"""Definitions for the `DiffusionCSM` class."""
from collections import OrderedDict
import numpy as np
from scipy.interpolate import interp1d
from mosfit.constants import C_CGS, DAY_CGS, M_SUN_CGS, AU_CGS
from mosfit.modules.transforms.transform import Transform
# Important: Only define one ``Module`` class per file.
[docs]class DiffusionCSM(Transform):
"""Photon diffusion transform for CSM model."""
N_INT_TIMES = 3000
MIN_LOG_SPACING = -3
[docs] def process(self, **kwargs):
"""Process module."""
Transform.process(self, **kwargs)
self._kappa = kwargs[self.key('kappa')]
self._mass = kwargs[self.key('mcsm')] * M_SUN_CGS
self._R0 = kwargs[self.key('r0')] * AU_CGS # AU to cm
self._s = kwargs[self.key('s')]
self._rho = kwargs[self.key('rho')]
self._mejecta = kwargs[self.key('mejecta')] * M_SUN_CGS # Msol to grms
# scaling constant for CSM density profile
self._q = self._rho * self._R0 ** self._s
# outer radius of CSM shell
self._Rcsm = (
(3.0 - self._s) / (4.0 * np.pi * self._q) * self._mass +
self._R0 ** (3.0 - self._s)) ** (1.0 / (3.0 - self._s))
# radius of photosphere (should be within CSM)
self._Rph = abs(
(-2.0 * (1.0 - self._s) / (3.0 * self._kappa * self._q) +
self._Rcsm ** (1.0 - self._s)) ** (1.0 / (1.0 - self._s)))
self._tau_diff = (
self._kappa * self._mass) / (13.8 * C_CGS * self._Rph) / DAY_CGS
# mass of the optically thick CSM (tau > 2/3).
self._Mcsm_th = np.abs(4.0 * np.pi * self._q / (3.0 - self._s) * (
self._Rph**(3.0 - self._s) - self._R0 **
(3.0 - self._s)))
beta = 4. * np.pi ** 3. / 9.
td2 = self._tau_diff**2
td = self._tau_diff
t0 = self._kappa * (self._Mcsm_th) \
/ (beta * C_CGS * self._Rph) / DAY_CGS
new_lums = np.zeros_like(self._times_to_process)
if len(self._dense_times_since_exp) < 2:
return {self.dense_key('luminosities'): new_lums}
min_te = min(self._dense_times_since_exp)
tb = max(0.0, min_te)
linterp = interp1d(
self._dense_times_since_exp, self._dense_luminosities, copy=False,
assume_sorted=True, bounds_error=False, fill_value=0.0)
uniq_times = np.unique(self._times_to_process[
(self._times_to_process >= tb) & (
self._times_to_process <= self._dense_times_since_exp[-1])])
lu = len(uniq_times)
num = int(round(self.N_INT_TIMES / 2.0))
lsp = np.logspace(
np.log10(t0 /
self._dense_times_since_exp[-1]) +
self.MIN_LOG_SPACING, 0, num)
xm = np.unique(np.concatenate((lsp, 1 - lsp)))
int_times = np.clip(
tb + (uniq_times.reshape(lu, 1) - tb) * xm, tb,
self._dense_times_since_exp[-1])
int_times = tb + (uniq_times.reshape(lu, 1) - tb) * xm
int_tes = int_times[:, -1]
int_lums = linterp(int_times) # noqa: F841
int_args = int_lums * np.exp((int_times) / t0)
int_args[np.isnan(int_args)] = 0.0
uniq_lums = np.trapz(int_args, int_times)
uniq_lums*= np.exp(-int_tes/t0)/t0
new_lums = uniq_lums[np.searchsorted(uniq_times,
self._times_to_process)]
return {self.dense_key('luminosities'): new_lums}