Interstellar propagation of cosmic rays: analysis of the ULYSSES primary and secondary elemental abundances.

Duvernois, M. A.; Simpson, J. A.; Thayer, M. R.

United States

Abstract

The purpose of this investigation is to study the validity of the "leaky box" propagation model and determine some of its parameters. The pathlength distributions and their energy dependencies are derived from measurements of secondary-to-primary abundance ratios. A special charged particle telescope carried on the Ulysses spacecraft has provided a complete separation of elements from hydrogen to nickel in the energy range ~40 to ~400MeV/nucleon in a time interval sufficient to analyze more than 4000 iron nuclei. The secondary to primary ratios of elemental abundances reported here are B/C, F/Ne, P/S, Sc/Fe, Ti/Fe, V/Fe, Cr/Fe and Mn/Fe. Two cases of the leaky box model were analyzed: 1) a single exponential, energy dependent, pathlength distribution, and 2) a double exponential energy dependent pathlength distribution. During the measurement period (1990-1995) the heliospheric level of solar modulation changed over more than a factor of two and was taken into account using a spherically symmetric model of solar modulation. The primary abundances used in our analysis are those derived from the Ulysses High Energy Telescope results obtained by DuVemois & Thayer (1996). These abundances are consistent with solar system abundances after correction for the first ionization potential (FIP) bias. Only the single exponential pathlength distribution satisfied all the measurements from boron to iron with a mean pathlength decreasing with decreasing energy below ~1GeV/nucleon. The double exponential pathlength distribution predicted much higher abundances of secondary elements than permitted by the measurements. No depletion of short path-lengths was required at any energy for the Sc-Mn or Fe nuclei. It is not certain whether the results reported here can accommodate any models of continual acceleration or reacceleration in the interstellar medium. At most, these models would have to have a very low level of acceleration.

1996 Astronomy and Astrophysics
Ulysses 51