LENSING NOISE IN MILLIMETER-WAVE GALAXY CLUSTER SURVEYS

Yashar Hezaveh, Keith Vanderlinde, G. P. Holder, T. de HaanThe Astrophysical Journal 2013 · DOI

TL;DR

This paper investigates how gravitational lensing by galaxy clusters distorts the signals from background dusty star-forming galaxies (DSFGs) and the Cosmic Microwave Background (CMB). The authors find that these lensing effects largely cancel each other out in current Sunyaev-Zel'dovich (SZ) cluster surveys, meaning lensing noise is a sub-dominant source of error compared to statistical noise. However, these effects could become more significant for future, larger surveys.

What Problem Does This Paper Address?

Sunyaev-Zel'dovich (SZ) surveys are a powerful tool for detecting galaxy clusters by observing the spectral distortion of the CMB caused by hot gas within the cluster. However, these surveys are susceptible to "confusion noise" from astrophysical sources that can mimic or obscure the SZ signal. Two primary sources of this confusion noise are the Cosmic Microwave Background (CMB) itself and distant, dusty star-forming galaxies (DSFGs), which appear as unresolved point sources in SZ observations.

Galaxy clusters, being massive gravitational lenses, can significantly distort the light from these background sources. This paper addresses the question of how this gravitational lensing specifically affects the statistical properties of the DSFG and CMB backgrounds at the locations of galaxy clusters, and what the implications are for the accuracy and completeness of SZ cluster catalogs. Understanding these lensing effects is crucial for accurately interpreting SZ survey data and for making robust cosmological inferences from cluster counts.

Methods

The authors employ a combination of analytical modeling and numerical simulations to study the impact of gravitational lensing.

  • Analytical Modeling: They first develop an analytical framework to understand how uniform magnification affects the probability distribution functions (PDFs) of background sources and the CMB. They then extend this to a simplified case of lensing by a Singular Isothermal Sphere (SIS) halo, considering point sources at a single redshift. This analytical approach helps to build intuition about the underlying physics.
  • Numerical Simulations: To account for more realistic scenarios, they use a ray-tracing code to simulate lensing effects. This code incorporates realistic galaxy cluster mass profiles (NFW profiles with an SIS core) and models for DSFG number counts based on the Durham semi-analytic model. They generate lensing deflection maps for multiple source redshifts and then simulate the lensed DSFG and CMB backgrounds.
  • Matched-Filter Analysis: To assess the impact on SZ surveys, they apply a matched-filter, a standard technique used in SZ cluster detection. This filter is designed to optimize the detection of SZ signals by correlating the observed map with expected SZ cluster morphologies. They then analyze the resulting filtered maps at the locations of simulated clusters.
  • Probability Distribution Functions (PDFs): They specifically examine how lensing alters the PDFs of the filtered DSFG and CMB signals. They fit Gaussian and log-logistic functions to these distributions to quantify the changes in variance and shape.
  • Cluster Number Density Estimation: Finally, they use these modified PDFs to estimate the expected number of detected galaxy clusters in a simulated single-band SPT-like survey, comparing the results with and without lensing effects.

Datasets

  • DSFG Number Counts: The paper uses number counts predicted by the Durham semi-analytic model (Baugh et al. 2005) for dusty star-forming galaxies.
  • CMB Power Spectra: Realizations of CMB fields are generated using power spectra calculated with CAMB (Lewis et al. 2000).
  • Cluster Mass Function: The paper utilizes the mass function prescription from Tinker et al. (2012) to generate expected number densities of galaxy clusters.
  • Simulated Survey Data: The study models a single-band 150 GHz cluster survey similar to that of Vanderlinde et al. (2010) and a multi-frequency SPT survey (e.g., Reichardt et al. 2012b) for comparison.

Key Insights and Ideas

  • Lensing Distorts Backgrounds: The core idea is that galaxy clusters, acting as gravitational lenses, don't just magnify background sources uniformly. They create complex distortions, altering the probability distribution of the flux from both DSFGs and the CMB.
  • DSFG Lensing Boosts Signal: Lensing of DSFGs tends to increase the variance of the background flux. This leads to a slight increase in the number of detected clusters due to an enhanced Eddington bias (making fainter sources appear brighter and thus detectable).
  • CMB Lensing Suppresses Signal: Conversely, lensing of the CMB at cluster locations leads to a reduction in CMB power. This makes it slightly harder to detect clusters, as the SZ signal is a decrement against the CMB.
  • Cancellation Effect: Crucially, the paper finds that the increase in DSFG noise and the decrease in CMB noise due to lensing largely cancel each other out. This means the net effect on cluster number counts in current surveys is relatively small.
  • "Filling-in" Effect: A rare but interesting consequence of strong lensing is the "filling-in" of SZ decrements by bright, strongly lensed background sources. This can cause a cluster signal to be completely obscured, leading to a missed detection. This occurs in about 2% of cases.
  • Redshift Dependence: The magnitude of these lensing effects is dependent on the redshift of the background sources and the mass and redshift of the lensing cluster.

Key Results

  • DSFG Lensing: Leads to an approximate 10% increase in detected cluster number counts due to a ~50% increase in the variance of the DSFG background.
  • CMB Lensing: Results in a ~55% reduction in CMB power at cluster locations, leading to a net decrease in detected cluster number counts.
  • Net Effect: For current single-frequency SZ surveys, the combined effect of DSFG and CMB lensing on cluster number counts is sub-dominant to Poisson errors.
  • "Filling-in" Rate: Approximately 2% of SZ cluster signals can be obscured by strongly lensed background sources.
  • Future Surveys: These lensing effects are expected to become more significant for future surveys aiming to detect thousands of clusters.

Limitations and Future Work

  • Simplified Lens Models: While the simulations use realistic NFW profiles with SIS cores, the analytical models are simpler (e.g., uniform magnification, SIS halo). The paper acknowledges that more complex lens profiles and the non-zero size of sources can have substantial effects on lensing statistics.
  • Single-Band Survey Focus: The primary analysis focuses on a single-band 150 GHz survey. The authors note that multi-frequency surveys can better mitigate these lensing effects by spectrally isolating the SZ signal.
  • Future Surveys: The paper highlights that for future, larger surveys (e.g., CCAT, SPT-3G) aiming for catalogs of ~10,000 clusters, these lensing effects could become more significant and require careful consideration. Multi-band data is suggested as a way to mitigate these issues.

Summary updated

Abstract

We study the effects of gravitational lensing by galaxy clusters of the background of dusty star-forming galaxies (DSFGs) and the Cosmic Microwave Background (CMB), and examine the implications for Sunyaev-Zel'dovich-based (SZ) galaxy cluster surveys. At the locations of galaxy clusters, gravitational lensing modifies the probability distribution of the background flux of the DSFGs as well as the CMB. We find that, in the case of a single-frequency 150 GHz survey, lensing of DSFGs leads to both a slight increase (~10%) in detected cluster number counts (due to a ~ 50% increase in the variance of the DSFG background, and hence an increased Eddington bias), as well as to a rare (occurring in ~2% of clusters) "filling-in" of SZ cluster signals by bright strongly lensed background sources. Lensing of the CMB leads to a ~55% reduction in CMB power at the location of massive galaxy clusters in a spatially-matched single-frequency filter, leading to a net decrease in detected cluster number counts. We find that the increase in DSFG power and decrease in CMB power due to lensing at cluster locations largely cancel, such that the net effect on cluster number counts for current SZ surveys is sub-dominant to Poisson errors.
BibTeX
@article{hezaveh2013lensing,
  author = {Yashar Hezaveh and Keith Vanderlinde and G. P. Holder and T. de Haan},
  title = {{LENSING NOISE IN MILLIMETER-WAVE GALAXY CLUSTER SURVEYS}},
  journal = {The Astrophysical Journal},
  year = {2013},
  doi = {10.1088/0004-637x/772/2/121},
}

Loading related content...

Figure