def three_fields_corner_plot():
survey = 'CCAT-p'
lines = ['6-5', '5-4', '4-3']
b = 3
bandwidth = 60
kmax = 1
zobs = 0.88
nuemit = np.array([CO_data.CO_lines[l] for l in lines])
nuobs = nuemit / (1.+zobs)
Vsurv = CO_data.calc_Vsurv(nuobs[0], nuemit[0], bandwidth, CO_data.survey_area[survey], cosmo)
Blist, Nlist = CO_data.gen_Blist_Nlist(3, bandwidth, kmax, zobs, lines, survey, cosmo)
tf = threefield(zobs, Blist, Nlist, kmax, Vsurv, cosmo)
labels = [r'$B_{6-5}\,[\,\text{Jy}/\text{str}\,]$', r'$B_{5-4}\,[\,\text{Jy}/\text{str}\,]$',
r'$B_{4-3}\,[\,\text{Jy}/\text{str}\,]$']
fig, ax, _ = corner_plot(zobs, Blist, Nlist, cosmo, 1.75, tf=tf, labels=labels, printtext=False)
fig.tight_layout()
fig.savefig('CO_tf.pdf')
def plot_all_SN(survey, lines, zbounds=[0, 10], nz=1000):
freq_range = CO_data.survey_freq_range[survey]
b = 3
bandwidth = 60
kmax = 1
# Vsurv = CO_data.calc_Vsurv(nuobs[0], nuemit[0], bandwidth, CO_data.survey_area[survey], cosmo)
# Blist, Nlist = CO_data.gen_Blist_Nlist(3, bandwidth, kmax, zobs, lines, survey, cosmo)
zextent_list = []
for l in lines:
nuemit = CO_data.CO_lines[l]
zextent = (nuemit - freq_range)/freq_range
zextent_list.append(np.flip(zextent))
zextent_list = np.array(zextent_list)
print(zextent_list)
zlist = np.linspace(zbounds[0], zbounds[1], nz)
lines_obs_list = []
tf_bool_list = []
for z in zlist:
bool1 = z > zextent_list[:,0]
bool2 = z < zextent_list[:,1]
keys = np.where(np.logical_and(bool1, bool2))[0]
lines_obs = [ lines[k] for k in keys ]
lines_obs_list.append(lines_obs)
if len(lines_obs) >= 3:
tf_bool_list.append(True)
else:
tf_bool_list.append(False)
mf_keys = np.where(tf_bool_list)[0]
# first generate the S/N Cramer-Rao bound for the auto-spectrum
# this is done at every redshift for each line that is in band
SN_auto_list = []
SN_mf_list = []
for z, lines_at_z, mfbool in zip(zlist, lines_obs_list, tf_bool_list):
SN = np.zeros(np.shape(lines))
SN_mf = np.zeros(np.shape(lines))
nuemit = CO_data.CO_lines[l]
nuobs = nuemit / (1.+z)
Vsurv = CO_data.calc_Vsurv(nuobs, nuemit, bandwidth, CO_data.survey_area[survey], cosmo)
Blist, Nlist = CO_data.gen_Blist_Nlist(3, bandwidth, kmax, z, lines_at_z, survey, cosmo)
mf = multifield(z, Blist, Nlist, kmax, Vsurv, cosmo)
Vk = _Vk_(mf.klist, mf.Pklist, Vsurv)
for i,l in enumerate(lines_at_z):
for key, lp in enumerate(lines):
if l == lp:
SN[key] = Blist[i]**2 * Vk / Nlist[i]
if mfbool:
cov = np.linalg.inv(mf.fmat)
noise = np.sqrt(np.diag(cov))
sn = Blist / noise
for i,l in enumerate(lines_at_z):
for key, lp in enumerate(lines):
if l == lp:
SN_mf[key] = sn[i]
SN_auto_list.append(SN)
SN_mf_list.append(SN_mf)
SN_auto_list = np.array(SN_auto_list)
SN_mf_list = np.array(SN_mf_list)
for i,l in enumerate(lines):
plt.plot(zlist, SN_auto_list[:,i], ls='dashed', c=tb_c[i])
plt.plot(zlist, SN_mf_list[:,i], label=l, c=tb_c[i])
# now generate the S/N Cramer-Rao bound for the auto-spectrum
# done at redshifts where we have >= 3 fields
plt.xlabel(r'$z$')
plt.ylabel(r'$\text{S}/\text{N}(B)$')
plt.yscale('log')
plt.xlim(zbounds)
plt.legend(frameon=False, title=r'$B$')
plt.show()
return zlist, SN_auto_list
'ns': 1.0
}
cosmo = cosmology.setCosmology('myCosmo', params)
survey = 'CCAT-p'
lines = ['6-5', '5-4', '4-3']
b = 3
bandwidth = 60
kmax = 1
zobs = 0.88
nuemit = np.array([CO_data.CO_lines[l] for l in lines])
nuobs = nuemit / (1. + zobs)
Blist, Nlist = CO_data.gen_Blist_Nlist(3, bandwidth, kmax, zobs, lines, survey,
cosmo)
Vsurv = CO_data.calc_Vsurv(nuobs[0], nuemit[0], bandwidth,
CO_data.survey_area[survey], cosmo)
fig, ax, _ = corner_plot(1,
np.array([1, 2, 3]),
np.array([100, 100, 100]),
cosmo,
0.25,
kmax=1,
Vk=200000,
norm=True,
intstr=True)
fig.tight_layout()
fig.savefig('corner_Nconstant.pdf')