MHD heliosphere with boundary conditions from a tomographic reconstruction using interplanetary scintillation data

Abstract

AbstractObservations of interplanetary scintillation (IPS) provide a set of data that is used in estimating the solar wind parameters with reasonably good accuracy. Various tomography techniques have been developed to deconvolve the line‐of‐sight integration effects ingrained in observations of IPS to improve the accuracy of solar wind reconstructions. Among those, the time‐dependent tomography developed at the University of California, San Diego (UCSD) is well known for its remarkable accuracy in reproducing the solar wind speed and density at Earth by iteratively fitting a kinematic solar wind model to observations of IPS and near‐Earth spacecraft measurements. However, the kinematic model gradually breaks down as the distance from the Sun increases beyond the orbit of Earth. Therefore, it would be appropriate to use a more sophisticated model, such as a magnetohydrodynamics (MHD) model, to extend the kinematic solar wind reconstruction beyond the Earth's orbit and to the outer heliosphere. To test the suitability of this approach, we use boundary conditions provided by the UCSD time‐dependent tomography to propagate the solar wind outward in a MHD model and compare the simulation results with in situ measurements and also with the corresponding kinematic solution. Interestingly, we find notable differences in proton radial velocity and number density at Earth and various locations in the inner heliosphere between the MHD results and both the in situ data and the kinematic solution. For example, at 1 AU, the MHD velocities are generally larger than the spacecraft data by up to 150 km s−1, and the amplitude of density fluctuations is also markedly larger in the MHD solution. We show that the MHD model can deliver more reasonable results at Earth with an ad hoc adjustment of the inner boundary values. However, we conclude that the MHD model using the inner boundary conditions derived from kinematic simulations has little chance to match IPS and in situ data as well as the kinematic model does unless it too is iteratively fit to the observational data and measurements.