skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: Deep Learning–based Fast Spectral Inversion of Hα and Ca ii 8542 Line Spectra
Abstract A multilayer spectral inversion (MLSI) model has recently been proposed for inferring the physical parameters of plasmas in the solar chromosphere from strong absorption lines taken by the Fast Imaging Solar Spectrograph (FISS). We apply a deep neural network (DNN) technique in order to produce the MLSI outputs with reduced computational costs. We train the model using two absorption lines, H α and Ca ii 8542 Å, taken by FISS, and 13 physical parameters obtained from the application of MLSI to 49 raster scans (∼2,000,000 spectra). We use a fully connected network with skip connections and multi-branch architecture to avoid the problem of vanishing gradients and to improve the model’s performance. Our test shows that the DNN successfully reproduces the physical parameters for each line with high accuracy and a computing time of about 0.3–0.4 ms per line, which is about 250 times faster than the direct application of MLSI. We also confirm that the DNN reliably reproduces the temporal variations of the physical parameters generated by the MLSI inversion. By taking advantage of the high performance of the DNN, we plan to provide physical parameter maps for all the FISS observations, in order to understand the chromospheric plasma conditions in various solar features.  more » « less
Award ID(s):
1821294
PAR ID:
10408428
Author(s) / Creator(s):
; ; ; ; ;
Date Published:
Journal Name:
The Astrophysical Journal
Volume:
940
Issue:
2
ISSN:
0004-637X
Page Range / eLocation ID:
147
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; (, Astronomy & Astrophysics)
    null (Ed.)
    The solar chromosphere can be observed well through strong absorption lines. We infer the physical parameters of chromospheric plasmas from these lines using a multilayer spectral inversion. This is a new technique of spectral inversion. We assume that the atmosphere consists of a finite number of layers. In each layer the absorption profile is constant and the source function varies with optical depth with a constant gradient. Specifically, we consider a three-layer model of radiative transfer where the lowest layer is identified with the photosphere and the two upper layers are identified with the chromosphere. The absorption profile in the photosphere is described by a Voigt function, and the profile in the chromosphere by a Gaussian function. This three-layer model is fully specified by 13 parameters. Four parameters can be fixed to prescribed values, and one parameter can be determined from the analysis of a satellite photospheric line. The remaining 8 parameters are determined from a constrained least-squares fitting. We applied the multilayer spectral inversion to the spectral data of the H α and the Ca  II 854.21 nm lines taken in a quiet region by the Fast Imaging Solar Spectrograph (FISS) of the Goode Solar Telescope (GST). We find that our model successfully fits most of the observed profiles and produces regular maps of the model parameters. The combination of the inferred Doppler widths of the two lines yields reasonable estimates of temperature and nonthermal speed in the chromosphere. We conclude that our multilayer inversion is useful to infer chromospheric plasma parameters on the Sun. 
    more » « less
  2. ; ; ; ; ; ; ; ; (, The Astrophysical Journal)
    Abstract Small-scale jets, such as chromospheric and transition region (TR) network jets, are of great interest regarding coronal heating and solar wind acceleration. Spectroscopic analysis based on multiple spectral lines with different formation temperatures is essential for understanding the physical properties and driving mechanisms of jets. Here, we conduct an investigation of the physical properties of a small-scale chromospheric jet in a quiet-Sun network region and its TR counterpart. This jet is recorded from formation to extinction using the Fast Imaging Solar Spectrograph at the Goode Solar Telescope and the Interface Region Imaging Spectrograph. The chromospheric component of the jet exhibits a high line-of-sight speed of up to 45 km s−1during its ascending phase, accompanied by spectral profiles akin to rapid blueshifted excursion and downflowing rapid redshifted excursion during the descending phase. Using a cloud model combined with a Multi-Layer Spectral Inversion, we quantify the jet’s temperature during its ascending phase, which starts at approximately 11,000 K and increases by only 1000 K over 1 minute, much smaller than a few 104K, the excess temperature expected in an ideal gas reconnection jet at an outflow speed of 45 km s−1. The TR counterpart exhibits a Siiv1394 Å line profile with a non-Gaussian shape, including a blueshifted component and a large nonthermal width. Our results suggest that if the jet is driven by magnetic reconnection in the chromosphere, the heat released by the reconnection may be mostly used to ionize the hydrogen rather than to increase the temperature so that the gas may appear almost isothermal. 
    more » « less
  3. ; ; ; ; ; ; ; ; ; ; et al (, Astronomy & Astrophysics)
    Aims. Our aim is to investigate the role of acoustic and magneto-acoustic waves in heating the solar chromosphere. Observations in strong chromospheric lines are analyzed by comparing the deposited acoustic-energy flux with the total integrated radiative losses. Methods. Quiet-Sun and weak-plage regions were observed in the Ca  II 854.2 nm and H α lines with the Fast Imaging Solar Spectrograph (FISS) at the 1.6-m Goode Solar Telescope on 2019 October 3 and in the H α and H β lines with the echelle spectrograph attached to the Vacuum Tower Telescope on 2018 December 11 and 2019 June 6. The deposited acoustic energy flux at frequencies up to 20 mHz was derived from Doppler velocities observed in line centers and wings. Radiative losses were computed by means of a set of scaled non-local thermodynamic equilibrium 1D hydrostatic semi-empirical models obtained by fitting synthetic to observed line profiles. Results. In the middle chromosphere ( h = 1000–1400 km), the radiative losses can be fully balanced by the deposited acoustic energy flux in a quiet-Sun region. In the upper chromosphere ( h  >  1400 km), the deposited acoustic flux is small compared to the radiative losses in quiet as well as in plage regions. The crucial parameter determining the amount of deposited acoustic flux is the gas density at a given height. Conclusions. The acoustic energy flux is efficiently deposited in the middle chromosphere, where the density of gas is sufficiently high. About 90% of the available acoustic energy flux in the quiet-Sun region is deposited in these layers, and thus it is a major contributor to the radiative losses of the middle chromosphere. In the upper chromosphere, the deposited acoustic flux is too low, so that other heating mechanisms have to act to balance the radiative cooling. 
    more » « less
  4. ; ; ; ; ; (, The Astrophysical Journal)
    Abstract Submillimeter emission lines produced by the interstellar medium (ISM) are strong tracers of star formation and are some of the main targets of line intensity mapping (LIM) surveys. In this work we present an empirical multiline emission model that simultaneously covers the mean, scatter, and correlations of [C ii ], CO J = 1–0 to J = 5–4, and [C i ] lines in the redshift range 1 ≤ z ≤ 9. We assume that the galaxy ISM line emission luminosity versus halo mass relations can be described by double power laws with redshift-dependent lognormal scatter. The model parameters are then derived by fitting to the state-of-the-art semianalytic simulation results that have successfully reproduced multiple submillimeter line observations at 0 ≤ z ≲ 6. We cross-check the line emission statistics predicted by the semianalytic simulation and our empirical model, finding that at z ≥ 1 our model reproduces the simulated line intensities with fractional error less than about 10%. The fractional difference is less than 25% for the power spectra. Grounded on physically motivated and self-consistent galaxy simulations, this computationally efficient model will be helpful in forecasting ISM emission-line statistics for upcoming LIM surveys. 
    more » « less
  5. ; ; ; ; ; ; ; (, Journal of Hydrometeorology)
    Abstract Raindrop size distributions (DSD) and rain rate have been estimated from polarimetric radar data using different approaches with the accuracy depending on the errors both in the radar measurements and the estimation methods. Herein, a deep neural network (DNN) technique was utilized to improve the estimation of the DSD and rain rate by mitigating these errors. The performance of this approach was evaluated using measurements from a two-dimensional video disdrometer (2DVD) at the Kessler Atmospheric and Ecological Field Station in Oklahoma as ground truth with the results compared against conventional estimation methods for the period 2006–17. Physical parameters (mass-/volume-weighted diameter and liquid water content), rain rate, and polarimetric radar variables (including radar reflectivity and differential reflectivity) were obtained from the DSD data. Three methods—physics-based inversion, empirical formula, and DNN—were applied to two different temporal domains (instantaneous and rain-event average) with three diverse error assumptions (fitting, measurement, and model errors). The DSD retrievals and rain estimates from 18 cases were evaluated by calculating the bias and root-mean-squared error (RMSE). DNN produced the best performance for most cases, with up to a 5% reduction in RMSE when model errors existed. DSD and rain estimated from a nearby polarimetric radar using the empirical and DNN methods were well correlated with the disdrometer observations; the rain-rate estimate bias of the DNN was significantly reduced (3.3% in DNN vs 50.1% in empirical). These results suggest that DNN has advantages over the physics-based and empirical methods in retrieving rain microphysics from radar observations. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email