Recent observations of r-process-enriched metal-poor star abundances reveal a non-uniform abundance pattern for elements $Z\leq47$. Based on non-correlation trends between elemental abundances as a function of Eu-richness in a large sample of metal-poor stars, it is shown that the mixing of a consistent and robust light element primary process (LEPP) and the r-process pattern found in r-II metal-poor stars explains such apparent non-uniformity. Furthermore, we derive the abundance pattern of the LEPP from observation and show that it is consistent with a missing component in the solar abundances when using a recent s-process model. As the astrophysical site of the LEPP is not known, we explore the possibility of a neutron capture process within a site-independent approach. It is suggested that scenarios with neutron densities $n_{n}\leq10^{13}$ $cm^{-3}$ or in the range $n_{n}\geq10^{24}$ $cm^{-3}$ best explain the observations.Comment: 28 pages, 7 Postscript figures. To be published in The Astrophysical Journa

We present first comparisons between Light Element Primary Process (LEPP) abundances observed in some ultra metal poor (UMP) stars and nucleosynthesis calculations based on long-time hydrodynamical simulations of core-collapse supernovae and their neutrino-driven wind. UMP star observations indicate Z≥38 elements include the contributions of at least two nucleosynthesis components: r-process nuclei that are synthesized by rapid neutron capture in a yet unknown site and LEPP elements (mainly Sr, Y, Zr). We show that neutrino-driven wind simulations can explain the observed LEPP pattern. We explore in detail the sensitivity of the calculated abundances to the electron fraction, which is a key nucleosynthesis parameter but poorly known due to uncertainties in neutrino interactions and transport. Our results show that the observed LEPP pattern can be reproduced in proton-and neutron-rich winds.