Source--Line Associations in JWST/NIRSpec Dense-Shutter Spectroscopy

@misc{wu2026sourceline,
  title = {{Source--Line Associations in JWST/NIRSpec Dense-Shutter Spectroscopy}},
  author = {Zihao Wu and Daniel J. Eisenstein and Francesco D'Eugenio and Erica Nelson and Benjamin D. Johnson and Andrew J. Bunker and Zhiyuan Ji and Pierluigi Rinaldi and Brant Robertson and Sandro Tacchella and Christopher N. A. Willmer and Joris Witstok},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.35423},
  eprint = {2609.35423},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.35423},
}

TY  - UNPB
TI  - Source--Line Associations in JWST/NIRSpec Dense-Shutter Spectroscopy
AU  - Zihao Wu
AU  - Daniel J. Eisenstein
AU  - Francesco D'Eugenio
AU  - Erica Nelson
AU  - Benjamin D. Johnson
AU  - Andrew J. Bunker
AU  - Zhiyuan Ji
AU  - Pierluigi Rinaldi
AU  - Brant Robertson
AU  - Sandro Tacchella
AU  - Christopher N. A. Willmer
AU  - Joris Witstok
PY  - 2026
DA  - 2026/09/28
PB  - arXiv
DO  - 10.48550/arxiv.2609.35423
UR  - https://arxiv.org/abs/2609.35423
AB  - We demonstrate the feasibility of source--line association in JWST/NIRSpec dense-shutter spectroscopy, with ∼2000 sources observed simultaneously per pointing regardless of spectral overlap. While JWST wide-field slitless spectroscopy (WFSS) can disentangle overlapping spectra using two grisms with perpendicular dispersion directions, NIRSpec can achieve this using gratings with different dispersions. Compared with WFSS, NIRSpec dense-shutter spectroscopy offers substantially higher sensitivity and broader wavelength coverage. We develop a Bayesian framework that combines cross-dispersion offsets, wavelength consistency, and changes in relative line positions between gratings to jointly determine source associations and redshifts. We test the method using mock observations based on the observed source distribution and emission-line fluxes from the JWST Advanced Deep Extragalactic Survey (JADES), accounting for the NIRSpec optical model, MSA operability, measurement noise, false-positive detections, and bad pixels. Although the spectral traces overlap extensively, fewer than 0.5% of the simulated emission lines are blended with another line on the detectors in NIRSpec grating spectroscopy. Applying our method to these simulations, we obtain redshifts, at an accuracy of 98%, for the ∼1800 sources that have detectable emission lines, of which ∼1500 are classified as confident and are 99.9% accurate. We further investigate how the association accuracy depends on the number of grating configurations. By combining high multiplexing, high sensitivity, and broad wavelength coverage, NIRSpec dense-shutter spectroscopy enables efficient, highly complete spectroscopic surveys for the high-redshift Universe.
ER  -