Research Article

Helio1D modeling of temporal variation of solar wind: Interfacing between MULTI-VP and 1D MHD for future operational forecasting at L1

¹ Institut de Recherche en Astrophysique et Planétologie, UPS, CNES, CNRS, Toulouse, France
² Departement d’Astrophysique/AIM, CEA/IRFU, CNRS/INSU, Univ. Paris-Saclay & Univ. de Paris, 91191 Gif-sur-Yvette, France
³ CmPA, KU Leuven, Leuven, Belgium
⁴ NASA, Goddard Space Flight Center, Greenbelt, MD, USA
⁵ National Institute of Information and Communications Technology (NICT), Koganei, Tokyo 184-8795, Japan
⁶ Laboratoire d’astrophysique de Bordeaux, Univ. Bordeaux, CNRS, Pessac, France
⁷ ONERA/DPHY, Université de Toulouse, F-31055 Toulouse, France
⁸ Department of Physics, National and Kapodistrian University of Athens, 15784 Athens, Greece
⁹ Royal Belgian Institute for Space Aeronomy (IASB-BIRA), 1180 Brussels, Belgium
¹⁰ Institute of Atmospheric Physics CAS, Dept of Space Physics, 14100 Prague, Czech Republic
¹¹ Hellenic Space Center, Athens, Greece

^* Corresponding author: rkieokaew@irap.omp.eu

Received: 15 March 2023
Accepted: 22 May 2024

Abstract

Developing an automated pipeline for solar-wind condition prediction upstream of Earth is an important step for transitioning from space weather research to operation. We develop a prototype pipeline called “Helio1D” to model ambient solar wind conditions comprising temporal profiles of wind speed, density, temperature, and tangential magnetic field at L1 up to 4 days in advance. The prototype pipeline connects the MULTI-VP coronal model that provides daily predictions of the solar wind at 0.14 AU and a 1D magnetohydrodynamics (MHD) model that propagates the solar wind to 1 AU. As a part of development towards a better-performing operational pipeline in the future, our present work focuses on the proof-of-concept, initial implementation, and validation of Helio1D. Here, we first benchmark Helio1D using the synoptic magnetograms provided by Wilcox Space Observatory as inputs to the coronal part of MULTI-VP for the intervals in 2004–2013 and 2017–2018. Using the classic point-to-point metrics, it is found that Helio1D underperforms the 27-day recurrence model for all time intervals while outperforming the 4-day persistence model in the late declining phase of the solar cycle. As a complementary analysis, we evaluate the time and magnitude differences between Helio1D and the observations by exploiting the Fast Dynamic Time Warping technique, which allows us to discuss Helio1D caveats and address calibration to improve the Helio1D performance. Furthermore, we model several solar wind conditions in parallel, for a total of 21 profiles corresponding to various virtual targets to provide uncertainties. Although our prototype pipeline shows less satisfactory results compared to existing works, it is fully automated and computationally fast, both of which are desirable qualities for operational forecasting. Our strategies for future improvements towards better-performing pipeline are addressed.