EPTA meeting, Paris - Jun 2026

Jun 1, 2026, 9:00 AM → Jun 5, 2026, 7:00 PM Europe/Paris

APC

EPTA meeting 2026 - Spring edition. This meeting gathers collaborators from the European Pulsar Timing Array collaboration to discuss Pulsar timing and Gravitational wave science.

 

It will be held a Laboratoire AstroParticule et Cosmologie - 10 Rue Alice Domon et Léonie Duquet, 75013 Paris.

Mon, June 1

Shared session between EPTA and Entretiens meetings

1 Introduction on the EPTA and the IPTA

Speaker: Gilles THEUREAU (LPC2E, CNRS-Orléans and Nançay Radio Observatory, Paris Observatory)

2 The principle of Pulsar Timing Arrays

Speaker: Mikel Falxa (LPC2E)

3 Studying the ionized ISM with pulsar timing

Speakers: Pauline Noé (LPC2E), Francesco Iraci (APC, CNRS)

3:00 PM Pause / Q&A

4 Q/A - Pause

5 Astrophysics of Massive Black Holes

Speaker: Hippolyte Jean Hugo Quelquejay-Leclere (Università degli Studi di Milano-Bicocca)

6 The intrinsic spin noise of MSPs

Speaker: Aditya Parthasarathy (ASTRON)

7 The theoritical aspects of the intrinsic spin noise

Speaker: Jérome Petri (Observatoire astronomique de Strasbourg)

8 Q/A - Pause

9 The Gamma-ray PTA

Speaker: Serena Valtolina

10 Updates from MeerKAT

Speaker: Beatrice Moreschi (Milano Bicocca)

11 The PTAs in the SKA era

Speaker: Antoine Petiteau (CEA/IRFU/DPhP)

Tue, June 2

DC & Timing session

10:30 AM Coffee break

DC & Timing session

Hands-on session and student projects

3:30 PM Coffee break

Group updates

12 Updates from Milano Bicocca group

Speaker: Alberto Sesana

13 Updates from APC-Orléans group

Speaker: Gilles THEUREAU (LPC2E, CNRS-Orléans and Nançay Radio Observatory, Paris Observatory)

14 Updates from MPIfR & Effelsberg

In this talk, I provide an update on behalf of the Max Planck Institute for Radio Astronomy EPTA Team. I briefly highlight contributions to DR3 as well as the work of individual members of the team. I summarize the status of EPTA observations taken by the Effelsberg Radio Telescope, describe the new ultra-broadband (UBB) receiver and discuss its potential in pulsar timing. I conclude by sharing recent developments of the “Coastguard3” pulsar timing software to support the new UBB receiver.

Speaker: Thomas Abbott (Max Planck Institute for Radio Astronomy)

15 Updates from the SHAO PTA group

An important science goal of the radio division at the Shanghai Astronomical Observatory (SHAO) is pulsar timing array (PTA). As such we are also actively taking part in the IPTA as well as regional PTA collaborations, like the European PTA and Chinese PTA. We are working on the combination and analysis of the IPTA DR3, as well as the EPTA DR3. Another focus of the group is on the astrophysical/new physics interpretation efforts. In my talk, I will give a short overview of the PTA science done at SHAO over the last half a year or so.

Speaker: Siyuan Chen (Shanghai Astronomical Observatory, Chinese Academy of Sciences)

16 Updates from the ASTRON group

Speaker: Dr Aditya Parthasarathy (ASTRON)

COCKTAIL DINNER

Barge du CROUS, Quai François Mauriac, Port de la Gare, 75013 Paris

Wed, June 3

Noise modelling session

17 Probing turbulence in the interstellar medium with long-term dispersion and rotation measure variations of young pulsars

Radio waves emitted by pulsars experience a frequency-dependence propagation effect caused by free elections in the ionized interstellar medium (IISM), parameterized by the dispersion measure (DM). Temporal variations in the DM of pulsars provide a powerful probe of the density fluctuations of the free electrons in the ionized interstellar medium. These variations provide insight into the turbulent and large-scale overdensities in the IISM in dynamic environments, such as supernova remnants and pulsar wind nebulae.

In this presentation, I will present the measurements of DM variations of 364 young pulsars measured at the Parkes radio telescope. By combining more than 15 years of observations with the Multibeam receiver and the recently installed ultrawideband receiver, which allows for simultaneous measurements across 3.4 GHz of bandwidth and an increase in the DM measurement precision of an order of magnitude. We are able to measure significant DM variations for more than 83 pulsars. We characterize these variations using a power-law wavenumber spectrum and find good agreement with a Kolmogorov scaling along different lines of sight. I will also highlight several pulsars that show deviations from this behaviour and discuss possible physical explanations.

Speaker: Selah Melfor

18 Investigating Interstellar Turbulence in HII Regions with LOFAR Pulsars

Interstellar turbulence critically influences the propagation of pulsar radio signals, affecting both the timing precision and the interpretation of interstellar plasma properties. While its statistical behaviour in the diffuse ionised medium is relatively well described by a Kolmogorov-like spectrum, the nature of turbulence within discrete structures such as HII regions remains poorly understood. The lines of sight of more than 600 pulsars are known to intersect HII regions, six of which are included in the pulsar pool observed with the LOFAR observatory. These lines of sight offer a rare opportunity to probe turbulence in more extreme plasma environments, where enhanced density and complex internal structure may lead to deviations from canonical models. To characterise the turbulent spectrum, I analyse dispersion measure time series using statistical tools such as the structure function, which enables inference of the power spectrum of electron density fluctuations. In parallel, I examine scintillation observables to extract complementary constraints on the structure of the plasma across various scales. This multi-probe approach aims to build a coherent physical picture of the turbulent conditions in HII regions and to assess their role in contributing to propagation-induced noise in high-precision pulsar timing experiments.

Speaker: Francesco Iraci (APC, CNRS)

19 An Elegant FD Formulation for a More Civilized PTA Age

Radio pulsars exhibit frequency-dependent variations in the shape of their pulse profiles that introduce systematic delays in the pulse times of arrival (TOA) across an observing band. Some Pulsar Timing Arrays model these delays using a frequency-dependent (FD) log-polynomial function whose coefficients are fit to TOA measurements. However, these FD parameters are prone to absorbing power from other chromatic effects, such as interstellar dispersion and scattering, leading to models that overestimate pulse-profile evolution and are not self-consistent across datasets. To better decouple these effects, we introduce two new mathematical formulations of frequency-dependent profile evolution, based on monomial and Legendre polynomial expansions, and test them using both observational data and targeted simulations. We find that the Legendre-based model generally predicts delays that more closely match the observed frequency-resolved pulse profiles, are more self-consistent across datasets, and better recover the injected profile evolution in simulated data.

Speaker: Sofia Sosa Fiscella (ASTRON)

20 Solar wind variations in DR2low

DR3 will add low-frequency data from LOFAR and NenuFAR. A characteristic of low-frequency pulsar observations is their high sensitivity to the diverse effects of propagation through a plasma. Adding these data is thereby useful for determining the various chromatic noises more accurately. In addition to being more sensitive to interstellar medium variations, low-frequency data become sensitive to smaller DM variations within the Solar System due to the solar wind contribution.
Contrary to purely stochastic DM and other chromatic variations, those from the solar wind are based on an annual variation, as well as an 11-year variation due to the solar cycle. The software run_enterprise allows for the fitting of this annual variation. For now, we have applied it to the DR2low data from J1022+1001 and J0030+0451. And we modified it to fit the solar cycle variations in addition to the annual ones. We will present some initial results for these two pulsars.
A better consideration of these various solar wind variations could significantly improve the PTA sensitivity. Moreover, in the reverse direction, the PTA analysis could help in defining the solar wind model more accurately, which will ultimately also improve the PTA sensitivity.

Speaker: Mark Brionne (INAF-OA Cagliari)

10:30 AM Coffee break

Gravitational waves and interpretation session

21 Continuous-wave candidate validation across PTAs

Building on our recent work on continuous-wave searches in PTA data, this talk will discuss possible strategies for testing the origin of low-frequency CW candidates identified in recent EPTA and NANOGrav analyses, including the persistent 4 nHz feature. I will outline ideas for studying how candidate significance and coherence change across datasets and array combinations, with the aim of distinguishing individual supermassive black hole binary signals from noise artifacts or stochastic-background-induced structure. The talk is intended both as a brief overview of this direction and as a starting point for broader collaboration on CW-related analyses within EPTA and jointly with NANOGrav/IPTA.

Speaker: Bence Becsy (University of Birmingham)

22 Eccentric binaries in PTA data: parameter estimation and interplay with the gravitational wave background

Pulsar timing array (PTA) experiments are designed to detect gravitational waves (GWs) at nano-hertz frequencies. Supermassive black hole binaries (SMBHBs) represent the most plausible astrophysical sources in this band. The superposition of GWs emitted by the entire SMBHB population produces a stochastic GW background (GWB). Particularly massive and nearby SMBHBs can produce signals strong enough to stand out above this background as individually resolvable sources. PTAs are sensitive to the early inspiral of these systems at large orbital separations, where circularisation due to GW emission may not yet be efficient, leading binaries to retain significant eccentricity when entering the PTA sensitivity band.

In this talk, I will present results of parameter estimation performed using a gravitational waveform model based on the Effective-One-Body (EOB) approach for individually resolvable binaries. I will focus on binaries relevant for an EPTA-like mock dataset, spanning the parameter space in eccentricity, chirp mass, and orbital frequency for various signal-to-noise ratios (SNRs). The main focus will be on the interplay between individual eccentric binaries and a common red noise process (representing the GWB), examining how eccentricity affects our ability to characterise both signals simultaneously.

Speaker: Sara Manzini (APC - Université Paris Cité)

23 Probing Supermassive Black Hole Mergers with Pulsar Timing Arrays

By monitoring the times of arrival of radio pulses from millisecond pulsars, Pulsar Timing Arrays (PTAs) serve as unique gravitational wave (GW) laboratories in the nanohertz band. To date, the primary astrophysical sources of GWs targeted in this frequency range have been inspiraling supermassive black hole binaries (SMBHBs) on circular and eccentric orbits. In this work, we demonstrate that, thanks to the so-called pulsar term in the timing residual waveform of GW signals, PTAs can probe individual SMBHBs that merged before timing observations began. We refer to the latter as \emph{zombie binaries}. Using SMBHB population models consistent with current PTA constraints, we find that while the probability of detecting such systems in existing PTA datasets remains low, the Square Kilometer Array observatory is expected to achieve sufficient sensitivity to have a few zombie binaries with a signal-to-noise ratio exceeding 3 in its data. Although their confident identification might be challenging, this new class of PTA sources opens a novel window for studying the most massive SMBHBs in our local universe.

Speaker: Hippolyte Jean Hugo Quelquejay-Leclere (Università degli Studi di Milano-Bicocca)

24 QuickGWecc: A fast Bayesian pipeline for searching continuous GWs from eccentric binaries in PTA data

Most searches for continuous gravitational waves (CGWs) from individual supermassive black hole binaries (SMBHBs) in pulsar timing array (PTA) data assume circular orbits, despite theoretical studies suggesting that these sources may retain significant eccentricity. In this talk, I will present QuickGWecc, a fast and efficient Bayesian pipeline for searching for CGWs from SMBHBs in relativistic eccentric orbits, extending the QuickCW framework. By exploiting the separation between projection and shape parameters, the method achieves orders-of-magnitude computational speedups, making eccentric searches tractable. I will show validation results using simulated datasets and discuss its prospects for application to upcoming PTA datasets.

Speaker: Lankeswar Dey (Institute of Astrophysics, FORTH)

25 Ongoing search for memory effect in IPTA-DR3

The search for the non-linear memory effect, which is a permanent strain in spacetime that remains after the passage of a gravitational wave (GW), has been ongoing in PTA data for years, using a simple burst model. With the upcoming IPTA-DR3, a working group has been formed to research memories (both linear and nonlinear). In this talk, we will present the preliminary work carried out to search for a realistic model of nonlinear memory in PTA data.

Speaker: Adrien Cogez (CEA/IRFU/DPhP)

Hands-on session and student projects

3:30 PM Coffee break

Contributed talks

26 Updates for the Interpretation WG

I would like to give an overview of activities undertaken in the Interpretation WG.

Speaker: Nataliya Porayko (MPIfR)

27 Inflationary interpretation of the gravitational-wave signal in the European Pulsar Timing Array DR2 data with constraints

The second data release of the European Pulsar Timing Array (EPTA) collaboration provides
evidence for the presence of a gravitational-wave (GW) background. In this work, we explore a
potential cosmological interpretation of this signal in terms of inflationary scenarios. We parametrize
the tensor power spectrum in terms of the tensor-to-scalar ratio r, the tensor spectral index nt, the
reheating temperature Trh and the cut-off frequency fend. We incorporate all relevant observational
constraints, including those from Cosmic Microwave Background, Big Bang Nucleosynthesis and
LIGO–Virgo–KAGRA observations. We demonstrate that imposing these constraints consistently
reduces the region of parameter space that provides a viable interpretation of the EPTA signal, to
log10 r > −12, 1.2 ≲ nt ≲ 2.5, 10 MeV ≲ Trh ≲ 50 GeV and 10−8 Hz ≲ fend ≲ 1012 Hz at 95%
confidence level. This favours
the scenario in which the GW spectrum in the EPTA frequency band originates from tensor modes
that re-entered the horizon during the radiation-dominated era, allowing for a higher r and a flatter
spectrum. However, Trh must take very low values, which are challenging to explain theoretically.

Speaker: Delphine Perrodin

28 Characterisation of Continuous Gravitational Waves with Deep Learning: Detection and Sky Localisation

Accurate localisation of continuous gravitational waves (CGWs) from supermassive black hole binaries (SMBHBs) remains one of the key challenges in Pulsar Timing Array (PTA) data analysis. Traditional searches based on the $\mathcal{F}_e$ statistic provide a robust analytic framework, but the resulting sky maps are strongly affected by the PTA antenna pattern, which redistributes signal power across the sky and generates secondary peaks that complicate the identification of the true source position. This degeneracy motivates the development of alternative approaches capable of disentangling instrumental artefacts from true localisation information.

In this work, we investigate whether Sequential Simulation-Based Inference (SBI) can improve CGW sky localisation by learning the mapping between $\mathcal{F}_e$ maps and true source positions. Using simulated PTA datasets spanning multiple configurations, isotropic and non-isotropic 25-pulsar arrays, with and without the pulsar term, and under non-uniform sensitivity, we demonstrate that our SBI pipeline effectively marginalises over antenna pattern search artefacts, providing reliable posterior distributions for the source coordinates. A key advantage of this framework is its extreme computational efficiency: our pipeline can generate hundreds of thousands of $\mathcal{F}_e$ statistic maps in approximately 3 minutes, with the subsequent network training completed in under 2 hours on a single GPU. This enables both the rapid generation of large training sets and near-instantaneous source characterisation. Our results show that the angular resolution ($\Delta\Omega$) achieved via SBI is consistent with the theoretical lower bounds predicted by the Fisher Information Matrix.

We further extend this framework to the detection problem, training a binary neural classifier on the same $\mathcal{F}_e$-statistic maps to distinguish signal-containing observations from noise-only realisations. Independent classifiers are trained for each GW frequency bin, covering the sub-threshold SNR regime ($\mathrm{SNR} \in [3, 10]$). We characterise detection performance across all tested configurations, finding consistent and robust performance.

Speaker: Ludovica Carbone (University of Milano-Bicocca)

29 Towards multi-messenger observations of supermassive black hole binaries

Supermassive black hole binaries (SMBHBs) are exceptional multi-messenger targets, since on top of their low-frequency gravitational waves (GW), they may emit bright electromagnetic (EM) emission. I will describe a recently developed joint likelihood method that combines EM and pulsar timing array (PTA) data, which was shown to improve the binary parameter estimation. I will also discuss the prospects and challenges in finding the host galaxy once PTAs detect the first individually resolved binary.

Speaker: Maria Charisi (IA-FORTH & Washington State University)

30 Impact of Dispersion Measure Noise Modeling on Gravitational-Wave Background

We investigated the impact of dispersion measure (DM) variations on the recovery of a stochastic Gravitational Wave Background (GWB) in Pulsar Timing Array (PTA) datasets, extending the chromatic noise analysis of Iraci et al. (2024). We generated multiple synthetic PTA datasets for different configurations of a 25-pulsar array, combining LOFAR-like low-frequency observations with Nançay L-band ToAs, modeling DM variations with the Discovery software suite to accelerate parameter estimation. The consistency of the recovered GWB posteriors, obtained through variational inference with normalizing flows, was evaluated using probability–probability (P–P) plots and Kolmogorov–Smirnov (KS) statistics. Several biases are identified, largely independent of the assumed DM variation level.

Speaker: Danilo Zecca (Università del Salento)

Thu, June 4

Contributed talks

31 Looking for non-gaussianity in PTA through the four-point correlator

Pulsar Timing Arrays have recently reported strong evidence for a stochastic gravitational wave background. In standard analyses, this is modeled through pulsar-dependent Fourier coefficients assumed to follow Gaussian statistics, so that the signal is fully characterized by its two-point function. However, if the background arises from a finite population of inspiralling supermassive black hole binaries, non-Gaussian features may emerge, making higher-order correlators essential. In this work, we compute the complete four-point correlator of the Fourier coefficients for four arbitrary pulsar positions, identifying it as the leading probe of non-Gaussianity. The result separates into a Gaussian contribution, proportional to the square of the two-point function, and a genuinely non-Gaussian connected component whose angular dependence generalizes the Hellings–Downs correlation to four pulsars. This angular structure depends only on averages of products of antenna pattern functions, and is therefore independent of the specific physical origin of the background. We further propose incorporating the four-point correlator into the parameter-estimation pipeline via a marginalized likelihood that perturbatively accounts for non-Gaussian effects. Our results provide the theoretical framework needed to search for non-Gaussian features in Pulsar Timing Array data, opening the way to a more complete characterization of gravitational-wave backgrounds.

Speaker: Massimo Vaglio (SISSA)

32 Modeling the nHz gravitational wave background with t-process

Unveiling the nature of the nanoHz gravitational wave signal reported by Pulsar Timing Array collaborations is the next goal of low-frequency GW astronomy. The signal likely originates from the incoherent superposition of GWs emitted by a population of supermassive black hole binaries (SMBHBs). The discrete nature of the population, environmental effects and eccentricity are expected to produce significant deviations of the GWB power spectrum from the smooth powerlaw approximation.
In this talk I explore the possibility of modeling the GWB using a t-process power spectral density. This model also takes into account possible extra power in specific frequency bins, which may arise from single sources dominating the GWB. I will present results from testing the model on real PTA datasets and on simulations of realistic populations to evaluate the ability to detect a more complex PSD that deviates from the powerlaw. I will discuss current limitations and potential improvements of the model.

Speaker: Beatrice Eleonora Moreschi (University of Milano-Bicocca)

33 Possible imprints of a clustered Universe in astrophysical PTA GW data

The compelling evidence for the presence of nHz gravitational waves (GWs) in pulsar timing data calls for a refinement of what can and cannot be expected beyond vanilla scenarios. Under the hypothesis of an astrophysical origin for the GWs, we discuss possible imprints left in the signal by the fact the Universe is inherently clustered, with anisotropic large scale structure.
At the background level, this should imply primary anisotropies in the GW signal. Distinguishing the latter from the Poisson anisotropies due to the discreteness of sources appears as a challenge for which cross-correlations with galaxy catalogues is a promising approach.
If time allows, we may also discuss secondary imprints induced by the clustered Universe on individual continuous wave signals from supermassive binaries.

Speaker: Martin Pijnenburg

34 Downsampling PTA data

Pulsar Timing Array (PTA) data are challenging to analyze due to their uneven sampling and the presence of observational gaps. In gravitational wave (GW) searches with PTA data, the signal is expected to dominate primarily at low frequencies, motivating the question of whether the data can be efficiently downsampled without significant loss of information. We propose a method to generate downsampled and evenly sampled PTA datasets using Gaussian process regression. We demonstrate that, when the GW signal is concentrated in the low-frequency band, downsampling results in only minimal loss of sensitivity. This enables fully time-domain analyses that would otherwise be computationally prohibitive because of expensive covariance matrix inversions. Furthermore, the production of evenly sampled data could allow the application of FFT techniques and facilitate analyses directly in the frequency domain.

Speaker: Mikel Falxa

10:30 AM Coffee break

Contributed talks

35 Methods for GW searches and pulsar noise characterisation from gamma-ray photons

Accurately modelling pulsar noise is one of the main challenges for current radio PTA experiments. For many pulsars, the data are not informative enough to efficiently solve the covariance between chromatic and achromatic processes. In gamma-ray PTA (GPTA) experiments instead, the sensitivity to stochastic processes is limited by the low photon flux of gamma-ray pulsar observations; thus, the radio folding procedures becomes inapplicable for most pulsars. However, because of the high frequency of the gamma rays, there is no interaction with the interstellar medium. This results in much simpler noise models, which include only the pulsar’s spin noise and Poisson statistics noise.
In this talk, I will discuss the new data analysis pipeline that the GPTA group has been developing for GW searches from gamma-ray data and present a method to consistently combine radio and gamma-ray pulsar observations. Being able to break the degeneracy between chromatic and achromatic signals would not only provide valuable information on the interstellar medium distribution, but would also improve our sensitivity to a GWB signal.

Speaker: Serena Valtolina

36 Observation Strategy Study of Pulsar Timing Array

This project aims to investigate optimal observational modes of radio interferometric arrays for diverse scientific goals (e.g., PTAs) using numerical simulation tools.
We develop an open-source software framework capable of generating simulated pulsar timing data and injecting multiple signals, including gravitational waves. Leveraging existing observational datasets, we use this framework to analyze how to rationally configure observational parameters—such as observing time, number of subarrays, and observing frequencies—for specific scientific objectives, thereby identifying optimal observing strategies.
The results will significantly improve observational efficiency and data quality, enhance the detectability of nanohertz gravitational wave signals, and provide direct decision-making support for future PTA strategies to be adopted by the Square Kilometre Array (SKA).

Speaker: Yang Liu (Shanghai Astronomical Observatory, Chinese Academy of Sciences)

37 Resolving signals with finite angular size in the GWB sky

In this work, we consider the case of a model GWB source with a Gaussian power spectrum on a sphere, and obtain an analytical expression of the overlap reduction function of the signals of two pulsars, which generalizes the previously obtained expressions for isotropic and delta-shaped model sources. Using the generalized correlation function allows us to determine the characteristic angular size of the source. Such signals can arise in some models of new physics, e.g. oscillating loops from a single cosmic string. The algorithm was tested on simulated datasets of pulsars with isotropic and anisotropic sky distribution.

Speakers: Egor Ledenev (MSU SAI), Viacheslav Chernosov (MIPT)

38 The First search for gravitational waves with the DR2low dataset

The EPTA+InPTA DR2new+ dataset has recently been combined with low-frequency measurements from LOFAR and NenuFAR (Iraci et al., 2025), with the aim of generating a dataset capable of disentangling the DM variation noise from achromatic red noise and the stochastic background. Simulations have shown that better distinguishing between these components is expected to improve sensitivity to the GWB signal and lead to more accurate and precise GWB parameter estimation (Ferranti et al., 2025). This talk presents and discusses the first results from a GWB search performed on this dataset. A key challenge in this analysis is addressing noise sources mostly neglected in EPTA DR2, such as solar wind and scattering variations noise. The very low frequency backends of LOFAR and NenuFAR make the new dataset extremely sensitive to these two sources of chromatic noise. Furthermore, the DR2low dataset has been employed to evaluate the performance of an analysis pipeline based on the Discovery package, coupled with an efficient sampling strategy exploiting variational inference. The ability of this new pipeline to accurately reproduce the posterior distribution of single pulsar noise analyses and GWB joint searches has been carefully tested to determine how much we can trust this pipeline for analysing future datasets such as EPTA DR3.

Speaker: Irene Ferranti (University of Milano-Bicocca)

Closed EPTA session : SC+EC

3:30 PM Coffee break

Closed EPTA session : EPTA DR3 & LEAP

Fri, June 5

Hands-on session and student projects

10:30 AM Coffee break

Hands-on session and student projects

Last discussions & End of conference