J/A+A/631/A32 3D dust extinction map of 600pc cude Sun centered (Leike+, 2019)
Charting nearby dust clouds using Gaia data only.
Leike R., Ensslin T.
<Astron. Astrophys. 631, A32 (2019)>
=2019A&A...631A..32L 2019A&A...631A..32L (SIMBAD/NED BibCode)
ADC_Keywords: Stars, nearby ; Interstellar medium ; Extinction
Keywords: dust, extinction - local interstellar matter -
methods: data analysis - solar neighborhood
Abstract:
Highly resolved maps of the local Galactic dust are an important
ingredient for sky emission models. In nearly the whole
electromagnetic spectrum one can see imprints of dust, many of which
originate from dust clouds within 300pc. Having a detailed 3D
reconstruction of these local dust clouds enables detailed studies,
helps to quantify the impact on other observables and is a milestone
necessary to enable larger reconstructions, as every sightline for
more distant objects will pass through the local dust.
To infer the dust density we use parallax and extinction estimates
published by the Gaia collaboration in their second data release. We
model the dust as a log-normal process using a hierarchical Bayesian
model. We also infer non-parametrically the kernel of the log-normal
process, which corresponds to the physical spatial correlation power
spectrum of the log-density.
Using only Gaia data of the second Gaia data release, we reconstruct
the 3D dust density and its spatial correlation spectrum in a 600pc
cube centered on the Sun. We report a spectral index of the
logarithmic dust density of 3.1 on Fourier scales with wavelengths
between 2pc and 125pc. The resulting 3D dust map as well as the power
spectrum and posterior samples are publicly available for download.
Description:
Two data files are provided;
'samples.h5' is containing 50 approximate posterior samples of the
dust density and its spatial correlation power spectrum 'mean_std.h5'
is only containing the sample mean and standard deviation. It can be
computed from the file 'samples.h5' and it is thus advised to only
download one of the files.
Both files are provided in the hdf5 format, allowing for fast and
flexible access in many programming languages.
samples.h5:
structure:
attributes:
'information': A string containing information about the origin
of the dataset
datasets:
'dust_samples': A four dimensional array of float64 numbers.
The ijkl-th data value contains the absorption per parsec in the
region where x reaches from (j*600/256-300)pc to
((j+1)*600/256-300)pc , analogously for (k,y) and (l,z).
The Galactic center lies in positive x direction, z is the
coordinate of axis perpendicular to the galactic disk. The index i
counts through the samples. Note that the absorption is not in
magnitudes but in natural logarithmic units, i.e. a value of 1
corresponds to an intensity reduction of e^{-1} for G-Band photons
traveling one parsec through that dust.
attributes:
'information': A string containing the description above
'power_samples': A 2D float64 array. The ij-th data value contains the
i-th sample of the j-th k-mode of the Power spectrum of the
logarithmic G-band dust absorption cross section density. The
dust_samples are a-priori assumed to be distributed like
F-1(sqrt(Pk*V)*xik)-np.log(0.001) where F-1 is the
inverse 3D Fourier transform, sqrt(P_k) is the square root
of the power at the Fourier coordinate k, V is the total
volume of the simulation box in pc, i.e. 600^3, and xi_k
are independent normally distributed values for every
Fourier mode k. The power is given in units of pc^3. The
dataset 'power_coordinates' contains the corresponding Fourier
coordinates to this dataset. Note that the power samples and the
dust samples are jointly distributed.
attributes:
'information': A string containing information about
'power_coordinates': A 1D float64 array containing the coordinate for
the power spectrum. Note that we follow the Fourier-convention
where the minimal non-zero k-coordinate is 1/L (without 2 Pi),
where L is the Length of the box (600pc). Coordinates are given in
units of 1/pc.
attributes:
'information': A string containing the description above
mean_std.h5:
structure:
attributes:
'information': A string containing information about the origin
of the dataset
datasets:
'mean': A 3D array of float64 number. The jkl-th data value contains
the sample averaged absorption per parsec in the region where x
reaches from (j*600/256-300)pc to ((j+1)*600/256-300)pc ,
analogously for (k,y) and (l,z). The galactic center lies in
positive x direction, z is the coordinate of axis perpendicular to
the galactic disk. Note that the absorption is not in magnitudes
but in natural logarithmic units, i.e. a value of 1 corresponds to
an intensity reduction of e^{-1} for G-Band photons traveling one
parsec through that dust.
attributes:
'information': A string containing the description above
'std': One sigma uncertainty estimates on the absorption values. Note
that the uncertainties of individual voxels are correlated, for a
detailed analysis using the samples in the full data cube is
recommended.
attributes:
'information': A string containing information about
'power_mean': A 1D array of float64 numbers. The i-th data value
contains the i-th k-mode of the sampled mean of the power spectrum
of the logarithmic G-band dust absorption cross section density.
The log-dust is a-priori assumed to be distributed like
F-1(sqrt(Pk*V)*xik)-np.log(0.001) where F-1 is the inverse
3D Fourier transform, sqrt(P_k) is the square root of the power at
the Fourier coordinate k, V is the total volume of the simulation
box in pc, i.e. 600^3, and xi_k are independent normally
distributed values for every Fourier mode k. The power is given in
units of pc^3. The dataset 'power_coordinates' contains the
corresponding Fourier coordinates to this dataset.
attributes:
'information': A string containing information about
'power_std': One sigma uncertainty estimates on the power spectrum
values. Note that the uncertainties of individual values are
correlated, for a detailed analysis using the samples in the full
data cube is recommended.
attributes:
'information': A string containing information about
'power_coordinates': A 1D float64 array containing the coordinate for
the power spectrum. Note that we follow the Fourier-convention
where the minimal non-zero k-coordinate is 1/L (without 2 Pi),
where L is the Length of the box (600pc). Coordinates are given in
units of 1/pc.
attributes:
'information': A string containing the description above
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
samples.h5 512 13134098 50 approximate posterior samples of the dust
density and its spatial correlation power spectrum
mean_std.h5 512 524832 Sample mean and standard deviation
license.txt 512 40 ODC Attribution License (ODC-By)
--------------------------------------------------------------------------------
Acknowledgements:
Reimar Leike, reimar(at)mpa-garching.mpg.de
(End) Patricia Vannier [CDS] 26-Aug-2019