Simula+ng Wide-­‐Field Slitless Spectroscopy of High-­‐Redshi= Galaxies Lensed by MACS J0647+7015 with JWST/NIRISS W. V. Dixon (STScI), C. J. Willou (NRC-­‐Herzberg, Canada), R. G. Abraham (Univ. of Toronto, Canada), M. Sawicki (Saint Mary's University, Canada), J. B. Hutchings (NRC-­‐Herzberg, Canada), A. W. Fullerton (STScI) The Near Infrared Imager and Slitless Spectrograph (NIRISS) aboard the James Webb Space Telescope (JWST) will offer wide-­‐field slitless spectroscopy (WFSS) with a resolving power R = 150 over the wavelength range 0.8 to 2.25 μm. In this band, NIRISS will be sensi+ve to Lyman α emission and con+nuum breaks from galaxies with redshi=s 6 < z < 17. To explore its ability to observe such high-­‐redshi= galaxies, we have modeled a NIRISS observa+on of the massive galaxy cluster MACS J0647+7015 and analyzed the synthe+c images using standard so=ware tools. In this poster, we present preliminary results of our analysis. NIRISS is provided to the JWST project by the Canadian Space Agency under the leadership of René Doyon of the Université de Montréal. The prime contractor is COM DEV Canada. For each galaxy in the catalog, we used the aXe so=ware package (Kümmel et al. 2009) to extract and calibrate its spectra from each combina+on of filter and grism. While analysis of these spectra is ongoing, we present a few examples here. Direct
Figure 2. Using Source Extractor (Ber+n & Arnouts 1996), we iden+fied 7200 galaxies in the F200W image, including 165 of the 180 high-­‐z galaxies. The 15 missing galaxies (le= panel) fell on bright foreground objects. 15 of the 165 detected galaxies are shown in the right panel. We combined these galaxies with 835 foreground galaxies to produce a catalog of 1000 objects. GR150R
GR150C
F090W
5
0
BEAM_681A GR150R
Redshift: 9.25
−5
: −4.0
2
: 1.5
−10
1.0
1.5
Wavelength (µm)
2.0
Figure 4. Extracted spectrum of galaxy #681. Black points are the data (from all six filters), the red curve represents the error array, and the blue curve is the best-­‐fit con+nuum model. The model does not include a Lyman α emission feature, which may explain the rather steep value of the power-­‐law slope. 10
5
0
BEAM_6334A GR150R
Redshift: 8.25
−5
: −3.7
2
: 1.5
−10
F115W
1.0
F140M
1.5
Wavelength (µm)
2.0
Figure 5. Extracted spectrum of galaxy #6334, mag(F200W) = 26.6. Black, red, and blue points represent the data, error, and best-­‐fit con+nuum, respec+vely. F150W
F158M
10
Flux (10−19 erg cm−2 s−1 Å−1)
F200W
10
Ly α 8
6
4
5
0
BEAM_7155A GR150R
Redshift: 11.00
−5
: −3.1
10
2
2
: 1.8
−10
0
0.8
1.0
1.2
1.4
1.6
Wavelength (µm)
1.8
2.0
2.2
Figure 3. Having two grisms allows NIRISS to disentangle many spectral overlaps. This figure presents all of the data for galaxy #681, for which mag(F200W) = 26.4 and z = 9.3. (Top) 2-­‐D postage stamps from the direct and dispersed images through all six filters. (BoKom) flux-­‐ and wavelength-­‐calibrated spectra from the GR150R (black) and GR150C (red) grisms (wide filters only). Note the galaxy’s disappearance at short wavelengths and the contamina+on by a foreground galaxy in the GR150C spectra. Flux units are 10-­‐19 erg cm-­‐2 s-­‐1 Å-­‐1. Lyman α emission features are circled. Flux (10−19 erg cm−2 s−1 Å−1)
Flux (10−19 erg cm−2 s−1 Å−1)
Figure 1. Simulated dispersed image through the F200W filter. Exposure +me is 10 hours. No bad pixels, cosmic rays, background, flat-­‐field irregulari+es, or edge effects are included. Dispersed images were constructed for each of the six NIRISS filters and for both the GR150R (disperses onto rows on the detector) and GR150C (disperses onto columns) grisms. Galaxy con+nua are derived from CLASH redshi=s and photometry. Emission lines were added based on correla+ons between the line equivalent width and specific SFR, mass, redshi=, and reddening. High-­‐redshi= galaxies have power-­‐law rest-­‐frame UV con+nua. 70% have Lyman α emission lines with an exponen+al equivalent-­‐width distribu+on, while 30% have no Lyman α. Flux (10−19 erg cm−2 s−1 Å−1)
10
Flux (10−19 erg cm−2 s−1 Å−1)
Background image: Synthe+c image of the galaxy cluster MACS J0647+7015 through the NIRISS F200W filter. Pixel scale is half that of NIRISS (0.0328”) to simulate sub-­‐pixel dithering. The HST/ACS PSF (FWHM = 0.065”) is used for all filters. Exposure +me is one hour per filter. Photometry and redshi=s are from the CLASH catalog (Postman et al. 2012). Direct images were constructed for each of the six NIRISS filters. To the CLASH catalog were added 180 high-­‐redshi= galaxies with a uniform, random distribu+on of redshi= (6 < z < 15) and magnitude (26.4 < F200W < 28.4) and a Gaussian distribu+on of shapes. (FWHM distribu+on has μ = 3 and σ = 1 pixel in each dimension, but with FWHM > 2.2 pixels.) 1.0
5
1.5
Wavelength (µm)
2.0
0
BEAM_7155A GR150C
Redshift: 11.00
−5
: −5.7
2
: 1.9
−10
1.0
1.5
Wavelength (µm)
2.0
Figure 6. GR150R (horizontal) and GR150C (ver+cal) spectra of galaxy #7155, for which mag(F200W) = 26.4 and z = 11.0. Black, red, and blue points represent the data, error, and best-­‐fit con+nuum, respec+vely. References: Ber+n, E. & Arnouts, S. 1996, A&AS, 317, 393 Ÿ Kümmel, M., Walsh, J. R., Pirzkal, N., Kuntschner, H., & Pasquali, A.J. 2009, PASP, 121, 59 Ÿ Postman, M. et al. 2012, ApJS, 199, 25