Measurements of the A dependence and pseudorapidity interval (δη) dependence of midrapidity ET distributions in a half-azimuth (Δφ=π) electromagnetic calorimeter are presented for p+Be, p+Au, O+Cu, Si+Au, and Au+Au collisions at the BNL-AGS (Alternating-Gradient Synchrotron). The shapes of the upper edges of midrapidity ET distributions as a function of the pseudorapidity interval δη in the range 0.3 to 1.3, roughly centered at midrapidity, are observed to vary with δη, like multiplicity—the upper edges of the distributions flatten as δη is reduced. At the typical fixed upper percentiles of ET distributions used for nuclear geometry characterization by centrality definition—7 percentile, 4 percentile, 2 percentile, 1 percentile, 0.5 percentile—the effect of this variation in shape on the measured projectile Ap dependence for 16O, 28Si, 197Au projectiles on an Au target is small for the ranges of δη and percentile examined. The ET distributions for p+Au and p+Be change in shape with δη; but in each δη interval the shapes of the p+Au and p+Be distributions remain indentical with each other—a striking confirmation of the absence of multiple-collision effects at midrapidity at AGS energies. The validity of the nuclear geometry characterization versus δη is illustrated by plots of the ET(δη) distribution in each δη interval in units of the measured 〈ET(δη)〉p+Au in the same δη interval for p+Au collisions. These plots, in the physically meaningful units of “number of average p+Au collisions,” are nearly universal as a function of δη, confirming that the reaction dynamics for ET production at midrapidity at AGS energies is governed by the number of projectile participants and can be well characterized by measurements in apertures as small as Δφ=π, δη=0.3.
Forum