Wide-Field Slitless Spectroscopy with Webb’s NIRISS

Van Dixon, dixon@stsci.edu, & Chris Willott, Chris.Willott@nrc-cnrc.gc.ca

The James Webb Space Telescope’s Near Infrared Imager and Slitless Spectrograph (NIRISS) will offer several innovative observing modes, one of which is wide-field slitless spectroscopy (WFSS). This mode provides a mean resolving power (λ/Δλ) of 150 over the wavelength range 0.8–2.25 microns. The wavelength coverage has been optimized for the detection of Lyα emission lines and breaks in the spectra of galaxies at redshifts 6 to 17, to probe the first stars and ionizing sources in the early universe.

In WFSS, a grism placed in the optical path renders each object in the field as a spectrum. The powerful multiplex capability of WFSS has recently been illustrated with the WFC3 instrument on Hubble. In crowded fields grism spectra often overlap, complicating their analysis. To alleviate this problem, NIRISS employs two identical grisms oriented with orthogonal dispersion directions: GR150R disperses spectra into rows on the detector, while GR150C disperses into columns.

To explore its ability to observe high-redshift galaxies—and our ability to identify them—we modeled a NIRISS observation of the massive lensing galaxy cluster MACS J0647+7015. Using published images, photometry, and redshifts from the CLASH survey (Postman et al. 2012), we constructed a series of simulated direct and dispersed images in the six filters available for WFSS with NIRISS. Figure 1 is a color composite of the direct images through the F090W, F150W, and F200W filters. Figure 2 presents dispersed images through the F200W filter. Direct images assume a total integration time of one hour per filter; dispersed images, ten hours. The NIRISS field of view is 2.2 arcminutes on a side.










We added 180 simulated high-redshift (6 < z < 15) galaxies distributed uniformly in space, redshift, and magnitude to each image. We used Source Extractor  (Bertin & Arnouts 1996) to identify all objects in the F200W direct image and perform photometry on each. We used the aXe software package (Kümmel et al. 2009) to extract and calibrate each spectrum and to estimate the contamination due to overlapping spectra. Figure 3 presents postage-stamp images of a z = 9.3 galaxy (marked with a crosshair in Figure 1) through each filter. The galaxy is relatively bright (mag = 26.4) through the F200W filter, but undetectable in F090W. While a nearby galaxy obliterates the GR150C spectra, the GR150R spectra are unscathed, demonstrating the utility of our two-grism design.

NIRISS-3NIRISS-4For each galaxy and each grism, we combine the data obtained through all six filters into a flux- and wavelength-calibrated spectrum, then fit the spectrum with a simple high-redshift model consisting of a power-law continuum and a Lyα emission line. The model flux drops to zero at wavelengths shorter than Lyα. We fit the spectra with a variety of other templates (ellipticals, spirals, starbursts, quasars) to identify lower-redshift objects, as well. The GR150R spectrum of the z = 9.3 galaxy is shown in Figure 4. The black points represent the data, the red curve represents the error array (computed by aXe), and the blue curve is our best-fit model.

Work is now underway to analyze a large sample of simulated galaxies and determine how best to mitigate the effects of spectral overlap for deep observations in crowded fields. Redshifts of simulated galaxies much fainter than that in Figure 4 can be determined via spectral breaks and/or emission lines. Targeting high-redshift galaxies magnified by foreground lensing clusters, such as those in the Hubble Frontier Fields, will enable a more detailed study of the intrinsically low-luminosity galaxies within the epoch of reionization. Our simulations demonstrate that WFSS with NIRISS will provide a powerful tool for the exploration of the high-redshift universe.

NIRISS is provided to the Webb project by the Canadian Space Agency. René Doyon (Université de Montréal) is the principal investigator, and COM DEV Canada is the prime contractor. More information about NIRISS, including tools to simulate and analyze WFSS images, is available on the NIRISS website at the Institute.


Bertin, E. & Arnouts, S. 1996, A&AS, 317, 393
Kümmel, M., Walsh, J. R., Pirzkal, N., Kuntschner, H., & Pasquali, A. J. 2009, PASP, 121, 59
Postman, M., et al. 2012, ApJS, 199, 25