Medium‐Range Forecasting of Solar Wind: A Case Study of Building Regression Model With Space Weather Forecast Testbed (SWFT). Issue 12 (10th December 2020)

• Record Type: Journal Article
• Authors: Wang, Chunming; Rosen, I. Gary; Tsurutani, Bruce T.; Verkhoglyadova, Olga P.; Meng, Xing; Mannucci, Anthony J.
• Is Part Of: Space weather. Volume 18:Issue 12(2020)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2020-12-10
• Subjects: forecasting -- space weather -- machine learning -- solar wind; Space environment -- Periodicals; 551.509992
• DOI: 10.1029/2019SW002433
• Languages: English
• ISSNs: 1542-7390

Abstract

Abstract: The Space Weather Forecast Testbed (SWFT) is developed by a team of space weather scientists and mathematicians at the University of Southern California (USC) and Jet Propulsion Laboratory (JPL) to foster the creation of models for space weather forecast by exploration of existing historic data using techniques of machine learning. As an example to demonstrate the potential power of SWFT, we present in this paper a multilinear regression‐based forecast model for solar wind. Solar wind is one of the key drivers for numerous physics‐based models for space weather including thermosphere and ionosphere models. Many attempts have been made to produce forecasts for the solar wind. SWFT provides a unified framework for forecast model formulation, training, and performance assessment. In particular, the preparation of training and validation data by SWFT takes into account the realistic constraints on data latency and forecast lead time. In developing a solar wind forecast model, SWFT allows fast exploration of many metaparameters such as the list of predictive variables and their time history used in constructing a model. We present the impact of metaparameter selection, as well as performance relative to existing solar wind forecast models. Key Points: Space Weather Forecast Testbed (SWFT) provides a useful tool for exploring forecasting space weather using data‐driven techniques Machine learning approaches can be beneficial for space weather forecast A case study of using SWFT to develop a data‐driven forecast model for solar wind speed demonstrates the utility of SWFT