Temperature comparison: pure ice vs. conductivity¶
Compare the along-track temperature distributions for the ASE attenuation dataset under the two computation modes. The two parquet files are produced by livist temperature (pure ice) and livist temperature --mode conductivity.
pure_ice = pandas.read_parquet("../data/temperature-ase-pure-ice.parquet")["temperature"]
conductivity = pandas.read_parquet("../data/temperature-ase-conductivity.parquet")["temperature"]
pure_ice.describe(), conductivity.describe()
(count 1.624351e+06
mean 2.551333e+02
std 3.333328e+00
min 2.036604e+02
25% 2.528742e+02
50% 2.542445e+02
75% 2.568683e+02
max 2.753755e+02
Name: temperature, dtype: float64,
count 1.624351e+06
mean 2.565409e+02
std 8.749606e+00
min 1.568602e+02
25% 2.506707e+02
50% 2.543089e+02
75% 2.608719e+02
max 3.145590e+02
Name: temperature, dtype: float64)
fig, ax = plt.subplots(figsize=(10, 6))
bins = 100
hist_range = (-40, 20)
ax.hist(pure_ice - 273.15, bins=bins, range=hist_range, alpha=0.5, label="pure ice", color="#3182ce")
ax.hist(conductivity - 273.15, bins=bins, range=hist_range, alpha=0.5, label="conductivity", color="#38a169")
ax.set_xlim(*hist_range)
ax.set_xlabel("temperature [°C]")
ax.set_ylabel("count")
ax.set_title("ASE along-track temperature: pure ice vs. conductivity")
ax.legend()
plt.tight_layout()
plt.show()
