Hertzsprung–Russell diagram¶

Stellar evolution data is stored in sev.83_[time] files, where the time is given in nbody units.

Let's have a look at the first sev.83_0 file. Note that purposely only the first 10 lines and the header are shown. Furthermore some columns are cut for the sake of brevity.

In [1]:
with open("example_1k/sev.83_0", "r") as myfile:
    # print header
    print(myfile.readline().replace("\n",""))
    # print first ten lines of stellar evolution data
    for i in range(0,10):
        print(myfile.readline()[:60].replace("\n","") + " [...]")

    1000      0.0
   0.00000E+00       1       1  1  2.51606E+00  5.25373E+01  [...]
   0.00000E+00       2       2  1  2.29434E+00  2.66005E+01  [...]
   0.00000E+00       3       3  1  8.72258E+00  1.97269E+01  [...]
   0.00000E+00       4       4  1  1.02014E+00  1.37391E+01  [...]
   0.00000E+00       5       5  1  1.54393E+00  1.16746E+01  [...]
   0.00000E+00       6       6  1  1.74069E+00  9.48461E+00  [...]
   0.00000E+00       7       7  1  1.45003E+00  9.24989E+00  [...]
   0.00000E+00       8       8  1  4.64537E+00  7.04640E+00  [...]
   0.00000E+00       9       9  1  8.03392E-01  6.10987E+00  [...]
   0.00000E+00      10      10  1  2.90117E+00  5.78393E+00  [...]

We do not know, what the columns mean, but looking at the manual we find

We can now extract the column names. As we are looking at just one sev.83 with all data at the same point in time, the TIME[NB] column will be skipped over.

In [2]:
# import some useful libraries
import pandas as pd
import matplotlib.pyplot as plt
In [3]:
# define column names
COLS = ["I", "NAME", "K*", "RI[RC]", "M[M*]", "log10(L[L*])", "log10(RS[R*])", "log10(Teff[K])"]

# read pandas DataFrame from csv
df = pd.read_csv("example_1k/sev.83_0",
                 delim_whitespace=True, # Make pandas use whitespace(s) as anbody does
                 skiprows = 1, # skip the header
                 names = COLS, # assign column names
                 usecols=range(1,len(COLS) + 1) # ignore the time coloumn
                 )
df
Out[3]:
I NAME K* RI[RC] M[M*] log10(L[L*]) log10(RS[R*]) log10(Teff[K])
0 1 1 1 2.516060 52.537300 5.57510 0.890588 4.71026
1 2 2 1 2.294340 26.600500 4.92389 0.723003 4.63125
2 3 3 1 8.722580 19.726900 4.59654 0.645161 4.58833
3 4 4 1 1.020140 13.739100 4.16783 0.549341 4.52907
4 5 5 1 1.543930 11.674600 3.96428 0.505943 4.49988
... ... ... ... ... ... ... ... ...
995 996 996 0 2.477800 0.080752 -3.26329 -0.816145 3.35403
996 997 997 0 3.337520 0.080735 -3.26366 -0.816115 3.35392
997 998 998 0 3.188000 0.080733 -3.26372 -0.816111 3.35390
998 999 999 0 0.684293 0.080726 -3.26387 -0.816099 3.35386
999 1000 1000 0 4.157180 0.080535 -3.26826 -0.815756 3.35259

1000 rows × 8 columns

Plotting¶

After extracting the data we can continue plot our HRD.

In [4]:
# Initialize figure
fig = plt.figure()
ax = fig.gca()

# plot
ax.scatter(df["log10(Teff[K])"], df["log10(L[L*])"])

# make the plot look nice
ax.set_title("Hertzsprung-Russel-Diagram")
# typical for HRDs
ax.invert_xaxis()

ax.set_xlabel(r"$\mathrm{log}_{10}(T_\mathrm{eff}$ [K])")
ax.set_ylabel(r"$\mathrm{log}_{10}(L$ [$L_\odot]$)")
ax.grid()