mass_function.py

from __future__ import print_function
import numpy as np
from scipy.interpolate import RegularGridInterpolator
from sigma import Sigma
import parameters as par

class MassFunction(object):
    
    def __init__(self):
        
        self.sigma = Sigma() #sigma(M)

        # read in MXXL mass function fit parameters
        snap, redshift, A, a, p = \
                   np.loadtxt(par.mf_fits_file, skiprows=1, unpack=True)

        # interpolate parameters
        self._A = RegularGridInterpolator((redshift,), A, bounds_error=False, 
                                          fill_value=None)

        self._a = RegularGridInterpolator((redshift,), a, bounds_error=False, 
                                          fill_value=None)

        self._p = RegularGridInterpolator((redshift,), p, bounds_error=False, 
                                          fill_value=None)

    def A(self, redshift):
        return self._A(redshift)

    def a(self, redshift):
        return self._a(redshift)

    def p(self, redshift):
        return self._p(redshift)

    def mass_function(self, log_mass, redshift):
        '''
        Fit to the MXXL mass function with the form of a Sheth-Tormen
        mass function
        '''
        sigma = self.sigma.sigma(log_mass, redshift)

        dc=1
        A = self.A(redshift)
        a = self.a(redshift)
        p = self.p(redshift)
        
        mf = A * np.sqrt(2*a/np.pi)
        mf *= 1 + (sigma**2 / (a * dc**2))**p
        mf *= dc / sigma
        mf *= np.exp(-a * dc**2 / (2*sigma**2))

        return mf

def test():
    import matplotlib.pyplot as plt
    mf = MassFunction()

    log_mass = np.arange(10, 16, 0.1)
    
    for z in np.arange(0, 1, 0.1):
        plt.plot(log_mass, mf.mass_function(log_mass, np.ones(len(log_mass))*z))

    plt.yscale('log')

    plt.show()

if __name__ == '__main__':

    test()