In [7]:

    

import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline


    

In [26]:

    

pulse_len = 300
k =3.
N1 = 0.01
eps = 0.2
Lx, Lz = (3000000, 10000)


    

In [13]:

    

def tau_approx(m): 
    tau = Lx*np.pi*m**2/(2.*Lz*eps*N1*k)
    return tau
def tau_exact(m): 
    tau = tau_approx(m) + eps * (Lx/Lz) * (2. * (m*np.pi)**2 - 3.)/(12. * N1 * k * np.pi)
    return tau
def tau_off(m):
    tau = Lx*(6 + np.pi**2*m**2*(1.+3.*eps**2))/(6.*eps*N1*k*np.pi*Lz)
    return tau


    

In [27]:

    

tau_exact(2)*1


    

    
Out[27]:





318188.9006206738

In [116]:

    

m = np.linspace(0,20)
plt.plot(m, tau_exact(m))
plt.plot(m, tau_off(m), marker = 'x')


    

    
Out[116]:





[<matplotlib.lines.Line2D at 0x1121c9198>]






    

In [21]:

    

plt.plot(m, tau_exact(m)/tau_off(m))


    

    
Out[21]:





[<matplotlib.lines.Line2D at 0x7fc7345750d0>]






    

In [186]:

    

import numpy as np
import h5py
import dedalus_plots as dp
from scipy import stats

sim_name = 'k2m1'
filepath = sim_name + "/" + sim_name + "_s1/" + sim_name + "_s1_p0.h5"
data = h5py.File(filepath, "r")
dict_vars = {'b3d':'buoyancy', 'tropenergy':'tropo energy'}

vars = dp.read_vars(data, dict_vars)
dims = dp.read_dims(data)
data.close()

#parameters
m = 1
k = 10.
N1 = 0.01
eps = 1
Lx, Lz = (2000000, 10000)


tau_approx = Lx*np.pi*m**2/(2.*Lz*eps*N1*k)
tau_exact = tau_approx + eps * (Lx/Lz) * (2. * (m*np.pi)**2 - 3.)/(12. * N1 * k * np.pi)

tau_off = Lx*(6 + np.pi**2*m**2*(1.+3.*eps**2))/(6.*eps*N1*k*np.pi*Lz)


energ_normed = vars['tropenergy'][:,0,0]/np.max(vars['tropenergy'][:,0,0])
energ_theory = np.exp(-(dims['t'] )/tau_exact)
energ_approx  = np.exp(-(dims['t'])/tau_approx)
energ_off  = np.exp(-(dims['t'] )/tau_off)


    

In [187]:

    

plt.plot(dims['t'], np.log(energ_normed))
plt.plot(dims['t'], np.log(energ_theory))


    

    
Out[187]:





[<matplotlib.lines.Line2D at 0x11452d7f0>]






    

In [153]:

    

plt.plot(dims['t'], vars['tropenergy'][:,0,0])


    

    
Out[153]:





[<matplotlib.lines.Line2D at 0x1122d1e10>]






    

In [1]:

    

from scipy import stats
import pickle


    

In [11]:

    

files = !ls


    

In [12]:

    

test = [i for i in files if i[-1] == 'p']


    

In [13]:

    

for a in test:
    print(a)


    

    




eps02.p
eps02_m2.p
eps03.p
eps03_m2.p
eps04.p
eps04_m2.p
eps05.p
eps05_m2.p
eps06.p
eps06_m2.p
eps07.p
eps07_m2.p
eps08.p
eps08_m2.p
eps09.p
eps09_m2.p
eps1.p
eps1_m2.p

In [24]:

    

biglist = []
for a in test:
    infile = open( a, 'rb')
    tmp = pickle.load(infile)
    biglist = biglist + tmp


    

In [26]:

    

biglist[0]


    

    
Out[26]:





{'eps': 0.2,
 'intercept': 0.03488821948901466,
 'k': 3,
 'm': 1,
 'r': -0.99348217945311446,
 'std_err': 1.6250419988868962e-07,
 'tau': 43454.098434801024,
 'tau_approx': 39269.90816987241,
 'tau_theory': 39713.92947392476}

In [27]:

    

x = np.linspace(8,12)
x = np.linspace(0,3)
plt.figure(figsize=(10,5))
for a in biglist:
       # plt.scatter(np.log(a['tau']), np.log(a['tau_theory']), color = 'green')
        plt.scatter(a['tau']/86400., a['tau_theory']/86400., color = 'green', marker = 'x')
        plt.errorbar(a['tau']/86400., a['tau_theory']/86400.,xerr = a['std_err']*a['tau']*a['tau']/86400)
plt.plot(x,x,color = 'black', ls ='--')
plt.xlim(0,2)
plt.ylim(0,2)
plt.xlabel('theoretical $\\tau$ (days)', fontsize = 16)
plt.ylabel('simulated $\\tau$ (days)', fontsize = 16)


    

    
Out[27]:





<matplotlib.text.Text at 0x10e4d40f0>






    

In [10]:

    

x = np.linspace(8,12)
x = np.linspace(0,2)
for run in biglist:
    for a in run: 
       # plt.scatter(np.log(a['tau']), np.log(a['tau_theory']), color = 'green')
       # plt.scatter(a['tau']/86400., a['tau_theory']/86400., color = 'green')
        plt.scatter(a['std_err']*a['tau'], a['k'])
#plt.plot(x,x,color = 'black', ls ='--')
#plt.xlim(0,2)
#plt.ylim(0,2)


    

In [38]:

    

x = np.linspace(8,12)
x = np.linspace(0,2)
plt.figure(figsize=(10,5))
for run in biglist:
    for a in run: 
       # plt.scatter(np.log(a['tau']), np.log(a['tau_theory']), color = 'green')
        plt.scatter(a['m'], a['tau']/a['tau_theory'])
        plt.errorbar(a['m'], a['tau']/a['tau_theory'],yerr = a['std_err']*a['tau'] )
plt.axhline(y=1, xmin = 0, xmax = 1, color = 'black')
#plt.xlim(0,2)
#plt.ylim(0,2)


    

    
Out[38]:





<matplotlib.lines.Line2D at 0x10fd2b6d8>






    

In [41]:

    

for a in archive_list1:
    plt.scatter(a['k'], a['tau'])
    plt.scatter(a['k'], a['tau_theory'], color= 'red')
for a in archive_list05:
    plt.scatter(a['k'], a['tau'])
    plt.scatter(a['k'], a['tau_theory'], color= 'red')
for a in archive_list05_m2:
    plt.scatter(a['k'], a['tau'])
    plt.scatter(a['k'], a['tau_theory'], color= 'red')
for a in archive_list1_m2:
    plt.scatter(a['k'], a['tau'], marker = 'x')
    plt.scatter(a['k'], a['tau_theory'],  color= 'red', marker = 'x')


    

In [54]:

    

x = np.linspace(0,80000)
for a in archive_list1:
    plt.scatter(a['tau'], a['tau_theory'])
for a in archive_list05:
    plt.scatter(a['tau'], a['tau_theory'], color = 'red')
for a in archive_list07:
    plt.scatter(a['tau'], a['tau_theory'], color = 'purple')
for a in archive_list05_m2:
    plt.scatter(a['tau'], a['tau_theory'], color = 'green')
for a in archive_list1_m2:
    plt.scatter(a['tau'], a['tau_theory'], color = 'orange', marker = 'x')
plt.plot(x,x, ls ='--')
plt.xlim(0,20000)
plt.ylim(0,20000)


    

    
Out[54]:





(0, 20000)






    

In [46]:

    

for a in archive_list1:
    plt.scatter(a['k'], a['intercept'])
for a in archive_list05:
    plt.scatter(a['k'], a['intercept'], color = 'red')
for a in archive_list07:
    plt.scatter(a['k'], a['intercept'], color = 'purple')
for a in archive_list05_m2:
    plt.scatter(a['k'], a['intercept'], color = 'green')
for a in archive_list1_m2:
    plt.scatter(a['k'], a['intercept'], color = 'orange')


    

In [44]:

    

for a in archive_list1:
    plt.scatter(a['k'], a['r']**2)
for a in archive_list05:
    plt.scatter(a['k'], a['r']**2, color ='red')
for a in archive_list05_m2:
    plt.scatter(a['k'], a['r']**2, color ='green')
for a in archive_list1_m2:
    plt.scatter(a['k'], a['r']**2, color ='orange')


    

In [45]:

    

for a in archive_list1:
    plt.scatter(a['k'], 1./a['std_err'])
for a in archive_list05:
    plt.scatter(a['k'], 1./a['std_err'], color ='red')
for a in archive_list05_m2:
    plt.scatter(a['k'], 1./a['std_err'], color ='green')
for a in archive_list1_m2:
    plt.scatter(a['k'], 1./a['std_err'], color ='orange')


    

In [ ]: