How to merge two differently multi-indexed dataframes

I have two multi-indexed dataframes df1 with three levels and df2 with two. The indices resulted from df1.groupby([col_1, col_2, col_3]) and df2.groupby([col_1, col_2]). col_1 and col_2 are the same in both dataframes, but, because of the third level in df1 of different lenghts; df1has 2425 rows and df2783.

What I'm trying to do is to merge both dataframes so that df2 gets spread up that the length of indices level 0 and 1 are of the same length in df1and df2 so that the resulting dataframe is also of 2425 rows.

I used df3 = df1.merge(df2, left_index=True, right_index=True) but the resulting dataframe remains with only 2385 rows. I used df3 = pd.concat([df1, df2], axis=1) but it raised a ValueError: operands could not be broadcast together with shapes.

Is there an elegant way to solve this? I appreciate every help

EDIT: data sample

df1:

                                                                    Areaclccat1990  ...  Areaclccat2012
FID_Weser_Catchments_134_WQ_Stations_FINAL_LAEA... SNR1 gridcode_1                  ...                
0                                                  3152 1                 0.002764  ...        0.007248
                                                        2                 0.980105  ...        0.972941
                                                        3                 0.005049  ...        0.017166
                                                        4                 0.012082  ...        0.002645
                                                   3155 1                      NaN  ...        0.000003
                                                        2                 1.000000  ...        0.996788
                                                        3                      NaN  ...        0.003209
                                                   3255 1                      NaN  ...        0.058950
                                                        2                 0.989654  ...        0.941050
                                                        4                 0.010346  ...             NaN
                                                   5958 1                      NaN  ...        0.004463
                                                        2                 0.955098  ...        0.958452
                                                        3                 0.014408  ...        0.027835
                                                        4                 0.030494  ...        0.009250
                                                   5966 1                 0.007184  ...        0.011448
                                                        2                 0.955668  ...        0.949824
                                                        3                 0.037148  ...        0.038728
                                                   5970 1                      NaN  ...        0.001141
                                                        2                 0.979750  ...        0.930495
                                                        3                 0.011281  ...        0.068364

df2:

                                                          Areaclccat1990  ...  Areaclccat2012
FID_Weser_Catchments_134_WQ_Stations_FINAL_LAEA... SNR1                   ...                
0                                                  3152      1654.636456  ...     1550.415658
                                                   3155      1820.433231  ...     1758.125539
                                                   3255        43.056576  ...       39.436385
                                                   5958      2306.806057  ...     2120.791289
                                                   5966         7.444977  ...        5.763853
                                                   5970      3087.717009  ...     2615.253450
                                                   6435       240.342745  ...      255.033888
                                                   6534       647.293171  ...      621.116222
                                                   6535      9929.136397  ...     9653.021903
                                                   6611       947.912232  ...      754.783147
                                                   6631     13528.073523  ...    13545.356498
                                                   6632     14023.097062  ...    13897.394309
                                                   6633      5913.895620  ...     5398.585720
                                                   6634     17463.795952  ...    17159.138628
                                                   6635     10791.618411  ...    10306.725199
                                                   6636      9664.138661  ...     9742.442935
                                                   9473       131.268559  ...      128.477078
                                                   9672       107.831005  ...      102.464959
                                                   9673        13.044806  ...       29.566828
                                                   16051      443.810802  ...      428.493495

Solution

Coonvert third level to column before merge with how='left' for left join:

df3 = df1.reset_index(level=2).merge(df2, left_index=True, right_index=True, how='left')