Mapping the Stellar Kinematics in the Central 240 Parsecs of M87 with the James Webb Space Telescope

arXiv:2510.02439 · astro-ph.GA · Submitted 2025-10-02 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Astrophysics Radio. Generated commentary on the latest astrophysics papers.

Vera: I'm Vera.

Jocelyn: And I'm Jocelyn, here with Subrahmanyan.

Vera: We're looking at the paper: "Mapping the Stellar Kinematics in the Central 240 Parsecs of M87 with the James Webb Space Telescope".

Paper summary: Vera: They set out to revisit the central three arcsecond region, which is about two hundred forty by two hundred forty parsecs, using JWST NIRSpec. Jocelyn The paper claims they're implementing modifications to the default reduction pipeline specifically to boost the quality of that data before they get to the science. Vera They manually flagged remaining bad pixels in their two-dimensional calibrated images and then assembled their final data cube by masking regions where all voxels had a good quality flag, a positive drizzle weight, and an intensity that was finite, positive, and non-zero.

Subrahmanyan: That sounds like they are being very careful about the noise they're including in the final map. Jocelyn They determined stellar kinematics using a method called penalized pixel fitting or pPXF to compare velocity template stars from the PHOENIX synthetic library against the observed galaxy spectrum in each spatial bin. Vera They report radial velocity, which is V, and velocity dispersion, sigma, plus higher-order moments like h3 through h8.

Subrahmanyan: Those higher moments tell us about how asymmetric or symmetric the line-of-sight velocity distribution is deviating from a simple Gaussian shape. Jocelyn They use a Monte Carlo procedure to figure out the uncertainties on these stellar kinematics by adding random Gaussian noise to their initial best-fit pPXF model.

Vera: They focused their analysis on a specific wavelength range, specifically targeting the CO bandheads located on the NRS1 detector, and they spectrally masked Ca I absorption that wasn't well matched by their PHOENIX template library. Jocelyn They then tested how robust their results were by comparing the fiducial stellar kinematics against different velocity template libraries, including PHOENIX, K2 OSIRIS, and Winge GNIRS.

Subrahmanyan: Interestingly, they found that the K2 OSIRIS and Winge GNIRS libraries could match the observed Ca I feature, and there was no difference in those higher moments when the feature was fitted versus when it was masked. Vera So they ended up deriving their final kinematics using the PHOENIX library with Ca I masked.

Jocelyn: The resulting maps show rotation at the center of M87, with a velocity V around plus or minus forty-five kilometers per second. Vera And the velocity dispersion sharply rises toward the center, reaching about four hundred twenty kilometers per second at a radius of zero point zero four fifty parsecs.

Subrahmanyan: Plus, they noted a slight elevation in h4 toward the nucleus, and other higher moments fluctuate around zero. Jocelyn They found that these small-scale NIRSpec kinematics transition smoothly into the results from the more recent large-scale kinematics published by KCWI.

Vera: This is significant because it shows a continuous coverage, spanning from central parsecs out to several kiloparsec scales, and it includes a continuation of that rapid rise in velocity dispersion toward the very center of the galaxy. Jocelyn This smooth transition between the small-scale NIRSpec data and the large-scale KCWI data is really neat.

Conclusion: Vera: So we’re wrapping up this look at M87, this paper titled "Mapping the Stellar Kinematics in the Central two hundred forty Parsecs of M87 with the James Webb Space Telescope" <ref:2510.02439#pg1,Mapping the Stellar Kinematics in the Central 240 Parsecs of M87 with>. Jocelyn The authors are Al-Amri, Walsh, Liepold, Ma, and Greene. Vera The core idea is that they provide stellar kinematics in that central two hundred forty by two hundred forty parsec region using JWST data.

Jocelyn: What this really means for us is that the stellar dynamics measurements for the supermassive black hole in M87 have been a bit inconsistent over the last few decades. Subrahmanyan That's because you have these disagreements, like one measurement being about two times larger than another, for example with Harms et al. one thousand nine hundred ninety-four or Gebhardt et al <ref:2510.02439#pg1>. two thousand eleven <ref:2510.02439#pg1>.

Vera: This JWST work gives us a new anchor point to try and resolve those differences by providing detailed measurements of the environment around the supermassive black hole. Jocelyn They suggest that using different template libraries, while paying attention to features like the Ca I line, is important for getting consistent results.

Subrahmanyan: For cosmology and structure formation, having a clearer picture of the dynamics in this extreme environment helps us constrain models of how galaxies grow around massive black holes. Vera This detailed mapping should enable tighter constraints on things like the stellar mass-to-light ratio, dark matter halo shape, and orbital distribution throughout the galaxy.

Jocelyn: The paper also points toward future work, suggesting that examining the polynomial degree of their fitting model could be useful for further analysis. Vera Ultimately, this study shows how these small-scale JWST kinematics connect seamlessly to the wider view we already have from KCWI data across central parsecs to several kiloparsec scales.

Refa M. Al-Amri, Jonelle L. Walsh, Emily R. Liepold, Chung-Pei Ma, Jenny E. Greene

George P. · Cynthia Woods Mitchell Institute for Fundamental Physics and Astronomy · Department of Physics and Astronomy, Texas A&M University · Department of Astronomy, University of California, Berkeley · Department of Physics, University of California, Berkeley · Department of Astrophysical Sciences, Princeton University

astro-ph.GA

Submitted: 2025-10-02

Updated: 2025-10-02

Comments: 16 pages, 8 figures, submitted to the AAS Journals

Journal ref: Refa M. Al-Amri et al 2026 ApJ 999 238

DOI: 0.3847/1538-4357/ae4233

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 91/100

The gist: The paper presents new stellar kinematic measurements of the central region of M87 using JWST NIRSpec IFU data and details modifications to the default reduction pipeline to enhance data quality…

Key concepts

pPXF
Penalized Pixel Fitting is a method used to determine stellar kinematics by comparing observed galaxy spectra against synthetic templates. It finds the best fit for the stellar motions, allowing researchers to measure key properties like rotation and velocity dispersion in different spatial regions of the galaxy.
LOSVD
Line-of-Sight Velocity Distribution refers to how the velocities of stars along a specific line of sight are distributed. The paper analyzes this distribution using a Gauss-Hermite (GH) series, which helps quantify deviations from a simple Gaussian shape, indicating asymmetries in stellar motion.
Ca I absorption masking
The analysis focused on fitting spectra over the CO bandheads while spectrally masking Ca I absorption features. This was done because the PHOENIX template library did not perfectly match these features. Testing showed that results remained consistent even when this feature was masked, confirming the robustness of their kinematic measurements.

Terminology

Summary

The paper presents new stellar kinematic measurements of the central region of M87 using JWST NIRSpec IFU data and details modifications to the default reduction pipeline to enhance data quality, which significantly improves upon prior ground-based studies and provides detailed stellar kinematics in the immediate environment around the SMBH

Data Acquisition and Pipeline Improvements

The researchers observed M87 using JWST NIRSpec IFU with specific grating and filter settings on a particular date as part of Cycle 1 GO-2228 The raw data processing involved several stages where modifications were made to the default pipeline They manually flagged remaining bad pixels in each of the 2D calibrated images before assembling the final cube to reduce artifacts Furthermore, they constructed their final data cube by masking regions where all voxels had a DQ flag of ‘GOOD’, a positive drizzle weight, and an intensity that is finite, positive, and nonzero

Kinematic Measurement Methodology

The stellar kinematics are determined using the penalized pixel fitting method (pPXF) to compare velocity template stars from the PHOENIX synthetic library with the observed galaxy spectrum in each spatial bin The LOSVD is parameterized in terms of a Gauss-Hermite (GH) series, and they report radial velocity V, velocity dispersion σ, and higher-order GH moments (h3 − h8), which quantify the asymmetric and symmetric deviations of the LOSVD from Gaussian They use a Monte Carlo procedure to determine uncertainties on the stellar kinematics by adding random Gaussian noise to the initial best-fit pPXF model The analysis focuses on fitting over a wavelength range focused on the CO bandheads located on the NRS1 detector and spectrally masking Ca I absorption that is not well matched by their PHOENIX template library

Robustness Testing and Template Selection

The robustness of the fiducial stellar kinematics was tested by comparing results across different velocity template libraries, including PHOENIX, K2 OSIRIS, and Winge GNIRS They found that the K2 OSIRIS and Winge GNIRS libraries were able to match the observed Ca I feature, and there is no difference in the inferred GH moments when the feature is fit versus when it is masked The final kinematics for M87 are derived using the PHOENIX library with Ca I masked

Kinematic Results and Comparison

The resulting maps of the first eight GH moments show rotation at the center of M87, with V ∼ ±45 km s−1, a velocity dispersion sharply rises towards the center reaching ∼420 km s−1 at r = 0.′′45, a slight elevation in h4 toward the nucleus, and other higher GH moments fluctuate about zero The NIRSpec stellar kinematics transition seamlessly to the most recently published large-scale kinematics from KCWI, providing continuous coverage from the central parsecs to several kiloparsec scales The results show that the small-scale NIRSpec kinematics nicely transition to the large-scale KCWI kinematics, including a continuation of the rapid rise in the velocity dispersion towards the center of the galaxy

Conclusion and Future Work

The study concludes that using different template libraries with attention paid to the fit of the Ca I line and there be an examination of the polynomial degree is recommended for future work The M87 NIRSpec stellar kinematics transition seamlessly to the most recently published large-scale kinematics from KCWI, providing continuous coverage from the central parsecs to several kiloparsec scales Together, they expect these datasets to enable tight constraints on the M87 MBH, stellar mass-to-light ratio, dark matter halo, galaxy intrinsic shape, and orbital distribution throughout the galaxy and especially near the SMBH Future work will apply the most general, triaxial stellar-dynamical models to both the NIRSpec small-scale and the KCWI widefield stellar kinematics in the future The M87 NIRSpec stellar kinematics transition seamlessly to the most recently published large-scale kinematics from KCWI, providing continuous coverage from the central parsecs to several kiloparsec scales The M87 NIRSpec stellar kinematics transition seamlessly to the most recently published large-scale kinematics from KCWI, providing continuous coverage from the central parsecs to several kiloparsec scales The M87 NIRSpec stellar kinematics transition seamlessly to the most recently published large-scale kinematics from KCWI, providing continuous coverage from the central parsecs to several kiloparsec scales

References

Agazie, G., Anumarlapudi, A., Archibald, A. M., et al. 2023, ApJL, 951, L50

Astropy Collaboration, Robitaille, T. P., Tollerud, E. J., et al. 2013, A&A, 558, A33

Astropy Collaboration, Price-Whelan, A. M., Sip˝ocz, B. M., et al. 2018, AJ, 156, 123

Astropy Collaboration, Price-Whelan, A. M., Lim, P. L., et al. 2022, ApJ, 935, 167

Bennert, V. N., Auger, M. W., Treu, T., Woo, J.-H., & Malan, M. A. 2011, ApJ, 742, 107

Bianchin, M., U, V., Song, Y., et al. 2024, ApJ, 965, 103

Bohn, T., Inami, H., Togi, A., et al. 2024, ApJ, 977, 36

Bushouse, H., Eisenhamer, J., Dencheva, N., et al. 2023, JWST Calibration Pipeline

Cappellari, M. 2023, MNRAS, 526, 3273

Cappellari, M., & Copin, Y. p. 5 2003, MNRAS, 342, 345

Cappellari, M., Emsellem, E., Krajnovi´c, D. p. 6 p. 6 p. 6 p. 6, et al. 2011, MNRAS, 413, 813

de Nicola, S., Thomas, J. p. 1 p. 1, Saglia, R. p. 0 p. 0 P., et al. 2024, MNRAS, 530, 1035

Dressler, A., & Richstone, D. p. 1 O. 1990, ApJ, 348, 120

Dumont, A., Neumayer, N., Seth, A. C., et al. 2025, arXiv e-prints

EHT Collaboration, Akiyama, K., Alberdi, A., et al. 2019, ApJL, 875, L6

Emsellem, E., Krajnovi´c, D., & Sarzi, M. 2014, Monthly Notices of the Royal Astronomical Society: Letters, 445, L79

Emsellem, E., Cappellari, M. p. 5 p. 5 p. 5, Peletier, R. F., et al. 2004, MNRAS, 352, 721

Event Horizon Telescope Collaboration, Akiyama, K., Alberdi, A., et al. 2019, ApJL, 875, L1

Fabian, A. C. 1999, MNRAS, 308, L39

Falc´on-Barroso, J., S´anchez-Bl´azquez, P., Vazdekis, A., et al. 2011, A&A, 532, A95

Gebhardt, K., Adams, J., Richstone, D. p. 1, et al. 2011, ApJ, 729, 119

Gebhardt, K., & Thomas, J. p. 1 2009, ApJ, 700, 1690

Greene, J. E., Peng, C. Y., Kim, M., et al. 2010, ApJ, 721, 26

Harms, R. J., Ford, H. C., Tsvetanov, Z. I., et al. 1994, ApJL, 435, L35

Harris, C. R., Millman, K. J., van der Walt, S. p. 1 J., et al. 2020, Nature, 585, 357

Hunter, J. D. 2007, Computing in Science & Engineering, 9, 90

Husser,, Wende-von Berg, S., Dreizler, S., et al. 2013, A&A, 553, A6

Hutchison, T. A., Welch, B. D., Rigby, J. R., et al. 2024, PASP, 136, 044503

Ishikawa, Y., Zakamska, N. L., Shen, Y., et al. 2025, ApJ, 982, 22

Kormendy, J., & Ho, L. C. 2013, ARA&A, 51, 511

Krajnovi´c, D., McDermid, R. M., Cappellari, M., & Davies, R. L. 2009, MNRAS, 399, 1839

Merrell, K. A., Vasiliev, E. p. 1 p. 1, Bentz, M. C., Valluri, M. p. 1 p. 1 p. 1, & Onken, C. A. 2023, ApJ, 949, 13

Murphy, J. D., Gebhardt, K. p. 1 p. 1 p. 1 p. 1, & Adams, J. J. 2011, The Astrophysical Journal

Neureiter, B., Thomas, J., Saglia, R., et al. 2021, MNRAS, 500, 1437

Nguyen, D. D., Ngo, H. N., Le, T. Q. T., et al. 2025, A&A, 698, L9

Osorno, J., Nagar, N., Richtler, T., et al. 2023, A&A, 679, A37

Pacucci, F., Nguyen, B. p. 6, Carniani, S., Maiolino, R., & Fan, X. 2023, ApJL, 957, L3

Perna, M., Arribas, S., Marshall, M. p. 5, et al. 2023, A&A, 679, A89

Pilawa, J., Liepold, E. p. 0 p. 0 p. 0 R., Ma, C.-P., Walsh, J. p. 0 L., & Greene, J. p. 0 E. 2025, The Astrophysical Journal

Pilawa, J. D., Liepold, E. R., Delgado Andrade, S. C., et al. 2022, ApJ, 928, 178

Quenneville, M. E., Liepold, E. R., & Ma, C.-P. 2021, ApJS, 254, 25

Rauscher, B. J. 2024, PASP, 136, 015001

Reines, A. E., Greene, J. p. 0 E., & Geha, M. 2013, ApJ, 775, 116

Roberts, C. A., Bentz, M. C., Vasiliev, E., Valluri, M., & Onken, C. A. 2021, ApJ, 916, 25

Ruffio, J.-B., Perrin, M. D., Hoch, K. K. W., et al. 2024, AJ, 168, 73

Saglia, R. P., Opitsch, M., Erwin, P., et al. 2016, ApJ, 818, 47

S´anchez-Bl´azquez, P., Peletier, R. F., Jim´enez-Vicente, J., et al. 2006, MNRAS, 371, 703

Sargent, W. L. W., Young, P. p. 1 J., Boksenberg, A., et al. 1978, ApJ, 221, 731

Sarzi, M., Spiniello, C., La Barbera, F., Krajnovi´c, D., & van den Bosch, R. p. 1 2018, Monthly Notices of the Royal Astronomical Society

Seth, A. C., van den Bosch, R. p. 1 p. 1 p. 1 p. 1, Mieske, S., et al. 2014, Nature, 513, 398

Shajib, A. J. 2025, arXiv e-prints

Shajib, A. J., Treu, T., Melo, A., et al. 2025, arXiv e-prints

Shankar, F., Weinberg, D. H., & Miralda-Escud´e, J. 2009, ApJ, 690, 20

Silk, J., & Rees, M. p. 0 J. 1998, A&A, 331, L1

Simon, D. A., Cappellari, M., & Hartke, J. 2024, MNRAS, 527, 2341

Tahmasebzadeh, B., Zhu, L., Shen, J., Gerhard, O., & van de Ven, G. p. 1 2022, ApJ, 941, 109

Tahmasebzadeh, B., Taylor, M. p. 6 A., Valluri, M., et al. 2025, The Astrophysical Journal Letters

Taylor, M. A., Tahmasebzadeh, B. p. 1, Thompson, S., et al. 2025, The Astrophysical Journal Letters

Thater, S., Lyubenova, M., Fahrion, K., et al. 2023, A&A, 675, A18

Thater, S., Krajnovi´c, D., Weilbacher, P. M., et al. 2022, MNRAS, 509, 5416

van der Marel, R. P. 1994, MNRAS, 270, 271

Vasiliev, E., & Valluri, M. p. 1 2020, ApJ, 889, 39

Virtanen, P., Gommers, R., Oliphant, T. E., et al. 2020, Nature Methods

Walsh, J. L., Barth, A. J., Ho, L. C., & Sarzi, M. 2013, ApJ, 770, 86

Walsh, J. L., van den Bosch, R. C. E., Barth, A. J., & Sarzi, M. 2012, ApJ, 753, 79

Walsh, J. L., van den Bosch, R. C. E., Gebhardt, K., et al. 2015, ApJ, 808, 183

Walsh, J. L., van den Bosch, R. C. E., Gebhardt, K., et al. 2017, ApJ, 835, 208

Winge, C., Riffel, R. A., & Storchi-Bergmann, T. 2009, The Astrophysical Journal Supplement Series

Wylezalek, D., Vayner, A., Rupke, D. S. N., et al. 2022, ApJL, 940, L7

Young, P. J., Westphal, J. A., Kristian, J., Wilson, C. P., & Landauer, F. P. 1978, ApJ, 221, 721

Zhang, Y., Ouchi, M., Gebhardt, K., et al.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improve stellar kinematic modeling by implementing a multiplicative Legendre polynomial of degree 3 when fitting over a wavelength range covering four CO bandheads before the detector gap to prevent biased even moments.

  2. Enhance template library selection by systematically comparing results across PHOENIX, K2 OSIRIS, and Winge GNIRS libraries to ensure robust kinematics, particularly regarding the masking of Ca I.

  3. Develop a data reduction pipeline modification that involves manually flagging remaining bad pixels in each 2D calibrated detector image before building the combined cube to minimize artifacts near the NRS1/NRS2 detector gap.

  4. Implement a spatial binning strategy using Voronoi binning repeated until all spectra meet a high signal-to-noise threshold, resulting in 2810 bins with 2232 composed of single spaxels for reliable kinematic measurement.

  5. Train AI systems to recognize and correct systematic offsets in even higher odd GH moments by calculating the median value over all spatial bins to determine the systemic velocity and attribute template mismatch offsets.

  6. Create a system that automatically selects the optimal multiplicative polynomial degree based on the number of CO bandheads before the detector gap, as varying this degree prevents an artificial gradient in velocity dispersion maps.

  7. Develop a method for comparing small-scale NIRSpec kinematics with large-scale KCWI kinematics to ensure seamless transitions and validate the continuity of velocity dispersion trends across different instruments and wavelength regimes.

Sources

Related papers