Modeling the Global Distribution of Solar Wind Parameters on the Source Surface Using Multiple Observations and the Artificial Neural Network Technique

Yang, Yi; Shen, Fang

doi:10.1007/s11207-019-1496-5

Cited by 10 publications

(7 citation statements)

References 75 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such methodology implies computing several fit performance metrics between the predicted indices values and the real ones. The metrics used are the RMSE and the coefficient of determination R 2 , which are also included in the survey by Camporeale (2019) for regression problems and commonly used other works in the field, for example, Yang et al (2018) and Yang and Shen (2019). Both have been computed after reverting the standardization process done to the variables prior to using them in the network.…”

Section: Metricsmentioning

confidence: 99%

Deep Neural Networks With Convolutional and LSTM Layers for SYM‐H and ASY‐H Forecasting

2021

View full text Add to dashboard Cite

Geomagnetic indices quantify the disturbance caused by the solar activity on a planetary scale or in particular regions of the Earth. Among them, the SYM‐H and ASY‐H indices represent the (longitudinally) symmetric and asymmetric geomagnetic disturbance of the horizontal component of the magnetic field at midlatitude with a 1‐min resolution. Their resolution, along with their relation to the solar wind parameters, makes the forecasting of the geomagnetic indices a problem that can be addressed through the use of Deep Learning, particularly using Deep Neural Networks (DNNs). In this work, we present two DNNs developed to forecast respectively the SYM‐H and ASY‐H indices. Both networks have been trained using the Interplanetary Magnetic Field (IMF) and the related index for the solar storms occurred in the last two solar cycles. As a result, the networks are able to accurately forecast the indices 2 h in advance, considering the IMF and indices values for the previous 200 min. The evaluation of both networks reveals a great forecasting precision, including good predictions for large storms that occurred during the solar cycle 23 and comparing with the persistence model for the period 2013–2020.

show abstract

Section: Metricsmentioning

confidence: 99%

Deep Neural Networks With Convolutional and LSTM Layers for SYM‐H and ASY‐H Forecasting

2021

View full text Add to dashboard Cite

show abstract

“…11, which demonstrates that the simulation retrieves most of the high-speed streams (HSSs), and the duration time and the magnitude of the HSSs are largely consistent with those of the observations. Later, Yang and Shen (2019) presented a new method to construct the global distribution of solar wind parameters at the source surface using multiple observations and the ANN (Artificial Neural Network) technique, which could be used to provide a more realistic boundary condition for 3D MHD solar wind modeling.…”

Section: Modeling the Background Solar Windmentioning

confidence: 99%

Earth-affecting solar transients: a review of progresses in solar cycle 24

Zhang

Temmer

Gopalswamy

et al. 2021

Prog Earth Planet Sci

Self Cite

View full text Add to dashboard Cite

This review article summarizes the advancement in the studies of Earth-affecting solar transients in the last decade that encompasses most of solar cycle 24. It is a part of the effort of the International Study of Earth-affecting Solar Transients (ISEST) project, sponsored by the SCOSTEP/VarSITI program (2014–2018). The Sun-Earth is an integrated physical system in which the space environment of the Earth sustains continuous influence from mass, magnetic field, and radiation energy output of the Sun in varying timescales from minutes to millennium. This article addresses short timescale events, from minutes to days that directly cause transient disturbances in the Earth’s space environment and generate intense adverse effects on advanced technological systems of human society. Such transient events largely fall into the following four types: (1) solar flares, (2) coronal mass ejections (CMEs) including their interplanetary counterparts ICMEs, (3) solar energetic particle (SEP) events, and (4) stream interaction regions (SIRs) including corotating interaction regions (CIRs). In the last decade, the unprecedented multi-viewpoint observations of the Sun from space, enabled by STEREO Ahead/Behind spacecraft in combination with a suite of observatories along the Sun-Earth lines, have provided much more accurate and global measurements of the size, speed, propagation direction, and morphology of CMEs in both 3D and over a large volume in the heliosphere. Many CMEs, fast ones, in particular, can be clearly characterized as a two-front (shock front plus ejecta front) and three-part (bright ejecta front, dark cavity, and bright core) structure. Drag-based kinematic models of CMEs are developed to interpret CME propagation in the heliosphere and are applied to predict their arrival times at 1 AU in an efficient manner. Several advanced MHD models have been developed to simulate realistic CME events from the initiation on the Sun until their arrival at 1 AU. Much progress has been made on detailed kinematic and dynamic behaviors of CMEs, including non-radial motion, rotation and deformation of CMEs, CME-CME interaction, and stealth CMEs and problematic ICMEs. The knowledge about SEPs has also been significantly improved. An outlook of how to address critical issues related to Earth-affecting solar transients concludes this article.

show abstract

“…The algorithm is easily transferable to other solar wind frameworks and is therefore an important contribution to the space weather community and can serve as a benchmark for future development of numerical ambient solar wind models. In a broader context, this study lays the foundation for future work on this subject, which will look into improving the modeling of solar wind conditions near the Sun (see e.g., Yang & Shen, 2019) and provide important input for MHD codes. P31659-N27, J4160-N27, P31521-N27, and P31265-N27.…”

Section: Discussionmentioning

confidence: 99%

Using Gradient Boosting Regression to Improve Ambient Solar Wind Model Predictions

et al. 2021

View full text Add to dashboard Cite

It was only in the 1970s, through the view of X-ray telescopes on the NASA Skylab satellite, that the dark patches in the Sun's polar regions were identified as coronal holes. Over the past decades, it has become clear that these are areas of open magnetic field lines, along which the solar plasma leaving the Sun in a continuous flow is accelerated to supersonic speeds into the vast reaches of our solar system. This fast component of the ambient solar wind flow and the magnetic field embedded within it corotate with the Sun in an ever expanding spiral, the Parker spiral. Understanding of the ambient solar wind is important in space weather forecasting to determine the evolution and possible impact of solar storms, particularly as they influence the evolution of transient events such as coronal mass ejections (CMEs), the catalysts for the strongest geomagnetic storms, on their path from Sun to Earth (M. J. Owens & Forsyth, 2013;Temmer et al., 2011). Ambient solar wind flows are themselves also an important driver of recurrent geomagnetic activity at Earth (Nakagawa et al., 2019;Verbanac et al., 2011;Zhang et al., 2007), determining critical properties in interplanetary space such as the solar wind speed, magnetic field strength and orientation (Luhmann et al., 2002). A clear picture and understanding and accurate modeling of the ambient solar wind are therefore essential in all aspects of space weather research. In today's ambient solar wind modeling approaches, the solar surface, corona and inner heliosphere are treated as a connected system to simulate the dynamics of the ambient solar wind flow from the Sun to the vicinity of Earth. These coupled modeling approaches commonly span the range from one solar radius (R ⊙ ) to Earth at 1 AU, with the coronal model from 1 R ⊙ to 5 R ⊙ (or 30 R ⊙ ) and the heliospheric model spanning the range from 5-30 R ⊙ to 1 AU. Despite the discovery of an empirical relationship between the configuration of open magnetic field lines on the Sun and the solar wind properties measured at the Sun-Earth Lagrange Point 1 (L1) near the Earth (Wang & Sheeley, 1990), it still proves challenging to develop and optimize empirical techniques for specifying the solar wind conditions at the inner boundary of the heliospheric domain (

show abstract

Modeling the Global Distribution of Solar Wind Parameters on the Source Surface Using Multiple Observations and the Artificial Neural Network Technique

Cited by 10 publications

References 75 publications

Deep Neural Networks With Convolutional and LSTM Layers for SYM‐H and ASY‐H Forecasting

Deep Neural Networks With Convolutional and LSTM Layers for SYM‐H and ASY‐H Forecasting

Earth-affecting solar transients: a review of progresses in solar cycle 24

Using Gradient Boosting Regression to Improve Ambient Solar Wind Model Predictions

Contact Info

Product

Resources

About