Four solar cycle ecliptic solar wind data set generated from the OMNI data set

Numerical modeling of the global heliosphere provides key insights into the physics of the outer heliosphere. The solar wind data utilized by these models impacts their accuracy and should be as close as possible to the actual solar wind. Ideally, the solar wind data would come directly from in situ measurements. Unfortunately the availability of such measurements is severely limited, mostly to points near Earth. A method to infer the solar wind at points not directly observed is developed such that a data set with four solar cycles of solar wind data which fills the ecliptic plane is generated, using the hourly OMNI data set as the seed data.;First the OMNI data are separated into four separate categories: fast, slow, CIR and ICME. The categories are characterized by relative coverage, durations, variable distributions, running averages, and step-size distributions. The category statistics allow for inferences as to which category is present during periods and for locations for which no solar wind data are available, generating a continuous 2D category map of the solar wind for the full four solar cycles covered by OMNI. The category maps are used to determine solar wind plasma and interplanetary magnetic field (IMF) characteristics. The plasma and IMF values are determined by local running averages coupled with a random walk technique. The averages provide baseline values and the random walk adds short-duration deviations from this baseline.;The resulting four solar cycle 2D data set is analyzed along various heliolongitudes at 1 AU. The statistics from the data along these trajectories are compared to the statistics of the original OMNI data set. The category durations and relative coverage are similar to those of the OMNI data, though with some discrepancies. The plasma and IMF variable distributions are also very similar to those of the OMNI data, again with some discrepancies. Additional generated data sets using different sets of plasma and IMF variables could possibly improve the agreement with OMNI, and the data set remains to be deployed in simulations.