Euclid Quick Data Release (Q1). The first Euclid view of Planck galaxy protocluster candidates at cosmic noon

A large catalogue of candidate galaxy protoclusters detected by Planck is being investigated using Euclid's first data release to understand their optical nature and evolutionary state at cosmic noon. This study utilizes Euclid's photometric data, combined with Spitzer photometry, to identify 20 counterparts to eight Planck protocluster candidates, revealing that these overdensities reside in dark matter halo mass ranges indicative of a transition between cold flows in hot media and the accretion of hot material.

Detection Methodology

The researchers employed a multi-step process to find the Euclid counterparts of the Planck candidates. The initial search involved using the DETECTIFz algorithm, an overdensity finder based on Delaunay tessellation that uses photometric redshift probability distributions through Monte Carlo simulations. This algorithm was applied to galaxies located less than 1° around each Planck protocluster candidate, searching for extended galaxy overdensities within specific redshift slices.

Key selection criteria included:

1. A stringent magnitude cut: "HE < 23.5, which was chosen to remain on the cautious side for the photometric redshift determination."

2. Source isolation: Sources had to be relatively isolated from bright stars and/or known remaining artefacts, selected by a spurious probability below 20%.

3. Redshift consistency: The photometric redshift estimated by NNPZ (Euclid's method) had to be within 1 σ error margin of the redshift estimated by NNPZ (Euclid's method).

4. Redshift error threshold: A constraint was imposed on the photometric redshift uncertainty: "δzph/(1+zph) < 0.20," as above this threshold, the detection was biased towards lower density values.

Counterpart Identification and Validation

The study successfully identified 20 Euclid counterparts corresponding to eight Planck protocluster candidates, with six candidates having multiple detections and two having a single counterpart. These detections were confirmed through cross-validation using two independent algorithms:

1. The Poisson Probability Method (PPM), which searches for high-z megaparsec-scale overdensities of galaxies around a given target using photometric redshift probability estimates.

2. The DIANAS detection strategy, which was specifically designed and fine-tuned for galaxy protoclusters in EWS.

The results showed that both alternative algorithms detected most (17 out of 20) structures at the same sky positions and photometric redshifts as those given by DETECTIFz. The purity of the final sample was assessed using simulated data, finding a mean purity of 95% when selecting overdensities with at least seven galaxy members.

Physical Characterization and Evolution

The physical properties of the 20 Euclid counterparts were quantified, including photometric redshifts (spanning 1.4 < zph < 2.7), total stellar masses, star-formation rates (SFRs), and halo mass lower limits.

(See Table 1 for detailed parameters.)

The mean values for these structures are:

(Stellar Mass: 1.3 × 1011M⊙; SFR: 300 M⊙ yr−1.)

The detected structures lie in the dark matter halo mass (Mh) / redshift plane within the region "12.6 < log10(Mh/M⊙) < 13.4, 1.4 < z < 2.7, suggesting that the halos are expected to have experienced a transition between cold flows in hot media to accretion of hot material."

The analysis of member galaxies indicated that SFRs remain within a factor 3 of the one expected for their stellar mass on the main sequence, meaning they are presenting at most a slightly enhanced star-formation, and none are clearly starbursting or quenching.

Discussion and Interpretation

The findings suggest that the Planck catalogue mainly contains protoclusters in their maturing phase. The structures detected in the Mh–z plane indicate that the accreted matter is hot and diffuse rather than channeled along cold streams inside the halo. This supports the hypothesis that these structures are not virialised, as calculated angular sizes were significantly smaller than the angular radius outputed by DETECTIFz, by about one order of magnitude.

Furthermore, a first search for X-ray or SZ clusters showed no association, suggesting that these halos do not contain gas hot and dense enough to emit in X-rays or cast shadows in the cosmic microwave background via the Sunyaev–Zeldovich effect. The paper concludes that the SFR is typical of galaxies that are active, but not starbursting, marking the onset of quenching driven by the stabilisation of the virial shock and the subsequent inability of the halo to retain cold inflows. Future observations are planned to obtain a clearer view as thousands of protoclusters will be detected in forthcoming Euclid DR1 data.

As a fastidious and diligent researcher, I have analyzed this manuscript, The first Euclid view of Planck galaxy protocluster candidates at cosmic noon. This paper provides a novel methodology for using the Euclid survey data to identify and characterize high-redshift protoclusters that were initially identified by Planck.

Here are the specific improvements that can be made to AI systems based on this scientific paper, and what those improved systems could achieve:

)

1. Identify Protocluster Candidates in Large-Scale Surveys:

High-level AI models can be trained on the detection algorithms described (DETECTIFz, PPM, DIANAS) and the selection criteria (photometric redshift distributions, S/N thresholds).

[Improved AI System Capability]: An autonomous pipeline capable of scanning massive survey datasets (like Euclid's) to automatically flag potential galaxy overdensities based on photometric redshift probability distributions, effectively performing blind protocluster discovery that mirrors the methodology in Section 3.2.

2. Quantify Physical Properties from Photometric Data:

The paper details how stellar masses and Star Formation Rates (SFRs) are derived using NNPZ and Kennicutt & Evans relations, even with uncertainties inherent in photometric redshift estimates.

[Improved AI System Capability]: A sophisticated regression model capable of taking multi-band photometric data (IE, YE, JE, HE) as input and predicting the physical properties of identified structures (Total Stellar Mass, Total SFR) with quantified uncertainty bounds derived from the NNPZ pipeline uncertainties described in Section 2.2.

3. Determine Halo Mass Estimates:

The system employs complex scaling relations (Method B) to estimate dark matter halo masses using stellar mass and literature-based relations (B13, L19, S22).

[Improved AI System Capability]: A physical modeling engine that can ingest member galaxy properties and apply hierarchical structure formation models (like those from Behroozi et al. 2013) to estimate the total dark matter halo mass of a candidate overdensity, providing multiple mass estimates (Method A vs. Method B) with quantified systematic biases and uncertainties as shown in Table 2.

4. Assess Evolutionary State and Quenching Mechanisms:

The analysis focuses on the location of objects in the Halo Mass–Redshift plane to infer physical processes like the transition between cold flows and hot accretion, leading to hypotheses about quenching (Section 6).

[Improved AI System Capability]: A predictive dynamical model that can map identified protoclusters onto cosmological evolution tracks. This system could predict whether a detected structure is in a maturing phase or approaching virialization, helping to identify the epoch of quenching driven by shock heating versus cold accretion, as suggested by the findings in Section 6.

5. Cross-Validation and Purity Assessment:

The paper rigorously tests its detection methods against independent algorithms (PPM, DIANAS) and uses simulations (GAEA lightcone) to calculate purity (95%) and completeness (86%).

[Improved AI System Capability]: A meta-validation module that can automatically cross-check results from multiple detection algorithms. It could quantify the confidence level of a detected object by comparing its performance across PPM, DIANAS, and DETECTIFz outputs, providing real-time purity metrics similar to the 95% purity reported in Section 5.3.

6. Handle Line-of-Sight Contamination (Merging Structures):

The paper explicitly addresses the possibility of multiple structures along the line of sight by analyzing compatible photometric redshifts and angular separations to infer merging events (Section 5.4.3).

[Improved AI System Capability]: A structure decomposition module that analyzes spatially close, statistically compatible detections to determine if they represent a single, larger protocluster or two distinct, merging components separated by physical distances comparable to cluster virial radii (e.g., 1–2 Mpc), refining the taxonomy of the detected structures.