In this study, we present a new method for forecasting arrival times and speeds of coronal mass ejections (CMEs) at any location in the inner heliosphere. This new approach enables the adoption of a highly flexible geometrical shape for the CME front with an adjustable CME angular width and an adjustable radius of curvature of its leading edge, i.e. the assumed geometry is elliptical. Using, as input, STEREO heliospheric imager (HI) observations, a new elliptic conversion (ElCon) method is introduced and combined with the use of drag-based model (DBM) fitting to quantify the deceleration or acceleration experienced by CMEs during propagation. The result is then used as input for the Ellipse Evolution Model (ElEvo). Together, ElCon, DBM fitting, and ElEvo form the novel ElEvoHI forecasting utility. To demonstrate the applicability of ElEvoHI, we forecast the arrival times and speeds of 21 CMEs remotely observed from STEREO/HI and compare them to in situ arrival times and speeds at 1 AU. Compared to the commonly used STEREO/HI fitting techniques (Fixed-φ, Harmonic Mean, and Self-similar Expansion fitting), ElEvoHI improves the arrival time forecast by about 2 hours to ±6.5 hours and the arrival speed forecast by ≈ 250 km s −1 to ±53 km s −1 , depending on the ellipse aspect ratio assumed. In particular, the remarkable improvement of the arrival speed prediction is potentially beneficial for predicting geomagnetic storm strength at Earth. ing coronagraph observations, leads to errors in CME arrival time predictions at Earth that lie in the range of ±12 to ±18 hrs (e.g. Mays et al. 2015; Vršnak et al. 2014). Note that for a selected sample of CMEs Millward et al. (2013) obtained errors of ±7.5 hrs. The reasons for these large forecasting errors are diverse. Firstly, the observations are limited. Currently, only the LASCO C2 and C3 coronagraphs (Brueckner et al. 1995) onboard the SOlar and Heliospheric Observatory (SOHO) and the COR1 and COR2 coronagraphs onboard the Ahead spacecraft of the twin satellite mission Solar Terrestrial Relations Observatory (STEREO; Kaiser et al. 2008) can be used to operationally forecast the arrival times of Earth-directed CMEs. Secondly, the structures, shapes, orientations, sizes, directions and speeds of CMEs are highly variable, i.e. it is quite diffi-arXiv:1605.00510v2 [astro-ph.SR]

We present a comprehensive statistical analysis spanning over a solar cycle of the properties and drivers of traveling fast forward and fast reverse interplanetary shocks. We combine statistics of 679 shocks between 1995 and 2013 identified from the near‐Earth (Wind and ACE) and STEREO‐A observations. We find that fast forward shocks dominate over fast reverse shocks in all solar cycle phases except during solar minimum. Nearly all fast reverse shocks are driven by slow‐fast stream interaction regions (SIRs), while coronal mass ejections (CMEs) are the principal drivers of fast forward shocks in all phases except at solar minimum. The occurrence rate and median speeds of CME‐driven fast forward shocks follow the sunspot cycle, while SIR‐associated shocks do not show such correspondence. The strength of the shock (characterized by the magnetosonic Mach number and by the upstream to downstream magnetic field and density ratio) shows relatively little variations over solar cycle. However, the shocks were slightly stronger during the ascending phase of a relatively weak solar cycle 24 than during the previous ascending phase. The CME‐ and SIR‐driven fast forward shocks and fast reverse shocks have distinct upstream solar wind conditions, which reflect to their relative strengths. We found that CME‐driven shocks are on average stronger and faster, and they show broader distributions of shock parameters than the shocks driven by SIRs.

Flux ropes ejected from the Sun may change their geometrical orientation during their evolution, which directly affects their geoeffectiveness. Therefore, it is crucial to understand how solar flux ropes evolve in the heliosphere to improve our space-weather forecasting tools. In this article we present a follow-up study of the concepts described by Isavnin, Vourlidas, and Kilpua (2013). We analyze 14 coronal mass ejections (CMEs), with clear flux rope signatures, observed during the decay of Solar Cycle 23 and rise of Solar Cycle 24. First, we estimate initial orientations of the flux ropes at the origin using extreme ultraviolet observations of post-eruption arcades and/or eruptive prominences. Then we reconstruct multiviewpoint coronagraph observations of the CMEs from ≈ 2 to 30 R ⊙ with a three-dimensional geometric representation of a flux rope to determine their geometrical parameters. Finally, we propagate the flux ropes from ≈ 30 R ⊙ to 1 AU through MHD-simulated background solar wind while using in-situ measurements at 1 AU of the associated magnetic cloud as a constraint for the propagation technique. This methodology allows us to estimate the flux-rope orientation all the way from the Sun to 1 AU. We find that while the flux-ropes' deflection occurs predominantly below 30 R ⊙ , a significant amount of deflection and rotation happens between 30 R ⊙ and 1 AU. We compare the flux-rope orientation to the local orientation of the heliospheric current sheet (HCS). We find that slow flux ropes tend to align with the streams of slow solar wind in the inner heliosphere. During the solar-cycle minimum the slow solar wind channel as well as the HCS usually occupy the area in the vicinity of the solar equatorial plane, which in the past led researchers to the hypothesis that flux ropes align with the HCS. Our results show that exceptions from this rule are explained by interaction with the Parker-spiraled background magnetic field, which dominates over the magnetic interaction with the HCS in the inner heliosphere at least during solar-minimum conditions.