Approximate Bayesian Computation for Estimating Parameters of Data-Consistent Forbush Decrease Model

Area of Science:

• Cosmic Ray Physics
• Heliophysics
• Computational Physics

Background:

• Modeling complex physical phenomena like Forbush decreases (Fd) is challenging due to uncertainties in input parameters.
• Galactic cosmic ray intensity is influenced by solar events, necessitating accurate modeling of Fd.
• Approximate Bayesian Computation (ABC) offers a method to estimate model parameters when complete information is unavailable.

Purpose of the Study:

• To present a framework applying ABC methodology for estimating parameters in Forbush decrease (Fd) modeling.
• To refine the modeling of galactic cosmic ray intensity variations.
• To improve the understanding of parameters governing Fd, particularly the diffusion coefficient.

Main Methods:

• Utilized Approximate Bayesian Computation (ABC) with a Sequential Monte Carlo algorithm.
• Modeled Forbush decrease (Fd) using numerical solutions of the Fokker-Planck equation in five-dimensional space.
• Scanned diffusion coefficient parameters in the heliosphere and compared model outputs with experimental galactic cosmic ray intensity data.

Main Results:

• Developed a framework to estimate parameters for Forbush decrease (Fd) models.
• Assessed parameter correctness by comparing model output with experimental galactic cosmic ray intensity data.
• Focused on the rigidity dependence of the rigidity spectrum exponent for Fd.

Conclusions:

• The proposed ABC framework provides a robust method for modeling Forbush decreases (Fd).
• The study successfully modeled the Fd observed in November 2004 using neutron monitors and muon telescope data.
• This approach enhances the realistic modeling of complex physical phenomena with uncertain parameters.