JWST ETC Source Spectral Energy Distributions
The JWST Exposure Time Calculator (ETC) offers a series of spectral energy distributions (SEDs) that may be used when building a source. Templates include many flavors of stellar and extragalactic spectra, as well as novae, nebulae, planetary, and analytic spectral distributions. A user-supplied spectrum may also be uploaded to the ETC for use in calculations.
When creating a new source in the Exposure Time Calculator (ETC), the user may choose to apply a template spectrum for the source's continuum. Many of the commonly used stellar and extragalactic templates have been provided, which cover a wide range of observed spectral energy distributions, along with spectral templates of other astronomical objects and several analytic functions. The Continuum tab is also where the redshift and extinction parameters may be entered. The available choices for source spectral energy distributions are described below. There are two ways to access the Continuum tab of the Source Editor, as shown in Figures 1 & 2 below.
The remainder of this article presents pertinent information on each of the spectral models found in the JWST ETC, including caveats and limitations. Additional information on each of the spectra, including references, can be found on the Astronomical Catalogs webpage maintained by STScI, and linked in the Additional information on spectral template files, below.
Analytic spectral energy distributions include a Flat Continuum, Power-law Continuum, and a Blackbody Spectrum.
The power-law continuum is defined as F(λ)~λn, where n is specified by the user in the exp box. Units of flam (wavelength units of erg/s/cm2/Å) or fnu (frequency units of erg/s/cm2/Hz) can be chosen from a drop-down menu. The Power-law Continuum is converted to fnu units (if necessary) and normalized to 1 mJy at 1 μm before being re-normalized to the value input by the user under the Renorm tab.
A flat continuum is a special case of the power law spectrum, where n=0. This distribution is so-named because the spectrum has constant energy per unit wavelength (flam) or unit frequency (fnu). The source spectrum plot will show a straight line profile for a Flat Continuum in fnu units but a curved profile when flam is selected due to the conversion factor between fnu and flam.
The Blackbody Spectrum is computed at the temperature of the blackbody specified by the user using the equation:
The units are ignored and the blackbody profile is normalized such that the value under the curve is the flux density (in mJy) at each wavelength of the input spectrum.
Phoenix Stellar Models
The pull-down menu of Phoenix stellar models contains synthetic spectra spanning spectral types from O3 to M5, obtained using the Star, Brown Dwarf, and Planet Simulator. They use static, spherically symmetric, 1D simulations to completely describe the atmospheric emission spectrum. The models account for the formation of molecular bands, such as those of water vapor, methane, or titanium dioxide, solving for the transfer equation over more than 20,000 wavelength points on average, resulting in synthetic spectra with 2 Angstrom resolution. The line selection is repeated for each iteration of the model until it has converged and the thermal structure obtained. The models here are calculated with a cloud model, valid across the entire parameter range. Each model's name contains a concatenation of the spectral type, effective temperature, and gravity (log(g) value).
HST Standard Stars
Several HST calibration standard star spectra are available. These spectra are stored in the Calibration Database System (CDBS) and were originally chosen from the paper Spectrophotometric Standards from the Far-UV to the Near-IR on the White Dwarf Flux Scale by Bohlin (1996) and later updated as new data became available. See also Comparison of White Dwarf Models with ACS Spectrophotometry by Bohlin et al. (2001).
More information, along with a list of the complete set of files, including older versions, can be found in the CALSPEC Calibration Database. This page provides a table with the available Flux Standards and their CDBS name. In this table the order of preference for the choice of a standard flux distribution is from left to right in the table, i.e., from the best in column 6 to the last choice with the lowest quality in column 9. In this case, models have higher fidelity and extend to longer wavelength ranges while the more outdated are those derived applying corrections to the original IUE and optical fluxes. Note that for the cases when the CALSPEC data are updated after the ETC software is released, the ETC will not be able to access the most recent files, but only those that were available at the time of the build. If the ETC produces an error when trying to access an HST Standard Star spectrum, review the update history at the bottom of the CALSPEC page to determine when the spectrum was updated. If it was updated after the current ETC version, you may want to use the previous version of the model, or download the most recent spectrum, and apply it as a user-supplied spectrum.
Each stellar model's spectral type and related identifying information is shown in the pull-down menu listing the star by name. The related information includes the spectral type and V magnitude in parentheses, and wavelength range in Angstroms in brackets.
Note that HST CALSPEC spectra do not extend to wavelengths shorter than 0.55 μm and therefore cannot be normalized in the very blue bandpasses (e.g., Johnson V and HST WFC3 F336W).
The planetary nebula spectrum models are all calculated using the CLOUDY photoionization code version 17.02 (Ferland et al., 2017). Each model assumes a constant density shell around the star and that the nebula is radiation bounded. Four models have gas abundances specific to the mean LMC abundances, because these are lower in metallicity than the average for the galaxy and so match the normal galactic planetary nebula abundance better than the default solar or ISM abundances in CLOUDY. Each of these models also include a silicate grain component with an assumed dust radius of 0.01 microns. These models span the temperature range from 30000 K to 100000 K. As the temperature increases the inner radius of the shell increases and the gas density decreases. The stellar spectra use the Rauch models for temperatures of 50000 K and higher (Rauch, 2003) and a Castelli/Kurucz model (Castelli & Kurucz, 2004) for the 30000 K case.
The final model is specifically tailored to match the properties of the specific nebula SMP LMC 58 (SIMBAD link). This nebula is carbon-rich and the model uses the measured abundances of the object. The density profile has a core-halo structure to match the radio and Hβ observations of the object. In this model there is carbon-based dust, radius 0.25 μm amorphous carbon grains adjusted to approximately match the overall spectral energy distribution of the object. While the overall amount of dust emission and the peak wavelength are similar to the observed values from the Spitzer IRS spectroscopy of this nebula, there are differences in detailed infrared spectral shape. As well, the discrete infrared emission features observed in the Spitzer spectrum were not modeled here.
For each model the line intensities and the continuum shape calculated by the CLOUDY code was used to simulate a high resolution spectrum assuming a Gaussian line shape having a velocity sigma of 10 km/s. For the hydrogen line series the CLOUDY line list was extended up to 200 lines in each series (Balmer, Paschen, Pfund, and so on to the N=12 low level series) using the expected relative line strengths for the higher series lines, since several of the line series have limits within the JWST wavelength range.
The selection of stellar novae spectral models include two varieties of Type II SNe (fast & slow decline), two Type Ia SNe, a Type Iax, a Type Ib, and a kilonova. All templates are labeled in the dropdown menu with the time since the event, with the majority at T+0 days and the kilonova at T+0.05 days. The full list of references, and other information on model parameters, used to create these models can be found in the STScI Astronomical Catalogs, specifically the catalog entry for Novae Spectra.
Be aware that the wavelength ranges for these models are extremely limited, with the expectation that they will be redshifted to the appropriate wavelength range for JWST observations. The kilonova has the broadest range and the longest red cut-off wavelength, covering 0.11-3 μm. The wavelength ranges are labeled in brackets in the dropdown menu for each template.
Galaxy Spectra from Brown et al. (2014)
Extragalactic template spectral energy distributions include models from Brown et al. (2014). Wavelength coverage for these models spans UV to mid-infrared wavelengths. The atlas includes a broad range of galaxy types, including ellipticals, spirals, merging galaxies, blue compact dwarfs and luminous infrared galaxies.
Galaxy Spectra from Brown et al. (2019)
Extragalactic template spectral energy distributions include models from Brown et al. (2019). Wavelength coverage for these models spans UV to mid-infrared wavelengths. The galaxy spectra from this work include QSOs and Seyfert galaxies.
This is a composite model made from a large number of quasar observations. This model was created using optical (visible) spectra of 2200 SDSS quasars from Vanden Berk et al. (2001) and 27 quasar spectra observed with the IRTF in the near-infrared (NIR) by Glikman et al. (2006).
Note that his template spectrum only covers the wavelength range between 0.086 and 3.52 μm and not the full JWST wavelength range.
Simple Stellar Populations
These are models of stellar groupings parametrized by the log(age) of the grouping and the metallicity (z/H). This set of template spectra is unique in the JWST ETC because it provides two dropdowns after selection of "Simple Stellar Populations", instead of just one (as seen in Figure 3). The first dropdown specifies the log(age) and the second dropdown specifies the metallicity. The list of available options in each dropdown is always the same, no matter which option is selected first, with the combination of parameters uniquely specifying the template spectrum. The list of references used to create these models, and a full list of all the available spectra, can be found in the STScI Astronomical Catalogs, specifically in the Simple Stellar Population Atlas.
The so-called "Normal Galaxies" are spectral models of standard elliptical and spiral galaxies, with more "exotic" galaxy types (QSOs, AGNs, LIRGs, Seyferts, etc.) covered in other sets. The available models are all from the SWIRE template library and cover elliptical galaxies at 2, 5, and 13 Gyr, as well as the following spiral galaxy types: S0, Sa, Sb, Sc, Sd, and Sdm. There are two options for spiral galaxy type Sc (#1 and #2), with #2 at a significantly reduced spectral resolution. (This is referred to as "Spi4" on the TRDS SWIRE Galaxies page.) For the JWST ETC, the wavelength ranges of these spectra were reduced to 0.2-32 μm.
Brown Dwarfs and Exoplanets spectra
Low-Temperature Phoenix Models
These template spectra were made using the ATMO 2020 code of Phillips et al. (2020) and represent the atmosphere and evolutionary models for cool T-Y brown dwarfs and giant exoplanets in radiative-convective equilibrium. The models available in the JWST ETC are limited to those with log(g) = 5.0. These models were generated using solar metallicity and with a temperature range between 200 K to 2000 K, in step sizes of 50 K between 200-600 K and 100 K between 600-2000 K. Wavelength coverage is from ~0.2-35 μm. Estimated spectral types are not provided in the dropdown menu labels, as they are for the Phoenix Stellar Models.
Solar System spectra
Sun and Giant Planets
Currently available solar system spectral models include Jupiter, Saturn, Uranus, Neptune, and the Sun. All spectral templates are a combination of observations (NIR) and thermal models (MIR) stitched together in the 4-8 μm region in such a way as to avoid discontinuities. The giant planet spectra cover ~0.55-28.75 μm and the solar model covers 0.2-30 μm. The full list of references used to create these models can be found in the STScI Astronomical Catalogs, specifically the catalog entry for Solar System Objects.
Note that the giant planet spectral templates have units of surface brightness (erg s-1 cm-2 Å-1 arcsec-2) whereas the solar spectrum has units of flux density (erg s-1 cm-2 Å-1). Accurate application of the giant planet spectral templates therefore requires an extended source to be specified.
Additional information on spectral template files
Links to additional information for all template spectra used in the JWST ETC can be found on the STScI Astronomical Catalogs webpage. The spectra themselves can be downloaded as binary FITS tables from the included links.
Go to the online JWST Exposure Time Calculator Tool
Bohlin et al. 1996, AJ, 111, 1743
Spectrophotometric Standards From the Far-UV to the Near-IR on the White Dwarf Flux Scale
Bohlin et al. 2001, AJ, 122, 2118
Spectrophotometric Standards from the Far-Ultraviolet to the Near-Infrared: STIS and NICMOS Fluxes
Brown et al. 2014, ApJS, 212,18
An Atlas of Galaxy Spectral Energy Distributions from the Ultraviolet to the Mid-infrared
Brown et al. 2019, MNRAS, 489, 3
The spectral energy distributions of active galactic nuclei
Castelli & Kurucz 2004, arXiv:astro-ph/0405087
New Grids of ATLAS9 Model Atmospheres
Ferland et al. 2017, RMxAA, 53, 385
The 2017 Release Cloudy
Glikman et al. 2006, ApJ, 640, 2
A near-infrared spectral template for quasars
Phillips et al. 2020, A&A, 637, id.A38
A new set of atmosphere and evolution models for cool T-Y brown dwarfs and giant exoplanets
Pontoppidan, K. M., Pickering, T. E., Laidler, V. G. et al., 2016, Proc. SPIE 9910, Observatory Operations: Strategies, Processes, and Systems VI, 991016
Pandeia: a multi-mission exposure time calculator for JWST and WFIRST
Rauch, T. 2003, A7A, 403, 709
A grid of synthetic ionizing spectra for very hot compact stars from NLTE model atmospheres
Vanden Berk et al. 2001, AJ, 122, 2
Composite Quasar Spectra from the Sloan Digital Sky Survey