ISW–tSZ Cross-Correlation in Cosmology

• ISW–tSZ cross-correlation is a statistical measure that quantifies the spatial correspondence between CMB temperature anisotropies from the ISW effect and distortions from the tSZ effect due to hot ionized gas.
• It employs angular cross-power spectrum techniques and leverages high-resolution data from Planck and complementary surveys to constrain both cosmological and astrophysical parameters.
• The measurement provides critical insights into dark energy dynamics, baryonic distribution, and structure growth while addressing challenges like foreground contamination and cluster masking.

The ISW–tSZ cross-correlation is the statistical measure of spatial correspondence between temperature anisotropies in the cosmic microwave background (CMB) due to the Integrated Sachs–Wolfe (ISW) effect and distortions arising from the thermal Sunyaev–Zeldovich (tSZ) effect. This cross-correlation emerges from the interaction of CMB photons with evolving large-scale gravitational potentials (ISW) and hot ionized gas (tSZ), both tracing the cosmic web's potential wells. Its detection provides a direct observational link between gravitational potential evolution driven by dark energy and the baryonic distribution traced by intracluster and intergalactic gas. The ISW–tSZ cross-spectrum offers a cosmological probe uniquely sensitive to the properties of dark energy, the thermal history and distribution of baryons, and the growth of structure at low redshift.

1. Physical Origins: ISW and tSZ Effects

The ISW effect arises when CMB photons transit time-varying gravitational potentials, acquiring net energy shifts when potentials decay during late-time cosmic acceleration. The temperature fluctuation in direction $\hat{n}$ due to the ISW effect is given by

$$\Delta T_{\rm ISW}(\hat n) = T_{\rm CMB} \int_{0}^{\chi_*} d\chi\, e^{-\tau(\chi)}\, \frac{\partial}{\partial\eta}\left[\Psi(\chi, \hat n\chi) + \Phi(\chi, \hat n\chi)\right],$$

where $T_{\rm CMB}$ is the blackbody temperature, $\chi$ is comoving distance, $\tau$ is the Thomson optical depth, and $\Psi, \Phi$ are Bardeen potentials.

The tSZ effect measures spectral distortions in the CMB introduced by inverse Compton scattering of CMB photons off hot electrons, typically in galaxy clusters and the warm-hot intergalactic medium (WHIM). The tSZ Compton-$y$ parameter along direction $\hat{n}$ is

$$y(\hat n) = \frac{\sigma_T}{m_e c^2}\int_{\rm los} dl\, n_e(l,\hat n) k_B T_e(l,\hat n),$$

where $\sigma_T$ is the Thomson cross section, $$\Delta T_{\rm ISW}(\hat n) = T_{\rm CMB} \int_{0}^{\chi_*} d\chi\, e^{-\tau(\chi)}\, \frac{\partial}{\partial\eta}\left[\Psi(\chi, \hat n\chi) + \Phi(\chi, \hat n\chi)\right],$$0 is the free electron density, $$\Delta T_{\rm ISW}(\hat n) = T_{\rm CMB} \int_{0}^{\chi_*} d\chi\, e^{-\tau(\chi)}\, \frac{\partial}{\partial\eta}\left[\Psi(\chi, \hat n\chi) + \Phi(\chi, \hat n\chi)\right],$$1 is electron temperature, and $$\Delta T_{\rm ISW}(\hat n) = T_{\rm CMB} \int_{0}^{\chi_*} d\chi\, e^{-\tau(\chi)}\, \frac{\partial}{\partial\eta}\left[\Psi(\chi, \hat n\chi) + \Phi(\chi, \hat n\chi)\right],$$2 is Boltzmann’s constant.

Both ISW and tSZ signals originate from the same potential wells; ISW encodes the time-variation of the potentials due to dark energy or modified gravity, while tSZ traces the pressure of the ionized gas within those structures (Taburet et al., 2010, Ibitoye et al., 2023).

2. Cross-Power Spectrum Formalism

The statistical relationship is quantified by the angular cross-power spectrum,

$$\Delta T_{\rm ISW}(\hat n) = T_{\rm CMB} \int_{0}^{\chi_*} d\chi\, e^{-\tau(\chi)}\, \frac{\partial}{\partial\eta}\left[\Psi(\chi, \hat n\chi) + \Phi(\chi, \hat n\chi)\right],$$3

where $$\Delta T_{\rm ISW}(\hat n) = T_{\rm CMB} \int_{0}^{\chi_*} d\chi\, e^{-\tau(\chi)}\, \frac{\partial}{\partial\eta}\left[\Psi(\chi, \hat n\chi) + \Phi(\chi, \hat n\chi)\right],$$4 and $$\Delta T_{\rm ISW}(\hat n) = T_{\rm CMB} \int_{0}^{\chi_*} d\chi\, e^{-\tau(\chi)}\, \frac{\partial}{\partial\eta}\left[\Psi(\chi, \hat n\chi) + \Phi(\chi, \hat n\chi)\right],$$5 are spherical-harmonic coefficients of the $$\Delta T_{\rm ISW}(\hat n) = T_{\rm CMB} \int_{0}^{\chi_*} d\chi\, e^{-\tau(\chi)}\, \frac{\partial}{\partial\eta}\left[\Psi(\chi, \hat n\chi) + \Phi(\chi, \hat n\chi)\right],$$6 and temperature maps, respectively.

Under the Limber approximation (accurate for $$\Delta T_{\rm ISW}(\hat n) = T_{\rm CMB} \int_{0}^{\chi_*} d\chi\, e^{-\tau(\chi)}\, \frac{\partial}{\partial\eta}\left[\Psi(\chi, \hat n\chi) + \Phi(\chi, \hat n\chi)\right],$$7), the cross-power spectrum in a flat $$\Delta T_{\rm ISW}(\hat n) = T_{\rm CMB} \int_{0}^{\chi_*} d\chi\, e^{-\tau(\chi)}\, \frac{\partial}{\partial\eta}\left[\Psi(\chi, \hat n\chi) + \Phi(\chi, \hat n\chi)\right],$$8CDM cosmology is

$$\Delta T_{\rm ISW}(\hat n) = T_{\rm CMB} \int_{0}^{\chi_*} d\chi\, e^{-\tau(\chi)}\, \frac{\partial}{\partial\eta}\left[\Psi(\chi, \hat n\chi) + \Phi(\chi, \hat n\chi)\right],$$9

with $T_{\rm CMB}$0 the matter power spectrum. The ISW window $T_{\rm CMB}$1 is proportional to the conformal-time derivative of the growth factor, and the tSZ window $T_{\rm CMB}$2 directly traces electron pressure, incorporating a bias parameter $T_{\rm CMB}$3. Halo model formalism allows for decomposition of the electron pressure power spectrum into one-halo (intra-halo) and two-halo (large-scale correlation) terms (Creque-Sarbinowski et al., 2016, Ibitoye et al., 24 Jan 2026). For the parameter estimation and model fitting, the likelihood function is assumed Gaussian in $T_{\rm CMB}$4, with the data vector and theoretical prediction compared under an empirically determined covariance.

3. Data Sets, Estimation, and Signal Extraction

The cross-correlation analysis utilizes high-resolution, full-sky Compton-$T_{\rm CMB}$5 maps from Planck (e.g., MILCA algorithm) and CMB temperature maps cleaned of foregrounds (e.g., SMICA/Commander). ISW maps are reconstructed by filtering CMB temperature at large angular scales, with galaxy-survey cross-correlation methods providing equivalent reconstructions. Power spectra computations employ pseudo-$T_{\rm CMB}$6 methods (e.g., XPol, PolSpice, MASTER in NaMaster), correcting for mask-induced mode coupling. Covariance matrices are empirically estimated from Monte Carlo realizations that match observed auto-power spectra and suppress sample variance. Multipole binning is typically in bands of $T_{\rm CMB}$7 for the range $T_{\rm CMB}$8–200, with $T_{\rm CMB}$9–0.7 sky coverage.

Foregrounds—specifically, cosmic infrared background (CIB) leakage and residual dust—are marginalized via template fitting. The signal-to-noise of the cross-correlation is increased by masking low-$\chi$0 massive clusters, reducing the tSZ shot noise, and optimizing dust mitigation, with requirements on the residual power at $\chi$1 set by the achievable foreground subtraction accuracy (Taburet et al., 2010).

4. Cosmological and Astrophysical Constraints

Statistically significant detection of the ISW–tSZ cross-correlation has been achieved at $\chi$2 confidence with Planck data (Ibitoye et al., 2023). Joint analysis of auto- and cross-spectra yields constraints on cosmological parameters: $\chi$3 and on parameters specific to gas physics: $\chi$4 yielding an electron temperature of $\chi$5 for $\chi$6. These values are consistent with the warm–hot intergalactic medium (WHIM), directly indicating its substantial baryon content.

The cross-spectrum amplitude is largely determined by the overlap of the ISW kernel (potential decay rate) and the tSZ kernel (pressure bias), with maximal sensitivity at $\chi$7–1.5 and halo masses of $\chi$8–$\chi$9 (Creque-Sarbinowski et al., 2016). This redshift sensitivity allows the cross-correlation to probe the epoch and scale of potential decay and baryonic heating relevant for structure formation.

Cosmological parameter estimation with ISW–tSZ data is competitive with, and complementary to, primary CMB and large-scale structure measurements, providing independent validation of the presence of dark energy—since ISW vanishes in a pure matter-dominated universe—and sensitivity to the growth rate parameter $\tau$0 and dark energy equation of state $\tau$1 (Taburet et al., 2010).

5. Implications for Dark Energy and Fundamental Physics

The ISW–tSZ cross-correlation is sensitive not only to the standard $\tau$2CDM framework but to a range of dynamical dark energy scenarios. Analyses comparing thawing, tracker, and scaling-freezing models under current cross-spectrum constraints show that all yield results statistically consistent with $\tau$3CDM within $\tau$4, but with a mild preference for thawing models based on minimum $\tau$5 (Ibitoye et al., 24 Jan 2026). The sensitivity arises because the ISW effect, nonzero only for decaying potentials, directly tracks deviations from matter domination at $\tau$6–0.6, redshifts at which the tSZ signal is strongest. Thawing models, with potentials that begin to decay only at low $\tau$7, yield enhanced ISW amplitude at relevant epochs, while tracker and scaling-freezing models imprint distinct signatures in the cross-spectrum via their particular $\tau$8 evolution.

Forecasts also highlight the cross-correlation as a diagnostic of the redshift distribution of tSZ sources, exotic early-universe contributions to $\tau$9, and scale-dependent primordial non-Gaussianity. For the latter, scale and redshift dependence of the cross-spectrum enables separation of late-time from primordial contributions, given sufficient S/N and cluster masking.

6. Experimental Challenges, Systematics, and Future Prospects

The principal limitations of current ISW–tSZ measurements are cosmic infrared background leakage into Compton-$\Psi, \Phi$0 maps, residual cluster masking effects, and uncertainties in ISW map reconstruction. The amplitude of foreground residuals, especially galactic and extragalactic dust at low multipoles, must be maintained below $\Psi, \Phi$1 at $\Psi, \Phi$2 for high-fidelity detection; current Planck-level subtraction approaches this threshold but does not attain it (Taburet et al., 2010). Cluster masking strategies (removing $\Psi, \Phi$3 clusters) substantially reduce tSZ noise without sacrificing ISW–tSZ signal.

Upcoming surveys with deeper and higher-resolution CMB and galaxy maps (e.g., AdvACT, SPT-3G, Simons Observatory, LSST, and Euclid) are projected to yield sub-percent level precision in gas physics, improved CIB modeling, and improved measures of potential decay (Ibitoye et al., 2023). A plausible implication is that refined measurements will distinguish among dark energy models, rigorously probe WHIM properties, and potentially isolate primordial contributions to $\Psi, \Phi$4.

7. Summary Table of Key Results

Measurement/Constraint	Value/Description	Reference
$\Psi, \Phi$5 detection significance	$\Psi, \Phi$6 (Planck MILCA + SMICA)	(Ibitoye et al., 2023)
Matter density $\Psi, \Phi$7	$\Psi, \Phi$8	(Ibitoye et al., 2023)
Hubble constant $\Psi, \Phi$9	$y$0	(Ibitoye et al., 2023)
Gas bias $y$1 $y$2 $y$3 $y$4	$y$5	(Ibitoye et al., 2023)
Electron temperature $y$6	$y$7	(Ibitoye et al., 2023)
Sensitivity to DE models	Mild preference for thawing	(Ibitoye et al., 24 Jan 2026)

These results confirm the ISW–tSZ cross-correlation as a direct cosmological and astrophysical probe, offering unique insights into the interplay between gravitational potential evolution and baryonic pressure, with expanding prospects for the study of dark energy and the cosmic baryon census.