Astron. Astrophys. 347, 572-582 (1999)

2. The NACRE compilation in a nutshell

A detailed information about the procedure adopted to evaluate each of the NACRE reaction rates and about the derived values can be found in Angulo et al. (1999), or in electronic form at http://astro.ulb.ac.be , which also offers the possibility of generating interactively tables of reaction rates for networks and temperature grids selected by the user ¹. It is clearly impossible to go here into the details of the NACRE procedure. Let us just emphasize some of its specificities:

(1) For each reaction, the non-resonant and broad-resonance contributions to its rate are evaluated numerically in order to avoid the approximations which are classically made (see Fowler et al. 1975 for details) in order to allow analytical rate evaluations;

(2) Narrow or subthreshold resonances are in general approximated by Breit-Wigner shapes, and their contributions to the reaction rates are approximated in the usual analytical way (e.g. Fowler et al. 1975). However, in some cases, the resonance data are abundant enough to allow a numerical calculation avoiding these approximations;

(3) For each reaction, NACRE provides a recommended "adopted" rate, along with realistic lower and upper limits. The adopted values of, and the limits on the resonance contributions are derived from weighted averages duly taking into account the uncertainties on individual measurements, as well as the different measurements that are sometimes available for a given resonance [see Eq. (15) of Angulo et al. 1999]. For non-resonant contributions, -fits to available data provide the recommended values along with the lower and upper limits, as the experimental uncertainties on one set of data and the differences between various sets, if available, are taken into account in the -procedure. It is worth stressing at this point that enough information is provided by NACRE for helping the user to tailor his own preferred rates if he wants.

The procedure just sketched in (1) - (3) is the selected standard methodology, and has the advantage of being easily reproducible and of avoiding any subjective renormalization of different experimental data sets.Quite clearly, however, the large variety of different situations makes unavoidable some slight modifications of the standard procedure in some cases. These specific adjustments are clearly identified and discussed in Angulo et al. (1999);

(4) A theoretical (Hauser-Feshbach) evaluation of the contribution to each rate of the thermally populated excited states of the target is also provided. It has to be noted that the widely used compilation of Caughlan & Fowler (1988, hereafter referred to as CF88) provides uncertainties for some rates only, while the contribution of excited target states is derived in most cases from a rough (referred to as "equal strength") approximation;

(5) It has to be emphasized that the major goal of the NACRE compilation is to provide numerical reaction rates in tabular form (see http://astro.ulb.ac.be ). This philosophy differs markedly from the one promoted by the previous widely used compilations (CF88, and references therein), and is expected to lead to more accurate rate data. However, for completeness, NACRE also provides analytical approximations (Angulo et al. 1999) that differ in several respects from the classically used expressions (CF88, and references therein).

© European Southern Observatory (ESO) 1999

Online publication: June 30, 1999
helpdesk@link.springer.de