MOMOS: The Multi-Object MKID Optical Spectrometer Simulator and Data Reduction Package numpy ( van der Walt et al. 2011 ) , scipy ( Virtanen et al. 2020 ) , astropy ( Astropy Collaboration et al. 2013 ) , synphot ( STScI Development Team 2018 ) , lmfit ( Newville et al. 2016 ) , PyReduce ( Piskunov
Crystal S. Kim, John I. Bailey, III, Ronald A. López, W. Hawkins Clay, Benjamin A. Mazin · 2025
Read MOMOS: The Multi-Object MKID Optical Spectrometer Simulator and Data Reduction Package numpy ( van der Walt et al. 2011 ) , scipy ( Virtanen et al. 2020 ) , astropy ( Astropy Collaboration et al. 2013 ) , synphot ( STScI Development Team 2018 ) , lmfit ( Newville et al. 2016 ) , PyReduce ( Piskunov by Crystal S. Kim, John I. Bailey, III, Ronald A. López, W. Hawkins Clay, Benjamin A. Mazin (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-00663.txt.
Introduction
Crystal Kim and co-authors present the software that will plan and later unpack data from a proposed starlight instrument, in a 2025 paper. The instrument, MOMOS, splits the light of many objects at once using a special detector that senses each photon's energy. Those detectors have no read-out noise, so faint objects can be studied cleanly and spectra reshaped after the fact. The paper covers the twin software package.
One half simulates the instrument under different designs, so builders can test choices before cutting metal. The other half is the reduction pipeline: it takes simulated output and practices the calibration and spectrum extraction that real data will need. Writing both before the hardware exists is deliberate. When the spectrometer is built, its data can flow through tested code from the first night.
Catalog record
| CREATOR | Crystal S. Kim, John I. Bailey, III, Ronald A. López, W. Hawkins Clay, Benjamin A. Mazin |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-00663
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
MOMOS: The Multi-Object MKID Optical Spectrometer Simulator and Data Reduction Package numpy ( van der Walt et al. 2011 ) , scipy ( Virtanen et al. 2020 ) , astropy ( Astropy Collaboration et al. 2013 ) , synphot ( STScI Development Team 2018 ) , lmfit ( Newville et al. 2016 ) , PyReduce ( Piskunov et al. 2021 ) Crystal S. Kim, John I. Bailey, III, Ronald A. López, W. Hawkins Clay, Benjamin A. Mazin Abstract MOMOS, the Multi-Object MKID Optical Spectrometer, is a proposed visible wavelength spectrometer that uses MKIDs (Microwave Kinetic Inductance Detectors) targeting an initial resolving power of 3500. With their modest wavelength-resolving abilities, MKIDs take the place of both the cross disperser and detector in the spectrometer. MKIDs lack read noise and dark current enabling noiseless post-observation rebinning and characterization of faint objects, as well as time-resolved photon-counting spectroscopy. This work presents an MOMOS simulator customizable for different MOMOS configurations. Treating simulator products as inputs, an algorithm was developed and implemented in the MOMOS data reduction package to calibrate and extract spectra. I Introduction Most astronomical spectrometers use an echelle grating followed by a cross-disperser and detector to analyze luminous sources for emission and absorption features, usually across multiple spectral orders to yield a wide wavelength coverage. They can be found at almost every major telescope facility ( Rayner et al. 1998 ; Baudrand & Vitry 2000 ; Dressler et al. 2003 ; Tokoku et al. 2003 ; Crane et al. 2005 ; Hill et al. 2006 ; Edelstein et al. 2007 ; Flaugher & Bebek 2014 ; Crause et al. 2016 ) and are a mainstay instrument in astronomy. An advantage of the echelle spectrometer is the ability to use rectangular detectors already widely used in other astrophysical applications, most commonly charge-coupled devices (CCDs). CCDs can scale to Gigapixel arrays with small, energy-efficient, and sensitive pixels. Proposals to use energy-resolving detectors in echelle spectrometers date back over 20 years ( Cropper et al. 2003 ) . The spectrometer described in this paper is based off of a similar conceptual instrument called KIDSpec (Kinetic Inductance Detector Spectrograph) from O’Brien et al. 2014 that uses an emerging detector technology known as Microwave Kinetic Inductance Detectors (MKIDs). These photon counting, superconducting detectors resolve both the energy and arrival time for each photon ( Day et al. 2003 ) , lack both read noise and dark current, and are largely unaffected by cosmic ray contamination after post-processing. Several low-resolution MKID integral field spectrographs (IFS) exist today ( Mazin et al. 2013 ; Meeker et al. 2018 ; Walter et al. 2020 ; Swimmer 2023 ) . These IFS’s return spectral information solely using the innate wavelength discrimination of MKIDs. With \mathcal{R}_{\textrm{MKID}}\equiv\lambda/d\lambda\lesssim 12 at 600 nm, these instruments largely behave as broadband integrated photometers with high temporal resolution. Although UVOIR MKID device fabrication remains a significant challenge, \mathcal{R}_{\textrm{MKID}}\approx 46 at 600 nm has been measured ( de Visser et al. 2021 ) with a maximum theoretical \mathcal{R}_{\textrm{MKID}}\sim 140 at 600 nm for PtSi ( Zobrist et al. 2022 ) . All further \mathcal{R}_{\textrm{MKID}} are referenced with respect to 600 nm. These higher \mathcal{R}_{\textrm{MKID}} values more than suffice to discriminate spectral orders when used in moderate to high-resolution ( \mathcal{R}_{\textrm{spec}}\sim 4000 to 100,000 ) echelle(tte) spectrometers and would eliminate the need for a cross-dispersing optical element. With \mathcal{R}_{\textrm{MKID}}=15 , an MKID can discriminate orders 5 through 9 in the 400 to 800 nm for an \mathcal{R}_{\textrm{spec}}\sim 3500 . Because light is dispersed in only one direction (no cross-disperser), only a single row of MKIDs is required for the read out of a spectrum. By strategically locating sources and employing multiple linear MKID pixel arrays, a highly pixel-efficient, compact, multi-object spectrometer capable of simultaneously obtaining all spectral orders becomes possible. This work introduces a simulator and data reduction package for MOMOS, an MKID spectrometer testbed for the exploration of multi-object echelle designs employing MKID detectors. Such instruments offer significant potential to astronomy as they capture the entire echellogram for every fiber without either the read noise or dark current of traditional CCD or CMOS arrays. This fundamentally alters the design-space by enabling digital re-binning to match resolving power to source brightness and science case while simultaneously eliminating the wavelength coverage/target-count trade of existing multi-object spectrometers. Moreover, the temporal resolution of MKIDs can enable fundamentally new spectroscopic analysis approaches for stellar astrophysics. As a testbed, MOMOS is expected to undergo grating and resolution upgrades. Its initial incarnation is as a medium-resolution MKID spectrometer operating from 400-800 nm with up to five fiber feeds, an echellette grating in Littrow configuration, an MKID device with five 2048-pixel linear MKID arrays with 20x200 micron pixels, and ability to discriminate orders 4-7 in a commercially-available off-blaze grating configuration. Further changes will further expand MOMOS’s scientific value and inform the design of future MKID-based spectrometers. The MOMOS simulator has been instrumental in determining how future MOMOS upgrades might impact data reduction capabilities. It also produced the realistic input data used to test the data reduction package. Likewise, real MOMOS output will be processed through the data reduction described in this paper. A simulation of KIDSpec, the instrument concept which MOMOS was based on, was recently developed to understand improvements KIDSpec could bring to low-SNR spectroscopy ( Hofmann & O’Brien 2023 ) . There are several key differences between this work and the KIDSpec simulation, known as KSIM. The MOMOS simulator needs to mimic realistic MKID data as much as possible. This means that all wavelengths are converted to an appropriate phase response and then multiplied with random offsets on a pixel to pixel basis, which makes wavelength or order information impossible to recover without further processing. Conversely, KSIM retains order and wavelength information throughout the simulation and reduction. Secondly, since real instruments are not always perfectly aligned like their optical models, it cannot be assumed which portion of the spectrum (i.e., which photon energies) are incident on which pixels. That is why the MOMOS simulator involves a wavelength calibration step that is not present in KSIM. Secondly, one of the concerns mentioned by Hofmann & O’Brien 2023 was the degree of order misidentification due to the overlap of order Gaussians, which limited KSIM to \mathcal{R}_{\textrm{MKID}}\approx 22 . Because of the order-bleeding (overlap) correction described here in Section III.2 , the MOMOS data reduction is suitable down to \mathcal{R}_{\textrm{MKID}}=15 . These major differences highlight the intended purpose of either simulator; KSIM characterizes achievable science goals as a high fidelity SNR calculator over a range of conditions and the MOMOS simulator provides realistic data to rigorously test the data reduction before it is used with real MOMOS data. The MOMOS simulator takes an input spectrum and yields photon events as time-tagged MKID resonator phase shifts, which the data reduction package processes into tabulated photon data and standard astrophysical spectral orders. The simulator is described in Section II and the MOMOS data reduction in Section III . II MOMOS Simulator The MOMOS simulator takes a model input spectrum and applies the following effects: telluric attenuation, addition of sky emission lines, multiplication with grating blaze function, convolution with the optical Line Spread Function, convolution with the MKID response function, conversion