Star Formation Scaling Relations

• Star formation scaling relations are empirical and theoretical frameworks that quantify how star formation rates relate to gas, stellar, and environmental properties in galaxies.
• Observations and models reveal near invariant power-law slopes in the SFMS and KS law, with deviations driven by dense gas fraction, gravitational instability, and feedback processes.
• These relations provide key diagnostics for galaxy evolution by constraining baryonic cycles, star formation efficiency, and the impacts of environmental conditions.

Star formation scaling relations are empirical and theoretical frameworks that describe how the global and local rates of star formation in galaxies relate to the underlying gas and stellar properties across a wide range of cosmic environments, galaxy types, spatial scales, and evolutionary stages. These relations are central to galaxy evolution theory and serve as key diagnostics for the regulation of baryonic cycles, the efficiency of star-forming processes, and the interplay between gas reservoirs, feedback, and environmental factors.

1. Classical Empirical Relations

Star-Forming Main Sequence (SFMS)

The SFMS captures the scaling between integrated star formation rate (SFR) and stellar mass ($M_*$) in star-forming galaxies: $\log\mathrm{SFR} = \alpha\,\log M_* + \beta$ Canonical values for main-sequence slope and intercept are $\alpha\sim 0.7-0.8$, $\beta\sim -6.6$ to $-7.9$ at $z\approx1-3$ with an intrinsic scatter of 0.2–0.4 dex, nearly invariant up to $z\sim5$ for $M_*\gtrsim 10^9\,M_\odot$ (Mérida et al., 26 Sep 2025). At lower masses ($M_*<10^{9.5}\,M_\odot$), there is increasing evidence for a turnover and steepening of the SFMS, consistent with bursty star formation and feedback-limited evolution in dwarf systems (Mérida et al., 26 Sep 2025, Hunt et al., 2020).

Kennicutt–Schmidt (KS) Law

The classical KS law parameterizes the relation between SFR surface density and gas (usually H$_2$ or total gas) surface density: $\Sigma_\mathrm{SFR} = A\,\Sigma_\mathrm{gas}^N$ Empirical fits yield $N\sim1.4\pm0.15$, $A\sim 2.5\times10^{-4}$ ($M_\odot$ yr$^{-1}$ kpc$^{-2}$ at $\Sigma_\mathrm{gas}=1\,M_\odot\,$pc$^{-2}$) (Lada, 2014, Christie et al., 24 Sep 2025). However, resolved observations within galaxies reveal that $\Sigma_\mathrm{SFR}$ correlates nearly linearly with $\Sigma_{\mathrm{H}_2}$ (N~1.0) in molecule-dominated regions, while atomic-dominated regions are offset to longer depletion times by an order of magnitude, with the relation steepening in starburst and merger environments (Bigiel et al., 2010, Dou et al., 2020).

2. Dense Gas and Molecular Scaling Laws

A fundamental revision of the KS paradigm recognizes the tight, linear scaling between SFR and the dense molecular gas mass: $$\mathrm{SFR} = k\,M_\mathrm{dense},\quad k\approx 4.5\times10^{-8}\,\mathrm{yr}^{-1}$$ This relation holds over nine decades in mass and unifies diverse environments from Galactic molecular clumps to high-$z$ ULIRGs (Lada et al., 2011, Lada, 2014). The dense-gas fraction ($f_\mathrm{DG}=M_\mathrm{dense}/M_\mathrm{mol}$) governs the offset in total-gas relations, with starbursts ($f_\mathrm{DG}\to1$) displaying much higher SFE compared to disks ($f_\mathrm{DG}\ll1$) (Lada et al., 2011). The observed ($\Sigma_\mathrm{SFR},\Sigma_\mathrm{gas}$) locus emerges as a superposition of linear tracks with $N=1$, offset by $f_\mathrm{DG}$.

3. Physical Drivers and Interpretation

Surface Density Thresholds and Gravitational Instability

Star formation is directly regulated by the ability of the gas to reach critical surface densities for gravitational instability (Toomre $Q<1$) and to assemble cold, dense molecular phases. In low surface brightness (LSB) galaxies, gas surface densities $\Sigma_\mathrm{HI}\sim3$–$5\,M_\odot\,\mathrm{pc}^{-2}$ keep $Q>1$, suppressing collapse and molecular cloud formation (Christie et al., 24 Sep 2025). The ISM midplane pressure sets the H$_2$ fraction via the scaling $R_\mathrm{mol} \propto (P_\mathrm{ext}/P_0)^\alpha$.

Star Formation Efficiency and Gas Depletion Time

The global efficiency per unit molecular gas mass (SFE) sets the gas depletion time $t_\mathrm{dep}=M_\mathrm{gas}/\mathrm{SFR}$. In normal spirals, $t_\mathrm{dep}(\mathrm{H}_2)\sim2$ Gyr and is weakly dependent on $M_*$, while LSB disks exhibit $t_\mathrm{dep}\sim10$–$30$ Gyr, reflecting low star formation efficiency (Christie et al., 24 Sep 2025, Dou et al., 2020). In parsec-scale cluster-forming clumps, depletion times drop to $\sim100$ Myr and the instantaneous $\epsilon_\mathrm{ff}\sim 0.13$—an order of magnitude above the extragalactic average (Rawat et al., 15 May 2025).

Scale Dependence and Breakdown of Universality

Power-law slopes for $SFR$–$M_\mathrm{gas}$ and $SFR$–$(M_\mathrm{gas}/t_\mathrm{ff})$ relations are remarkably invariant (N~1.4–1.5; N'~0.8–1.0) over scales from kpc disks to $\sim$1 pc clumps, but the normalization (i.e., star formation efficiency per free-fall time) increases by more than an order of magnitude for smaller, denser, more actively collapsing systems (Rawat et al., 15 May 2025, Das et al., 2020). Thus, scaling relations are not truly universal in normalization but are modulated by local gas dynamics, feedback state, and environment.

4. Multidimensional and "Fundamental Plane" Relations

Three-Parameter Fundamental Plane

A combined analysis of stellar mass, SFR, and gas-phase metallicity (O/H) demonstrates that galaxies occupy a thin, redshift-independent "fundamental plane" in $(M_*,\,\mathrm{SFR},\,\mathrm{O/H})$ space: $$12+\log(\mathrm{O/H}) = -0.12\,\log(\mathrm{SFR}) + 0.33\,\log(M_*) + 5.20$$ with an rms dispersion $\lesssim 0.17$ dex for $M_*\lesssim3\times10^{10}\,M_\odot$ (Hunt et al., 2012). This FP unites the SFMS and the mass–metallicity relation and includes outlier populations such as low-metallicity starbursts, which deviate from 2D projections but lie on the same 3D plane, explained by different star-formation modes (active starburst versus passive quiescent) (Magrini et al., 2012).

Fundamental Formation Relation (FFR)

A nearly scatter-free relation links sSFR, gas-to-stellar mass ratio ($\mu=M_\mathrm{H_2}/M_*$), and molecular SFE: $\mathrm{sSFR} = \mu \times \mathrm{SFE}$ with tight correlations in the $\mu$–sSFR and SFE–sSFR planes (scatter $\sim0.19$–0.20 dex), more fundamental than the classical SFR–$M_*$ or $M_\mathrm{gas}$–SFR relations. The FFR provides a minimal description of star formation and quenching across the full range of galaxy properties (Dou et al., 2020).

5. Spatially Resolved and Environmental Variations

Resolved Scaling Laws

On $\sim100$–$1000$ pc scales, resolved versions of the SFMS (rSFMS), KS law (rKS), and molecular gas main sequence (rMGMS) are recovered with near-unity slopes:

• rSFMS: $a\sim1.0$–$1.1$
• rKS: $a\sim1.0$–$1.1$
• rMGMS: $a\sim1.2$

The tightest relation at 100 pc is the rMGMS (scatter $\sim0.34$ dex), followed by rKS ($\sim0.41$ dex) and rSFMS ($\sim0.51$ dex), reflecting the timescales and coupling between these quantities (Pessa et al., 2021). At coarser scales, scatter decreases further, with the rKS maintaining the tightest and most environment-independent correlation.

Variation across Galaxy Types and Environments

Comparisons between green valley and main sequence galaxies reveal systematic offsets in sSFR, SFE, and molecular gas fraction at kpc scales, indicative of simultaneous depletion of molecular gas and reduction in SFE across both bulge and disk, consistent with global (as opposed to purely central) quenching (Lin et al., 2022). LSB galaxies follow the same SFMS and atomic gas-to-stellar relations as their high surface brightness counterparts but are poorly efficient at converting HI reservoirs to stars due to high Toomre $Q$, low midplane pressures, and low molecular fractions (Christie et al., 24 Sep 2025).

6. Size–Luminosity and Turbulence-Dependent Relations

H II regions and star-forming clumps follow distinct size–luminosity scaling depending on SFR surface density:

• Low $Σ_\mathrm{SFR}$: $L_{H\alpha}\propto r^{2.77}$
• High $Σ_\mathrm{SFR}$: $L_{H\alpha}\propto r^{1.74}$

This transition, at $$Σ_\mathrm{SFR}\sim1\,M_\odot\,\mathrm{yr}^{-1}\,\mathrm{kpc}^{-2}$$, reflects a change from ionization-bounded (volume-limited) to density- or surface-bounded (disk-thickness–limited) regimes, tightly connected to turbulence and feedback regulation in the ISM (Cosens et al., 2018, Wisnioski et al., 2012).

7. Limitations, Open Questions, and Theoretical Implications

Current star formation "laws" are fundamentally relations between the amount of cold, dense gas and the star formation activity, but their normalization is sensitive to local physical conditions, sub-kiloparsec ISM structure, and evolutionary phase. The primary sources of scatter and systematic deviation are uncertainties in gas tracers (e.g., CO-to-H$_2$ conversion in low metallicity), variations in the dense gas fraction, and the effects of feedback, turbulence, and dynamical environment (Das et al., 2020, Rawat et al., 15 May 2025).

A comprehensive theoretical framework must therefore not only reproduce the empirically tight power-law scalings across spatial and mass scales but also explain the environmental dependence of star formation efficiency, dense-gas assembly, and depletion times. The scaling relations provide stringent observational benchmarks for models of baryonic cycling, cosmic star-formation history, and galaxy quenching processes.

---

Relation	Power-Law Slope (N)	Scatter (dex)	Characteristic Scale/Regime
SFMS (global)	0.7–0.8 (up to $z\sim5$)	0.2–0.4	$M_*\gtrsim10^9M_\odot$, integrated
SFMS (LSB)	0.59	0.30	LSB galaxies (Christie et al., 24 Sep 2025)
KS (gas–SFR)	1.4 (canonical); 1.0 (H$_2$-dominated)	0.2–0.3	$>750$ pc, normal spirals (Bigiel et al., 2010)
Dense-gas SFR	1.0	0.1–0.2	Milky Way to ULIRGs (Lada et al., 2011)
Clump-scale $\Sigma_\mathrm{SFR}$–$\Sigma_\mathrm{gas}$	1.46	$\sim0.28$	1 pc clumps (Rawat et al., 15 May 2025)
Size–Luminosity ($L$–$r$)	2.8 (low $\Sigma_\mathrm{SFR}$); 1.7 (high $\Sigma_\mathrm{SFR}$)	$<0.1$	HII, clumps (50 pc–3 kpc) (Cosens et al., 2018)

All quantitative and qualitative claims are directly extracted from the referenced literature (Christie et al., 24 Sep 2025, Rawat et al., 15 May 2025, Cosens et al., 2018, Lada et al., 2011, Bigiel et al., 2010, Dou et al., 2020, Hunt et al., 2012, Wisnioski et al., 2012, Pessa et al., 2021, Lin et al., 2022, Hunt et al., 2020, Magrini et al., 2012, Mérida et al., 26 Sep 2025, Lada, 2014).