Skip to content

Latest commit

 

History

History
60 lines (50 loc) · 5.01 KB

README.md

File metadata and controls

60 lines (50 loc) · 5.01 KB
Raw

symphony

Overview

symphony calculates synchrotron emissivities and absorptivities, polarized in the Stokes basis {I, Q, U, V}, for any arbitrary gyrotropic momentum space distribution function. As of 2018, symphony also computes Faraday rotation coefficients in the Stokes basis and the full relativistic magnetized plasma susceptibility tensor, provided the distribution function is isotropic. Three distribution functions are built in: a relativistic thermal (Maxwell-Juettner) distribution, a nonthermal power-law distribution, and a kappa distribution.

symphony was first described in Pandya et al., 2016 ApJ 822 34, and extended to compute Faraday rotation coefficients in Pandya et al., 2018. If you use this code in an academic context, cite these papers.

Features

  • C code to calculate synchrotron emissivities via the j_nu() function and absorptivities via the alpha_nu() function.
  • C code to calculate Faraday rotation coefficients via the rho_nu() function
  • C code to evaluate approximate fitting function values for the emissivity and absorptivity, via the j_nu_fit() and alpha_nu_fit() functions, respectively.
  • CMake configure system, which helps during the build process to find all necessary libraries and files.
  • Python interface for j_nu(), alpha_nu(), j_nu_fit(), and alpha_nu_fit().
  • Python interface to rho_nu() and spline fits
    • This combines the speed of C when evaluating emissivities and absorptivities with Python's user-friendly syntax. It also allows for interfacing with larger Python codes.

How to use symphony

To download and build symphony:

  1. Clone symphony from github. Navigate into the "symphony" folder, and create a folder named "build"; navigate into it.
  2. Type cmake followed by the location of the "src/" folder (in the "symphony" folder, above). Altogether, this line should look something like: cmake /location/to/symphony/src. You can add the argument -DCMAKE_INSTALL_PREFIX=/name/of/dir to set the name of the directory to install to.
  3. Type make.
  4. Optionally, run make install to install the library and Python module onto your system.
  5. In the "build" folder, navigate into the newly created "susceptibility_tensor" folder. Make an empty file called "_init_.py" and save it.
  6. Unzip the "kernel_samples_datafiles.zip" file in place within the "symphony" folder.
  7. Navigate into the "src" folder, then into the "susceptibility_tensor" folder. Open the file "susceptibilityPy.pyx" and go to the line (line 289) that says main_directory = ; after the equals sign, add the absolute directory pointing to the "kernel_samples_datafiles/" folder.

To use symphony's Python interface:

  1. Navigate to the "build/" folder created in step 1., above. Open Python in the command line or by writing a ".py" file.
  2. Import symphony by typing import symphonyPy.
  • This allows one to call the functions: j_nu_py(), alpha_nu_py(), j_nu_fit_py(), alpha_nu_fit_py(), rho_nu_py(), and spline fit functions alpha_{I, Q, V}_spline(), rho_{Q, V}_spline(), and chi_ij().
  • The first two provide calculated values of the emissivity and absorptivity for the input parameters, and the second two provide the corresponding approximate fitting formula results.
  • The function rho_nu_py() computes Faraday rotation coefficients using the symphony integrator; this is slow, and should only be used as a check of the spline fits.
  1. The arguments of these functions can be found by accessing the associated docstrings. This can be done in the Python command line using the following:
import symphonyPy
symphonyPy.j_nu_py?

Arguments for Emissivity, Absorptivity, and Faraday Rotation Coefficient Functions

  • Arguments for all emissivity, absorptivity, and Faraday rotation coefficient functions in both C and Python take nearly the same arguments and output a double. The only difference is that the C version has an error_message parameter for handling evaluation errors, and the absorptivity coefficients include a switch to choose the integration method. The arguments are:
j_nu(nu, magnetic_field, electron_density, observer_angle, distribution, polarization, 
     theta_e, power_law_p, gamma_min, gamma_max, gamma_cutoff, kappa, kappa_width,
	 error_message)

Sample values:

j_nu_py(): j_nu_py(230e9, 30, 1, 1.047, symphonyPy.MAXWELL_JUETTNER, symphonyPy.STOKES_I,
                   10, 2.5, 1, 1000, 1e10, 3.5, 10)
  • Note: All parameters with units are in CGS.
  • Note: In C, the keys symphonyPy.MAXWELL_JUETTNER and symphonyPy.STOKES_I are members of a struct called params. They can be used with: params->MAXWELL_JUETTNER and params->STOKES_I.

TODO

  1. Add an anisotropic DF
  2. Put in warnings for frequencies outside the intended frequency regime (much greater than the plasma, relativistic cyclotron frequencies)
  3. Make cmake automatically add kernel_samples_datafiles/ directory to susceptibilityPy.pyx