (NASA, 2008).
3. METHOD
To determine crustal density we benefit from the fact that the
short wavelength gravity field (corresponding to high degree co-
efficients) strongly correlates with the topography (Turcotte and
Schubert, 2014). From the given topography and some adopted
bulk density ρ, we compute the Bouguer correction, i.e. the grav-
ity attraction of the terrain related to a certain reference radius.
The Bouguer correction was then subtracted from the observed
gravity field (truncated to degree and order 150 - 700) to deter-
mine the best-fit density ρ.
For a given surface point a simple model of the Bouguer correc-
tion is the infinite plate of thickness hwith constant density ρ,
where Gis the gravitational constant (Hofmann-Wellenhof and
Moritz, 2005)
g= 2πGρh (2)
however, the method is not applicable if the topography of the
surrounding terrain is rough. Instead, the Bouguer correction was
calculated applying the finite amplitude method of Wieczorek and
Phillips (1998), which – like the observed gravity and topography
– may be given in terms of spherical harmonic coefficients, and
which may be computed from the expression
Clm =4πD3
M(2l+ 1)
l+3
X
n=1
(ρhn)lm
Dnn!
Qn
j=1(l+ 4 −j)
(l+ 3) (3)
where his the terrain above a reference surface of radius Dwith a
constant density ρ(Wieczorek, 2009). Mdenotes the mass of the
planetary body. To account for the high resolution of our dataset,
we estimated coefficients using power series of topography up to
n= 9.
Bouguer correction was computed for 20 distinct values of den-
sities varying between 2000 and 3000 kg m−3. Next, Bouguer
correction was applied to the observed gravity field, which was
downward continued (free-air correction) to the mean elevation
of each analysis region. Subtracting normal gravity, we obtain 20
sets of Bouguer anomalies. Assuming a constant density in the
vertical direction the resulting signal would be zero if the correct
crustal density was applied. If a wrong density was used, gravity
attraction from topography is mapped in the Bouguer anomalies
(see sketch in Fig. 2).
The computation of the bulk density was realized in the spatial
domain. Hence, topography and the 20 Bouguer anomaly mod-
els were converted to spatial grid space. Due to the high degree
and order of the data, classical models for the synthesis as e.g.
presented by Hofmann-Wellenhof and Moritz (2005) take a large
computational effort and may suffer from numerical instabilities
(Gruber et al., 2011). The transformation from the spectral do-
main to a spatial grid was performed using the SHTOOLS archive
for Python (Wieczorek and Meschede, 2018). The routine is fast
and accurate up to degree 2800 (Wieczorek and Meschede, 2018).
We chose a regular sampled grid with equally spaced points in
latitude and longitude, that conforms the sampling theorem by
Driscoll and Healy (1994). The grid, chosen to match the reso-
lution of the given gravity field, has 2lmax + 2 points in latitude
and twice the number in longitude direction, i.e. 3002 x 6004
grid points.
The analysis was carried out over small circular analysis regions
with radii of 3.0◦. We used the weighted Pearson correlation
scheme (Pozzi et al., 2012) to account for distortion of the circu-
lar regions with proximity to the poles. Points within the analysis
region were weighted regarding their latitudinal position, to com-
pensate for their different contributions.
A significant fraction of around 17% (Head, 1976) of the surface
of the Moon is characterized by basaltic mare regions. Since mare
basalt deposits were most likely formed by partial melting of lu-
nar upper mantle cumulates (Smith et al., 1970) and therefore do
not represent crustal material, we omitted those regions from our
analysis. We used maps of lunar maria by Nelson et al. (2014)
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume IV-2/W5, 2019
ISPRS Geospatial Week 2019, 10–14 June 2019, Enschede, The Netherlands
This contribution has been peer-reviewed. The double-blind peer-review was conducted on the basis of the full paper.
https://doi.org/10.5194/isprs-annals-IV-2-W5-527-2019 | © Authors 2019. CC BY 4.0 License.
528
to identify and exclude locations, having a contribution of more
than 2.5% basalts in the analysis circles.
The uncertainties were estimated by considering the average vari-
ation around each individual density value. Since the variation
mostly depends on different geological characteristics within the
considered region (Besserer et al., 2014), the uncertainties are
therefore probably on the pessimistic side.
Figure 2. Profiles selected from global data sets, showing
topography (“Terrain”) and 3 out of 20 Bouguer anomalies,
computed using different bulk densities. (a) Applied bulk
density too high (b) Applied bulk density just about right. (c)
Applied bulk density too low. The best-fit density is found by
two different techniques, minimizing the correlation between
topography and Bouguer anomalies (sec. 3.1) or by maximizing
smoothness of the Bouguer anomalies (sec. 3.2).
3.1 Correlation analysis between Bouguer anomalies and
topography
The 20 sets of Bouguer anomalies were analyzed to find the best-
fit bulk density in each analysis circle using two distinct tech-
niques. Using the technique of Wieczorek et al. (2013), we com-
puted correlation coefficients between Bouguer anomalies and
the terrain. Correct crustal density was found where correlation
was at a minimum (see Fig. 2 and 3 for sketches).
Figure 3. Search for best-fit density for an arbitrary analysis
region in the lunar highlands. The black curve shows the
determined coefficients of correlation between topography and
Bouguer anomalies; best-fit density is found where the curve
crosses zero (near 2480 kg m−3). The blue curve shows the
estimated standard deviations of the Bouguer anomalies; best-fit
density is found where standard deviation is smallest (and thus
anomaly is smoothest) (near 2460 kg m−3).
3.2 Roughness of Bouguer anomalies
In the second approach, we computed the standard deviation of
the 20 Bouguer anomalies for each analysis circle. Correct den-
sity was adopted where standard deviation, i.e., anomaly rough-
ness, was at a minimum (Fig. 2 and 3).
4. RESULTS
4.1 Bulk density map from correlation analysis
The global bulk density map of the upper lunar crust (Fig. 4) has
480 x 240 grid elements, corresponding to a grid size of 0.75◦
(approx. 22.5 km at the equator). The map was derived from
gravity and topography data as described above, having a spatial
resolution of 0.06◦.
Figure 4. Bulk density of the lunar crust from correlation
analysis. The data is mapped in lambert azimuthal equal-area
projection, covering a 240◦latitude and longitude range
(nearside: left; farside: right). Prominent impact basins are
marked with black circles. White areas represent lunar maria
and are not mapped in our investigation.
Based on degree and order of the truncated spherical harmonic
series and the vertical extent of each analysis circle, we estimate
bulk density to a depth of several kilometers.
We found an average bulk density of 2536 kg m−3with an uncer-
tainty of ±21 kg m−3. A histogram of globally estimated densi-
ties is given in 5A. The majority (99.6%) of determined densities
are within a range from 2300 to 2900 kg m−3, with only a small
number of points beyond.
For estimating the correlation between topography and Bouguer
anomalies, we used a circular analysis region of 3◦radius for each
grid point. For finding the minimum dimension for the analysis
circle we tested different radii. Since our estimated errors are in-
fluenced by geological characteristics of the investigated region,
the uncertainties do not give an indication for the right dimension.
Huang and Wieczorek (2012) demonstrated that grain densities
of the lunar highlands do not exceed 3000 kg m−3. We applied a
minimum radius of the analysis circle, where the resulting densi-
ties in the highlands remain below the value of 3000 kg m−3.
4.2 Bulk density from Bouguer anomaly analysis
We tested a modified method, where, instead of examining the
correlation between topography and truncated Bouguer anoma-
lies, the roughness of the truncated Bouguer anomalies were
considered. We applied the same conditions as in the correlation
approach, using circular analysis region of 3.0◦radius on a
regular grid of 0.75◦. An average bulk density of 2503 kg m−3
with an uncertainty of ±22 kg m−3was estimated.
Results for most areas agree with those of the correlation
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume IV-2/W5, 2019
ISPRS Geospatial Week 2019, 10–14 June 2019, Enschede, The Netherlands
This contribution has been peer-reviewed. The double-blind peer-review was conducted on the basis of the full paper.
https://doi.org/10.5194/isprs-annals-IV-2-W5-527-2019 | © Authors 2019. CC BY 4.0 License.
529
approach, which supports the validity of the two methods.
Overall, the Bouguer roughness approach exhibits slightly lower
densities, compared to the correlation approach (Fig. 5). For 3%
of the areas on the global map larger differences occur, which
will be examined in the discussion section.
Figure 5. Frequency of the globally determined bulk densities.
(A) Histogram of global crustal density estimates from
correlation analysis. (B) Histogram of global crustal density
estimates from Bouguer anomaly roughness.
4.3 Porosity
Porosity is the fraction of the volume made up of pore space (Tur-
cotte and Schubert, 2014). For calculating the porosity, bulk den-
sities as well as the grain densities need to be known. We ben-
efit from the fact that average grain densities of lunar rocks are
proportional to abundances of their main chemical constituents.
Following Huang and Wieczorek (2012) we used titanium abun-
dances from gamma ray spectrometer data (Prettyman et al.,
2006) obtained by the Lunar Prospector mission to map the grain
density of the upper crust. The porosity φwas then determined
from the relation
φ= 1 −ρbulk
ρgrain
.(4)
Maps were prepared in matching formats as those in Fig. 4.
For the highlands upper crust of the Moon an average of 13%
porosity for the correlation approach and 14% porosity for the
Bouguer roughness approach was determined.
Figure 6. Porosity of the lunar crust (bulk densities were
estimated applying the correlation approach). Projection and
labels as in Fig. 4.
As it can be seen in Huang and Wieczorek (2012), grain densities
are quite similar over the globe, with the only exception of Pro-
cellarum KREEP Terrane (PKT). Appart from PKT densities may
vary locally only by up to 2%. Therefore, we may conclude that
our observed lateral variations in bulk density are mainly induced
by the variations in porosity of lunar rock.
5. DISCUSSION
Our results agree with those from earlier studies by Wieczorek
et al. (2013) and Goossens et al. (2017), who determined an
average bulk density of the upper crust of 2550 ±18 kg m−3
(GL0420A gravity model) and 2587 ±54 kg m−3(GRGM900C
gravity model), respectively. The porosity of 12% in the study
by Wieczorek et al. (2013) is slightly smaller than our estimated
average of 13%, using the correlation approach, and 14%,
considering the Bouguer anomaly roughness.
The derived maps, showing lateral variations in bulk density
and porosity of the upper lunar highland crust, have a resolution
two times higher compared to the maps of Wieczorek et al.
(2013). Due to the high level of detail we found evidence for
significant regional variety in near-surface composition and
physical properties. The high-resolution maps reveal a coherence
between impact basins and the porosity of the lunar crust.
An example is given in Fig. 7, showing the Korolev farside
basin, with a diameter of 417 km (Neumann et al., 2015). Within
the inner ring of the basin, the upper crust possesses low porosity
of 5%, likely caused by compaction of the target rock and the
formation of dense impact melt (Melosh, 1989). In contrast,
the region towards the outer rim holds high porosities of around
15%. These areas were probably affected by the shock wave
during the contact and compression stage (Melosh, 1989), which
caused fracturing and brecciation of the rock (Collins, 2014).
These findings confirm earlier investigations suggesting that
impact induced fracturing is the primary mechanism forming
the porous upper crust of the Moon and other terrestrial planets
(Soderblom et al., 2015; Collins, 2014). The high spatial
resolution of the maps will enable further studies focusing on the
geophysical characteristics of single impact basins.
Figure 7. Korolev impact basin. The maps are presented in
Mercator projection, with its meridian placed in the center of the
basin. (A) Digital terrain model derived from LOLA data (Smith
et al., 2017), elevations refer to mean topography radius of
1737.151 km, (B) Porosity of the upper lunar crust.
To obtain the best-fit density, we used two distinct methods. For
most areas (97%) results are similar, not different by more than
100 kg m−3. However, in particular regions (e.g. at the edge to
the Procellarum KREEP Terrane, especially at the northern rim)
large differences of up to 700 kg m−3are seen.
To find out the reason for the differences, we mapped topo-
graphic roughness, using the standard deviation of topographic
data points within the same circular analysis regions, which were
used for determining the bulk density. We found that different
bulk density solutions are typically found in areas of low surface
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume IV-2/W5, 2019
ISPRS Geospatial Week 2019, 10–14 June 2019, Enschede, The Netherlands
This contribution has been peer-reviewed. The double-blind peer-review was conducted on the basis of the full paper.
https://doi.org/10.5194/isprs-annals-IV-2-W5-527-2019 | © Authors 2019. CC BY 4.0 License.
530
roughness. An example is demonstrated in Fig. 8, for a region at
the northern boundary of the PKT.
A smooth surface, lacking topographic features, is a challange for
both methods. Usually, if a wrong density was applied, surface
attributes are mapped in the Bouguer anomalies. The anomalies
become rough and a correlation with the topography is found. If
topographic features lack, any difference can be made between a
wrong and the correct density which was used. The situation is
similar for regions of lunar maria: also here topography exhibits
only few landforms and craters, so that most of the areas are flat.
In the present work, mare regions were omitted from the analysis
in particular due to their different composition and origin com-
pared to lunar highland crust (see earlier explanations). But the
surface characteristics also make it difficult to achieve meaning-
ful results in those areas.
Figure 8. Region of large inconsistence of the two tested
methods at the northern rim of PKT. (A) Bulk density difference
(bulk density from correlation between topography and Bouguer
anomalies minus bulk density from Bouguer roughness) with
shaded topography in the background (B) Surface roughness,
given as the standard deviation of the topography.
The highest bulk densities can be found in the region of South
Pole-Aitken basin (SPA), the oldest impact basin on the Moon
(Wilhelms, 1987). With a diameter of about 2400 km (Neumann
et al., 2015) it represents the largest impact basin in the solar sys-
tem. Unfortunateley, SPA shows a number of distinct patches
of mare basalts, which were excluded from our mapping, why
a statement regarding its properties is only possible to a limited
extent. As remote sensing data reveals, SPA exhibits a high abun-
dance in pyroxenes, even though the proportions vary within the
basin (Moriarty and Pieters, 2018). This mineral most likely orig-
inates from the lower crust and/or upper mantle and has a higher
density than the anorthite highland crust (Kiefer et al., 2012),
which may explain the high bulk density found in SPA.
6. CONCLUSION AND OUTLOOK
We mapped lateral variations of bulk density and porosity of the
lunar highland crust using most recent high-resolution GRAIL
gravity and LOLA topography data. For calculating the density,
two methods were demonstrated, both of which yield similar re-
sults for most of the studied areas. Both methods suffer in the
case of a smooth topography. Consequently, the two methods
show significantly different results of up to 700 kg m−3in such
areas.
Using global grain densities derived from remote sensing data,
we also estimated the porosity of the upper lunar crust. From
inspection of the data, we conclude, that differences in the bulk
density are mostly determined by differences in crustal poros-
ity, not by differences in the statistics of chemical compounds.
We demonstrate that high porosity correlates with areas around
impact basins, which suggests that the fracturing is most likely
impact-induced.
With the present high-resolution datasets, it is possible to study
impact structures in more detail. We will take a closer look at
single basins regarding their variations in porosity and investi-
gate their characteristics in terms of dimension and age.
Accurate bulk densities near the surface are important for studies
of deeper crustal structures. Rather than using one average value,
the high-resolution bulk density maps can be used to calculate
mass variations at the crust-mantle boundary with individual val-
ues for different areas.
ACKNOWLEDGEMENTS
We gratefully acknowledge Mark Wieczorek for helpful discus-
sions and advice. We also would like to thank two anonymous re-
viewers for critical comments and suggestions, which helped im-
proving the manuscript. This work was funded by the Deutsche
Forschungsgemeinschaft (SFB-TRR 170, subproject A04-60).
REFERENCES
Besserer, J., Nimmo, F., Wieczorek, M. A., Weber, R. C., Kiefer,
W. S., McGovern, P. J., Andrews-Hanna, J. C., Smith, D. E. and
Zuber, M. T., 2014. GRAIL gravity constraints on the vertical and
lateral density structure of the lunar crust. Geophysical Research
Letters 41(16), pp. 5771–5777.
Collins, G. S., 2014. Numerical simulations of impact crater for-
mation with dilatancy. Journal of Geophysical Research 119(12),
pp. 2600–2619.
Driscoll, J. R. and Healy, D. M., 1994. Computing fourier trans-
forms and convolutions on the 2-sphere.
Goossens, S., Sabaka, T. J., Genova, A., Mazarico, E., Nicholas,
J. B. and Neumann, G. A., 2017. Evidence for a low bulk crustal
density for Mars from gravity and topography. Geophysical Re-
search Letters 44(15), pp. 7686–7694.
Gruber, C., Nov´
ak, P. and Sebera, J., 2011. Fft-based high-
performance spherical harmonic transformation. Studia Geo-
physica et Geodaetica 55(3), pp. 489–500.
Head, J. W., 1976. Lunar volcanism in space and time. Reviews
of Geophysics and Space Physics 14(2), pp. 265–300.
Hofmann-Wellenhof, B. and Moritz, H., 2005. Physical Geodesy.
Springer-Verlag, Wien.
Huang, Q. and Wieczorek, M. A., 2012. Density and porosity of
the lunar crust from gravity and topography. Journal of Geophys-
ical Research E: Planets 117(5), pp. 1–9.
Kaula, W., 1966. Theory of Satellite Geodesy. Blaisdell Puplish-
ing Company, Waltham, Mass. (republished by Dover, New York,
2000).
Kiefer, W. S., Macke, R. J., Britt, D. T. and Irving, A. J., 2012.
The density and porosity of lunar rocks. Geophysical Research
Letters 39, pp. 1–5.
Lemoine, F. G., Goossens, S., Sabaka, T. J., Nicholas, J. B.,
Mazarico, E., Rowlands, D. D., Loomis, B. D., Chinn, D. S., Neu-
mann, G. A., Smith, D. E. and Zuber, M. T., 2014. GRGM900C:
A degree 900 lunar gravity model from GRAIL primary and
extended mission data. Geophysical Research Letters 41(10),
pp. 3382–3389.
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume IV-2/W5, 2019
ISPRS Geospatial Week 2019, 10–14 June 2019, Enschede, The Netherlands
This contribution has been peer-reviewed. The double-blind peer-review was conducted on the basis of the full paper.
https://doi.org/10.5194/isprs-annals-IV-2-W5-527-2019 | © Authors 2019. CC BY 4.0 License.
531
Melosh, H. J., 1989. Impact cratering: A Geologic Process. Ox-
ford University, New York.
Moriarty, D. P. and Pieters, C. M., 2018. The Character of South
Pole-Aitken Basin: Patterns of Surface and Subsurface Compo-
sition. Journal of Geophysical Research: Planets 123, pp. 729–
747.
NASA, 2008. A Standardized Lunar Coordinate System for the
Lunar Reconnaissance Orbiter and Lunar Datasets. Vol. 5,
NASA Goddard Space Flight Center, Greenbelt, Md.
Nelson, D. M., Koeber, S. D., Daud, K., Robinson, M. S., Wat-
ters, T., Banks, M. and Williams, N. R., 2014. Mapping Lunar
Maria Extents and Lobate Scarps Using LROC Image Products.
Lunar and Planetary Science Conference.
Neumann, G. A., Zuber, M. T., Wieczorek, M. A., Head, J. W.,
Baker, D. M. H., Solomon, S. C., Smith, D. E., Lemoine, F. G.,
Mazarico, E., Sabaka, T. J., Goossens, S. J., Melosh, H. J.,
Phillips, R. J., Asmar, S. W., Konopliv, A. S., Williams, J. G.,
Sori, M. M., Soderblom, J. M., Miljkovi´
c, K., Andrews-Hanna,
J. C., Nimmo, F. and Kiefer, W. S., 2015. Lunar impact basins
revealed by Gravity Recovery and Interior Laboratory measure-
ments. Science Advances 9(1), pp. 1–10.
Park, R. S., Konopliv, A. S., Yuan, D., Asmar, S., Watkins, M.,
Williams, J., Smith, D. and Zuber, M., 2015. A high-resolution
spherical harmonic degree 1500 lunar gravity field from the grail
mission. AGU Fall Meeting Abstracts pp. G41B–01.
Pozzi, F., Di Matteo, T., and Aste, T., 2012. Exponential smooth-
ing weighted correlations. European Physical Journal B 85(6),
pp. 1–21.
Prettyman, T. H., Hagerty, J. J., Elphic, R. C., Feldman, W. C.,
Lawrence, D. J., McKinney, G. W. and Vaniman, D. T., 2006. El-
emental composition of the lunar surface: Analysis of gamma ray
spectroscopy data from Lunar Prospector. Journal of Geophysi-
cal Research E: Planets 111(12), pp. 1–41.
Smith, D. E., Zuber, M. T., Neumann, G. A., Mazarico, E.,
Lemoine, F. G., Head, J. W., Lucey, P. G., Aharonson, O., Robin-
son, M. S., Sun, X., Torrence, M. H., Barker, M. K., Oberst, J.,
Duxbury, T. C., Mao, D., Barnouin, O. S., Jha, K., Rowlands,
D. D., Goossens, S., Baker, D., Bauer, S., Gl¨
aser, P., Lemelin,
M., Rosenburg, M., Sori, M. M., Whitten, J. and Mcclanahan, T.,
2017. Summary of the results from the lunar orbiter laser altime-
ter after seven years in lunar orbit. Icarus 283, pp. 70–91.
Smith, J. V., Anderson, A. T., Newton, R. C., Olsen, E. J., Crewe,
A. V., Isaacson, M. S., Johnson, D. and Wyllie, P. J., 1970. Petro-
logic history of the moon inferred from petrography, mineralogy
and petrogenesis of apollo 11 rocks. Geochimica et Cosmochim-
ica Acta Supplement, Volume 1. Proceedings of the Apollo 11 Lu-
nar Science Conference held 5-8 January, 1970 in Houston, TX.:
Mineraolgy and Petrology. Edited by A. A. Levinson. New York:
Pergammon Press, 1970. 1, pp. 897–925.
Soderblom, J. M., Evans, A. J., Johnson, B. C., Melosh, H. J.,
Miljkovi, K., Phillips, R. J., Andrews-hanna, J. C., Bierson, C. J.,
Iii, J. W. H., Milbury, C., Neumann, G. A., Nimmo, F., Smith,
D. E., Solomon, S. C. and Sori, M. M., 2015. The fractured
Moon: Production and saturation of porosity in the lunar high-
lands from impact cratering. pp. 6939–6944.
Turcotte, D. and Schubert, G., 2014. Geodynamics, 3rd edition.
Cambridge University Press, Cambridge.
Wieczorek, M. A., 2009. Gravity and topography of the terres-
trial planets. In: G. Schubert and T. Spohn (eds), Treatise on
Geophysics, Vol. 10, Elsevier, Amsterdam, pp. 165–206.
Wieczorek, M. A. and Meschede, M., 2018. SHTools: Tools for
Working with Spherical Harmonics. Geochemistry, Geophysics,
Geosystems 19(8), pp. 2574–2592.
Wieczorek, M. A. and Phillips, R. J., 1998. Potential anoma-
lies on a sphere: Applications to the thickness of the lunar crust.
Journal of Geophysical Research 103(97), pp. 1715–1724.
Wieczorek, M. A., Neumann, G. A., Nimmo, F., Kiefer, W. S.,
Taylor, J. G., Melosh, H. J., Phillips, R. J., Solomon, S. C.,
Andrews-Hanna, J. C., Asmar, S. W., Konopliv, A. S., Lemoine,
F. G., Smith, D. E., Watkins, M. M., Williams, J. G. and Zuber,
M. T., 2013. The Crust of the Moon as Seen by GRAIL. Science
339, pp. 671–675.
Wilhelms, D. E., 1987. The geological history of the Moon.
Zuber, M. T., Smith, D. E., Watkins, M. M., Asmar, S. W., Kono-
pliv, A. S., Lemoine, F. G., Melosh, H. J., Neumann, G. A.,
Phillips, R. J., Solomon, S. C., Wieczorek, M. A., Williams, J. G.,
Goossens, S. J., Kruizinga, G., Mazarico, E., Park, R. S. and
Yuan, D.-n., 2013. Gravity Field of the Moon from the Grav-
ity Recovery and Interior Laboratory (GRAIL) Mission. Science
339, pp. 2011–2014.
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume IV-2/W5, 2019
ISPRS Geospatial Week 2019, 10–14 June 2019, Enschede, The Netherlands
This contribution has been peer-reviewed. The double-blind peer-review was conducted on the basis of the full paper.
https://doi.org/10.5194/isprs-annals-IV-2-W5-527-2019 | © Authors 2019. CC BY 4.0 License.
532
