sSFR Thresholds in Galaxy Evolution

• sSFR is defined as the ratio of the star formation rate to stellar mass, categorizing galaxies into quenched, steady, and burst regimes.
• Empirical sSFR thresholds vary with redshift and mass, revealing key scaling relations and the influence of environment on galaxy evolution.
• Methodological consistency, including SFR tracer selection and stacking techniques, is crucial for robust sSFR measurements in modern surveys.

Specific star formation rate (sSFR) thresholds delineate core regimes of galaxy evolution, offering a quantitative framework for distinguishing between steady, burst-dominated, and quenched star-formation activity. Defined as the ratio of star formation rate (SFR) to stellar mass (M_*), sSFR traces the relative mass growth rate of star-forming systems. Its empirical thresholds, functional dependencies, and redshift evolution underpin much of the contemporary classification of galaxies across cosmic time, as constrained by multiwavelength surveys from $z \sim 0$ to the reionization era. The precise definition and application of these thresholds, along with their methodological and physical implications, are foundational in the interpretation of galaxy main-sequence structure, quenched fractions, and the transition to starburst activity.

1. Definitions and Functional Dependencies

The specific star formation rate is formally defined by

$\mathrm{sSFR}(z) = \frac{\mathrm{SFR}(z)}{M_*(z)}$

with typical units of Gyr⁻¹ (1 Gyr⁻¹ ≃ 10⁻⁹ yr⁻¹). This parameter encapsulates the star-formation activity normalized to accumulated stellar mass, rendering it insensitive to absolute SFR scaling and directly comparable across a wide mass and redshift range [(Karim et al., 2010); (Topping et al., 2022)].

Observed galaxy populations exhibit well-characterized, continuous relations of sSFR with both redshift ($z$) and stellar mass ($M_*$):

• Redshift dependence: Empirically, $\langle \mathrm{sSFR}\rangle$ at fixed mass typically follows a power-law in $(1+z)$:

$\mathrm{sSFR}(z) \propto (1+z)^n$

with $n$ varying from $n \simeq 3.5$ for star-forming (SF) galaxies to $n \simeq 4.3$ for all galaxies in the COSMOS field for $\mathrm{sSFR}(z) = \frac{\mathrm{SFR}(z)}{M_*(z)}$0 (Karim et al., 2010). For UV-bright galaxies at $\mathrm{sSFR}(z) = \frac{\mathrm{SFR}(z)}{M_*(z)}$1, the best-fit exponent is shallower, $\mathrm{sSFR}(z) = \frac{\mathrm{SFR}(z)}{M_*(z)}$2 (Topping et al., 2022).

• Stellar mass dependence: At fixed $\mathrm{sSFR}(z) = \frac{\mathrm{SFR}(z)}{M_*(z)}$3, the average sSFR for SF galaxies declines with mass as

$\mathrm{sSFR}(z) = \frac{\mathrm{SFR}(z)}{M_*(z)}$4

where $\mathrm{sSFR}(z) = \frac{\mathrm{SFR}(z)}{M_*(z)}$5 (Karim et al., 2010). For mass-selected samples, the slope steepens to $\mathrm{sSFR}(z) = \frac{\mathrm{SFR}(z)}{M_*(z)}$6 due to the inclusion of quiescent galaxies.

2. Empirical sSFR Thresholds and Binning Schemes

A precise thresholding scheme enables stratification of galaxy populations by star-forming activity regime, supporting main-sequence discrimination and facilitating the study of environmental effects and quenching. The thresholds and bins vary by redshift, survey wavelength, and sample selection:

sSFR Range [units]	Regime (z context)	Source
$\mathrm{sSFR}(z) = \frac{\mathrm{SFR}(z)}{M_*(z)}$7 0.1 Gyr⁻¹	Quenched (z~0)	(Weinmann et al., 2011)
2 Gyr⁻¹	Plateau (2 $\mathrm{sSFR}(z) = \frac{\mathrm{SFR}(z)}{M_*(z)}$8 z $\mathrm{sSFR}(z) = \frac{\mathrm{SFR}(z)}{M_*(z)}$9 7)	(Weinmann et al., 2011)
5–15 Gyr⁻¹	Steady/main-sequence (z~7)	(Topping et al., 2022)
15–30 Gyr⁻¹	Typical UV-bright (z~7)	(Topping et al., 2022)
30–100 Gyr⁻¹	Starburst	(Topping et al., 2022)
$z$0 100 Gyr⁻¹	Extreme outlier	(Topping et al., 2022)

In the low-redshift universe ($z$1), sSFR is commonly binned logarithmically:

• Main sequence: log sSFR (yr⁻¹) $z$2 –9 ($z$3 Gyr⁻¹).
• Star-forming: log sSFR $z$4.
• Passive: log sSFR $z$5 (Kim et al., 2015). Upper envelopes (extreme starburst) correspond to log sSFR $z$6.

3. Physical Interpretation and Upper Limits

The physical origin of sSFR thresholds is closely tied to dynamical and gas-regulation processes:

• Dynamical ceiling: At $z$7 and for $z$8, empirical sSFR relations flatten at $z$9 Gyr⁻¹, a value associated with the inverse disk dynamical time ($M_*$0); this upper bound reflects the gravitational regulation of star formation efficiency (Karim et al., 2010).
• High-redshift plateau: $M_*$1 UV-selected galaxies display an sSFR plateau at $M_*$22 Gyr⁻¹, distinct from both theoretical expectations based on dark matter accretion $M_*$3 and the local universe, where typical sSFRs are an order of magnitude lower (Weinmann et al., 2011).
• Burst-dominated regime: At $M_*$4, ALMA REBELS finds median sSFR values of $M_*$5–$M_*$6 Gyr⁻¹, with the burst-dominated tail extending up to $M_*$7100 Gyr⁻¹ (Topping et al., 2022).

4. Methodological Considerations and Systematics

Robust determination of sSFR thresholds requires consistent methodologies for SFR and $M_*$8 estimation:

• SFR measurements: Incorporation of obscured SFR via FIR continuum (ALMA) systematically increases sSFR values by $M_*$90.3 dex relative to UV+optical SED-only techniques (Topping et al., 2022).
• Stellar mass estimates: Non-parametric SFHs yield higher $\langle \mathrm{sSFR}\rangle$0 and thus lower sSFR ($\langle \mathrm{sSFR}\rangle$1 dex; $\langle \mathrm{sSFR}\rangle$2 dex) compared to constant-SFH fits (Topping et al., 2022).
• Sample completeness: Application of mass-completeness constraints is critical. Resulting sSFR values below these limits should be regarded as upper limits due to Malmquist bias (Karim et al., 2010).
• Stacking techniques: When individual SFR detections are sparse in dust-unbiased tracers (e.g., FIR, radio), median stacking with noise-weighting is recommended (Karim et al., 2010).

5. sSFR Thresholds and Environment: Halo Mass and Clustering

sSFR-based bins reveal strong environmental trends:

• Low-$\langle \mathrm{sSFR}\rangle$3 clustering: At $\langle \mathrm{sSFR}\rangle$4, main-sequence (log sSFR $\langle \mathrm{sSFR}\rangle$5) galaxies preferentially reside in $\langle \mathrm{sSFR}\rangle$6 $\langle \mathrm{sSFR}\rangle$7 haloes as centrals, exhibiting the lowest clustering amplitude. Both high-sSFR ("starbursts", log sSFR $\langle \mathrm{sSFR}\rangle$8) and low-sSFR galaxies inhabit more massive ($\langle \mathrm{sSFR}\rangle$9 $(1+z)$0) haloes, commonly as satellites (Kim et al., 2015). This U-shaped trend in bias and halo mass with sSFR persists across stellar-mass sub-bins.
• Passive transition: log sSFR $(1+z)$1 galaxies (passive) mostly occupy haloes with masses similar to, or slightly above, the main sequence, indicating environmental quenching predominantly operates in group-scale halos.

6. Redshift Evolution and Theoretical Tension

The redshift evolution of sSFR thresholds reveals substantive theoretical challenges:

• Plateau and decline: The observation of a prolonged sSFR plateau at $(1+z)$2 Gyr⁻¹ for $(1+z)$3, with a sharp drop at lower $(1+z)$4, is at odds with expectations from cosmological accretion. The predicted rise in sSFR is steeper than observed, prompting refinements in semi-analytic models:
  • Suppressed star formation efficiency at $(1+z)$5.
  • Enhanced feedback (mass loading factors $(1+z)$6–$(1+z)$7).
  • Delayed gas reincorporation to sustain high sSFR at $(1+z)$8.
  • Accelerated growth of massive galaxies via mergers or starbursts (Weinmann et al., 2011).

A consensus is that successful models require a composite of rapid early mass assembly, suppressed SFR at $(1+z)$9, delayed star formation post-feedback, and enhanced late starbursts to reproduce the observed plateau (Weinmann et al., 2011).

7. Practical Application and Classification in Modern Surveys

sSFR thresholds are integrated into contemporary survey classification schemes:

• Color–color separation: Rather than hard sSFR cuts, rest-frame (NUV–r) color (e.g., $\mathrm{sSFR}(z) \propto (1+z)^n$0 for star-forming) is used to mitigate dust contamination up to $\mathrm{sSFR}(z) \propto (1+z)^n$1 (Karim et al., 2010).
• Main-sequence identification: sSFR–mass scaling relations, with thresholds derived from power-law fits, demarcate the main-sequence for mass-selected samples at all accessible redshifts (Karim et al., 2010).
• High-redshift categorization:
  • At $\mathrm{sSFR}(z) \propto (1+z)^n$2, galaxies with $\mathrm{sSFR}(z) \propto (1+z)^n$3 Gyr⁻¹ are classified as steady, main-sequence systems.
  • sSFRs of $\mathrm{sSFR}(z) \propto (1+z)^n$4 Gyr⁻¹ typify UV-bright or "bursty" galaxies.
  • sSFR $\mathrm{sSFR}(z) \propto (1+z)^n$5 Gyr⁻¹ indicate extreme, possibly outlier, starburst episodes, with hidden stellar populations implied by spectral modeling (Topping et al., 2022).

These empirically-derived thresholds and scaling relations are directly portable to other mass-selected, multiwavelength galaxy samples, provided completeness, SFR tracers, and classification methodologies are rigorously controlled. Maintaining internal consistency in sSFR definitions and thresholding is essential to mitigating selection biases and ensuring robust cosmic star-formation history mapping across epochs.