interestingly enough, this happens even after switching to the "normal" colorbar implementation (#13397), even though this doesn't happen for non-imagegrid axes...

It seems like it could be related to #12155 as when ImageGrid is called with the above parameters it creates a default colorbar on each row which then has to be updated. I tried to adapt the solutions in #12155 but no luck yet. I also came across #13397 but haven't had the time to install locally to test, I guess now I don't

Actually this kinda works cb1 = grid.cbar_axes[1].colorbar(im1, ticks=[0.001,10, 1000]) so seems like there is an extra axis somewhere ...

The code also errors out when run interactively with the qt backend and moving the mouse over the axes.

AttributeError: 'Colorbar' object has no attribute 'formatter'
Traceback (most recent call last):
  File "d:\***\lib\matplotlib\cbook\__init__.py", line 196, in process
    func(*args, **kwargs)
  File "d:\***\lib\matplotlib\backend_bases.py", line 2805, in mouse_move
    data_str = a.format_cursor_data(data)
  File "d:\***\lib\matplotlib\image.py", line 972, in format_cursor_data
    + "]")
AttributeError: 'Colorbar' object has no attribute 'formatter'

Have you tried to set vmin when performing plt.imshow?

You can see an example of this here

https://stackoverflow.com/questions/56439837/colorbar-scaling-in-imagegrid

Yea it doesn't help unfortunately

Have you tried something without Imagegrid? Because I think the problem lies there...

Here, I tried something without Imagegrid. Works better to say the least

import numpy as np
from matplotlib import pyplot as plt
from matplotlib.colors import LogNorm
from mpl_toolkits.axes_grid1.axes_grid import ImageGrid

x,y = np.ogrid[-4:4:31j,-4:4:31j]
z = 120000*np.exp(-x**2-y**2)


fig, ax = plt.subplots(2,1)
map_ax0 = ax[0].imshow(z)
map_ax1 = ax[1].imshow(z,  norm=LogNorm(z.min(), z.max()))
plt.colorbar(mappable=map_ax0, ax=ax[0])
plt.colorbar(mappable=map_ax1, ax=ax[1])
plt.tight_layout()

Here is an example