Answers are grounded in this paper and cite their sources.
Demonstrates causal mechanisms in turbulent flows, highlighting the need for intervention in analysis.
Information-theoretic tools are now standard for inferring causality in turbulent flows, and the synergistic component of causality decompositions is widely read as evidence of joint, nonlinear mechanisms. We prove that in the regime where such analyses operate this reading is unavailable: whenever one source dominates the target state by state, a checkable condition that held in every dataset of this study, the synergy component equals the classical transfer entropy exactly, and is therefore identical for additive and multiplicative couplings. We verify the identity to machine precision on the SURD method's own benchmark systems, where the published synergies survive full linearization, and on its published application data. We then supply what observation cannot: intervention. In Rayleigh-Bénard convection with temperature-dependent viscosity, in-solver causal surgery deletes the fluctuating coupling while preserving the mean state, removing the coupling's beyond-linear information while sparing the linear channel, whereas a linear buoyancy control reproduces the observational signal in full. Attributing information flow to a physical coupling requires intervention.
No takes yet. Share an insight, caveat, or question.
tafakor et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: