| DATA_SET_DESCRIPTION |
Data Set Overview
=================
This dataset contains derived data products from the Atmospheres
Science Theme Team of the New Horizons project. It includes the
wavelength-dependent solar atmospheric occultation count rate and
opacities, a model of the unocculted solar count rate, atmospheric
composition profiles for the species N2, CH4, C2H2, C2H4, and C2H6
based on data taken from the Alice UV imaging spectrograph, lower
atmospheric temperature and pressure profiles from the Radio
EXperiment (REX) instrument, and vertical I/F haze profiles from
the LOng Range Reconnaissance Imager (LORRI) instrument.
Version
=======
This is VERSION 2.0 of this data set. It is identical to Version 1.0
except for the addition of the starocc/ subdirectory, which contains
derived data from the Alice UltraViolet Imaging spectrograph aboard
the New Horizons spacecraft acquired during the occultation and
appulse of UV bright stars 69 Ori and 72 Ori by Pluto.
Processing
==========
The data generated by the Atmospheres Science Theme Team of the
New Horizons project can be grouped into a number of sub-specialities,
and the data for these is grouped into subdirectories of:
ALICEOCC/ - solar occultations and an unocculted model
ATMOSCOMP/ - atmospheric composition profiles
HAZE/ - details of the Pluto haze layers
REXATMOS/ - lower atmospheric temperature and pressure profiles
STAROCC/ - stellar occultation and appulse data
THERMSCAN/ - diametric and polar thermscans of Pluto
The data content that resides in each subdirectory is detailed below.
ALICEOCC
========
Solar Atmospheric Occultation Count Rate and Opacities
------------------------------------------------------
Data from the Alice UltraViolet Imaging spectrograph aboard the
New Horizons spacecraft was acquired during the occultations of
the Sun by Pluto and Charon. The data have been binned in time to
1-second (MET - Mission Elapsed Time) resolution. The observed solar
spectrum has been extracted and corrected for a variety of
instrumental effects.
For a detailed description of the data analysis process, please
see [YOUNGETAL2017].
The list of Raw files used in creating this product can be found
in the first header of the fits file.
ATMOSCOMP
=========
Atmospheric Composition Profiles
--------------------------------
The Alice instrument on NASA's New Horizons spacecraft observed an
ultraviolet solar occultation by Pluto's atmosphere on 2015 July 14.
The transmission vs. altitude was sensitive to the presence of N2,
CH4, C2H2, C2H4, C2H6, and haze.
Line-of-sight abundances and local number densities for the 5 molecular
species, and line-of-sight optical depth and extinction coefficients
for the haze, were derived. The UV occultation occured from
approximately 2015 July 14 12:15 to 13:32 UTC (spacecraft time).
The line-of-sight abundance spreadsheet in data/atmoscomp holds data
for N2, CH4, C2H2, C2H4, C2H6, and haze as a function of radius.
The line-of-sight abundance N_s of species s as a function of tangent
radius r' can be found in Section 3, Equation (6) in [YOUNGETAL2017].
Data acquisition
----------------
Data was acquired using the Alice instrument. Alice is an imaging
spectrograph that has a bandpass from 52 to 187 nm, with a
photocathode gap from 118 to 125 nm designed to decrease the count
rate near Ly-alpha. For the Pluto occultation, the 'pixel list'
data collection mode was used for higher time cadence. During the
observation, the Sun was placed in the 'box' of the Solar
Occultation Channel (SOCC) to avoid slit losses, to avoid
oversaturation, and to observe the UV solar occultation
simultaneously with the radio Earth occultation. The SOCC is
roughly co-aligned with the field of view of REX [TYLERETAL2008]
to allow for simultaneous observations of the solar and the
uplink radio occultations. During the Pluto solar occultation,
the REX field of view was centered on Earth, which placed the Sun
within a few tenths of a degree from the center of the 2x2 degree
'box'. Thrusters were fired to keep REX centered on the Earth
within 0.0143 degrees (deadband half-width). Thus, the Sun moved
only slightly within the 2x2 degree 'box' during the solar
occultation observation. The Sun's diameter as seen from New
Horizons in July 2015 was 0.016 degrees, which was much smaller
than the size of the 'box' and slightly smaller than the Alice pixel
size (Alice pixels subtend 0.019 degrees in the spectral axis, and
0.308 degrees in the spatial axis). Pluto, by contrast, was large
compared to the box, varying from 3.0 degrees at 2015 July 14 12:41
UT, when Pluto first entered the box, to 2.2 degrees at 13:01 UT,
when Pluto exited the box.
The locations probed by the solar occultation depended only on the
relative positions of the Sun, Pluto, and New Horizons, and not on
the pointing of the Alice field of view. Pluto passed across the
Sun at a sky-plane velocity (that is, the component of the velocity
perpendicular to the spacecraft-Sun line) of 3.586 km/s, so it took
~11 minutes for the solid body of Pluto to pass across the Sun.
This is slightly faster than the Earth's sky-plane velocity of
3.531 km/s during the Earth occultation, [HINSONETAL2017]. Solar
ingress occurred at 195.3 degrees E longitude, 15.5 degrees S
latitude, while egress occurred at 13.3 degrees E longitude,
16.5 degrees N latitude [GLADSTONEETAL2016]. Thus, ingress probed
the atmosphere just off of the southern tip of the left-hand side
of the bright heart-shaped feature, named Sputnik
Planitia, and egress probed the atmosphere near the transition
between dark equatorial regions and the mid-latitude areas.
Observations were taken in 'pixel list' mode, in which each
detected photon is tagged with its location on the detector.
Effectively, this location was a measure of which of the 1024
spectral and 32 spatial pixels was stimulated by each detected
photon. The pixel resolution was 0.177-0.183 nm/pixel, which
Nyquist samples the instrumental spectral resolution (0.35 nm
when operated in pixel list mode). Timing was determined by the
insertion of special 'time hack' values into the instrument's
memory buffer every 4 ms. This 4 ms timing was much finer than
that required for the analysis presented here, and the counts
were summed into 1-second time bins. This resulted in a 1024 by 32
image of counts per second at each 1 second interval, called a
count rate image.
Data analysis
--------------
In order to extract 1-second count-rate spectra from the pixel
list data, the following analysis steps were performed.
1) Dead time correction in the raw pixel list stream were
calculated. The detector electronics took a finite amount of time
to process each count. During this time, the detector was 'dead'
i.e., it was insensitive to any additional counts. Therefore each
detected photon was weighted by a factor of 1 / (1 - tau_d * C),
where tau_d = 18 microseconds, the dead time constant of the
electronics, and C is the count rate measured over a 4 ms interval.
2) The pixel stream was summed to construct 2-D count-rate images
at 1-second resolution. During one second, the tangent altitude
probed by the Sun moved ~3.586 km through Pluto's atmosphere. The
choice of 1-second binning was a balance between increasing
signal-to-noise ratio per image and sub-sampling the 4 to 5 second
(~16 km) smoothing caused by the Sun's finite size.
3) A 1-D solar spectrum was extracted from each 2-D count rate
image. The Sun varied in its deadband by 0.0143 deg (half-width).
Since this was much smaller than the pixel size of 0.308 deg/pixel
in the spatial direction, the variation within the deadband did
not change which detector row contained the counts from the Sun.
The solar spectra was extracted by a simple sum of rows 19-22,
inclusive, which accounted for the width of the spatial
point-spread function ([STERNETAL2008], Figure 12) and the motion
within the deadband. The contribution of Pluto's nightside to the
UV signal was negligible compared to the direct solar flux.
4) Alice 'stim pixels' [STERNETAL2008] was used to correct the
wavelength scale for temperature effects in the Alice detector.
The mapping between the physical location of an event on the
detector and its pixel number in data space depended on the
resistivity of the readout anode, which itself depended on
temperature. Essentially, the detector electronics produced counts
at two known physical locations on opposite ends of the detector.
These counts were then mapped into data space, allowing for a
linear correction to the apparent position of detected photons.
5) Each one-second count-rate spectrum was then corrected for the
wavelength dependence on the location of the Sun within the 2x2
degree 'box' portion of the slit. The Sun was offset slightly from
the center line of the slit, which introduced an overall wavelength
shift of 0.396 nm. There was some variation in wavelength as the
Sun's position moved in the +/-0.0143 degree deadband, since a pixel
subtended 0.019 degrees in the dispersion direction. For unocculted
spectra, the wavelength shift was determined by fitting a Gaussian
line profile to five solar lines, including the Lyman-Alpha line
at 121.6 nm. The shift was also calculated from the spacecraft
attitude, and the two methods agreed to within 0.10 pixels; the
shift calculated from the spacecraft altitude was used for spectra
taken when the solar lines were obscured by Pluto's atmosphere.
The resulting spectra were placed on a common wavelength grid
using a sinc interpolation. Doppler shifts were ignored in this
analysis. Pluto's heliocentric motion, which affected the
interaction of the solar lines and absorption in Pluto's
atmosphere, was at most 1.075 km/s during this observation
(0.00035 nm shift at 100 nm). This was much less than the width
of the solar lines [CURDTETAL2001]. The rate at which the
spacecraft receded from Pluto, which affected how the spectrum
is recorded on the Alice spectrograph, ranged from 11.66 to
13.60 km/s over the POCC observation (0.0038 to 0.0045 nm shift
at 100 nm). This led to a shift of only ~2% of a pixel.
6) The gradual decrease in sensitivity of regions of the detector
that saw the highest solar flux had to be corrected for. This
localized phenomenon, known as 'gain sag', is a function of the
total amount of charge extracted from the micro-channel plates,
per unit area, over the lifetime of the detector [STERNETAL2008].
The magnitude of the sensitivity loss due to gain sag varied from
zero at the short wavelength end of the detector, where the solar
flux is low, up to several percent at the long wavelength end of
the detector, where the solar flux is greatest. A correction for
the gain sag during the occultation observation was derived, on
a pixel by pixel basis, by fitting a line to the observed count
rate when the sun was unocculted by Pluto or its atmosphere as
a function of the total integrated number of counts during the
observation, using both ingress and egress spectra.
7) After correcting for gain sag, the dark rate spectrum was
subtracted. This was calculated from a composite dark spectrum,
shifted (with sinc interpolation) to match the wavelength of
each count-rate spectrum. The resulting extracted one-second
count-rate spectra vs. tangent radius are shown in Figure 4A of
[YOUNGETAL2017]. The tangent radius was the closest distance
between Pluto's center and the ray connecting the spacecraft
and the center of the solar disk, and was calculated at each
second using SPICE kernels supplied by the New Horizons
KinetX navigation team for science analysis:
NH_PRED_20141201_20190301_OD122.BSP
NAVSE_PLU047_OD122.BSP
NAVPE_DE433_OD122.BSP
These SPICE kernels are archived with NAIF at JPL and can be found
with the search keywords 'JPL NAIF data New Horizons.'
8) A reference solar spectrum for each one-second spectrum was
created. Because of the small entrance aperture of the SOCC, the
repeller grid, a lattice-like pattern of wires designed to remove
stray ions ([STERNETAL2008] figure 5) produced discrete shadows
on the surface of the detector. Even a sub-pixel change in the
Sun's position on the focal plane affected the observed shadow
pattern on the detector, and caused up to a 40% variation in
apparent throughput at certain regions of the spectrum. Failure
to correct for this effect would leave periodic reductions in
counts that mimic absorption features. To correct for this, the
spacecraft attitude information derived from the star trackers
was used to derive the position of the Sun in the focal plane.
For each one-second count rate spectrum, other spectra (in either
the Pluto or Charon solar occultations) that (i) were obtained
when the Sun was within 0.002 deg of the spectrum in question,
and (ii) were unocculted, were identified. This selection gave a
median of 60 reference spectra for each one-second sample. These
were averaged to produce a reference solar spectrum for each
one-second count-rate spectrum. Figure 4B in [YOUNGETAL2017]
shows the result of dividing each one-second count rate spectrum
by its reference solar spectrum at full one-second, 0.17 nm
resolution.
Absorption cross sections
-------------------------
For the ultraviolet solar occultation by Pluto observed by New
Horizons, the refraction of Pluto's atmosphere can be ignored
[HINSONETAL2017], simplifying the geometry of the occultation.
The ray connecting the Sun and the New Horizons spacecraft has
a minimum distance to the body center, called the tangent radius,
r', which can be defined by the surface radius, r_s, and the
height of the tangent point above the surface (the tangent
height), h, by r' = r_s + h. At a distance along the ray, x,
defined as 0 at the tangent point, the relationship between the
radius in the atmosphere along the ray, r, and the tangent radius
is simply r^2 = r'^2 + x^2. We can define the radius in the
atmosphere as the sum of the surface radius and the altitude, z,
by r = r_s + z.
The incident UV solar flux is diminished by the absorption of
an occulting atmosphere (e.g., Smith and Hunten 1990). The
line-of-sight transmission, Tr, is a function of the line-of-sight
optical depth, which is itself a function of tangent radius, r',
and wavelength, lambda. The cross-sections can be approximated
as constant along the line-of-sight.
Equations and further details regarding the determination of
cross-sections for N2, CH4, CO, C2H6, C2H2, C2H4, and Hazes
can be found in Section 3, [YOUNGETAL2017].
Determination of line-of-sight abundances
-----------------------------------------
The retrieval of N2, CH4, C2H6, C2H2, C2H4, and haze line-of-sight
abundances was performed individually at each altitude, in a method
very similar to that used in [GLADSTONEETAL2016]. Fitting for the
hazes and hydrocarbons was first performed using only the
wavelengths 100-180 nm, where the signal-to-noise of the
occultation data was highest, and N2 did not contribute. In a later
step, N2 was included to analyze the wavelengths below 65 nm. HCN
and CO, while known to be present in Pluto's atmosphere, were not
detected in the Alice solar occultation.
Retrievals were performed separately for ingress and egress. The
retrieval started at a tangent height 2000 km altitude, above the
first measurable absorption, and progressed downward toward the
surface. To retrieve line-of-sight abundances from each spectrum,
the spectrum at the next higher altitude step acted as the initial
condition in a weighted Levenberg-Marquardt least-squares fit to
minimize the weighted sum of squared residuals,
(Press et al. 2007). Fitting to individual one-second spectra gave
unacceptably large errors and unstable solutions to the non-linear
least-squares fit, so multiple spectra were averaged before
retrieving line-of-sight abundances. Because of the 3.586 km/s
sky-plane velocity of the Sun, averaging in time corresponded to
averaging in altitude. To investigate the impact of different
averaging lengths, one retrieval was performed that averaged 23
seconds (82.5 km) for altitudes > 1000 km, 11 seconds (39.5 km)
for 1000 to 800 km, and 5 seconds (17.9 km) below 800 km,
scaling the errors per spectrum appropriately. A second retrieval
averaged 23 seconds throughout the entire span. The 5-second
averaging scale was matched to the angular size of the Sun as
seen by New Horizons. At altitudes below 700 km, the 23-second
averaging sampled less than one point per scale height. Using
Eq. 8 [YOUNGETAL2017], line-of-sight abundances were retrieved
that produced a transmission, Tr, that best matched the
time-averaged data, D, when weighted by the solar rate, R,
convolved by the line-spread function, k, normalized by the
convolved solar rate, k*R, and multiplied by the time-averaged
reference spectrum, Dref. For each averaged spectrum, the log of
the line-of-sight abundance was fitted for, because that quantity
guaranteed a positive N_s (N_s is the line-of-sight abundance of
species s), and because the surfaces of chi^2 were more symmetric
in ln(N) than in N itself.
Determination of local number densities
---------------------------------------
The line-of-sight abundance, N, is the integral of the local
number density, n, along the line-of-sight. Under certain
assumptions, one can invert this relationship to derive n given N
(for details on how N were determined, see [YOUNGETAL2017]). By far,
the most common assumption is that of local spherical symmetry
(that is, within the region where the ray path intersects Pluto's
atmosphere). For Pluto's upper atmosphere, spherical symmetry
appeared to be a very good assumption. The Alice Pluto solar
occultation data had close agreement between the ingress and egress
lightcurves, and nearly identical profiles for ingress and egress
for the line-of-sight abundances of all species considered here.
Similarly, the REX Pluto Earth radio occultation [HINSONETAL2017]
showed similar density profiles for ingress and egress above 30 km
altitude.
For these two New Horizons datasets, the ingress and egress
latitudes differed by only ~30 deg, raising the possibility that
the similarity is simply due to ingress and egress probing similar
latitudes. Analyses of the main occultation drop in ground-based
stellar occultations (probing ~10 km to 400 km altitude) do not
show evidence for statistically significant ellipticity [PERSON2001]
or only very rare cases of statistically significant dawn/dusk or
summer/winter differences [ZANGARI2013]. There are robust
theoretical reasons to expect very small horizontal variations in
temperature over all the body (at a given pressure level) in the
part of the atmosphere not influenced by topography because of the
very long radiative timescale of the atmosphere. This is predicted
by 3D Global Circulation Models even when methane is not well
mixed horizontally [TOIGOETAL2015], [FORGETETAL2017].
Data processing was performed with the assumption of spherical
symmetry, with the caveat that this might lead to inaccuracies
in derived number densities in the lowest 30 km. Given a
spherically symmetric atmosphere, the classic method for deriving
local number density from line-of-sight abundance is the Abel
transform [ROBLE&HAYS1972]. This method includes a derivative and
an integral which amplify the noise in the profile. There are
various methods for dealing with this noise, for example by
imposing functional forms [ROBLE&HAYS1972] or by imposing a
smoothness constraint using a Tikhonov regularization
[QUEMERAISETAL2006]. Because data was already smoothed over
23-second (82.5-km) intervals, and a quality constraint of < 30%
errors to define the valid altitudes was imposed, the unmodified
Abel transform technique was used.
For the haze analysis, the relationship between line-of-sight
optical depth, tau, and local extinction coefficient, xi, was
mathematically equivalent to the relationship between N and n.
The Abel transform formally includes an integral to infinity.
Specifying the atmosphere above the highest valid data point was
done by defining an altitude region over which a functional form
N(r) is fitted, and then the function was extrapolated to a
radius where the contributions to the integral were negligible
(in practice, this was taken to be 2000 km altitude). The higher
altitude of the fitting region, h_1, was near the top of the
region of valid retrievals of the line-of-sight abundance
(determined by the errors in the lower panel of Fig. 9 of
[YOUNGETAL2017]).
Altitudes sampled every 23 seconds (~82.5 km) were used for the
fit, to avoid fitting to correlated points, so the lower altitude
of the fitting region, h_0, was a multiple of 82.5 km below h_1.
This step required some judgment, to balance using a larger region
to better constrain the fits, while using a smaller region to avoid
non-exponential changes in the profile. Within the extrapolation
altitudes, a simple function that assumed that each species has a
constant ratio of temperature, T, to molecular weight \mu, was used
for the fit.
Two parameters were fitted for at the lower extrapolation altitude:
the line-of-sight abundance, N_0, and the scale height H_0. The scale
height increases proportionally with the square of the radius,
r = r_s + z, due to variable gravity. The number density, n, is an
exponential in geopotential (Eq. 13 of [YOUNGETAL2017]).
In order to define the scale height and number density at r0, a new
geopotential referenced to r_0= r_s + h_0 was defined,
xi' = (r - r_0)r_0/r (or, for N(r'), substitute r' for r).
Because the scale height was not small compared with the radius on
Pluto, the line-of-sight column, N, was nearly but not quite an
exponential in geopotential (Eq. 13, [YOUNG2009]).
The functional form established from the fitting altitude range was
used to calculate the densities up to 2000 km for use with the Abel
transform. Errors on the density were calculated by propagation of
errors on the line-of-sight abundances. As expected, the errors on
the density were larger than those on the line-of-sight abundance.
Whereas the altitudes on the line-of-sight abundance were selected
to hold fractional errors to < 30%, the fractional errors on the
densities for many of the species were near one. This means they
are only measured to within a factor of ~2.7, or exp(1).
Quality Flags
-------------
The quality flags were defined for all species as:
Quality flag values:
0 = good
1 = out of HEIGHT/ALTITUDE range
2 = out of VALUE or ERROR CRITERIA range
4 = solution did not CONVERGE
and if multiple flags applied to a given value, the flags are added
together (e.g., a flag of 3 = both 1 and 2 apply).
HAZE
====
Summary of Haze in Pluto's atmosphere:
--------------------------------------
Atmospheric haze was detected in images by both the Long Range
Reconnaissance Imager (LORRI) and the Multispectral Visible Imaging
Camera (MVIC) on New Horizons. LORRI observed haze up to altitudes of
at least 200 km above Pluto's surface at solar phase angles from ~20
degrees to ~169 degrees. The haze is structured with about ~20 layers,
and the extinction due to haze is greater in the northern hemisphere
than at equatorial or southern latitudes. However, more haze layers
are discerned at equatorial latitudes. A search for temporal
variations found no evidence for motions of haze layers (temporal
changes in layer altitudes) on time scales of 2 to 5 hours, but did
find evidence of changes in haze scale height above 100 km altitude.
An ultraviolet extinction attributable to the atmospheric haze was
also detected by the ALICE ultraviolet spectrograph on New Horizons.
The haze particles are strongly forward-scattering in the visible, and
a microphysical model of haze is presented which reproduces the
visible phase function just above the surface with 0.5 micrometer
spherical particles, but also invokes fractal aggregate particles to
fit the visible phase function at 45 km altitude and account for UV
extinction. A model of haze layer generation by orographic excitation
of gravity waves is presented. This model accounts for the observed
layer thickness and distribution with altitude. Haze particles settle
out of the atmosphere and onto Pluto's surface, at a rate sufficient
to alter surface optical properties on seasonal time scales.
Additional details of Pluto haze observations are presented in the
literature [GLADSTONEETAL2016 and CHENGETAL2017]. LORRI is the
panchromatic, long focal length visible imager on New Horizons mission
to Pluto and Charon, and a full instrument description is given in
[CHENGETAL2008].
This dataset includes several categories of Pluto haze profile data:
- Azimuthally averaged profiles
- Temporal variation search
- Latitude variation search
- Phase angle variation
Azimuthally averaged profile data: ----------------------------------
Product ID: AZIMUTHAL_AVERAGE_PROFILE
Departure images of Pluto were obtained by New Horizons LORRI in
forward scatter geometries at high solar phase angle, in order to
observe atmospheric aerosols over the night side limb of Pluto and the
thin illuminated crescent Pluto. The four images from the sequence
P_LORRI_FULLFRAME_DEP included the full disk of Pluto at 3.85 km/px
resolution and are shown in Figure 2 of [CHENGETAL2017]. These images
show a bright ring of scattered light emission from atmospheric haze
all around Pluto's limb, over the night side and the illuminated
crescent. The present data appeared in Figure 3 of [CHENGETAL2017].
The four LORRI images were obtained with METs of
299236719, 299236749, 299236779 and 299236809, each exposure at 0.15 s
in 1x1 unbinned mode, at a range of 775,278 km from Pluto and at a
solar phase angle 166 degrees. The sub-observer longitude and
latitude, or Pluto (longitude, latitude) at the center of the observed
disk, is (288.7 degrees, -43.9 degrees). The sub-solar (longitude,
latitude) is (91.1 degrees, 51.6 degrees). The images were scaled,
shifted and co-added, after which a smoothly varying image background
was subtracted to remove solar stray light (this subtraction did not
completely remove the stray light, but left fine linear features
directed toward the sun and low level ghost features). The adopted
Pluto radius was 1190 km [GLADSTONEETAL2016]. The image brightness
from a pixel in data units (DN) was converted into I/F at the pivot
wavelength 607.6 nm [CHENGETAL2008], where I is the scattered radiance
and Pi*F is the solar irradiance 1.76 W/m2/nm, according to: I/F =
[DN/sec] * [7.5 e-5].
The azimuthally averaged I/F versus radius is obtained as in aperture
photometry, using concentric circular apertures of successively larger
radii from Pluto center. The I/F versus radius is found from the
average DN per pixel within each of the annuli between successive
apertures. This curve has been differentiated to find the scale height
by H = -1/(dln[I/F]/dr) with r the radius. The center of Pluto is
found by fitting a circle to the limb, with limb radius placed at the
maximum brightness gradient.
The three columns of the table are, from left to right, the radius
from Pluto center in km, the I/F value, and the scale height in km.
The noise floor in I/F is reached at ~1440 km.
Temporal variation search data: -------------------------------
The New Horizons departure images, at sufficiently high resolution to
characterize haze layering, covered a time base of several hours
including the image sequences named P_MULTI_DEP_LONG_1, P_LORRI_DEP_0,
P_MULTI_DEP_LONG_2, and P_LORRI_ALICE_DEP_1. These sequences (see
Table 1 of [CHENGETAL2017] were obtained at pixel scales of 2.3
km/pixel or less and phase angles in the range 166.6 degrees to 169
degrees. The New Horizons observations of haze above the limb of
Pluto, obtained at different times by different image sequences, also
measured haze over different locations on Pluto. The Pluto longitudes
and latitudes seen at the limb in these sequences are shown in the
limb traces of Figure 8 [CHENGETAL2017]. The intersection points of
the limb traces corresponding to different image sequences indicated
Pluto locations which were observed at the limb in more than one
sequence. These limb trace intersections afforded an opportunity to
study temporal variations in the haze layers, by comparing the haze
seen over the same locations on Pluto at different times. Four such
comparisons are shown in Figure 9 of [CHENGETAL2017].
Product IDs: TEMPVAR_A_MULTI_DEP_LONG_2, TEMPVAR_A_MULTI_DEP_LONG_1
The present data (from Figure 9a of [CHENGETAL2017]) compares two haze
I/F profiles obtained over the location (longitude, latitude) = (334
degrees, 42.0 degrees) with a time interval between observations of
3.46 hours. The comparison is made between the image sequences
P_MULTI_DEP_LONG_1 and P_MULTI_DEP_LONG_2. The adopted Pluto radius
was 1190 km [GLADSTONEETAL2016]. The image brightness from a pixel in
data units (DN) was converted into I/F at the pivot wavelength 607.6
nm [CHENGETAL2008], where I is the scattered radiance and Pi*F is the
solar irradiance 1.76 W/m2/nm, according to: I/F = [DN/sec] * [7.5
e-5].
From the sequence P_MULTI_DEP_LONG_2, the specific location on Pluto
was found in the images MET 299206659 and MET 299206660 (these were
both 0.15 s exposures, obtained one second apart), which were rescaled
to a common range of 360779 km, shifted and co-added. The pixel scale
became 1.79 km/pixel. A smoothly varying image background was
subtracted to remove solar stray light (this subtraction did not
completely remove the stray light, but left fine linear features
directed toward the sun and low level ghost features). After a counter
clock-wise rotation of the co-added image through 56.7 degrees, a
rectangular 600 x 12 pixel selection box was defined within which a
column-average profile was measured (the brightness was averaged over
12 rows). The three columns of the table are, from left to right, the
radius from Pluto center in km, the I/F, and the detrended I/F. The
detrended I/F is the fractional deviation of I/F from a trend, defined
as [(I/F) / trend] - 1, where the trend is a 6th order polynomial.
From the sequence P_MULTI_DEP_LONG_1, the specific location on Pluto
was found in the images MET 299194487 and MET 299194497 (these were
both 0.15 s exposures, obtained one second apart), which were rescaled
to a common range of 193342 km, shifted and co-added. The pixel scale
became 0.96 km/pixel. A smoothly varying image background was
subtracted to remove solar stray light (this subtraction did not
completely remove the stray light, but left fine linear features
directed toward the sun and low level ghost features). After a counter
clock-wise rotation of the co-added image through 51.5 degrees, a
rectangular 960 x 20 pixel selection box was defined within which a
column-average profile was measured (the brightness was averaged over
20 rows). The three columns of the table are, from left to right, the
radius from Pluto center in km, the I/F, and the detrended I/F. The
detrended I/F is the fractional deviation of I/F from a trend, defined
as [(I/F) / trend] - 1, where the trend is a 4th order polynomial.
Product IDs: TEMPVAR_B_LORRI_ALICE_DEP_1, TEMPVAR_B_MULTI_DEP_LONG_1
The present data (from Figure 9b of [CHENGETAL2017] compares two haze
I/F profiles obtained over the location (longitude, latitude) = (326
degrees, 43.2 degrees) with a time interval between observations of
5.43 hours. The comparison is made between the image sequences
P_MULTI_DEP_LONG_1 and P_LORRI_ALICE_DEP_1. The adopted Pluto radius
was 1190 km [GLADSTONEETAL2016]. The image brightness from a pixel in
data units (DN) was converted into I/F at the pivot wavelength 607.6
nm [CHENGETAL2008], where I is the scattered radiance and Pi*F is the
solar irradiance 1.76 W/m2/nm, according to: I/F = [DN/sec] * [7.5
e-5].
From the sequence P_LORRI_ALICE_DEP_1, the specific location on Pluto
was found in the images MET 299214015 and MET 299214045 (these were
both 0.15 s exposures, obtained 30 s apart), which were rescaled to a
common range of 458186 km, shifted and co-added. The pixel scale
became 2.27 km/pixel. A smoothly varying image background was
subtracted to remove solar stray light (this subtraction did not
completely remove the stray light, but left fine linear features
directed toward the sun and low level ghost features). After a counter
clock-wise rotation of the co-added image through 46.6 degrees, a
rectangular 600 x 12 pixel selection box was defined within which a
column-average profile was measured (the brightness was averaged over
12 rows). The three columns of the table are, from left to right, the
radius from Pluto center in km, the I/F, and the detrended I/F. The
detrended I/F is the fractional deviation of I/F from a trend, defined
as [(I/F) / trend] - 1, where the trend is exponential with a scale
height of 50.96 km.
From the sequence P_MULTI_DEP_LONG_1, the specific location on Pluto
was found in the images MET 299194487 and MET 299194497 (these were
both 0.15 s exposures, obtained one second apart), which were rescaled
to a common range of 193342 km, shifted and co-added. The pixel scale
became 0.96 km/pixel. A smoothly varying image background was
subtracted to remove solar stray light (this subtraction did not
completely remove the stray light, but left fine linear features
directed toward the sun and low level ghost features). After a counter
clock-wise rotation of the co-added image through 51.5 degrees, a
rectangular 960 x 20 pixel selection box was defined within which a
column-average profile was measured (the brightness was averaged over
20 rows). The three columns of the table are, from left to right, the
radius from Pluto center in km, the I/F, and the detrended I/F. The
detrended I/F is the fractional deviation of I/F from a trend, defined
as [(I/F) / trend] - 1, where the trend is a 4th order polynomial.
Product IDs: TEMPVAR_C_LORRI_ALICE_DEP_1, TEMPVAR_C_MULTI_DEP_LONG_2
The present data (from Figure 9c of [CHENGETAL2017] compares two haze
I/F profiles obtained over the location (longitude, latitude) = (314
degrees, 45 degrees) with a time interval between observations of 1.97
hours. The comparison is made between the image sequences
P_MULTI_DEP_LONG_2 and P_LORRI_ALICE_DEP_1. The adopted Pluto radius
was 1190 km [GLADSTONEETAL2016]. The image brightness from a pixel in
data units (DN) was converted into I/F at the pivot wavelength 607.6
nm [CHENGETAL2008], where I is the scattered radiance and Pi*F is the
solar irradiance 1.76 W/m2/nm, according to: I/F = [DN/sec] * [7.5
e-5].
From the sequence P_LORRI_ALICE_DEP_1, the specific location on Pluto
was found in the images MET 299214015, MET 299214045 and MET 299214075
(these were all 0.15 s exposures, obtained 30 s apart), which were
rescaled to a common range of 458186 km, shifted and co-added. The
pixel scale became 2.27 km/pixel. A smoothly varying image background
was subtracted to remove solar stray light (this subtraction did not
completely remove the stray light, but left fine linear features
directed toward the sun and low level ghost features). After a counter
clock-wise rotation of the co-added image through 37.8 degrees, a
rectangular 600 x 12 pixel selection box was defined within which a
column-average profile was measured (the brightness was averaged over
12 rows). The three columns of the table are, from left to right, the
radius from Pluto center in km, the I/F, and the detrended I/F. The
detrended I/F is the fractional deviation of I/F from a trend, defined
as [(I/F) / trend] - 1, where the trend was a 3rd order polynomial.
From the sequence P_MULTI_DEP_LONG_2, the specific location on Pluto
was found in the images MET 299206659, MET299206660, and MET 299206661
(these were all 0.15 s exposures, obtained one second apart), which
were rescaled to a common range of 360779 km, shifted and co-added.
The pixel scale became 1.79 km/pixel. A smoothly varying image
background was subtracted to remove solar stray light (this
subtraction did not completely remove the stray light, but left fine
linear features directed toward the sun and low level ghost features).
After a counter clock-wise rotation of the co-added image through 41.9
degrees, a rectangular 600 x 12 pixel selection box was defined within
which a column-average profile was measured (the brightness was
averaged over 12 rows). The three columns of the table are, from left
to right, the radius from Pluto center in km, the I/F, and the
detrended I/F. The detrended I/F is the fractional deviation of I/F
from a trend, defined as [(I/F) / trend] - 1, where the trend is a 3rd
order polynomial.
Product IDs: TEMPVAR_D_LORRI_ALICE_DEP_1, TEMPVAR_D_LORRI_DEP_0
The present data (from Figure 9d of [CHENGETAL2017]) compares two haze
I/F profiles obtained over the location (longitude, latitude) = (133
degrees, -45.4 degrees) with a time interval between observations of
2.61 hours. The comparison is made between the image sequences
P_LORRI_ALICE_DEP_1 and P_LORRI_DEP_0. The adopted Pluto radius was
1190.6 km [GLADSTONEETAL2016]. The image brightness from a pixel in
data units (DN) was converted into I/F at the pivot wavelength 607.6
nm [CHENGETAL2008], where I is the scattered radiance and Pi*F is the
solar irradiance 1.76 W/m2/nm, according to: I/F = [DN/sec] * [7.5
e-5].
From the sequence P_LORRI_ALICE_DEP_1, the specific location on Pluto
was found in the images MET 299213682 and MET 299213712 (these were
both 0.15 s exposures, obtained 30 s apart), which were rescaled to a
common range of 458186 km, shifted and co-added. The pixel scale
became 2.27 km/pixel. A smoothly varying image background was
subtracted to remove solar stray light (this subtraction did not
completely remove the stray light, but left fine linear features
directed toward the sun and low level ghost features). After a counter
clock-wise rotation of the co-added image through 37 degrees, a
rectangular 600 x 12 pixel selection box was defined within which a
column-average profile was measured (the brightness was averaged over
12 rows). The three columns of the table are, from left to right, the
radius from Pluto center in km, the I/F, and the detrended I/F. The
detrended I/F is the fractional deviation of I/F from a trend, defined
as [(I/F) / trend] - 1, where the trend is a 6th order polynomial.
From the sequence P_LORRI_DEP_0, the specific location on Pluto was
found in the images MET 299204282 and MET 299204283 (these were both
0.15 s exposures, obtained 1 s apart), which were rescaled to a common
range of 328830 km, shifted and co-added. The pixel scale became 1.63
km/pixel. A smoothly varying image background was subtracted to remove
solar stray light (this subtraction did not completely remove the
stray light, but left fine linear features directed toward the sun and
low level ghost features). After a counter clock-wise rotation of the
co-added image through 40.2 degrees, a rectangular 600 x 12 pixel
selection box was defined within which a column-average profile was
measured (the brightness was averaged over 12 rows). The three columns
of the table are, from left to right, the radius from Pluto center in
km, the I/F, and the detrended I/F. The detrended I/F is the
fractional deviation of I/F from a trend, defined as [(I/F) / trend] -
1, where the trend was a 5th order polynomial.
Product IDs: TEMPVAR_E_LORRI_ALICE_DEP_1, TEMPVAR_E_MULTI_DEP_LONG_2,
TEMPVAR_E_MULTI_DEP_LONG_1
The present data (from Figure 10 of [CHENGETAL2017]) compares three
haze I/F profiles obtained over equatorial locations, one profile from
each of the image sequences P_MULTI_DEP_LONG_1, P_MULTI_DEP_LONG_2,
and P_LORRI_ALICE_DEP_1. The times relative to closest approach for
these sequences were 3.78 hr, 7.25 hr, and 9.21 hr, respectively (from
Table 1 of [CHENGETAL2017]). The phase angles for these sequences were
169.0 degrees, 167.1 degrees, and 166.6 degrees respectively (from
Table 3 of [CHENGETAL2017]. The adopted Pluto radius was 1190 km
(Gladstone et al. 2016). The image brightness from a pixel in data
units (DN) was converted into I/F at the pivot wavelength 607.6 nm
[CHENGETAL2008], where I is the scattered radiance and Pi*F is the
solar irradiance 1.76 W/m2/nm, according to: I/F = [DN/sec] * [7.5
e-5].
From the sequence P_LORRI_ALICE_DEP_1, a profile was obtained over the
specific location on Pluto (longitude, latitude) = (33 degrees,-0.2
degrees), using the images MET 299213793, MET 299213823 and MET
299213853 (these were all 0.15 s exposures, obtained 30 s apart).
These images were rescaled to a common range of 458186 km, shifted and
co-added. The pixel scale became 2.27 km/pixel. A smoothly varying
image background was subtracted to remove solar stray light (this
subtraction did not completely remove the stray light, but left fine
linear features directed toward the sun and low level ghost features).
After a clock-wise rotation of the co-added image through 60 degrees,
a rectangular 60020 pixel selection box was defined within which a
column-average profile was measured (the brightness was averaged over
20 rows). The three columns of the table are, from left to right, the
radius from Pluto center in km, the I/F, and the detrended I/F. The
detrended I/F is the fractional deviation of I/F from a trend, defined
as [(I/F) / trend] - 1, where the trend is an exponential with scale
height 43.0 km.
From the sequence P_MULTI_DEP_LONG_2, a profile was obtained over the
specific location on Pluto (longitude, latitude) = (32 degrees, 5
degrees), using the images MET 299206714, MET 299206715 and MET
299206716 (these were all 0.15 s exposures, obtained 1 s apart). These
images were rescaled to a common range of 360779 km, shifted and co-
added. The pixel scale became 1.79 km/pixel. A smoothly varying image
background was subtracted to remove solar stray light (this
subtraction did not completely remove the stray light, but left fine
linear features directed toward the sun and low level ghost features).
After a clock-wise rotation of the co-added image through 60 degrees,
a rectangular 600 x 16 pixel selection box was defined within which a
column-average profile was measured (the brightness was averaged over
16 rows). The three columns of the table are, from left to right, the
radius from Pluto center in km, the I/F, and the detrended I/F. The
detrended I/F is the fractional deviation of I/F from a trend, defined
as [(I/F) / trend] - 1, where the trend is an exponential with scale
height 45.1 km.
From the sequence P_MULTI_DEP_LONG_1, a profile was obtained over the
specific location on Pluto (longitude, latitude) = (43 degrees, -0.5
degrees), using the images MET 299194661 and MET 299194671 (these were
both 0.15 s exposures, obtained 10 s apart). These images were
rescaled to a common range of 193342 km, shifted and co-added. The
pixel scale became 0.960 km/pixel. A smoothly varying image background
was subtracted to remove solar stray light (this subtraction did not
completely remove the stray light, but left fine linear features
directed toward the sun and low level ghost features). After a clock-
wise rotation of the co-added image through 54 degrees, a rectangular
600 x 20 pixel selection box was defined within which a column-average
profile was measured (the brightness was averaged over 20 rows). The
three columns of the table are, from left to right, the radius from
Pluto center in km, the I/F, and the detrended I/F. The detrended I/F
is the fractional deviation of I/F from a trend, defined as [(I/F) /
trend] - 1, where the trend is a 6th order polynomial.
Latitude variation data: ---------------------------
Haze brightness profiles were compared from the same observation
sequence P_MULTI_DEP_LONG_1 and obtained over the most northern
latitudes imaged (latitude 44 degrees) and over equatorial latitudes
at the same resolution of 0.96 km/px and at the same solar phase
angle. The New Horizons departure images from P_MULTI_DEP_LONG_1 were
at sufficiently high resolution to characterize haze layering, at a
pixel scale of 0.96 km/pixel, and the solar phase angle was 169
degrees (Table 3 of [CHENGETAL2017]). The Pluto longitudes and
latitudes seen at the limb in the New Horizons haze observation
sequences are shown in the limb traces of Figure 8 [CHENGETAL2017].
The present data (from Figure 11 of [CHENGETAL2017]) compares two haze
I/F profiles from the image sequence P_MULTI_DEP_LONG_1, one from an
equatorial latitude and one from a northern latitude, using images
obtained over a 184 s time span. The adopted Pluto radius was 1190 km
[GLADSTONEETAL2016]. The image brightness from a pixel in data units
(DN) was converted into I/F at the pivot wavelength 607.6 nm (Cheng et
al. 2008), where I is the scattered radiance and Pi*F is the solar
irradiance 1.76 W/m2/nm, according to: I/F = [DN/sec] * [7.5 e-5].
Product ID: LATVAR_N_P_MULTI_DEP_LONG_1
From the sequence P_MULTI_DEP_LONG_1, a profile was obtained over the
northern location on Pluto (longitude, latitude) = (312 degrees, 44
degrees), using the images MET 299194487 and MET 299194497 (these were
both 0.15 s exposures, obtained 10 s apart). These images were
rescaled to a common range of 193342 km, shifted and co-added. The
pixel scale became 0.96 km/pixel. A smoothly varying image background
was subtracted to remove solar stray light (this subtraction did not
completely remove the stray light, but left fine linear features
directed toward the sun and low level ghost features). After a
counter-clockwise rotation of the co-added image through 36 degrees, a
rectangular 600 x 20 pixel selection box was defined within which a
column-average profile was measured (the brightness was averaged over
20 rows). The three columns of the table are, from left to right, the
radius from Pluto center in km, the I/F, and the detrended I/F. The
detrended I/F is the fractional deviation of I/F from a trend, defined
as [(I/F) / trend] - 1, where the trend is a 6th order polynomial.
Product ID: LATVAR_E_P_MULTI_DEP_LONG_1
Also from the sequence P_MULTI_DEP_LONG_1, a profile was obtained over
the specific location on Pluto (longitude, latitude) = (43 degrees,
-0.5 degrees), using the images MET 299194661 and MET 299194671 (these
were both 0.15 s exposures, obtained 10 s apart). These images were
rescaled to a common range of 193342 km, shifted and co-added. The
pixel scale became 0.96 km/pixel. A smoothly varying image background
was subtracted to remove solar stray light (this subtraction did not
completely remove the stray light, but left fine linear features
directed toward the sun and low level ghost features). After a clock-
wise rotation of the co-added image through 54 degrees, a rectangular
600 x 20 pixel selection box was defined within which a column-average
profile was measured (the brightness was averaged over 20 rows). The
three columns of the table are, from left to right, the radius from
Pluto center in km, the I/F, and the detrended I/F. The detrended I/F
is the fractional deviation of I/F from a trend, defined as [(I/F) /
trend] - 1, where the trend is a 6th order polynomial.
Phase Angle Variation Data: ---------------------------
Day-side Limb phase angle variation data: Haze brightness profiles
were compared from observations obtained over the 19 hr time span from
3.6 hour before closest approach to 15.6 hr afterwards, including back
scatter and forward scatter geometries. The solar phase function was
compiled using haze observations over the day side limb at similar
northern latitudes > 40 degrees.
At low solar phase angles in approach imaging of Pluto, the haze above
the limb of Pluto is much fainter than the sunlit surface of Pluto,
and the instrumental stray light from the bright limb of Pluto must be
removed to measure haze. This removal was accomplished empirically,
using observations obtained in similar back scatter viewing geometries
and at similar image resolution by the P_LORRI_L1 and PELR_C_LORRI
sequences before closest approach (Table 1 of [CHENGETAL2008]). A haze
profile at 19.6 degrees solar phase angle was extracted using images
of Charon and Pluto obtained in similar viewing geometries, where the
Charon image from C_LORRI was used to remove stray light from the
P_LORRI Pluto image, and the excess brightness in the Pluto image was
characterized and attributed to haze. This method takes advantage of
the absence of an atmosphere, as well as absence of haze, at Charon
[STERNETAL2017].
The present data appeared in Fig. 12 of [CHENGETAL2017]. The adopted
Pluto radius was 1190.4 km [GLADSTONEETAL2016]. The image brightness
from a pixel in data units (DN) was converted into I/F at the pivot
wavelength 607.6 nm [CHENGETAL2008], where I is the scattered radiance
and Pi*F is the solar irradiance 1.76 W/m2/nm, according to: I/F =
[DN/sec] * [7.5 e-5].
Product ID: PHASE_DAY_P_LORRI_L1
From the sequence P_LORRI_L1, a profile was obtained over the
northern location on Pluto (longitude, latitude) = (312 degrees, 44
degrees), using the image MET 299168039 (an 0.15 s exposure). This
image was rescaled to a range of 171200 km. The pixel scale was 0.85
km/pixel. A hard stretch of the image shows low level ghost features;
a column-oriented artifact of readout smear removal is also visible. A
rectangular 725 x 75 pixel selection box was defined within which a
column-average profile was measured (the brightness was averaged over
75 rows). The three columns of the table are, from left to right, the
radius from Pluto center in km, the column-average DN profile before
removal of the scaled Charon profile (PHASE_DAY_PELR_C_LORRI), and the
corrected I/F after removal of the scaled Charon profile. This Charon
profile is scaled by a constant factor and is then subtracted from the
'column-average DN profile' of PHASE_DAY_P_LORRI_L1 to generate the
corrected I/F profile.
Product ID: PHASE_DAY_PELR_C_LORRI
From the sequence PELR_C_LORRI, the images MET 299169015 and MET
299169016 (these were both 0.15 s exposures, obtained 1 s apart) were
rescaled to a common range of 172233 km, shifted and co-added. The
pixel scale became 0.855 km/pixel. A hard stretch of the image shows
low level ghost features; a column-oriented artifact of readout smear
removal is also visible. A rectangular 725 x 75 pixel selection box
was defined within which a column-average profile was measured (the
brightness was averaged over 75 rows). The Charon DN values are scaled
by a factor 0.7 to account for the different average brightness
within the sunlit portions of the profiles. The result is the DN
profile attributed to stray light versus the number of pixels x from
the bright limb to be subtracted from the Pluto profile. (The number
of pixels x is defined as the profile of average DN versus column
number in pixel units from the PHASE_DAY_P_LORRI_L1 data product.) The
three columns of the table are, from left to right, the radius from
Pluto center in km, the pixel number x in the Charon profile (x has an
arbitrary zero; the limb is at x=675), and the scaled Charon DN
profile, where radius is the radius in the Pluto image at the
corresponding pixel number relative to the bright limb. The Charon
profile in PHASE_DAY_PELR_C_LORRI is scaled by a factor 0.7 and then
subtracted from the column-average DN profile of PHASE_DAY_P_LORRI_L1
to generate the corrected I/F after removal of stray light.
Night-side Limb phase angle variation data: At low or moderate solar
phase angles in approach imaging of Pluto, the haze above the limb of
Pluto is much fainter than the sunlit surface of Pluto, and the
instrumental stray light from the bright limb of Pluto must be removed
to measure haze. This removal was accomplished empirically, using
observations obtained in similar back scatter viewing geometries and
at similar image resolution by the P_LORRI_L1 and PELR_C_LORRI
sequences before closest approach (Table 1 of [CHENGETAL2017]). A haze
profile at 19.5 degrees solar phase angle over the night side limb of
Pluto was extracted using images of Charon and Pluto obtained in
similar viewing geometries, where the Charon image from C_LORRI was
used to remove stray light from the P_LORRI Pluto image, and the
excess brightness in the Pluto image was characterized and attributed
to haze. This method takes advantage of the absence of an atmosphere,
as well as absence of haze, at Charon [STERNETAL2017].
The present data appeared in Fig. 13 of [CHENGETAL2017]. The adopted
Pluto radius was 1190 km [GLADSTONEETAL2016]. The image brightness
from a pixel in data units (DN) was converted into I/F at the pivot
wavelength 607.6 nm [CHENGETAL2008], where I is the scattered radiance
and Pi*F is the solar irradiance 1.76 W/m2/nm, according to: I/F =
[DN/sec] * [7.5 e-5].
Product ID: PHASE_NIGHT_P_LORRI_L1
From the sequence P_LORRI_L1, a profile was obtained over the
northern location on Pluto (longitude, latitude) = (260 degrees, 13
degrees), using the images MET 299167703 and MET 299167704 (both are
0.15 s exposures). These images were rescaled to a range of 175814 km,
shifted and co-added. The pixel scale was 0.873 km/pixel. A hard
stretch of the image shows low level ghost features. After a
clockwise rotation of 56 degrees, a rectangular 341100 pixel selection
box was defined within which a column-average profile was measured
(the brightness was averaged over 100 rows). The three columns of the
table are, from left to right, the radius from Pluto center in km, the
column-average DN profile before removal of the scaled Charon profile,
and the corrected I/F after removal of the scaled Charon profile. If
the radius is <1900 km, then the brightness is a haze-lit Pluto
surface brightness, and the radius given is that at closest approach
of the line of sight to Pluto center.
Product ID: PHASE_NIGHT_PELR_C_LORRI
From the sequence PELR_C_LORRI, the images MET 299169015 and MET
299169016 (these were both 0.15 s exposures, obtained 1 s apart) were
rescaled to a common range of 172233 km, shifted and co-added. The
pixel scale became 0.855 km/pixel. A hard stretch of the image shows
low level ghost features. The Charon image was flipped horizontally to
match the geometry of the Pluto image, and a rectangular 40585 pixel
selection box was defined within which a column-average profile was
measured (the brightness was averaged over 85 rows). The bright limb
of the Charon profile was aligned to the terminator of the Pluto
profile, and the Charon profile DN values were scaled by a factor 0.70
to account for the average brightness within the sunlit portions. The
Charon profile gives the numbers of DN attributed to stray light
versus the number of pixels x from the bright limb (x has an arbitrary
zero; the image has been flipped; Charon limb is at x=62). The three
columns of the table are, from left to right, the radius from Pluto
center in km, the pixel number x in the Charon profile, and the scaled
Charon DN, where radius is the radius in the Pluto image at the
corresponding pixel number relative to the bright limb.
Product ID: PHASE_P_MVIC_LORRI_CA, Product ID:
PHASE_PELR_C_MVIC_LORRI_CA
At low or moderate solar phase angles in approach imaging of Pluto,
the haze above the limb of Pluto is much fainter than the sunlit
surface of Pluto, and the instrumental stray light from the bright
limb of Pluto must be removed to measure haze. This removal was
accomplished empirically, using observations obtained in similar back
scatter viewing geometries and at similar image resolution by the
P_MVIC_LORRI_CA and C_MVIC_LORRI_CA sequences before closest approach
(Table 1 of Cheng et al. 2017). A Pluto haze profile at 67.3 degrees
solar phase angle was extracted using images of Charon and Pluto
obtained in similar viewing geometries, where the Charon image from
PELR_C_MVIC_LORRI_CA was used to remove stray light from the
P_MVIC_LORRI_CA Pluto image, and the excess brightness in the Pluto
image was characterized and attributed to haze. This method takes
advantage of the absence of an atmosphere, as well as absence of haze,
at Charon [STERNETAL2017].
The present data appeared in Fig. 15 of [CHENGETAL2017]. The adopted
Pluto radius was 1190.5 km [GLADSTONEETAL2016]. The image brightness
from a pixel in data units (DN) was converted into I/F at the pivot
wavelength 607.6 nm [CHENGETAL2008], where I is the scattered radiance
and Pi*F is the solar irradiance 1.76 W/m2/nm, according to: I/F =
[DN/sec] * [7.5 e-5].
From the sequence P_MVIC_LORRI_CA, a profile was obtained over the
approximate northern location on Pluto (longitude, latitude) = (104
degrees, 55 degrees), using the image MET 299179658 (a 0.01 s
exposure). This image was rescaled to a range of 17297 km. The pixel
scale was 0.0859 km/pixel. A hard stretch of the image shows artifacts
of readout smear removal and low level ghost features. A rectangular
77290 pixel selection box was defined within which a column-average
profile was measured (the brightness was averaged over 90 rows). The
three columns of the table are, from left to right, the radius from
Pluto center in km, the column-average DN profile before removal of
the scaled Charon profile, and the corrected I/F after removal of the
scaled Charon profile.
From the sequence PELR_C_MVIC_LORRI_CA, the image MET 299180406 (a
0.01s exposure) was used, for which the target range was 31724 km, and
the pixel scale was 0.1575 km/pixel. A hard stretch of the image shows
artifacts of readout smear removal. A rectangular 77290 pixel
selection box was defined within which a column-average profile was
measured (the brightness was averaged over 90 rows). The limb of the
Charon profile was aligned to the limb of the Pluto profile, and the
Charon profile DN values were scaled by a factor 3.08 to account for
the average brightness within the sunlit portions. The Charon profile
gives the numbers of DN attributed to stray light versus the number of
pixels x from the bright limb (x has an arbitrary zero; Charon limb is
at x=218). The three columns of the table are, from left to right, the
radius from Pluto center in km, the pixel number x in the Charon
profile, and the scaled Charon DN, where radius is the radius in the
Pluto image at the corresponding pixel number relative to the bright
limb.
Product ID: PHASE_P_HIPHASE_HIRES
The sequence P_HIPHASE_HIRES obtained images after closest approach
(Table 1 of [CHENGETAL2017]). Image MET 299181359, a 0.01 s exposure
obtained at a range of 18758 km and apixel scale 0.0931 km/px, was
used to obtain a Pluto haze profile at 148.3 degrees solar phase
angle. The present data appeared in Fig. 16 of [CHENGETAL2017]. The
adopted Pluto radius was 1190 km [GLADSTONEETAL2016]. The image
brightness from a pixel in data units (DN) was converted into I/F at
the pivot wavelength 607.6 nm [CHENGETAL2008], where I is the
scattered radiance and Pi*F is the solar irradiance 1.76 W/m2/nm,
according to: I/F = [DN/sec] * [7.5 e-5].
The image MET 299181359 was obtained over the approximate location on
Pluto (longitude, latitude) = (160 degrees, -7 degrees). A rectangular
894 x 110 pixel selection box was defined within which a column-
average profile was measured (the brightness was averaged over 110
rows). The four columns of the table are, from left to right, the
radius from Pluto center in km, the I/F, the detrended I/F, and a
notation as to whether the brightness is a Pluto surface brightness or
a haze brightness. If a Pluto surface brightness is noted, the radius
given is that at closest approach of the line of sight to Pluto
center. The detrended I/F is the fractional deviation of I/F from a
trend, defined as [(I/F) / trend] - 1, where the trend is an
exponential with scale height 61.95 km.
REXATMOS
========
Lower Atmospheric Temperature and Pressure Profiles
---------------------------------------------------
On 14 July 2015 New Horizons performed a radio occultation (RO) that
sounded Pluto's atmosphere down to the surface. This file contains
the atmospheric pressure-temperature profile derived from measurements
at occultation entry, which occurred at sunset near the center of the
anti-Charon hemisphere. The sensitivity of the measurements was
enhanced by a unique configuration of ground equipment and spacecraft
instrumentation. Signals were transmitted simultaneously by four
antennas of the NASA Deep Space Network, each radiating 20 kW at a
wavelength of 4.2 cm. The polarization was right circular for one
pair of signals and left circular for the other pair. New Horizons
received the four signals and separated them by polarization for
processing by two independent receivers, each referenced to a
different ultra-stable oscillator. The two data streams were
digitized, filtered, and stored on the spacecraft for later
transmission to Earth. All subsequent steps of analysis were
performed after the data had been received on the ground. We
calibrated each signal to remove effects not associated with Pluto's
atmosphere, including the limb diffraction pattern. We then applied a
specialized method of analysis to retrieve profiles of number density,
pressure, and temperature from the combined phase measurements. See
[HINSONETAL2017] for a detailed discussion of the procedure used for
data analysis and profile retrieval.
STAROCC
=======
Stellar Atmospheric Occultation and Appulse Count Rates
-------------------------------------------------------
A few hours after its encounter with Pluto, Alice observed the
simultaneous stellar occultation and appulse of two UV-bright stars,
69 Ori and 72 Ori, respectively. As during the solar occultation that
occurred immediately prior, the transmission of starlight through
Pluto's atmosphere was sensitive to absorption by N2, CH4, C2H6, C2H2,
C2H4, and haze. The line of sight to each star passed over different
areas of Pluto from those probed by the solar occultation, providing
insight into the degree of spatial and diurnal variability of
atmospheric composition on Pluto.
The data have been binned in time to 1-second (MET) resolution. The
observed stellar spectra have been extracted and corrected for a
variety of instrumental effects.
For a detailed description of the data analysis process, please
see [KAMMERETAL2020].
THERMSCAN
=========
Diametric and Winter Pole Thermscans Across Pluto By REX
--------------------------------------------------------
The data contained in the thermscan directory gives the values in
Kelvin for the diametric and winter pole scans across Pluto for both
the Right Circularly Polarized (RCP) and Left Circularly Polarized
(LCP) channels of the REX (Radio EXperiment) instrument on New
Horizons. The document section of this dataset contains three REX
documents that can be used to help interpret the data.
Note the calibration document (v4.7) has the most recent calibration
constants which were used to generate the Calibrated REX dataset
in the PDS archive and the values given in that paper should be
used as the best known calibration values. The Radio Brightness
Temperature Measurement document [LINSCOTTETAL2017] uses older values
for these constants. See [BIRDETAL2019] for a discussion of the
uncertainties.
Contact Information
===================
For any questions regarding the data format of the archive,
contact
New Horizons Principal Investigator:
Alan Stern, Southwest Research Institute
S. Alan Stern
Southwest Research Institute
Department of Space Studies
1050 Walnut Street, Suite 400
Boulder, CO 80302
USA
|
.png)
.png)
