Search

Three helioseismic instruments on the Solar and Heliospheric Observatory have observed the Sun almost continuously since early 1996. This has led to detailed study of the biases induced by the instruments that measure intensity or Doppler velocity variation. Photospheric turbulence hardly influences the tiny signature of conditions in the energy-generating core in the low-order modes, which are therefore very informative. We use sound-speed and density profiles inferred from GOLF and MDI data including these modes, together with recent improvements to stellar model computations, to build a spherically symmetric seismically adjusted model in agreement with the observations. The model is in hydrostatic and thermal balance and produces the present observed luminosity. In constructing the model, we adopt the best physics available, although we adjust some fundamental ingredients, well within the commonly estimated errors, such as the p-p reaction rate (+1%) and the heavy-element abundance (+3.5%); we also examine the sensitivity of the density profile to the nuclear reaction rates. Then, we deduce the corresponding emitted neutrino fluxes and consequently demonstrate that it is unlikely that the deficit of the neutrino fluxes measured on Earth can be explained by a spherically symmetric classical model without neutrino flavor transitions. Finally, we discuss the limitations of our results and future developments.

Observations made by the Michelson Doppler Imager (MDI) and Variability of solar IRradiance and Gravity Oscillations (VIRGO) on the Solar and Heliospheric Observatory (SOHO) and by the ground-based Birmingham Solar Oscillations Network (BiSON) and Global Oscillations Network Group (GONG) have been used in a concerted effort to search for solar gravity oscillations. All spectra are dominated by solar noise in the frequency region from 100 to 1000 μHz, where g-modes are expected to be found. Several methods have been used in an effort to extract any g-mode signal present. These include (1) the correlation of data—both full-disk and imaged (with different spatial-mask properties)—collected over different time intervals from the same instrument, (2) the correlation of near-contemporaneous data from different instruments, and (3) the extraction—through the application of complex filtering techniques—of the coherent part of data collected at different heights in the solar atmosphere. The detection limit is set by the loss of coherence caused by the temporal evolution and the motion (e.g., rotation) of superficial structures. Although we cannot identify any g-mode signature, we have nevertheless set a firm upper limit to the amplitudes of the modes: at 200 μHz, they are below 10 mm s-1 in velocity, and below 0.5 parts per million in intensity. The velocity limit corresponds very approximately to a peak-to-peak vertical displacement of δR/R☉ = 2.3 × 10-8 at the solar surface. These levels which are much lower than prior claims, are consistent with theoretical predictions.

As a result of an error at the Press, the second panel of Figure 9 was repeated twice in the top row of the printed, black-andwhite version of this figure, and the first panel was omitted. This error appears in the print edition and the PDF and postscript (PS) versions available with the electronic edition of the journal, although the panels of the color figure displayed in the electronic article itself are correct. Please see below for the corrected print version of Figure 9. The Press sincerely regrets the error.