New Solar Opacities, Abundances, Helioseismology, and Neutrino Fluxes

We construct solar models with the newly calculated radiative opacities from the Opacity Project (OP) and with recently determined (lower) heavy-element abundances. We compare the results from the new models with the predictions of a series of models that use OPAL radiative opacities, older determinations of the surface heavyelement abundances, and refinements of nuclear reaction rates. For all the variations we consider, solar models that are constructed with the newer and lower heavy-element abundances advocated by Asplund et al. disagree by much more than the estimated measuring errors with the helioseismological determinations of the depth of the solar convective zone, the surface helium composition, the internal sound speeds, and the density profile. Using the new OP radiative opacities, the ratio of the B neutrino flux calculated with the older and larger heavyelement abundances (or with the newer and lower heavy-element abundances) to the total neutrino flux measured by the Sudbury Neutrino Observatory is 1.09 (0.87) with a 9% experimental uncertainty and a 16% theoretical uncertainty, 1 j errors. Subject headings: atomic processes — neutrinos — nuclear reactions, nucleosynthesis, abundances — Sun: abundances — Sun: interior Recent, refined determinations of the surface heavy-element abundances of the Sun have led to lower than previously believed heavy-element abundances (see Asplund et al. 2005 and references therein). A number of authors have pointed out that these lower heavy-element abundances lead to solar models that conflict with different aspects of helioseismological measurements (e.g., Bahcall & Pinsonneault 2004; Basu & Antia 2004; Bahcall et al. 2005). If the radiative opacity in the temperature range of K were to be increased by of 6 (2–4.5) # 10 order 10% relative to the standard OPAL opacity (Iglesias & Rogers 1996), then the discrepancy between new abundances and helioseismology could be resolved (Bahcall et al. 2005; see also Basu & Antia 2004). The Opacity Project (OP) has recently performed more precise and more physically complete calculations of the radiative opacities with the goal of determining if these new calculations could eliminate the discrepancy between helioseismology and solar modeling that uses the new (lower) heavy-element abundances (see Badnell et al. 2004, Seaton & Badnell 2004, and Seaton 2004). The Opacity Project refinements result in only a small increase (less than 2.5% everywhere of interest) relative to the OPAL opacity. In this Letter, we present a series of precise solar models that were calculated using the new OP opacities as well as the familiar OPAL opacities. We also present models that were constructed with the recently determined heavy-element abundances (Asplund et al. 2005) as well as with the previously standard abundances (Grevesse & Sauval 1998). In addition, we introduce refinements in the nuclear physics used in the solar models. We compare the results of each of our series of solar models with helioseismological and neutrino observations of the Sun. As a side product of this investigation, we determine the remarkable precision with which two very different stellar evolution codes reproduce the same solar model parameters. Table 1 gives the principal characteristics of seven precise solar models that we use in this Letter to investigate the helioseismological and neutrino flux implications of the recent redeterminations of heavy-element abundances and of radiative opacities. Table 2 presents the neutrino fluxes calculated for each of the seven solar models represented in Table 1. At the end of the Letter, we summarize in Figure 1 and the related discussion the comparison between the helioseismologically determined sound speeds and densities and the predictions of the various solar models. We begin by describing the differences between the various solar models and by commenting on how these differences affect the calculated properties of the models, including the helioseismological parameters and neutrino fluxes. The model BP04(Yale) was calculated by Bahcall & Pinsonneault (2004) and is their preferred standard solar model. BP04(Yale) uses the Grevesse & Sauval (1998) solar abundances and the best other input data available at the time the model was constructed. The model was constructed as described in Bahcall & Pinsonneault (1992) and Bahcall & Ulrich (1988) and uses the Yale–Ohio State–Princeton stellar evolution code (Pinsonneault et al. 1989; Bahcall & Pinsonneault 1992; Bahcall et al. 1995) as modified by iterations of the Bahcall-Ulrich nuclear energy generation subroutine. The model BP04(Garching) was derived using the Garching Stellar Evolution code (see, e.g., Schlattl et al. 1997 and Schlattl 2002 for details of the code) using the same procedures and input data as the BP04(Yale) solar model. The first two rows of Table 1 and Table 2 show that the principal characteristics of solar models are independent, to practical accuracy, of the evolutionary code used for their calculation. For example, the initial helium abundance is the same in the BP04(Yale) and BP04(Garching) models to an accuracy of 0.04%, and the depth of the convective zone is the same to 0.01%. In a more stringent test, the Be, B, F, and pep neutrino fluxes in the two models agree to 0.4% or better and the p-p, hep, N, and O neutrino fluxes to better than L86 BAHCALL, SERENELLI, & BASU Vol. 621