ajdawson
commented
10 years ago A bit of preliminary diagnosis, expert assistance is required! Using a dummy data set to reproduce without iris and adding a matplotlib plot:
import numpy as np
import matplotlib.pyplot as plt
import cartopy.crs as ccrs
x = np.arange(0, 360, 2.5)
y = np.arange(-89, 90, 2)
X, Y = np.meshgrid(*[np.deg2rad(c) for c in (x, y)])
Z = np.cos(Y) + 0.375 * np.sin(2. * X)
Z[:, 1:3] = 2.
Z[:, -4:-2] = -2.
plt.pcolormesh(x, y, Z)
plt.figure()
ax = plt.axes(projection=ccrs.InterruptedGoodeHomolosine())
ax.coastlines()
ax.pcolormesh(x, y, Z, transform=ccrs.PlateCarree())
So as @mikewalker92 said there is a discontinuity, but if I had my data defined on the longitude interval -180 to 180:
import numpy as np
import matplotlib.pyplot as plt
import cartopy.crs as ccrs
x = np.arange(-180, 180, 2.5)
y = np.arange(-89, 90, 2)
X, Y = np.meshgrid(*[np.deg2rad(c) for c in (x, y)])
Z = np.cos(Y) + 0.375 * np.sin(2. * X)
Z[:, 1:3] = 2.
Z[:, -4:-2] = -2.
plt.pcolormesh(x, y, Z)
plt.figure()
ax = plt.axes(projection=ccrs.InterruptedGoodeHomolosine())
ax.coastlines()
ax.pcolormesh(x, y, Z, transform=ccrs.PlateCarree())
Things look better (aside from data going outside the map boundary, which is likely a separate issue). Problem is I don't know how to fix it! But hopefully this might be helpful at least.
pelson
The following code produces a plot containing a vertical discontinuity which appears to run through 0 longitude.
import iris import cartopy import matplotlib.pyplot as plt import iris.quickplot as qplt import cartopy.crs as ccrs
cube = iris.load_cube(iris.sample_data_path('A1B.2098.pp'))
plt.axes(projection=ccrs.InterruptedGoodeHomolosine())
qplt.pcolormesh(cube)
plt.gca().coastlines()
plt.show()
The plot also appears not to be plotting the correct data to the right of the discontinuity, as the uniform horizontal stripes do not look correct.