Author name code: milic
ADS astronomy entries on 2022-09-14
author:"Milic, Ivan"
------------------------------------------------------------------------
Title: Investigating magnetic field inference from the spectral region
around the Mg I b2 line using the weak-field approximation
Authors: Vukadinović, D.; Milić, I.; Atanacković, O.
Bibcode: 2022A&A...664A.182V
Altcode: 2022arXiv220504236V
Context. The understanding of the magnetic field structure in the
solar atmosphere is important in assessing both the dynamics and
the energy balance of the solar atmosphere. Our knowledge about
these magnetic fields comes predominantly from the interpretation
of spectropolarimetric observations. Simpler approaches based on
approximations such as the weak-field approximation (WFA) deserve
special attention as key methods in the interpretation of large,
high-resolution datasets.
Aims: We investigate the applicability
of the WFA for retrieving the depth-dependent line-of-sight
(LOS) magnetic field from the spectral region containing the Mg I
b2 spectral line and two photospheric Ti I and Fe I lines in
its wings.
Methods: We constructed and applied a 12-level model
for Mg I atom that realistically reproduces the b2 line
profile of the mean quiet Sun. We tested the applicability of the WFA
to the spectra computed from the FAL C atmospheric model with different
magnetic and velocity fields added on an ad hoc basis . Then we extended
the analysis to the spectra computed from two 3D magneto-hydrodynamic
(MHD) MURaM simulations of the solar atmosphere. The first MHD cube
was used to estimate the Stokes V formation heights of each spectral
line. These heights correspond to optical depths at which the standard
deviation of the difference between the WFA-inferred magnetic field
and the magnetic field in the MHD cube is minimal. The estimated
formation heights were verified using the second MHD cube.
Results: The LOS magnetic field retrieved by the WFA is reliable for
the magnetic field strength up to 1.4 kG even when moderate velocity
gradients are present. The exception is the Fe I line, for which we
found a strong discrepancy in the WFA-inferred magnetic fields because
of the line blend. We estimated the Stokes V formation heights of each
spectral line to be: logτFe = −2.6, logτMg
= −3.3, and logτTi = −1.8. We were able to estimate
the LOS magnetic field from the MURaM cube at these heights with the
uncertainty of 150 G for the Fe I and Ti I lines and only 40 G for
the Mg I b2 line.
Conclusions: Using the WFA, we can
quickly get a reliable estimate of the structure of the LOS magnetic
field in the observed region. This offers a significant advantage in
comparison with otherwise time-consuming classical spectropolarimetric
inversions. We find that the Mg I b2 line profile calculated
from the quiet Sun MURaM simulation agrees very well with the observed
mean spectrum of the quiet Sun.
Title: Observational Approach to Computing the Poynting Flux in the
Quiet Sun Photosphere
Authors: Tilipman, Dennis; Martínez Pillet, Valentin; Tremblay,
Benoit; Kazachenko, Maria; Milic, Ivan; Yadav, Rahul
Bibcode: 2022cosp...44.2516T
Altcode:
Understanding magnetically-driven processes in the quiet Sun is
crucial for understanding chromospheric and coronal heating. The main
goal of our study is to quantify the energy output of the quiet Sun
photosphere. The amount of magnetic energy being transported upwards
from the photosphere can be expressed in terms of the vertical component
of Poynting flux, which is a cross-product of magnetic and electric
fields. While magnetic fields and energy fluxes within active regions
and plage have been evaluated before, quiet Sun magnetograms have only
recently become available with the launch of missions such as Hinode
and Sunrise and the Daniel K. Inouye Solar Telescope (DKIST) coming
online early this year. In this presentation, we present estimates
of Poynting flux using IMaX data. As the electric field E is one
of the two principal quantities required to compute Poynting flux,
we use two distinct approaches to infer E. In the first approach, we
derive the electric field using ideal plasma assumption with horizontal
velocities obtained from the convolutional neural network (DeepVel,
Asensio Ramos et al. 2017). In the second approach, we derive E using
the PDFI-SS approach uncurling Faraday's law (Fisher et al. 2020). We
discuss the distribution of Poynting flux and whether it is sufficient
to explain chromospheric and coronal heating.
Title: Evaluating Non-LTE Spectral Inversions with ALMA and IBIS
Authors: Hofmann, Ryan A.; Reardon, Kevin P.; Milic, Ivan; Molnar,
Momchil E.; Chai, Yi; Uitenbroek, Han
Bibcode: 2022ApJ...933..244H
Altcode: 2022arXiv220508760H
We present observations of a solar magnetic network region in the
millimeter continuum with the Atacama Large Millimeter/submillimeter
Array (ALMA) and in the Ca 8542 and Na 5896 Å spectral lines with
the Interferometric Bidimensional Spectrometer (IBIS). Our goal is
to compare the measurement of local gas temperatures provided by ALMA
with the temperature diagnostics provided by non-LTE inversions using
the STockholm inversion Code (STiC). In performing these inversions,
we find that using column mass as the reference height scale, rather
than optical depth, provides more reliable atmospheric profiles above
the temperature minimum and that the treatment of non-LTE hydrogen
ionization brings the inferred chromospheric temperatures into better
agreement with the ALMA measurements. The Band 3 brightness temperatures
are higher but well correlated spatially with the inversion-derived
temperatures at the height of formation of the Ca 8542 line core. The
Band 6 temperatures instead do not show good correlations with the
temperatures at any specific layer in the inverted atmospheres. We then
performed inversions that included the millimeter-continuum intensities
as an additional constraint. Incorporating Band 3 generally resulted in
atmospheres showing a strong temperature rise in the upper atmosphere,
while including Band 6 led to significant regions of anomalously low
temperatures at chromospheric heights. This is consistent with the
idea that the Band 6 emission can come from a combination of heights
ranging from the temperature minimum to upper chromosphere. The
poor constraints on the chromospheric electron density with existing
inversion codes introduces difficulties in determining the height(s)
of formation of the millimeter continuum as well as uncertainties in
the temperatures derived from the spectral lines.
Title: Quantifying Magnetic Energy Flux in the Quiet Sun Photosphere
using Sunrise/IMaX Observations
Authors: Tilipman, Dennis; Kazachenko, Maria; Tremblay, Benoit;
Martinez-Pillet, Valentin; Milic, Ivan
Bibcode: 2021AGUFMSH42B..04T
Altcode:
Understanding magnetically-driven processes in the quiet Sun is
crucial for understanding chromospheric and coronal heating. The main
goal of our study is to quantify the energy output of the quiet Sun
photosphere. The amount of magnetic energy can be expressed in terms of
the Poynting flux, which is a cross-product of magnetic and electric
fields. Poynting flux has been computed for active regions and plage,
but the weakness of polarimetric signal in the quiet Sun presents
a bigger challenge. Quiet Sun magnetic fields have only recently
become observable with the launch of missions such as Hinode and
Sunrise. The Daniel K. Inouye Solar Telescope (DKIST) is expected to
further improve the quality of these observations -- both spatial and
temporal resolutions, as well as polarimetric signal, are expected to
improve significantly, allowing us to perform reliable inversions of
magnetic, electric, and velocity fields, all of which are required
to compute the Poynting flux. We test different inversion methods
on Sunrise/IMaX data in order to streamline this process once DKIST
becomes operational and to understand the limitations of these inversion
techniques. In this work, we present our results obtained from velocity
and electric field inversions of photospheric images, magnetograms and
Doppler velocities from Sunrise/IMaX, the challenges associated with
these inversions, and implications for future DKIST observations. We
also discuss approaches to extend this analysis from photosphere to
overlying layers of the atmosphere, which would allow us to study how
these layers respond to magnetic energy injections from below.
Title: Limitations of the Ca II 8542 Å Line for the Determination
of Magnetic Field Oscillations
Authors: Felipe, Tobias; Socas Navarro, Hector; Sangeetha, C. R.;
Milic, Ivan
Bibcode: 2021ApJ...918...47F
Altcode: 2021arXiv210702160F
Chromospheric umbral oscillations produce periodic brightenings
in the core of some spectral lines, known as umbral flashes. They
are also accompanied by fluctuations in velocity, temperature, and,
according to several recent works, magnetic field. In this study, we
aim to ascertain the accuracy of the magnetic field determined from
inversions of the Ca II 8542 Å line. We have developed numerical
simulations of wave propagation in a sunspot umbra. Synthetic Stokes
profiles emerging from the simulated atmosphere were computed and
then inverted using the NICOLE code. The atmospheres inferred from
the inversions have been compared with the original parameters from
the simulations. Our results show that the inferred chromospheric
fluctuations in velocity and temperature match the known oscillations
from the numerical simulation. In contrast, the vertical magnetic field
obtained from the inversions exhibits an oscillatory pattern with a ~300
G peak-to-peak amplitude, which is absent in the simulation. We have
assessed the error in the inferred parameters by performing numerous
inversions with slightly different configurations of the same Stokes
profiles. We find that when the atmosphere is approximately at rest,
the inversion tends to favor solutions that underestimate the vertical
magnetic field strength. On the contrary, during umbral flashes,
the values inferred from most of the inversions are concentrated at
stronger fields than those from the simulation. Our analysis provides
a quantification of the errors associated with the inversions of the
Ca II 8542 Å line and suggests caution with the interpretation of
the inferred magnetic field fluctuations.
Title: The Lightweaver Framework for Nonlocal Thermal Equilibrium
Radiative Transfer in Python
Authors: Osborne, Christopher M. J.; Milić, Ivan
Bibcode: 2021ApJ...917...14O
Altcode: 2021arXiv210700475O
Tools for computing detailed optically thick spectral line profiles
out of local thermodynamic equilibrium have always been focused on
speed, due to the large computational effort involved. With the
Lightweaver framework, we have produced a more flexible, modular
toolkit for building custom tools in a high-level language, Python,
without sacrificing speed against the current state of the art. The
goal of providing a more flexible method for constructing these complex
simulations is to decrease the barrier to entry and allow more rapid
exploration of the field. In this paper we present an overview of the
theory of optically thick nonlocal thermodynamic equilibrium radiative
transfer, the numerical methods implemented in Lightweaver including
the problems of time-dependent populations and charge-conservation,
as well as an overview of the components most users will interact with,
to demonstrate their flexibility.
Title: Machine learning initialization to accelerate Stokes profile
inversions
Authors: Gafeira, R.; Orozco Suárez, D.; Milić, I.; Quintero Noda,
C.; Ruiz Cobo, B.; Uitenbroek, H.
Bibcode: 2021A&A...651A..31G
Altcode: 2021arXiv210309651G
Context. At present, an exponential growth in scientific data
from current and upcoming solar observatories is expected. Most of
the data consist of high spatial and temporal resolution cubes of
Stokes profiles taken in both local thermodynamic equilibrium (LTE)
and non-LTE spectral lines. The analysis of such solar observations
requires complex inversion codes. Hence, it is necessary to develop
new tools to boost the speed and efficiency of inversions and reduce
computation times and costs.
Aims: In this work we discuss
the application of convolutional neural networks (CNNs) as a tool to
advantageously initialize Stokes profile inversions.
Methods:
To demonstrate the usefulness of CNNs, we concentrate in this paper on
the inversion of LTE Stokes profiles. We use observations taken with
the spectropolarimeter on board the Hinode spacecraft as a test bench
mark. First, we carefully analyse the data with the SIR inversion code
using a given initial atmospheric model. The code provides a set of
atmospheric models that reproduce the observations well. These models
are then used to train a CNN. Afterwards, the same data are again
inverted with SIR but using the trained CNN to provide the initial
guess atmospheric models for SIR.
Results: The CNNs allow us
to significantly reduce the number of inversion cycles when used to
compute initial guess model atmospheres (`assisted inversions'),
therefore decreasing the computational time for LTE inversions by
a factor of two to four. CNNs alone are much faster than assisted
inversions, but the latter are more robust and accurate. CNNs also
help to automatically cluster pixels with similar physical properties,
allowing the association with different solar features on the solar
surface, which is useful when inverting huge datasets where completely
different regimes are present. The advantages and limitations of machine
learning techniques for estimating optimum initial atmospheric models
for spectral line inversions are discussed. Finally, we describe a
python wrapper for the SIR and DeSIRe codes that allows for the easy
setup of parallel inversions. The tool implements the assisted inversion
method described in this paper. The parallel wrapper can also be used
to synthesize Stokes profiles with the RH code.
Conclusions:
The assisted inversions can speed up the inversion process, but the
efficiency and accuracy of the inversion results depend strongly on
the solar scene and the data used for the CNN training. This method
(assisted inversions) will not obviate the need for analysing individual
events with the utmost care but will provide solar scientists with
a much better opportunity to sample large amounts of inverted data,
which will undoubtedly broaden the physical discovery space.
Title: Sparse Representation of HINODE/SOT/SP Spectra Using
Convolutional Neural Networks
Authors: Flint, S.; Milic, I.
Bibcode: 2021AAS...23821304F
Altcode:
A fundamental problem in solar spectropolarimetry is relating observed
spectra and their polarization to the physical parameters of the
underlying atmosphere. One of the difficulties in this process is the
fact that the spectra usually can be represented with a much smaller
number of hyperparameters than what is suggested by the number of
wavelength points used for sampling. Said differently, spectra can
usually be compressed or described in a sparser basis. In this work,
we use the neural networks to investigate the dimensionality of
photospheric spectra, and to compare the compressed spectra with the
maps of physical parameters used to generate the said spectra.
Title: ALMA and IRIS Observations Highlighting the Dynamics and
Structure of Chromospheric Plage
Authors: Hofmann, R. A.; Reardon, K.; Milic, I.
Bibcode: 2021AAS...23820505H
Altcode:
Studies of the thermal structure of the solar chromosphere are typically
hampered by the complexities of non-LTE radiative transfer. This issue
can be addressed using observations of the millimeter continuum, which
directly probes the electron temperatures in the chromosphere. In recent
years, the Atacama Large Millimeter/submillimeter Array (ALMA) has made
it possible, for the first time, to obtain millimeter observations of
sufficient spatial resolution to supplement spectral line observations
and inversions. Here, we present observations of a plage in the 3.0
mm and 1.2 mm continua with ~2 arcsecond resolution, combined with
simultaneous imaging spectroscopy observations from the Interferometric
Bidimensional Spectrometer (IBIS) at the Dunn Solar Telescope. We
compare the observed ALMA brightness temperatures with temperatures
inferred from spectral inversions using the Na D1 5896 Å and Ca II 8542
Å lines, and investigate the wide range of physical heights probed by
the millimeter continuum. We find that the millimeter emission arises
from a range of heights both above and below the chromospheric calcium
line, depending on the local temperature profile and electron densities.
Title: Are the Magnetic Fields Radial in the Solar Polar Region?
Authors: Sun, Xudong; Liu, Yang; Milić, Ivan; Griñón-Marín,
Ana Belén
Bibcode: 2021RNAAS...5..134S
Altcode: 2021arXiv210601461S
We investigate the orientation of the photospheric magnetic fields
in the solar polar region using observations from the Helioseismic
and Magnetic Imager (HMI). Inside small patches of significant
polarization, the inferred magnetic field vectors at 1″ scale appear
to systematically deviate from the radial direction. Most tilt toward
the pole; all are more inclined toward the plane of sky compared to the
radial vector. These results, however, depend on the "filling factor"
f that characterizes the unresolved magnetic structures. The default,
uninformative f ≡ 1 for HMI will incur larger inclination and less
radial fields than f < 1. The observed trend may be a systematic
bias inherent to the limited resolution.
Title: Sparse Representation of HINODE/SOT/SP Spectra Using
Convolutional Neural Networks
Authors: Flint, Serena; Milic, Ivan
Bibcode: 2021csss.confE.189F
Altcode:
A fundamental problem in solar spectropolarimetry is relating observed
spectra and their polarization to the physical parameters of the
underlying atmosphere. One of the difficulties in this process is the
fact that the spectra usually can be represented with a much smaller
number of hyperparameters than what is suggested by the number of
wavelength points used for sampling. Said differently, spectra can
usually be compressed or described in a sparser basis. In this work,
we use the neural networks to investigate the dimensionality of
photospheric spectra, and to compare the compressed spectra with the
maps of physical parameters used to generate the said spectra.
Title: Looking for Changes in Photospheric Temperature Gradients
over Solar Cycle 24 Using Hinode/SP
Authors: Crowley, James; Milic, Ivan
Bibcode: 2021csss.confE.206C
Altcode:
<strong>The intent of this project is to study the effects of the
solar magnetic cycle on the thermal structure of the solar atmosphere
in the quiet Sun. Using data from the spectropolarimeter onboard
the Hinode satellite, four datasets were selected from throughout
Solar Cycle 24; all datasets selected were near the disk center and
without any obvious magnetic signatures. Using an inversion based on
the Milne-Eddington model, a quarter-million pixels were inverted from
each dataset, using two different inversion schemes. By inverting the
data and analyzing the differences in the inverted parameters between
the datasets, we attempt to see if the resolution of the Hinode data
combined with a Milne-Eddington approach is able to detect meaningful
differences in photospheric structure throughout the solar cycle,
primarily the source function and its gradient. Our results so far
suggest that more detailed inversion and /or data preprocessing is
needed to detect eventual presence of the changes. </strong>
Title: ALMA observations and spectral inversions - what can we learn
about the Sun and our techniques?
Authors: Hofmann, R.; Reardon, K.; Milic, I.
Bibcode: 2020AGUFMSH0010002H
Altcode:
Studies of the thermal structure of the solar chromosphere are typically
hampered by the complexities of non-LTE radiative transfer. This issue
can be addressed using observations of the millimeter continuum, which
directly probes the electron temperatures in the chromosphere. In recent
years, the Atacama Large Millimeter/submillimeter Array (ALMA) has made
it possible, for the first time, to obtain millimeter observations of
sufficient spatial resolution to supplement spectral line observations
and inversions. Here, we present observations of a plage in the 3.0
mm and 1.2 mm continua with ~2 arcsecond resolution, combined with
simultaneous imaging spectroscopy observations from the Interferometric
Bidimensional Spectrometer (IBIS) at the Dunn Solar Telescope. We
compare the observed ALMA brightness temperatures with temperatures
inferred from spectral inversions using the Na D1 5896 Å and Ca II 8542
Å lines, and investigate the wide range of physical heights probed by
the millimeter continuum. We find that the millimeter emission arises
from a range of heights both above and below the chromospheric calcium
line, depending on the local temperature profile and electron densities.
Title: Mimicking spectropolarimetric inversions using convolutional
neural networks
Authors: Milić, I.; Gafeira, R.
Bibcode: 2020A&A...644A.129M
Altcode: 2020arXiv200602005M
Context. Interpreting spectropolarimetric observations of the solar
atmosphere takes much longer than the acquiring the data. The most
important reason for this is that the model fitting, or "inversion",
used to infer physical quantities from the observations is extremely
slow, because the underlying models are numerically demanding.
Aims: We aim to improve the speed of the inference by using a neural
network that relates input polarized spectra to the output physical
parameters.
Methods: We first select a subset of the data
to be interpreted and infer physical quantities from corresponding
spectra using a standard minimization-based inversion code. Taking
these results as reliable and representative of the whole data set, we
train a convolutional neural network to connect the input polarized
spectra to the output physical parameters (nodes, in context of
spectropolarimetric inversion). We then apply the neural network to
the various other data, previously unseen to the network. As a check,
we apply the referent inversion code to the unseen data and compare
the fit quality and the maps of the inferred parameters between the
two inversions.
Results: The physical parameters inferred by
the neural network show excellent agreement with the results from
the inversion, and are obtained in a factor of 105 less
time. Additionally, substituting the results of the neural network back
in the forward model, shows excellent agreement between inferred and
original spectra.
Conclusions: The method we present here is
very simple for implementation and extremely fast. It only requires a
training data set, which can be obtained by inverting a representative
subset of the observed data. Applying these (and similar) machine
learning techniques will yield orders of magnitude acceleration in
the routine interpretation of spectropolarimetric data.
Title: Chromospheric Resonances above Sunspots and Potential
Seismological Applications
Authors: Felipe, Tobias; Kuckein, Christoph; González Manrique,
Sergio Javier; Milic, Ivan; Sangeetha, C. R.
Bibcode: 2020ApJ...900L..29F
Altcode: 2020arXiv200810623F
Oscillations in sunspot umbrae exhibit remarkable differences
between the photosphere and chromosphere. We evaluate two competing
scenarios proposed for explaining those observations: a chromospheric
resonant cavity and waves traveling from the photosphere to upper
atmospheric layers. We have employed numerical simulations to
analyze the oscillations in both models. They have been compared with
observations in the low (Na I D2) and high (He I 10830 Å)
chromosphere. The nodes of the resonant cavity can be detected as
phase jumps or power dips, although the identification of the latter
is not sufficient to claim the existence of resonances. In contrast,
phase differences between velocity and temperature fluctuations reveal
standing waves and unequivocally prove the presence of an acoustic
resonator above umbrae. Our findings offer a new seismic method to probe
active region chromospheres through the detection of resonant nodes.
Title: Chromospheric resonant cavities in umbrae: unequivocal
detection and seismic applications
Authors: Felipe, T.; Kuckein, C.; González Manrique, S. J.; Milic,
I.; Sangeetha, C. R.
Bibcode: 2020sea..confE.196F
Altcode:
Umbral chromospheric oscillations exhibit significant differences
compared to their photospheric counterparts. We evaluate two competing
scenarios proposed for explaining those observations: a chromospheric
resonant cavity and waves traveling from the photosphere to upper
atmospheric layers. The oscillatory signatures of both models have been
determined from numerical simulations, and they have been compared to
observations. We find that a high-frequency peak in the He I 10830 Å
power spectra cannot discriminate between both theories, contrary to the
claims of Jess et al. (2019). In contrast, phase differences between
velocity and temperature fluctuations reveal a standing pattern and
unequivocally prove the presence of an acoustic cavity above umbrae. Our
findings offer a new seismic method to probe sunspot chromospheres
through the identification of resonant nodes in phase spectra.
Title: Spectral deconvolution with deep learning: removing the
effects of spectral PSF broadening
Authors: Molnar, Momchil; Reardon, Kevin P.; Osborne, Christopher;
Milić, Ivan
Bibcode: 2020FrASS...7...29M
Altcode: 2020arXiv200505529M
We explore novel methods of recovering the original spectral line
profiles from data obtained by instruments that sample those profiles
with an extended or multipeaked spectral transmission profile. The
techniques are tested on data obtained at high spatial resolution from
the Fast Imaging Solar Spectrograph (FISS) grating spectrograph at the
Big Bear Solar Observatory and from the Interferometric Bidimensional
Spectrometer (IBIS) instrument at the Dunn Solar Telescope. The method
robustly deconvolves wide spectral transmission profiles for fields of
view sampling a variety of solar structures (granulation, plage and
pores) with a photometrical precision of less than 1%. The results
and fidelity of the method are tested on data from IBIS obtained
using several different spectral resolution modes. The method, based
on convolutional neural networks (CNN), is extremely fast, performing
about 10^5 deconvolutions per second on a single CPU for a spectrum with
40 wavelength samples. This approach is applicable for deconvolving
large amounts of data from instruments with wide spectral profiles,
such as the Visible Tunable Filter (VTF) on the DKI Solar Telescope
(DKIST). We also investigate the application to future instruments
by recovering spectral line profiles obtained with a theoretical
multi-peaked spectral transmission profile. We further discuss the
limitations of this deconvolutional approach through the analysis of
the dimensionality of the original and multiplexed data.
Title: Sun-as-a-star observations of the 2017 August 21 solar eclipse
Authors: Dineva, Ekaterina; Denker, Carsten; Verma, Meetu; Strassmeier,
Klaus G.; Ilyin, Ilya; Milic, Ivan
Bibcode: 2020IAUS..354..473D
Altcode:
The Potsdam Echelle Polarimetric and Spectroscopic Instrument (PEPSI)
is a state-of-the-art, thermally stabilized, fiber-fed, high-resolution
spectrograph for the Large Binocular Telescope (LBT) at Mt. Graham,
Arizona. During daytime the instrument is fed with sunlight from the
10-millimeter aperture, fully automated, binocular Solar Disk-Integrated
(SDI) telescope. The observed Sun-as-a-star spectra contain a multitude
of photospheric and chromospheric spectral lines in the wavelength
ranges 4200-4800 Å and 5300-6300 Å. One of the advantages of PEPSI
is that solar spectra are recorded in the exactly same manner as
nighttime targets. Thus, solar and stellar spectra can be directly
compared. PEPSI/SDI recorded 116 Sun-as-a-star spectra during the
2017 August 21 solar eclipse. The observed maximum obscuration was
61.6%. The spectra were taken with a spectral resolution of ≈ 250000
and an exposure time of 0.3 s. The high-spectral resolution facilitates
detecting subtle changes in the spectra while the Moon passes the solar
disk. Sun-as-a-star spectra are affected by changing contributions due
to limb darkening and solar differential rotation, and to a lesser
extend by supergranular velocity pattern and the presence of active
regions on the solar surface. The goal of this study is to investigate
the temporal evolution of the chromospheric Na D doublet during the
eclipse and to compare observations with synthetic line profiles
computed with the state-of-the-art Bifrost code.
Title: Using the infrared iron lines to probe solar subsurface
convection
Authors: Milić, I.; Smitha, H. N.; Lagg, A.
Bibcode: 2019A&A...630A.133M
Altcode: 2019arXiv190407306M
Context. Studying the properties of solar convection using
high-resolution spectropolarimetry began in the early 1990s with
the focus on observations in the visible wavelength regions. Its
extension to the infrared (IR) remains largely unexplored.
Aims: The IR iron lines around 15 600 Å, most commonly known for
their high magnetic sensitivity, also have a non-zero response to
line-of-sight (LOS) velocity below log(τ) = 0.0. In this paper we
explore the possibility of using these lines to measure subsurface
convective velocities.
Methods: By assuming a snapshot of a
three-dimensional magnetohydrodynamic simulation to represent the quiet
Sun, we investigate how well the iron IR lines can reproduce the LOS
velocity in the cube and to what depth. We use the recently developed
spectropolarimetric inversion code SNAPI and discuss the optimal node
placements for the retrieval of reliable results from these spectral
lines.
Results: We find that the IR iron lines can measure the
convective velocities down to log(τ) = 0.5, below the photosphere,
not only at the original resolution of the cube, but also when degraded
with a reasonable spectral and spatial PSF and stray light. Instead, the
commonly used Fe I 6300 Å line pair performs significantly worse.
Conclusions: Our investigation reveals that the IR iron lines can
probe the subsurface convection in the solar photosphere. This paper
is a first step towards exploiting this diagnostic potential.
Title: The specific property of motion of resonant asteroids with
very slow Yarkovsky drift speeds
Authors: Milić; Žitnik, Ivana
Bibcode: 2019MNRAS.486.2435M
Altcode:
This work examines the specific characteristics of the motion of
asteroids with very slow Yarkovsky drift speeds (da/dt) across the
two-body mean-motion resonances (MMRs) with Jupiter, whose strengths
cover a wide range. Only asteroids that crossed a resonance completely
were observed. The investigation was carried out using numerical
integrations performed with the public-domain integrator ORBIT9. It was
found that the test asteroids with very small Yarkovsky drift speeds
moved extremely rapidly across MMRs (order of magnitude 10-5
au Myr-1 or less). This result may indicate that, below
a certain boundary value of da/dt, asteroids typically move quickly
across MMRs. From the obtained results, it is concluded that the
boundary value of the Yarkovsky drift speed is 7 × 10-5
au Myr-1.
Title: Mimicking spectropolarimetric inversion using convolutional
neural networks
Authors: Milic, Ivan; Gafeira, Ricardo
Bibcode: 2019AAS...23422605M
Altcode:
State of the art approach for the interpretation of
spectropolarimetric observations of the solar atmosphere are the so
called spectropolarimetric inversions. These methods fit a model
atmosphere to the observed polarized spectrum and provide us with
the maximum-likelyhood solution for the parameters of the underlying
atmosphere. Inversions are extremely numerically demanding, because
they fully take into account all the physical processes involved in
the spectral line formation. This is especially pronounced in the case
of spectral lines formed in the solar chromosphere. With the advent of
next generation telescopes, such as DKIST, standard, minimization-based,
inversions will simply be too slow. In this contribution we propose
a way to accelerate the inversions by means of convolutional neural
networks. We invert a small sub-set of the data using standard
inversion approach and then train a convolutional neural network to
generalize the results to the full data set. We analyze this method
on different synthetic and observed data sets and compare the results
with the results obtained by applying standard inversion methods. We
find that, given an extensive enough data set, convolutional neural
networks provide results that are very close to the ones obtained by
standard inversion methods, in a fraction of time.
Title: Center-to-Limb Continuum Polarization in Solar and Stellar
Atmospheres
Authors: Kostogryz, N. M.; Berdyugina, S. V.; Yakobchuk, T. M.;
Milić, I.
Bibcode: 2019ASPC..526..139K
Altcode:
The center-to-limb variation of the intensity (CLVI) and of the linear
polarization (CLVP) of stellar radiation arise when the scattering
and absorption processes are important in the stellar atmosphere. We
model the CLVI and CLVP of continuum radiation, taking into account
different contributions of scattering and absorption opacity for
a variety of spectral type stars with plane-parallel and spherical
PHOENIX atmosphere models. We show how the polarization depends on
the effective temperature and surface gravity of a star and how the
considered geometry of the stellar atmosphere affects the polarization
signal. For the Sun, we compare existing measurements with our
theoretical predictions for different solar models (FALA, FALC, FALP,
HSRA, and Phoenix). The CLVI and CLVP of stellar atmospheres are also
needed to interpret the light curves of transiting exoplanets. Here we
present the variation of the polarization in exoplanetary systems caused
by transits and grazing transits and discuss how the considered geometry
of stellar atmosphere models affect the transit curves of exoplanets.
Title: Response Functions for NLTE Lines
Authors: Milić, I.; van Noort, M.
Bibcode: 2019ASPC..526..179M
Altcode:
Response functions quantify the sensitivity of the emergent polarized
spectrum to perturbations in the atmospheric quantities. They are
an important diagnostics tool and an essential ingredient of the
so-called inversion codes, widely used in solar spectropolarimetry. The
computation of response functions for spectral lines formed out of
local thermodynamic equilibrium is complicated because of strong
spatial and non-linear couplings of the atomic populations. We have
recently proposed a novel, analytic approach for the computation of
NLTE response functions, and in this short contribution we discuss
the possibilities of computing response functions for scattering
polarization. We explicitly show the procedure for a two level atom
line (normal Zeeman triplet), and discuss the "formation heights"
of intensity and scattering polarization for a prototype line.
Title: Department of astronomy at Petnica science center: 2013-2017
Authors: Boskovic, M.; Obuljen, A.; Vukadinovic, D.; Milosevic, S.;
Milic, I.; Bozic, N.
Bibcode: 2018POBeo..98..101B
Altcode:
No abstract at ADS
Title: Department of astronomy at Petnica science center: 2013-2017
Authors: Boskovic, M.; Obuljen, A.; Vukadinovic, D.; Milosevic, S.;
Milic, I.; Bozic, N.
Bibcode: 2018POBeo..98...101
Altcode:
No abstract at ADS
Title: Spectropolarimetric NLTE inversion code SNAPI
Authors: Milić, I.; van Noort, M.
Bibcode: 2018A&A...617A..24M
Altcode: 2018arXiv180608134M
Context. Inversion codes are computer programs that fit a model
atmosphere to the observed Stokes spectra, thus retrieving the relevant
atmospheric parameters. The rising interest in the solar chromosphere,
where spectral lines are formed by scattering, requires developing,
testing, and comparing new non-local thermal equilibrium (NLTE)
inversion codes.
Aims: We present a new NLTE inversion code that
is based on the analytical computation of the response functions. We
named the code SNAPI, which is short for spectropolarimetic NLTE
analytically powered inversion.
Methods: SNAPI inverts full
Stokes spectrum in order to obtain a depth-dependent stratification of
the temperature, velocity, and the magnetic field vector. It is based
on the so-called node approach, where atmospheric parameters are free
to vary in several fixed points in the atmosphere, and are assumed to
behave as splines in between. We describe the inversion approach in
general and the specific choices we have made in the implementation.
Results: We test the performance on one academic problem and on two
interesting NLTE examples, the Ca II 8542 and Na I D spectral lines. The
code is found to have excellent convergence properties and outperforms
a finite-difference based code in this specific implementation by at
least a factor of three. We invert synthetic observations of Na lines
from a small part of a simulated solar atmosphere and conclude that
the Na lines reliably retrieve the magnetic field and velocity in the
range -3 < logτ < -0.5.
Title: Sweep-by-sweep implicit Lambda iteration for non-LTE radiative
transfer in 2D Cartesian coordinates
Authors: Milic, I.; Atanackovic, O.
Bibcode: 2017POBeo..96..147M
Altcode:
No abstract at ADS
Title: Polarimetry of transiting planets: Differences between
plane-parallel and spherical host star atmosphere models
Authors: Kostogryz, N. M.; Yakobchuk, T. M.; Berdyugina, S. V.;
Milic, I.
Bibcode: 2017A&A...601A...6K
Altcode:
Context. To properly interpret photometric and polarimetric observations
of exoplanetary transits, accurate calculations of center-to-limb
variations of intensity and linear polarization of the host star are
needed. These variations, in turn, depend on the choice of geometry of
stellar atmosphere.
Aims: We want to understand the dependence
of the flux and the polarization curves during a transit on the choice
of the applied approximation for the stellar atmosphere: spherical
and plane-parallel. We examine whether simpler plane-parallel models
of stellar atmospheres are good enough to interpret the flux and the
polarization light curves during planetary transits, or whether more
complicated spherical models should be used.
Methods: Linear
polarization during a transit appears because a planet eclipses
a stellar disk and thus breaks left-right symmetry. We calculate
the flux and the polarization variations during a transit with given
center-to-limb variations of intensity and polarization.
Results:
We calculate the flux and the polarization variations during transit for
a sample of 405 extrasolar systems. Most of them show higher transit
polarization for the spherical stellar atmosphere. Our calculations
reveal a group of exoplanetary systems that demonstrates lower maximum
polarization during the transits with spherical model atmospheres of
host stars with effective temperatures of Teff = 4400-5400
K and surface gravity of log g = 4.45-4.65 than that obtained with
plane-parallel atmospheres. Moreover, we have found two trends of the
transit polarization. The first trend is a decrease in the polarization
calculated with spherical model atmosphere of host stars with effective
temperatures Teff = 3500-5100 K, and the second shows an
increase in the polarization for host stars with Teff =
5100-7000 K. These trends can be explained by the relative variation
of temperature and pressure dependences in the plane-parallel and
spherical model atmospheres.
Conclusions: For most cases of
known transiting systems the plane-parallel approximation of stellar
model atmospheres may be safely used for calculation of the flux and
the polarization curves because the difference between two models
is tiny. However, there are some examples where the spherical model
atmospheres are necessary to get proper results, such as the systems
with grazing transits, with Earth-size planets, or for the hot host
stars with effective temperatures higher than 6000 K.
Title: Line response functions in nonlocal thermodynamic
equilibrium. Isotropic case
Authors: Milić, I.; van Noort, M.
Bibcode: 2017A&A...601A.100M
Altcode:
Context. Response functions provide us with a quantitative measure
of sensitivity of the emergent spectrum to perturbations in the
solar atmosphere and are thus the method of choice for interpreting
spectropolarimetric observations. For the lines formed in the solar
chromosphere, it is necessary to compute these responses taking into
account nonlocal thermodynamic equilibrium (NLTE) effects.
Aims: We show how to analytically compute the response of the level
populations in NLTE to a change of a given physical quantity at a
given depth in the atmosphere. These responses are then used to compute
opacity and emissivity responses, which are then propagated to obtain
the response of the emergent intensity.
Methods: Our method is
based on the derivative of the rate equations, where we explicitly
incorporate spatial coupling in the radiative rate terms. After
considering and collecting all interdependencies, the problem reduces
to a linear system of equations with a dimension equal to the product
of the number of spatial points and the number of energy levels.
Results: We compare analytically computed response functions with
those obtained using a finite difference approach and find very good
agreement. In addition, a more accurate way of propagating opacity
and emissivity perturbations through the numerical solution of the
radiative transfer equation was developed.
Conclusions: This
method allows for the fast evaluation of the response of the emergent
spectrum to perturbations of a given quantity at a given depth, and
thus is a significant step towards more efficient NLTE inversions.
Title: Inference of magnetic fields in inhomogeneous prominences
Authors: Milić, I.; Faurobert, M.; Atanacković, O.
Bibcode: 2017A&A...597A..31M
Altcode: 2016A&A...597A..31M; 2016arXiv160904954M
Context. Most of the quantitative information about the magnetic
field vector in solar prominences comes from the analysis of the
Hanle effect acting on lines formed by scattering. As these lines can
be of non-negligible optical thickness, it is of interest to study
the line formation process further.
Aims: We investigate the
multidimensional effects on the interpretation of spectropolarimetric
observations, particularly on the inference of the magnetic field
vector. We do this by analyzing the differences between multidimensional
models, which involve fully self-consistent radiative transfer
computations in the presence of spatial inhomogeneities and velocity
fields, and those which rely on simple one-dimensional geometry.
Methods: We study the formation of a prototype line in ad hoc
inhomogeneous, isothermal 2D prominence models. We solve the NLTE
polarized line formation problem in the presence of a large-scale
oriented magnetic field. The resulting polarized line profiles are
then interpreted (I.e. inverted) assuming a simple 1D slab model.
Results: We find that differences between input and the inferred
magnetic field vector are non-negligible. Namely, we almost universally
find that the inferred field is weaker and more horizontal than the
input field.
Conclusions: Spatial inhomogeneities and radiative
transfer have a strong effect on scattering line polarization in the
optically thick lines. In real-life situations, ignoring these effects
could lead to a serious misinterpretation of spectropolarimetric
observations of chromospheric objects such as prominences.
Title: Response functions for NLTE lines
Authors: Milic, Ivan
Bibcode: 2017psio.confE..22M
Altcode:
No abstract at ADS
Title: Center-to-limb variation of intensity and polarization in
continuum spectra of FGK stars for spherical atmospheres
Authors: Kostogryz, N. M.; Milic, I.; Berdyugina, S. V.; Hauschildt,
P. H.
Bibcode: 2016A&A...586A..87K
Altcode: 2015arXiv151107213K
Aims: One of the necessary parameters needed for the
interpretation of the light curves of transiting exoplanets or
eclipsing binary stars (as well as interferometric measurements of a
star or microlensing events) is how the intensity and polarization
of light changes from the center to the limb of a star. Scattering
and absorption processes in the stellar atmosphere affect both the
center-to-limb variation of intensity (CLVI) and polarization (CLVP). In
this paper, we present a study of the CLVI and CLVP in continuum
spectra, taking into consideration the different contributions of
scattering and absorption opacity for a variety of spectral type stars
with spherical atmospheres.
Methods: We solve the radiative
transfer equation for polarized light in the presence of a continuum
scattering, taking into consideration the spherical model of a stellar
atmosphere. To cross-check our results, we developed two independent
codes that are based on Feautrier and short characteristics methods,
respectively,
Results: We calculate the center-to-limb variation
of intensity (CLVI) and polarization (CLVP) in continuum for the
Phoenix grid of spherical stellar model atmospheres for a range of
effective temperatures (4000-7000 K), gravities (log g = 1.0-5.5), and
wavelengths (4000-7000 Å), which are tabulated and available at the
CDS. In addition, we present several tests of our codes and compare our
calculations for the solar atmosphere with published photometric and
polarimetric measurements. We also show that our two codes provide
similar results in all considered cases.
Conclusions: For
sub-giant and dwarf stars (log g = 3.0-4.5), the lower gravity and
lower effective temperature of a star lead to higher limb polarization
of the star. For giant and supergiant stars (log g = 1.0-2.5), the
highest effective temperature yields the largest polarization. By
decreasing the effective temperature of a star down to 4500-5500 K
(depending on log g), the limb polarization decreases and reaches a
local minimum. It increases again with a corresponding decrease in
temperature down to 4000 K. For the most compact dwarf stars (log g =
5.0-5.5), the limb polarization degree shows a maximum for models with
effective temperatures in the range 4200-4600 K (depending on log g) and
decreases toward higher and lower temperatures.