We investigate the effects of thermonuclear reaction rate uncertainties on nova nucleosynthesis. One-zone nucleosynthesis calculations have been performed by adopting temperature-density-time profiles of the hottest hydrogen-burning zone (i.e., the region in which most of the nucleosynthesis takes place). We obtain our profiles from 7 different, recently published, hydrodynamic nova simulations covering peak temperatures in the range from T peak =0.145-0.418 GK. For each of these profiles, we individually varied the rates of 175 reactions within their associated errors and analyzed the resulting abundance changes of 142 isotopes in the mass range below A=40. In total, we performed ≈7350 nuclear reaction network calculations. We use the most recent thermonuclear reaction rate evaluations for the mass ranges A=1-20 and A=20-40. For the theoretical astrophysicist, our results indicate the extent to which nova nucleosynthesis calculations depend on presently uncertain nuclear physics input, while for the experimental nuclear physicist our results represent at least a qualitative guide for future measurements at stable and radioactive ion beam facilities. We find that present reaction rate estimates are reliable for predictions of Li, Be, C and N abundances in nova nucleosynthesis. However, rate uncertainties of several reactions have to be reduced significantly in order to predict more reliable O, F, Ne, Na, Mg, Al, Si, S, Cl and Ar abundances.Results are presented in tabular form for each adopted nova simulation.

A method based on Monte Carlo techniques is presented for evaluating thermonuclear reaction rates. We begin by reviewing commonly applied procedures and point out that reaction rates that have been reported up to now in the literature have no rigorous statistical meaning. Subsequently, we associate each nuclear physics quantity entering in the calculation of reaction rates with a specific probability density function, including Gaussian, lognormal and chi-squared distributions. Based on these probability density functions the total reaction rate is randomly sampled many times until the required statistical precision is achieved. This procedure results in a median (Monte Carlo) rate which agrees under certain conditions with the commonly reported recommended "classical" rate. In addition, we present at each temperature a low rate and a high rate, corresponding to the 0.16 and 0.84 quantiles of the cumulative reaction rate distribution. These quantities are in general different from the statistically meaningless "minimum" (or "lower limit") and "maximum" (or "upper limit") reaction rates which are commonly reported. Furthermore, we approximate the output reaction rate probability density function by a lognormal distribution and present, at each temperature, the lognormal parameters µ and σ. The values of these quantities will be crucial for future Monte Carlo nucleosynthesis studies. Our new reaction rates, appropriate for bare nuclei in the laboratory, are tabulated in the second paper of this series (Paper II). The nuclear physics input used to derive our reaction rates is presented in the third paper of this series (Paper III). In the fourth paper of this series (Paper IV) we compare our new reaction rates to previous results.

We investigate the effects of thermonuclear reaction rate variations on 26 Al production in massive stars. The dominant production sites in such events were recently investigated by using stellar model calculations: explosive neon-carbon burning, convective shell carbon burning, and convective core hydrogen burning. Post-processing nucleosynthesis calculations are performed for each of these sites by adopting temperature-density-time profiles from recent stellar evolution models. For each profile, we individually multiplied the rates of all relevant reactions by factors of 10, 2, 0.5 and 0.1, and analyzed the resulting abundance changes of 26 Al. In total, we performed ≈ 900 nuclear reaction network calculations. Our simulations are based on a next-generation nuclear physics library, called STARLIB, which contains a recent evaluation of Monte Carlo reaction rates. Particular attention is paid to quantifying the rate uncertainties of those reactions that most sensitively influence 26 Al production. For stellar modelers our results indicate to what degree predictions of 26 Al nucleosynthesis depend on currently uncertain nuclear physics input, while for nuclear experimentalists our results represent a guide for future measurements. We also investigate equilibration effects of 26 Al. In all previous massive star investigations, either a single species or two species of 26 Al were taken into account, depending on whether thermal equilibrium was achieved or not. These are two extreme assumptions and in a hot stellar plasma the ground and isomeric state may communicate via γ-ray transitions involving higher-lying 26 Al levels.We tabulate the results of our reaction rate sensitivity study for each of the three distinct massive star sites referred to above. It is found that several current reaction rate uncertainties influence the production of 26 Al. Particularly important reactions are 26 Al(n,p) 26 Mg, 25 Mg(α,n) 28 Si, 24 Mg(n,γ) 25 Mg and 23 Na(α,p) 26 Mg. These reactions should be prime targets for future measurements. Overall, we estimate that the nuclear physics uncertainty of the 26 Al yield predicted by the massive star models explored here amounts to about a factor of 3. We also find that taking the equilibration of 26 Al levels explicitly into account in any of the massive star sites investigated here has only minor effects on the predicted 26 Al yields. Furthermore, we provide for the interested reader detailed comments regarding the current status of certain reactions, including 12 C( 12 C,n) 23 Mg, 23 Na(α,p) 26 Mg, 25 Mg(α,n) 28 Si, 26 Al m (p,γ) 27 Si, 26 Al(n,p) 26 Mg and 26 Al(n,α) 23 Na.