Speaker
Description
Long-range correlations observed in high-multiplicity p+p and p+Pb collisions exhibit features similar to those measured in heavy-ion collisions, where they are commonly associated with hydrodynamic collective response. Before extending this interpretation to smaller systems, however, correlations that can arise without hydrodynamic evolution must be quantified. Exact global conservation provides an important contribution, whose relative importance increases toward lower multiplicity. We develop a conditional-probability framework that yields the leading finite-system correlations induced by exact additive conservation laws and use its structure to construct observables orthogonal to selected conserved quantities. Rather than calculating and subtracting a conservation background, the observable weights are projected such that their leading conservation-induced long-range covariance vanishes by construction, without requiring full phase-space coverage. We validate the method through a controlled closure test using PYTHIA8/Angantyr p+Pb events at $\sqrt{s_{\rm NN}}=5.02$ TeV. Projection against transverse-momentum conservation removes the expected first-harmonic recoil correlation while leaving the second harmonic and mean-$p_T$ covariance essentially unchanged. We further show that the long-range baryon-number covariance is removed only when baryon number is included among the conserved charges used in the projection. The method therefore provides a direct way to isolate conservation-aligned components of long-range correlations, particularly in small and low-multiplicity collision systems.