Full Diffusion Code (FDC; Ni et al., 2008; Shprits and Ni, 2009; Orlova and Shprits, 2011) is used to calculate the quasi-linear bounce-averaged diffusion coefficients using a dipole magnetic field in the form given by Glauert and Horne (2005).
The equations used to calculate local diffusion coefficients can be found here.
The inputs for FDC are stored as .txt files in the Input folder. More information on input files can be found here.
Before running FDC, you must load the mathematical packages and compile the Fortran code. This process differs on MP machines and GLIC. Instructions for both are available here.
The code is run using a bash script. More details about the bash script for running FDC, how to execute it, and the outputs can be found here.
The FDC code is written in Fortran 90 and located in Src/. The pipeline is illustrated in the following diagram.
Pipeline of FDC. Credits: Katja Stoll
Descriptions of each module can be found here.
This documentation covers adding whistler-mode waves and changing density models. All additions involve adding new switch cases to the respective .f90 files.
GOLDEN RULE: Any change you make must be reflected in
FDC_MPI.f90, or the code will throw errors.
For whistler-mode waves, you might need to incorporate the following models:
Spatial Dependent Intensity: Add the model toBw_wave.f90.Frequency Dependent PSD: Add the model toPowerSpectralDensity.f90. Also add the upper and lower frequency cutoffs for the respective models toFreq_Spectral.f90.
To change density models, edit ElectronDensity.f90.
This documentation is based on the document created by Katja Stoll, which can be found here.