The role of velocity dispersion in the Kennicutt-Schmidt relation

The resolved Kennicutt-Schmidt (KS) relation describes how star formation surface density relates to molecular gas surface density in galaxies, but its significant scatter suggests that physical parameters beyond gas density are influencing star formation efficiency. This study investigates whether molecular gas velocity dispersion acts as a hidden parameter in this relation, revealing a systematic dependence for a subset of galaxies.

Key Findings on Velocity Dispersion

The research identifies two distinct groups of galaxies based on the direction of their molecular gas velocity dispersion (∆v) gradient relative to the best-fit KS line: group A, where the gradient is completely aligned with the KS line and does not contribute to scatter, and group B, where it has a component perpendicular to the line and contributes to scatter. For group B galaxies, a weak but statistically significant correlation between Res [residuals] and ∆v was found across all surveys. Specifically, in group B, the residuals show a systematic decrease with log (∆v), which we quantify by (least-squares) fitting a linear relation of the form Res = a0 log (∆v) + b0. This indicates that at fixed Σmol, regions with higher ∆v exhibit systematically lower ΣSFR, suggesting that velocity dispersion acts as an additional hidden parameter modulating star formation efficiency.

Methodology for Classification and Fitting

The study utilized spatially resolved measurements of ΣSFR, Σmol, and ∆v from the ALMaQUEST, EDGE–CALIFA, and PHANGS–ALMA surveys. To objectively classify galaxies into group A or B, a three-dimensional fit including ∆v was performed using an unsupervised Gaussian-mixture model (GMM) applied to combined data from ALMaQUEST and EDGE–CALIFA. This classification relied on three multiscale diagnostics (features): S3, S4, and S5, constructed from the strongest local scores derived by fitting residuals against log(Σmol) at different binning resolutions. The component with the largest score was defined as group B because it contains the galaxies with the strongest and most persistent negative residual–∆v slopes.

Quantifying the Extended Relation

For group B galaxies, a three-dimensional fit including ∆v yielded an extended KS relation: ΣSFR ∝ Σmol1.15 mol3 −0.59. This result was consistent across surveys, with the fitted slope coefficient 'a' being similar between ALMaQUEST and EDGE–CALIFA. The inclusion of ∆v resulted in only a marginal reduction of the overall scatter (σ) relative to the classical KS relation, as the new variance satisfied σ2new = σ2KS − σ2KS,∆v. The final combined scaling relation is expressed as ΣSFR ∝ Σmol1.15 mol3 −0.59.

Consistency and Interpretation

The similarity of fitted coefficients between surveys strongly supports the conclusion that ∆v plays a role in regulating star formation at fixed Σmol in certain galaxies. Despite this coherent trend, the authors found no single property—such as ∆v distribution or AGN content—that clearly differentiates group B from group A. While mergers are twice as frequent in group B compared to group A in the ALMaQUEST sample, the similarity in ∆v distributions suggests this difference does not reflect systematically higher velocity dispersions. The physical origin of this behavior remains uncertain, but it points toward a distinction between gas motions that suppress star formation and those that do not.

Constraints and Limitations

The study addressed potential confounding factors by performing tests to rule out selection effects; a permutation null test showed that the observed group B signal is unlikely to arise from random fluctuations alone. Furthermore, analysis of galaxy inclination revealed no systematic dependence on cosi, suggesting the observed trend is not driven by viewing angle. The authors also noted that the measured velocity dispersion at 1 kpc may reflect cloud–cloud motions rather than internal turbulence, and that uncertainties in extinction corrections are mitigated because group B behavior is found throughout the range of ΣSFR. The final scaling relation for group B is presented as log (ΣSFR) = n log (Σmol) + a log (∆v) + C, with the negative exponent 'a' confirming that ΣSFR decreases with increasing ∆v at fixed Σmol. The similarity in these coefficients across surveys provides a strong indication that ∆v plays a role in regulating star formation. The final derived scaling relation is written as ΣSFR ∝ Σmol1.15 mol3 −0.59.

The gist

For group B galaxies, the residuals of the KS relation systematically decrease with increasing log (∆v), indicating that velocity dispersion acts as a secondary parameter regulating star formation efficiency at fixed molecular gas surface density.

How it works

1. The researchers utilized spatially resolved measurements of ΣSFR, Σmol, and ∆v from ALMaQUEST, EDGE–CALIFA, PHANGS–ALMA surveys, and M51.

Based on the provided scientific paper, here are specific improvements that could be made to AI systems, focusing on leveraging the findings related to galaxy evolution and interstellar medium (ISM) physics:

The following improvements focus on integrating the physical constraints derived from the Kennicutt-Schmidt relation (KS) and velocity dispersion measurements into machine learning models for astrophysical data analysis.

1. Enhance Galaxy Classification Models with Kinematic Signatures

2. Develop Predictive Models for Star Formation Efficiency (SFE) based on Resolved Kinematics

3. Improve Parameter Estimation in Galaxy Surveys using Multi-Parameter Regression

A specific breakdown of what these improved AI systems can do:

1. Enhance Galaxy Classification Models with Kinematic Signatures

2. Instead of relying solely on photometric or simple surface density features, train supervised learning models (e.g., Gaussian Mixture Models, as used in the paper) to classify galaxies into Group A and Group B based on their derived kinematic signatures (the three-scale features: S3, S4, S5).

3. This improved system can distinguish between galaxies where velocity dispersion acts as a neutral parameter (Group A) versus those where it actively suppresses star formation efficiency (Group B).

4. Develop Predictive Models for Star Formation Efficiency (SFE) based on Resolved Kinematics

5. Create regression models that explicitly incorporate molecular gas velocity dispersion as an input feature, allowing the model to predict the star formation surface density (ΣSFR) given fixed molecular gas surface density (Σmol) and a specific kinematic state (Group A or B).

6. This system can quantify the hidden parameter effect: for any galaxy, it can predict if its SFE will be lower than predicted by the standard KS relation based on its measured velocity dispersion, providing a quantitative measure of this suppression.

7. Improve Parameter Estimation in Galaxy Surveys using Multi-Parameter Regression

8. Implement a hierarchical or multi-level regression framework that simultaneously fits the global KS relation and an extended relation (as derived in Section 7: log ΣSFR = n log Σmol + a log ∆v + C).

9. This system can robustly estimate the physical parameters of galaxy populations by accounting for the known hidden parameter effect, leading to more accurate determinations of the scaling exponents (like 'a' or '-0.59' in Equation 2) across different observational surveys (ALMaQUEST, EDGE–CALIFA).

10. The improved system can perform cross-survey validation by comparing the estimated coefficients from different surveys, identifying systematic biases or survey-specific effects in the underlying physics.

The resolved Kennicutt-Schmidt (KS) relation is a power-law relation between the surface densities of the star formation rate (Σ SFR) and molecular gas (Σ mol) in star-forming galaxies on kiloparsec scales. Despite its apparent simplicity, it exhibits substantial scatter, suggesting the influence of additional physical parameters beyond gas surface density. We investigate whether molecular gas velocity dispersion (Δv) acts as a hidden parameter in the resolved KS relation in nearby galaxies. We use spatially resolved measurements of Σ SFR, Σ mol, and Δv from the ALMaQUEST, EDGE-CALIFA, and PHANGS-ALMA surveys and from observations of the single galaxy M51. We identify two groups of galaxies based on the direction of the Δv gradient with respect to the best-fit KS line: one in which the Δv gradient is completely aligned with the KS line (group A) and thus does not contribute to the scatter, and one in which it has a component perpendicular to the line (group B) and contributes to the scatter. For each group, we fit the KS relation and examine the dependence of the residuals (Res) of the fit on Δv. In group B, we find a weak but statistically significant correlation between Res and Δv, which is consistent across all surveys. A three-dimensional fit including Δv yields Σ SFR proportional to Σ mol 1.15 Δv-0.59. We conclude that there is a set of star-forming galaxies for which the velocity dispersion acts as a secondary parameter in the KS relation, with higher Δv corresponding to lower Σ SFR at fixed Σ mol. We find no obvious correlation between membership in this set and the other properties examined.