Create scatter plot where the color is based on year without adding extra column

You can do it without creating an additional column, but it won't be more elegant. DataFrame.plot.scatter supports different values for its color argument:

c: str, int or array-like, optional

The color of each point. Possible values are:

• A single color string referred to by name, RGB or RGBA code, for instance ‘red’ or ‘#a98d19’.

• A sequence of color strings referred to by name, RGB or RGBA code, which will be used for each point’s color recursively. For instance [‘green’,’yellow’] all points will be filled in green or yellow, alternatively.

• A column name or position whose values will be used to color the marker points according to a colormap.

So instead of doing:

df['Year'] = df['TIME'].dt.year
df.plot.scatter('X', 'Y', c='Year')

You could then manually derive colors using a matplotlib's color map objects:

def scale_to_unity(s):
    return (s - s.min()) / (s.max() - s.min())

def series_to_colors(s, cmap='viridis'):
    import matplotlib.pyplot as plt
    color_map = plt.get_cmap(cmap)
    return scale_to_unity(s).apply(color_map)

df.plot.scatter('X', 'Y', c=series_to_colors(df.TIME))

Like before, you can discard the precise time information by only using df.TIME.dt.year in the last line. However, the first variant looks much less cumbersome to me.

A 'cheap' way to color your scatter dots according to the year is to directly use matplotlib's scatter plot and pass the year as argument for the c parameter of scatter.

Here is a minimal example:

import datetime
import pandas as pd
from matplotlib import pyplot as plt
mydf = pd.DataFrame((
               (datetime.datetime(2020, 1, 1, 23, 00), 0.08, 0.3),
               (datetime.datetime(2021, 2,2,1,2,55), 0.2, 0.12), 
               (datetime.datetime(2022, 2,3,11, 00), 0.12, 0.7)
           ),
           columns=("TIME", 'X', 'Y')
       )

plt.scatter(mydf.X, mydf.Y, c=mydf.TIME.dt.year)

plt.show()

Then you can use the cmap parameter to specify what colormap you want to apply.

c = pd.to_datetime(df.TIME).dt.year
c = c.tolist()
X = df.X
Y = df.Y
plt.scatter(X,Y,c=c)

Don't have to create a column. Just convert the year to a list and then you can map whatever colour based on the list.