Plotting heat data (with pcolormesh or similar) fails in the edges of map (cartopy)

Question

I want to plot heat data on a map (Cartopy projection), but on the edges of the map, a frame of cells is created regardless of the size of the array. It was noticed, that the columns on the far left and far right have identical values.

Code:

import cartopy.crs as ccrs
import matplotlib.pyplot as plt
import numpy as np

ax = plt.axes(projection=ccrs.PlateCarree())
ax.set_global()
ax.coastlines()

heat_data = np.random.rand(10, 10)
lat = np.linspace(-90, 90, heat_data.shape[1])
lon = np.linspace(-180, 180, heat_data.shape[0])
ax.pcolormesh(lon, lat, np.transpose(heat_data), alpha = 0.2)
plt.show()

Outcome No1: Result without transparency of the array (first and last columns have identical values)

Outcome No2: Result with transparent array

Answer 1

Running the code you provided worked fine for me, producing a map with transparent pcolormesh. I believe this is the desired result?

You also have a fencepost issues which may be causing the problem. Note that while you have made a 10 x 10 grid of cells, you specify lon and lat in a 10 x 10 grid that includes the endpoints, mot the centers of the grid. This will confuse meshgrid. Could you try:

lat = np.linspace(-90, 90, heat_data.shape[1] +1)
lon = np.linspace(-180, 180, heat_data.shape[0] +1)

ie setting the edges of the blocks in the desired places.