Pith. Sign in

REVIEW

Optimal Pulsar Timing Array Strategies with the Deep Synoptic Array

PaperPDF from arXiv

data files

ReviewThe objections and verdict, as JSON Paper breakdownThe authors' claims, conditions and measurements, as JSON Signed receiptProof this review has not been altered
X Bluesky LinkedIn Reddit Hacker News
Tyler Cohen, Paul B. Demorest, Jeremy G. Baier, Ryan S. Lynch, Scott M. Ransom, Michael T. Lam, Gabriella Agazie, Akash Anumarlapudi
show 45 more authors
Anne M. Archibald, Zaven Arzoumanian, Paul T. Baker, Ava L. Battaglia, Paul R. Brook, Anabelle L. Brueggemann, H. Thankful Cromartie, Kathryn Crowter, Megan E. DeCesar, Timothy Dolch, Elizabeth C. Ferrara, William Fiore, Emmanuel Fonseca, Gabriel E. Freedman, Nate Garver-Daniels, Peter A. Gentile, Joseph Glaser, Deborah C. Good, Jeffrey S. Hazboun, Ross J. Jennings, Megan L. Jones, David L. Kaplan, Matthew Kerr, Duncan R. Lorimer, Jing Luo, Hugo E. Marquis, Alexander McEwen, Maura A. McLaughlin, Natasha McMann, Bradley W. Meyers, Cherry Ng, David J. Nice, Timothy T. Pennucci, Benetge B. P. Perera, Nihan S. Pol, Henri A. Radovan, Paul S. Ray, Ann Schmiedekamp, Carl Schmiedekamp, Brent J. Shapiro-Albert, Ingrid H. Stairs, Kevin Stovall, Abhimanyu Susobhanan, Joseph K. Swiggum, Haley M. Wahl

High Energy Astrophysical Phenomena · Submitted 30 Sep 2026 · 89 references

Pith's verdict: Genuinely useful, honestly argued PTA planning paper: the framework is real, the 5% result is well-supported against scatter around the assumed red-noise model, and it deserves peer review.

T0 · AI review

How this review was made: An AI referee (deepseek/deepseek-v4-flash) read the paper on October 2, 2026 and answered a fixed list of questions about it. T0 is the starting level every paper gets. The levels describe how deeply Pith has checked the paper, never the authors.

  1. T0 AI review An AI read the paper and answered a fixed list of questions.
  2. T1 Audit Automatic checks that the references and files the paper relies on exist and match.
  3. T2 Compute The paper's code or calculations re-run with the same results.
  4. T3 Formal statement The claims written out in a formal math language a computer can read.
  5. T4 Proof Those claims proved, and the proof checked by computer.

How Pith checks papers →

· 2 major objections · 4 minor · Ask about this paper · The compiled record · Read the full review →
Challenge this review Comment (0) Read on arXiv ↗
Cite ▾
arXiv ID2610.00836 (version 1)
Pith IDMJVWT5LX, Pith's permanent record of this paper and its review
CitationAbhimanyu Susobhanan, Akash Anumarlapudi, Alexander McEwen, Anabelle L. Brueggemann, Anne M. Archibald, Ann Schmiedekamp, Ava L. Battaglia, Benetge B. P. Perera, Bradley W. Meyers, Brent J. Shapiro-Albert, Carl Schmiedekamp, Cherry Ng, David J. Nice, David L. Kaplan, Deborah C. Good, Duncan R. Lorimer, Elizabeth C. Ferrara, Emmanuel Fonseca, Gabriel E. Freedman, Gabriella Agazie, Haley M. Wahl, Henri A. Radovan, H. Thankful Cromartie, Hugo E. Marquis, Ingrid H. Stairs, Jeffrey S. Hazboun, Jeremy G. Baier, Jing Luo, Joseph Glaser, Joseph K. Swiggum, Kathryn Crowter, Kevin Stovall, Matthew Kerr, Maura A. McLaughlin, Megan E. DeCesar, Megan L. Jones, Michael T. Lam, Natasha McMann, Nate Garver-Daniels, Nihan S. Pol, Paul B. Demorest, Paul R. Brook, Paul S. Ray, Paul T. Baker, Peter A. Gentile, Ross J. Jennings, Ryan S. Lynch, Scott M. Ransom, Timothy Dolch, Timothy T. Pennucci, Tyler Cohen, William Fiore, Zaven Arzoumanian. "Optimal Pulsar Timing Array Strategies with the Deep Synoptic Array." arXiv:2610.00836 (2026). https://pith.science/paper/MJVWT5LX
Cite this reviewBibTeX for Pith's review of this paper
Claim paper

Why Pith says this

Highlight any text to comment on it.

Referee report

How Pith checks papers → · Full audit of this paper →

The paper

The paper claims

For the Deep Synoptic Array, fine-tuning per-pulsar observing time buys only about 5% in gravitational-wave background sensitivity — equal-time scheduling is nearly as good.

The reading

This paper argues that how the monthly telescope time is divided among pulsars barely matters for detecting the nanohertz gravitational-wave background with the Deep Synoptic Array, a dish array under construction in Nevada. The authors build a simulation framework that predicts, for each of 85 millisecond pulsars, the optimal instrument, observing frequency, and per-pulsar integration time. Their central result is that moving from an equal-time-per-pulsar allocation to a time-optimized allocation improves the background signal-to-noise from $\rho(t_{\rm eq}) \approx 14.28$ to $\rho(t_{\rm int}^*) \approx 14.85$, about 5%, and this gain persists across 100 Monte Carlo realizations of pulsar noise. A secondary claim is that a 15-year timing program begun fresh with DSA is considerably more sensitive than one using current instruments, roughly $\rho \approx 14.4$ versus $6.85$. If correct, the community can adopt simple equal-time DSA scheduling without sacrificing much sensitivity, and DSA becomes the clearly preferred instrument.

What carries the argument

The argument runs on two pieces of machinery. The per-pulsar figure of merit is the infinite-frequency TOA uncertainty $\sigma_{\rm TOA}$, the quadrature sum of a DM-fit standard error that carries the achromatic template-fitting, jitter, and diffractive-scintillation covariances, plus scattering-delay, DM-difference, and polarization-calibration errors, with receiver temperature and gain made frequency-dependent. The PTA-level figure of merit is the optimal cross-correlation GWB signal-to-noise $\rho^2$, written in a pseudo-quadratic form $F = \tilde{p}^T \tilde{Q} \tilde{p}$, where $\tilde{Q}$ is a pre-computable block-diagonal matrix of Hellings-Downs cross-pulsar weights per frequency bin and $\tilde{p}$ collects the discretized inverse pulsar power-spectral densities; the integration-time vector is then searched with a trust-region quasi-Newton method under a fixed time budget and per-pulsar bounds, with stationarity checked against the Karush-Kuhn-Tucker conditions. What carries the result is the geometry of this objective: the GWB and intrinsic red noise dominate the lowest frequencies, where integration time is irrelevant, so the optimum is shallow and the ~5% gain is robust to the Monte Carlo draws.

Extended reading notes

Core claim

The paper's central claim is that the gravitational-wave background signal-to-noise ratio of a DSA-based pulsar timing array is nearly flat with respect to how the fixed monthly observing budget is distributed across pulsars, so there is little to gain from per-pulsar integration-time optimization. On the authors' model, the equal-time allocation gives $\rho(t_{\rm eq}) = 14.28$ and the time-optimized allocation gives $\rho(t_{\rm int}^*) = 14.85$; across 100 Monte Carlo realizations in which fluxes, spectral indices, jitter, scattering, and red noise are drawn from distributions, the two values are tightly correlated with slope 1.0 and a constant offset of 0.9. The reason is the intermediate-signal regime: at the lowest gravitational-wave frequencies the noise is set by the background itself and by intrinsic pulsar red noise, neither of which responds to integration time, while at high frequencies most of these pulsars are not jitter-dominated. The paper's secondary claim is that a 15-year timing program begun fresh with DSA is considerably more sensitive than one using current instruments, roughly $\rho \approx 14.4$ versus $6.85$.

Load-bearing premise

The load-bearing premise is the intrinsic red-noise model applied to all 85 pulsars — a fixed spectral index $\gamma_{\rm red} = 5.6$ and amplitudes drawn from a population relation that may itself be contaminated by the gravitational-wave background — because red noise sets which pulsars keep observing time and which are pushed to the 60-second floor.

Editorial extensions

If this is right

  • Equal-time DSA scheduling, which is far simpler to operate, loses at most ~5% of the GWB signal-to-noise relative to a fully time-optimized allocation.
  • A 15-year PTA begun fresh with DSA roughly doubles GWB sensitivity compared with current instruments, $\rho \approx 14.4$ versus $6.85$.
  • Most of the 85 pulsars are not jitter-dominated at DSA sensitivities, so subarraying or multi-pulsar beam sharing brings little benefit.
  • Pulsars pushed to the 60-second floor, J1903+0327 and J0621+2514, contribute essentially no GWB sensitivity, and the brightest red-noise pulsar B1937+21 is timed for only 93 s.
  • The relative ~5% gain from time optimization is insensitive to the assumed noise models: Monte Carlo realizations keep the signal-to-noise at a constant ~0.9 offset.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the true intrinsic red-noise population is stronger or weaker than the assumed relation, the identity of the pulsars kept in the optimized array would change even though the paper shows the ~5% gain itself is stable to Monte Carlo draws.
  • The same pseudo-quadratic machinery could be repurposed, as the authors note, to optimize continuous-wave sensitivity or GWB spectral-parameter estimation, where integration time affects the science in a less flat way.
  • A concrete early test: during DSA commissioning, compare measured per-pulsar red-noise amplitudes and jitter fractions against the assumed values; deviations would directly predict whether equal-time scheduling stays near-optimal for the real array.
  • Because the high-frequency white-noise floor is set by a few best pulsars, an equal-time allocation slightly raises that floor; the authors find the effect vanishes when the top five pulsars are held fixed, so equal-time scheduling is a trade against high-frequency sensitivity rather than a free lunch.
Original abstract

Pulsar timing array (PTA) experiments have seen evidence for a nanohertz-frequency gravitational wave background (GWB) through coordinated radio timing observations of millisecond pulsars. The significance of this evidence is expected to grow and the GWB's progenitor(s) uncovered with continued observations, especially as the next generation of radio telescopes comes online. We present a framework for simulating timing observations of a set of pulsars with a set of telescopes to predict the per-pulsar optimal instrument, observing frequency, and integration time. We apply this methodology to the NANOGrav source list of 85 pulsars observed with the Green Bank Telescope, Very Large Array, Canadian Hydrogen Intensity Mapping Experiment (CHIME), and the Deep Synoptic Array (DSA), a radio dish array under construction in Nevada. We discuss the dominant noise contributions to the timing precision of these pulsars and show that most are not dominated by intrinsic pulse phase jitter noise. We also determine which pulsars should be timed with DSA and CHIME or only DSA and show that an ab initio DSA PTA is more sensitive than one which uses current instruments. Finally, we solve for the optimal integration time per pulsar, subject to a fixed time budget, that maximizes the GWB signal-to-noise. We show that there is little benefit to optimizing over integration time for these pulsars observed with DSA. Through Monte Carlo realizations of the PTA, we show that the GWB sensitivity improves by $\sim 5\%$ for time-optimized observations compared to the equal-time-per-source scenario with an early-science, fixed source list.

The compiled record

The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-10-02: 1 request, 7,134 input tokens.

What the abstract itself states, as probabilities: claims first 0.04 · states limitation 0.66 · human participants 0.02 · controlled comparison 0.82 · randomized assignment 0.02 · states quantitative result 0.98 · compares with prior results 0.90 · mentions code or data availability 0.03.

The abstract, sentence by sentence

arXiv listing abstract v1. Results and the main claim are underlined; .

Pulsar timing array (PTA) experiments have seen evidence for a nanohertz-frequency gravitational wave background (GWB) through coordinated radio timing observations of millisecond pulsars. The significance of this evidence is expected to grow and the GWB's progenitor(s) uncovered with continued observations, especially as the next generation of radio telescopes comes online. We present a framework for simulating timing observations of a set of pulsars with a set of telescopes to predict the per-pulsar optimal instrument, observing frequency, and integration time. We apply this methodology to the NANOGrav source list of 85 pulsars observed with the Green Bank Telescope, Very Large Array, Canadian Hydrogen Intensity Mapping Experiment (CHIME), and the Deep Synoptic Array (DSA), a radio dish array under construction in Nevada. We discuss the dominant noise contributions to the timing precision of these pulsars and show that most are not dominated by intrinsic pulse phase jitter noise. We also determine which pulsars should be timed with DSA and CHIME or only DSA and show that an ab initio DSA PTA is more sensitive than one which uses current instruments. Finally, we solve for the optimal integration time per pulsar, subject to a fixed time budget, that maximizes the GWB signal-to-noise. We show that there is little benefit to optimizing over integration time for these pulsars observed with DSA. Through Monte Carlo realizations of the PTA, we show that the GWB sensitivity improves by $\sim 5\%$ for time-optimized observations compared to the equal-time-per-source scenario with an early-science, fixed source list.

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#sentencerolethe authors' own workstates the main conclusionstrength
s0 Pulsar timing array (PTA) experiments have seen evidence for a nanohertz-frequency gravitational wave background (GWB) through coordinated radio timing observations of millisecond pulsars. background 1.00 0.07 0.02 asserted 0.92
s1 The significance of this evidence is expected to grow and the GWB's progenitor(s) uncovered with continued observations, especially as the next generation of radio telescopes comes online. background 0.98 0.06 0.03 hedged possibility 0.84
s2 We present a framework for simulating timing observations of a set of pulsars with a set of telescopes to predict the per-pulsar optimal instrument, observing frequency, and integration time. method 0.55 0.94 0.06 asserted 0.99
s3 We apply this methodology to the NANOGrav source list of 85 pulsars observed with the Green Bank Telescope, Very Large Array, Canadian Hydrogen Intensity Mapping Experiment (CHIME), and the Deep Synoptic Array (DSA), a radio dish array under construction in Nevada. method 0.72 0.91 0.04 asserted 0.97
s4 We discuss the dominant noise contributions to the timing precision of these pulsars and show that most are not dominated by intrinsic pulse phase jitter noise. result 0.95 0.92 0.05 asserted 1.00
s5 We also determine which pulsars should be timed with DSA and CHIME or only DSA and show that an ab initio DSA PTA is more sensitive than one which uses current instruments. result 0.78 0.96 0.23 asserted 1.00
s6 Finally, we solve for the optimal integration time per pulsar, subject to a fixed time budget, that maximizes the GWB signal-to-noise. method 0.59 0.96 0.11 asserted 0.94
s7 We show that there is little benefit to optimizing over integration time for these pulsars observed with DSA. result 0.98 0.96 0.22 asserted 1.00
s8 Through Monte Carlo realizations of the PTA, we show that the GWB sensitivity improves by $\sim 5\%$ for time-optimized observations compared to the equal-time-per-source scenario with an early-science, fixed source list. result 1.00 0.96 0.28 asserted 1.00

What kind of paper the classifier reports this is: method or tool 0.58 · evidence from simulation or computation 1.00.

What was examined, and what was not

sources read
  • arXiv listing abstract v1
not examined
  • full text
  • figures and tables
  • references and citing sentences
  • relations to other papers
levels run
  • P abstract
not run
  • S sections
  • T body sentences
  • C citations
  • R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

Figures
Figure 1. Receiver temperature and gain vs. observing frequency models for (a) GBT 820 MHz and L-band (b) VLA L-band and S-band (c) DSA including contributions from spillover, LNA, and feed (J. Flygare 2024) and (d) CHIME for a 0◦ Dec. pulsar. The dip in the GBT gain curve illustrates the low gain channel inserted between non-adjacent bands. where alt. is the minimum telescope elevation, δ is the Dec. of the pulsar, and λ is the telescope latitude. This translates to a maximum time of tmax = 86400 × ℓ × tepoch 2π s. (33) 4.2.1. DSA The Deep Synoptic Array is a 1650-dish array under construct
Figure 2. Strain-amplitude sensitivity curves for a status quo NANOGrav 15-year PTA using current instruments (solid, orange curve) and a 15-year PTA using the optimal instruments (dashed, purple curve), primarily CHIME and DSA. The dotted line is a GWB power-law spectrum using the NG15gwb Bayesian posterior amplitude. 6. JITTER-DOMINATED PULSARS When designing a pulsar timing program, it is often useful to ask whether a pulsar is dominated by jitter. Some authors (e.g. J. M. Cordes & R. M. Shannon 2010; P. Gitika et al. 2025) define jitter-dominated as a pulsar which has σS/N < σJ and found…
Figure 3. Individual epoch-averaged noise components (top panels) and total TOA uncertainties (bottom panels) vs. optimal integration factor for an (a) not jitter-dominated pulsar, J0610–2100, (b) a mildly jitter-dominated pulsar, B1855+09, and (c) a highly jitter-dominated pulsar, J1022+1001. Recall that σS/N and σJ are weighted by the frequency-frequency covariances and contained within σDMd, which is added in quadrature to σδtC and σDM(ν) in Eq. (1), so the curves in the top panels do not simply sum to the bottom panels. are somewhat arbitrary they do have some quantitative support. If the…
Figure 4. Histogram of optimal integration factor and jitter-dominated classification across the 83 pulsars in our dataset accessible to DSA. J1903+0327 is the sole pulsar with x ∗ > 16 despite being dominated by σδtC , not σJ. The majority of pulsars are not jitter-dominated. (2008) measured τd = 0.55 ms) and σδtC is the dominant noise component by several orders of magnitude. Recall from Equation (6) that the achromatic noise covariance contains contributions from template-fitting and DISS. Decreasing Aeff and increasing tint results in a net increase in Sν1ν2 relative to Dν1ν2 so the…
Figure 5. Scatter plot of jitter fraction, σJ/σTOA, (at tint = 1.75 hr and full collecting area) vs. scattering timescale for all DSA-timed pulsars. The color scale represents the optimal integration factor, with the light-blue point representing the outlier J1903+0327, which is dominated by σδtC . DSA will have θFWHM = 2.99◦ . The result is only three beams, each containing two pulsars, none of which are highly jitter-dominated
Figure 6. Sky locations for DSA-timed pulsars, colored by their optimal integration factor. Optimal DSA beams containing more than one pulsar are identified with a red circle. From left to right, the beams contain: J0636+5128 and J0645+5158, J1630+3550 and J1630+3734, J1910+1256 and J1911+1347. Beam size and shape not to scale. predicted red noise. Interestingly, four of five the dropped Arecibo pulsars are encouraged to be timed with DSA because they are predicted to have relatively low red noise
Figure 7. Optimal time shifts away from ⃗teq = 6300 s (dotted line). Pulsars which gain time in ⃗tint ∗ have a right-pointing arrow, while those that lose time have a left-pointing arrow
Figure 8. Changes in σTOA from adding or subtracting time from a pulsar according to ⃗tint ∗ . Pulsars are listed in the same order as
Figure 9. Optimal tint vs red noise power. Points are colored by their scattering timescale. the components of σTOA that improve by increasing tint. In fact, many of the pulsars which would normally be considered the best timed (e.g. see J1909–3744 in
Figure 10. Strain-amplitude sensitivity curves for a NANOGrav 15-year PTA with DSA observing pulsars for the optimal time (solid, black curve) and equal time (dashed, red curve) per epoch. The dotted line is a GWB power-law spectrum using the NG15gwb Bayesian posterior amplitude. curve. In this intermediate GWB signal limit, tint does not strongly impact ρ, leading to the shallow optimum. One could reasonably ask: what if the pulsar models are incorrect? Scintillation can lead to inaccurate flux measurements and spectral index estimates and NANOGrav TOA selection biases flux measurements…
Figure 11. Optimal vs equal-time allocation GWB S/N for 100 Monte Carlo realizations of the PTA and marginal histograms for ρ(⃗teq) and ρ(⃗tint ∗ ). The dashed line identifies no difference between the optimal and equal-time S/N. The correlation between ρ(⃗teq) and ρ(⃗tint ∗ ) is r = 0.997 and ρ(⃗tint ∗ ) is consistently higher by 0.9. source list observed with the DSA, but this methodology will be broadly applicable to other PTA experiments and radio telescopes, expecially dish arrays like ngVLA and SKA. Our main results are summarized below: 1. An ab initio, 15-year DSA PTA with a 25% tot
Figure 12. 95% confidence bands for hc(f;⃗teq) (red) and hc(f;⃗tint ∗ ) (black) along with the assumed GWB power-law spectrum (dotted line). 2. Most pulsars in our source list will not be highly jitter-dominated with DSA, in the sense that their timing precision does not improve significantly with reduced instrumental sensitivity and increased integration time. The benefits of subarraying or strategic beam placement are highly limited due to the scarcity and distribution of highly jitter-dominated pulsars on the sky. This could change if new discoveries and/or techniques like cyclic…

Figures reproduced from arXiv:2610.00836 by the authors.

References

Discussion

Disagree with the review, or know something it missed? Leave a comment, highlight any passage above to comment on it, or challenge this review formally.