Skip to content

Zinterp #80

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
zhuochen-ucla wants to merge 9 commits into
base: dev
Choose a base branch
from
Open

Zinterp #80

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
7793bb9
<test>
Feb 10, 2021
7d19237
<update z_interpolation>
May 3, 2021
05b2147
<edit MIST model folder name>
May 4, 2021
29a3c7b
<update synthetic.py naming>
May 10, 2021
d94027d
update
Sep 16, 2021
957457e
<update evo.py>
Sep 16, 2021
7698b2f
Merge branch 'zinterp' of https://github.com/astropy/SPISEA into zinterp
Sep 16, 2021
da3c97a
Merge branch 'dev' into zinterp
Dec 9, 2021
ac82704
<test badremove>
Jul 5, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
347 changes: 272 additions & 75 deletions spisea/evolution.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,8 @@
from scipy import interpolate
import pylab as py
from spisea.utils import objects
import matplotlib.pyplot as plt


logger = logging.getLogger('evolution')

Expand Down Expand Up @@ -942,22 +944,22 @@ class MISTv1(StellarEvolution):
and 4/2019 (other metallicities). Default is 1.2.
"""
def __init__(self, version=1.2):
# define metallicity parameters for Parsec models
self.z_list = [0.0000015, # [Fe/H] = -4.00
0.0000047, # [Fe/H] = -3.50
0.000015, # [Fe/H] = -3.00
0.000047, # [Fe/H] = -2.50
0.00015, # [Fe/H] = -2.00
0.00026, # [Fe/H] = -1.75
0.00047, # [Fe/H] = -1.50
0.00084, # [Fe/H] = -1.25
0.0015, # [Fe/H] = -1.00
0.0026, # [Fe/H] = -0.75
0.0046, # [Fe/H] = -0.50
0.0082, # [Fe/H] = -0.25
0.015, # [Fe/H] = 0.00
0.024, # [Fe/H] = 0.25
0.041] # [Fe/H] = 0.50
# define metallicity parameters for MIST models
self.z_list = [0.0000014, # [Fe/H] = -4.00
0.0000045, # [Fe/H] = -3.50
0.000014, # [Fe/H] = -3.00
0.000045, # [Fe/H] = -2.50
0.00014, # [Fe/H] = -2.00
0.00025, # [Fe/H] = -1.75
0.00045, # [Fe/H] = -1.50
0.00080, # [Fe/H] = -1.25
0.0014, # [Fe/H] = -1.00
0.0025, # [Fe/H] = -0.75
0.0045, # [Fe/H] = -0.50
0.0080, # [Fe/H] = -0.25
0.014, # [Fe/H] = 0.00
0.025, # [Fe/H] = 0.25
0.045] # [Fe/H] = 0.50

# populate list of isochrone ages (log scale)
self.age_list = np.arange(5.01, 10.30+0.005, 0.01)
Expand All @@ -976,21 +978,21 @@ def __init__(self, version=1.2):

# Specifying metallicity
self.z_solar = 0.0142
self.z_file_map = {0.0000015: 'z0000015/',
0.0000047: 'z0000047/',
0.000015: 'z000015/',
0.000047: 'z000047/',
0.00015: 'z00015/',
0.00026: 'z00026/',
0.00047: 'z00047/',
0.00084: 'z00084/',
0.0015: 'z0015/',
0.0026: 'z0026/',
0.0046: 'z0046/',
0.0082: 'z0082/',
0.015: 'z015/',
0.024: 'z024/',
0.041: 'z041/'}
self.z_file_map = {0.0000014: 'z0000014/',
0.0000045: 'z0000045/',
0.000014: 'z000014/',
0.000045: 'z000045/',
0.00014: 'z00014/',
0.00025: 'z00025/',
0.00045: 'z00045/',
0.00080: 'z00080/',
0.0014: 'z0014/',
0.0025: 'z0025/',
0.0045: 'z0045/',
0.0080: 'z0080/',
0.014: 'z014/',
0.025: 'z025/',
0.045: 'z045/'}


def isochrone(self, age=1.e8, metallicity=0.0):
Expand All @@ -999,9 +1001,12 @@ def isochrone(self, age=1.e8, metallicity=0.0):
collection.
"""
# convert metallicity to mass fraction
## here z_defined is converted z from metallicity numbering
z_defined = self.z_solar * (10.**metallicity)

#print('input_metal and z:', metallicity, z_defined)

log_age = math.log10(age)
#print('logage', log_age)

# check age and metallicity are within bounds
if ((log_age < np.min(self.age_list)) or (log_age > np.max(self.age_list))):
Expand All @@ -1019,51 +1024,241 @@ def isochrone(self, age=1.e8, metallicity=0.0):
iso_file = 'iso_{0:.2f}.fits'.format(self.age_list[age_idx])

# find closest metallicity value
z_idx = searchsorted(self.z_list, z_defined, side='left')
if z_idx == len(self.z_list): # in case just over last index
z_idx = z_idx - 1
z_dir = self.z_file_map[self.z_list[z_idx]]

# generate isochrone file string
full_iso_file = self.model_dir + 'iso/' + z_dir + iso_file

# return isochrone data. Column locations depend on
# version
iso = Table.read(full_iso_file, format='fits')
#z_idx = np.where(abs(self.z_list - z_defined) == min(abs(self.z_list - z_defined)))[0][0]
z_idx = np.where(abs(np.array(self.z_list) - z_defined) == min(abs(np.array(self.z_list) - z_defined)))[0][0]
#print('z_idx:', z_idx)

def iso_extract(full_iso_file, version):
iso = Table.read(full_iso_file, format='fits')
if version == 1.0:
iso.rename_column('col7', 'Z')
iso.rename_column('col2', 'logAge')
iso.rename_column('col3', 'mass')
iso.rename_column('col4', 'logT')
iso.rename_column('col5', 'logg')
iso.rename_column('col6', 'logL')
iso.rename_column('col65', 'phase')
elif version == 1.2:
iso.rename_column('col1', 'EEP') ### add EEP here to do metallicity interpolation
iso.rename_column('col2', 'logAge')
iso.rename_column('col3', 'mass')
iso.rename_column('col4', 'mass_current')
iso.rename_column('col9', 'logL')
iso.rename_column('col14', 'logT')
iso.rename_column('col17', 'logg')
iso.rename_column('col79', 'phase')

# For MIST isochrones, anything with phase = 6 is a WD.
# Following our IFMR convention, change the phase designation
# to 101
isWD = np.where(iso['phase'] == 6)[0]
iso['phase'][isWD] = 101

# Define "isWR" column based on phase info
isWR = Column([False] * len(iso), name='isWR')
idx_WR = np.where(iso['phase'] == 9)[0]
isWR[idx_WR] = True
iso.add_column(isWR)

return iso

## if version = 1.0, no interpolation
if self.version == 1.0:
iso.rename_column('col7', 'Z')
iso.rename_column('col2', 'logAge')
iso.rename_column('col3', 'mass')
iso.rename_column('col4', 'logT')
iso.rename_column('col5', 'logg')
iso.rename_column('col6', 'logL')
iso.rename_column('col65', 'phase')
#z_dir = self.z_file_map[self.z_list[z_idx]]
z_dir = 'z015/'
# generate isochrone file string
full_iso_file = self.model_dir + 'iso/' + z_dir + iso_file
iso = iso_extract(full_iso_file, self.version)
iso.meta['log_age'] = log_age
iso.meta['metallicity_in'] = metallicity
iso.meta['metallicity_act'] = np.log10(self.z_list[z_idx] / self.z_solar)

elif self.version == 1.2:
iso.rename_column('col2', 'logAge')
iso.rename_column('col3', 'mass')
iso.rename_column('col4', 'mass_current')
iso.rename_column('col9', 'logL')
iso.rename_column('col14', 'logT')
iso.rename_column('col17', 'logg')
iso.rename_column('col79', 'phase')

# For MIST isochrones, anything with phase = 6 is a WD.
# Following our IFMR convention, change the phase designation
# to 101
isWD = np.where(iso['phase'] == 6)[0]
iso['phase'][isWD] = 101

# Define "isWR" column based on phase info
isWR = Column([False] * len(iso), name='isWR')
idx_WR = np.where(iso['phase'] == 9)[0]
isWR[idx_WR] = True
iso.add_column(isWR)

iso.meta['log_age'] = log_age
iso.meta['metallicity_in'] = metallicity
iso.meta['metallicity_act'] = np.log10(self.z_list[z_idx] / self.z_solar)

### start interpolation!!
##find the closest two isochrones and do interpolation
z_diff = float(z_defined - self.z_list[z_idx])
#print('z_diff', z_diff)
if z_diff < 0:
#print('z_diff < 0')
z_low = self.z_list[z_idx - 1]
z_high = self.z_list[z_idx]
elif z_diff > 0:
z_low = self.z_list[z_idx]
z_high = self.z_list[z_idx + 1]

## For testing
##=====================================================================##
test_z = False
if test_z:
z_low = self.z_list[12] ###z = 0
z_high = self.z_list[13] ###z = 0.25
#full_iso_file_test = self.model_dir + 'iso/' + 'ztest/feh013_iso_6.50.fits'
full_iso_file_test = self.model_dir + 'iso/' + 'ztest/feh013_iso_9.50.fits'
z_diff < 0
iso_test = iso_extract(full_iso_file_test, self.version)
print('iso_test_mass', iso_test['mass'])
##=====================================================================##

z_dir_low = self.z_file_map[z_low]
z_dir_high = self.z_file_map[z_high]

#print('z_low, z_high:', z_low, z_high)
#print('z_dir_low, z_dir_high:', z_dir_low, z_dir_high)
full_iso_file_low = self.model_dir + 'iso/' + z_dir_low + iso_file
full_iso_file_high = self.model_dir + 'iso/' + z_dir_high + iso_file
iso_low = iso_extract(full_iso_file_low, self.version)
iso_high = iso_extract(full_iso_file_high, self.version)
#print('Two_iso_numStars:', len(iso_low['mass']), len(iso_high['mass']))

common_EEP, ind_low, ind_high = np.intersect1d(iso_low['EEP'], iso_high['EEP'], return_indices = True)
iso_low_common = iso_low[ind_low]
iso_high_common = iso_high[ind_high]

mass_interp = []
mass_cu_interp = []
logage_interp = []
logT_interp = []
logL_interp = []
logg_interp = []
phase_interp = []
isWR_interp = []
print('Begin looping over each star')

for i in range(len(common_EEP)):
# Do interpolations in linear space
model_metals = [z_low, z_high]
target_metal = z_defined
#print('insideloop')
#print('model_metals, target', model_metals, target_metal)

logL_arr = [iso_low_common['logL'][i], iso_high_common['logL'][i]]
logT_arr = [iso_low_common['logT'][i], iso_high_common['logT'][i]]
logg_arr = [iso_low_common['logg'][i], iso_high_common['logg'][i]]
mass_arr = [iso_low_common['mass'][i], iso_high_common['mass'][i]]
mass_cu_arr = [iso_low_common['mass_current'][i], iso_high_common['mass_current'][i]]

f_logL = interpolate.interp1d(model_metals, logL_arr, kind = 'linear')
f_logT = interpolate.interp1d(model_metals, logT_arr, kind = 'linear')
f_logg = interpolate.interp1d(model_metals, logg_arr, kind = 'linear')
f_mass = interpolate.interp1d(model_metals, mass_arr, kind = 'linear')
f_mass_cu = interpolate.interp1d(model_metals, mass_cu_arr, kind = 'linear')

logL_interp.append(f_logL(target_metal))
logT_interp.append(f_logT(target_metal))
logg_interp.append(f_logg(target_metal))
mass_interp.append(f_mass(target_metal))
mass_cu_interp.append(f_mass_cu(target_metal))
logage_interp.append(log_age)

## choose the closest to copy the grids
if z_diff > 0:
phase_interp.append(iso_low['phase'][i])
isWR_interp.append(iso_low['isWR'][i])

else:
phase_interp.append(iso_high['phase'][i])
isWR_interp.append(iso_high['isWR'][i])
print('End looping over each star')

##iso_test
if test_z:
common_EEP_test, ind_test, ind_interp = np.intersect1d(iso_test['EEP'], common_EEP, return_indices = True)
iso_test_common = iso_test[ind_test]

py.figure(1, figsize = (20, 10))
py.subplot(121)
py.plot(logT_interp, logL_interp, 'b-', label = 'Interp')
py.plot(iso_test['logT'], iso_test['logL'], 'r-', label = 'Interp_predicted')
py.plot(iso_low['logT'], iso_low['logL'], 'k--', alpha = 0.5, label = 'metal_low')
py.plot(iso_high['logT'], iso_high['logL'], 'k--', alpha = 0.5, label = 'metal_high')
rng = py.axis()
py.xlim(rng[1], rng[0])
py.xlabel('log Teff')
py.ylabel('log L')
py.legend()

py.subplot(122)
py.plot(logT_interp, logg_interp, 'b-', label = 'Interp')
py.plot(iso_test['logT'], iso_test['logg'], 'r-', label = 'Interp_predicted')
py.plot(iso_low['logT'], iso_low['logg'], 'k--', alpha = 0.5, label = 'metal_low')
py.plot(iso_high['logT'], iso_high['logg'], 'k--', alpha = 0.5, label = 'metal_high')
rng = py.axis()
py.xlim(rng[1], rng[0])
py.xlabel('log Teff')
py.ylabel('log g')
py.legend()
#py.savefig('/u/zhuochen/software/zhuocode/code_fitting/plots/interp_iso_EEP.png')


diff_logT = []
diff_logL = []
diff_logg = []
diff_logT_frac = []
diff_logL_frac = []
diff_logg_frac = []
for i in range(len(common_EEP_test)):
diff_logT.append(logT_interp[i] - iso_test_common['logT'][i])
diff_logL.append(logL_interp[i] - iso_test_common['logL'][i])
diff_logg.append(logg_interp[i] - iso_test_common['logg'][i])
diff_logT_frac.append((logT_interp[i] - iso_test_common['logT'][i])/iso_test_common['logT'][i])
diff_logL_frac.append((logL_interp[i] - iso_test_common['logL'][i])/iso_test_common['logL'][i])
diff_logg_frac.append((logg_interp[i] - iso_test_common['logg'][i])/iso_test_common['logg'][i])

py.figure(3, figsize = (20, 10))
py.subplot(121)
py.plot(iso_test_common['logT'], diff_logT, 'ko')
rng = py.axis()
py.xlim(rng[1], rng[0])
py.xlabel('log Teff (True)')
py.ylabel('Difference (logT_interp - logT_true)')
py.ylim(-0.02, 0.02)

py.subplot(122)
py.plot(iso_test_common['logT'], diff_logT_frac, 'ko')
rng = py.axis()
py.xlim(rng[1], rng[0])
py.xlabel('log Teff (True)')
py.ylabel('Fraction (logT_interp - logT_true)/logT_true')
py.ylim(-0.02, 0.02)
py.savefig('/u/zhuochen/software/zhuocode/code_fitting/plots/logT_diff_interpEEP.png')

py.figure(4, figsize = (20, 10))
py.subplot(121)
py.plot(iso_test_common['logL'], diff_logL, 'ko')
py.xlabel('log L (True)')
py.ylabel('Difference (logL_interp - logL_true)')
py.ylim(-0.12, 0.12)

py.subplot(122)
py.plot(iso_test_common['logL'], diff_logL_frac, 'ko')
py.xlabel('log L (True)')
py.ylabel('Fraction (logL_interp - logL_true)/logL_true')
py.ylim(-0.5, 0.5)
py.savefig('/u/zhuochen/software/zhuocode/code_fitting/plots/logL_diff_interpEEP.png')

py.figure(5, figsize = (20, 10))
py.subplot(121)
py.plot(iso_test_common['logg'], diff_logg, 'ko')
py.xlabel('log g (True)')
py.ylabel('Difference (logg_interp - logg_true)')
py.ylim(-0.1, 0.1)

py.subplot(122)
py.plot(iso_test_common['logg'], diff_logg_frac, 'ko')
py.xlabel('log g (True)')
py.ylabel('Fraction (logg_interp - logg_true)/logg_true')
py.ylim(-10, 10)
py.savefig('/u/zhuochen/software/zhuocode/code_fitting/plots/logg_diff_interpEEP.png')


iso = Table([logage_interp, mass_interp, mass_cu_interp, logL_interp, logT_interp, logg_interp, phase_interp, isWR_interp],
names=['logAge', 'mass', 'mass_current', 'logL', 'logT', 'logg', 'phase', 'isWR'])
iso.meta['log_age'] = log_age
iso.meta['metallicity_in'] = metallicity
iso.meta['metallicity_act'] = np.log10(self.z_list[z_idx] / self.z_solar)

return iso


def format_isochrones(self, input_iso_dir, metallicity_list):
r"""
Expand Down Expand Up @@ -1124,11 +1319,13 @@ def format_isochrones(self, input_iso_dir, metallicity_list):
os.chdir(start_dir)
return



#==============================#
# Merged model classes
#==============================#
class MergedBaraffePisaEkstromParsec(StellarEvolution):
"""
r"""
This is a combination of several different evolution models:

* Baraffe (`Baraffe et al. 2015 <https://ui.adsabs.harvard.edu/abs/2015A%26A...577A..42B/abstract>`_)
Expand Down
Loading