loading page

New Findings from Explainable SYM-H Forecasting using Gradient Boosting Machines
  • +5
  • Daniel Iong,
  • Yang Chen,
  • Gabor Toth,
  • Shasha Zou,
  • Tuija I. Pulkkinen,
  • Jiaen Ren,
  • Enrico Camporeale,
  • Tamas I. I. Gombosi
Daniel Iong
University of Michigan, Ann Arbor, University of Michigan, Ann Arbor, University of Michigan, Ann Arbor

Corresponding Author:daniong@umich.edu

Author Profile
Yang Chen
University of Michigan, University of Michigan, University of Michigan
Author Profile
Gabor Toth
University of Michigan-Ann Arbor, University of Michigan-Ann Arbor, University of Michigan-Ann Arbor
Author Profile
Shasha Zou
University of Michigan-Ann Arbor, University of Michigan-Ann Arbor, University of Michigan-Ann Arbor
Author Profile
Tuija I. Pulkkinen
University of Michigan, University of Michigan, University of Michigan
Author Profile
Jiaen Ren
University of Michigan - Ann Arbor, University of Michigan - Ann Arbor, University of Michigan - Ann Arbor
Author Profile
Enrico Camporeale
University of Colorado, University of Colorado, University of Colorado
Author Profile
Tamas I. I. Gombosi
University of Michigan-Ann Arbor, University of Michigan-Ann Arbor, University of Michigan-Ann Arbor
Author Profile

Abstract

In this work, we develop gradient boosting machines (GBMs) for forecasting the SYM-H index multiple hours ahead using different combinations of solar wind and interplanetary magnetic field (IMF) parameters, derived parameters, and past SYM-H values. Using Shapley Additive Explanation (SHAP) values to quantify the contributions from each input to predictions of the SYM-H index from GBMs, we show that our predictions are consistent with physical understanding while also providing insight into the complex relationship between the solar wind and Earth’s ring current. In particular, we found that feature contributions vary depending on the storm phase. We also perform a direct comparison between GBMs and neural networks presented in prior publications for forecasting the SYM-H index by training, validating, and testing them on the same data. We find that the GBMs have a comparable root mean squared error as the best published black-box neural network schemes.