Recent data from RHIC for high- $p_T$ hadrons raised again the long-standing problem of quantitatively understanding the Cronin effect, i.e., nuclear enhancement of high- $p_T$ hadrons. All existing models for the Cronin effect rely on a fit to the data to be explained. We develop a phenomenological description based on the light-cone QCD-dipole approach which allows one to explain data without fitting to them and to provide predictions for $\mathrm{pA}$ collisions at RHIC and LHC. We point out that the underlying mechanism drastically changes with energy, from incoherent production of high- $p_T$ hadrons on different nucleons at low energies, to an entirely coherent process at very high energies.

Recent measurements by the BRAHMS Collaboration of high-$p_T$ hadron production at forward rapidities at the BNL Relativistic Heavy Ion Collider found the relative production rate $(d\text{−}\mathrm{Au})/(p\text{−}p)$ to be suppressed rather than enhanced. Examining other known reactions (forward production of light hadrons, the Drell-Yan process, heavy flavor production, etc.), one notes that all of these display a similar property, namely, their cross sections in nuclei are suppressed at large $x_F$. Since this is the region where $x_2$ is minimal, it is tempting to interpret this as a manifestation of coherence or of a color glass condensate, whereas it is actually a simple consequence of energy conservation and takes place even at low energies. We demonstrate that in all these reactions there is a common suppression mechanism that can be viewed, alternatively, as a consequence of a reduced survival probability for large rapidity gap processes in nuclei, a Sudakov suppression, an enhanced resolution of higher Fock states by nuclei, or an effective energy loss that rises linearly with energy. Our calculations agree with the data.