Abstract. The Wang-Sheeley model is an empirical model that can predict the background solar wind speed and interplanetary magnetic field (IMF) polarity. We make a number of modifications to the basic technique that greatly improve the performance and reliability of the model. First, we establish a continuous empirical function that relates magnetic expansion factor to solar wind velocity at the source surface. Second, we propagate the wind from the source surface to the Earth using the assumption of radial streams and a simple scheme to account for their interactions. Third, we develop and apply a method for identifying and removing problematic magnetograms from the Wilcox Solar Observatory (WSO).

We have found an improved technique for empirically specifying solar wind flow speed near the Sun (~0.1 AU) using a set of three simple inter-linked coronal/solar wind models. In addition to magnetic field expansion factor, solar wind speed also appears to be influenced by the minimum angular distance that an open field footpoint lies from a coronal hole boundary. We conduct our study using polar field corrected Mount Wilson Solar Observatory Carrington maps from 1995. During this period, the Sun was in the declining phase of the solar cycle and the solar wind had relatively simple global structure.

Occasionally, anomalously low values of the solar wind proton temperature Tp are observed when the solar wind velocity v is high. A large fraction of such measurements by the Vela 3 satellites follow the passage of interplanetary shocks by some 20-60 hours. Of 24 postshock events in which v exceeded 400 km sec -x and for which Vela 3 measurements are available, 12 exhibited plasma states of anomalously low Tp, high v. The proton density at the time of these observations typically was depressed below normal, and the velocity tended to be constant or falling. A very strong association with abnormally high (_• 15%) concentrations of He++/H + in the solar wind is noted for these anomalous proton temperatures, the usual temporal sequence of events at 1 AU being: (1) shock wave, (2) helium enrichment, and (3) low Tp, high v. It is suggested that these observations are consistent with a model for some shock wave disturbances that includes the ejection of new material (distinguished by the helium enrichment at 1 AU) into the solar wind at the time of large solar flares and the formation of a magnetic bottle configuration in the solar wind behind and within the ejecta. The anomalo.usly low proton temperatures then result from the adiabatic cooling of the plasma within the magnetic bottle. The positive correlation between solar wind flow speed v and proton tempei•ature T• is. well documented in the literature [Strong et al., 1966; Hundhausen e,t al., 1967; Burlaga and Ogilvie, 1970]. A demonstration of this correlation is provided by Figure 1, which is a scatter plot in the v, T• plane of the 3-hour averages of the Vela 3 solar wind data obtained between July 1965 and December 1967. Several recent papers [Burlaga and Ogilvie, 1970; Hartle and Barnes, 1970; Hundhause.n, 1973] have concentrated on an understanding of the general increase in proton temperature with increasing flow s.peed. Although differences in interpretation exist, there is agreement that the average increase in proton temperature with increasing flow speed probably represents a mapping outward from the corona of increasingly high coronal proton temperatures. One of our concerns recently has been the interpretation of the relatively large scatter of plasma states evident in Figure I particularly at high flow speeds. That is, we have been concerned with the deviations from the average flow speed-proton temperature relationship. It is now clear that compressional heating caused by high-speed streams overtaking slower-moving plasma is responsible for a major part of the plasma states of excessively high proton temperature at moderate and high flow speeds [Burlaga and Ogilvie, 1970; Burlaga et al., 1971; Gosling et al., 1972; V. Pizzo, J. T. Gosling, and A. J. Hundhausen, unpublished manuscript, 1972; Hundhausen, 1973]. On the other hand, rarefactional cooling in the trailing part of high-speed streams does not appear to be a major factor in explaining the anomalously low proton temperature states sometimes observed at high flow speeds [Gosling et al., 1972; V....

We develop a numerical model that treats the three‐dimensional (3‐D) magnetohydrodynamic interactions taking place at interplanetary corotating stream fronts near the heliographic equator during those periods of the solar cycle when the large‐scale coronal magnetic structure is in the tilted‐dipole configuration. The dynamic simplicity of the tilted‐dipole geometry permits the formulation of a local approximation valid over a limited range of heliographic latitudes and longitudes centered about the stream front. We show that one component of the spatial orientation of the stream interface is fixed by the implied coronal configuration and that this orientation varies in a systematic and predictable way with heliocentric distance. This is important, since the orientation of the interface regulates the pace of the interplanetary evolution by determining the obliquity of the dynamical interaction between fast and slow flows at the stream front. We find that the appearance of 3‐D tilted‐dipole interfaces near the Sun (0.3–0.6 AU) may differ considerably from the classic 2‐D structures described previously in the literature: they should be subtle features, in which the amplitudes of the nonradial flow deflections are only a few degrees and the north‐south component dominates the east‐west component. In particular, we expect that the amplitude of the total nonradial deflection should go roughly as sin α (where α is the dipole tilt angle), whereas the ratio between the north‐south and east‐west deflections should vary as cot α. We also show that magnetic torques should have strong influence on the interface structure near the Sun (inside ∼0.5 AU); the amplitudes and phases of the nonradial flows there are subject to details of the large‐scale magnetic properties of the flow, which can only be assessed observationally. In addition, magnetic stresses carried in the solar wind may inhibit the formation of discontinuous interfaces in this region. The effects of the tilted‐dipole geometry on the evolution of corotating interaction regions and the formation of shocks at larger heliocentric distances will be addressed in a subsequent paper.

[1] Recently, we simulated the 12 May 1997 coronal mass ejection (CME) event with a numerical three-dimensional magnetohydrodynamic model , in which the background solar wind was determined from the Science Applications International Corporation (SAIC) coronal model (Riley et al., 2001) and the transient disturbance was determined from the cone model (Zhao et al., 2002). Although we reproduced with some fidelity the arrival of the shock and interplanetary CME at Earth, detailed analysis of the simulations showed a poorly defined shock and discrepancies in the standoff distance between the shock and the driving ejecta and in the inclination of the shock with respect to the Sun-Earth line. In this paper, we investigate these problems in more detail. First, we use an alternative coronal outflow model, the so-called Wang-Sheeley-Arge-Mount Wilson Observatory (WSA-MWO) model (Arge and Pizzo, 2000;Arge et al., 2002;Arge et al., 2004), to assess the effect of using synoptic, full rotation coronal maps that differ in method of preparation. Second, we investigate how differences in the presumed evolution of the coronal stream structure affect the propagation of the disturbance. We incorporate two time-dependent boundary conditions for the ambient solar wind as determined by the WSA model, one derived from pseudo daily updated maps and one derived from artificially modified full rotation maps. Numerical results from these different scenarios are compared with solar wind observations at Earth. We find that heliospheric simulations with the SAIC and WSA full rotation models provide qualitatively similar parameters of the background solar wind and transient disturbances at Earth. Improved agreement with the observations is achieved by artificially modified maps that simulate the rapid displacement of the coronal hole boundary after the CME eruption. We also consider how multipoint temporal profiles of solar wind parameters and multiperspective synthetic white light images emulating upcoming STEREO spacecraft observations might be used to differentiate between different event scenarios.