Mar-21-2024, 09:36 AM
Hi all,
I have a flow field with horizontal (realxh) and vertical (realzh) coordinates and ice flow velocities (vertical (wi) and horizontal (ui)), and I want to plot 2D velocity vectors within the ice sheet. I can do it with quiver:
which results in the plot in attachment. I think it looks okay, but the vectors at the right are quite big (so big they go out the ice sheet outlines) and at the bottom they are quite small. I therefore wanted to try streamlines instead of quiver (which should plot continuous lines of the ice flow instead of vectors), to make it look more aesthetic/realistic:
But it results in the following error:
Does anyone know how to resolve this? Apparently the shape of my coordinate files are not valid in this case. I have looked online for hours but did not find the answer. It works fine for quiver, but not for streamlines. I put the .mat files of the matrices in attachment (zip) so you can test it out (can be done by from scipy.io import loadmat). You can put nx= 31, nz = 15, and domainsize/(np.max(sur)-np.min(bed)) = 175. The latter scale factor is done to scale the y-axis with the size of the x-axis (the vertical ice thickness and its discretization scheme is lots thinner than the ice sheet horizontal extent).
I have a flow field with horizontal (realxh) and vertical (realzh) coordinates and ice flow velocities (vertical (wi) and horizontal (ui)), and I want to plot 2D velocity vectors within the ice sheet. I can do it with quiver:
fig, ax = plt.subplots()
ax.quiver(realxh[0:nz,0:nx-1], realzh[0:nz,0:nx-1], ui[0:nz,0:nx-1], wi[0:nz,0:nx-1]*(domainsize/(np.max(sur)-np.min(bed))),color='b')
ax.set_title('Ice sheet 2D flow field')
plt.ylabel('Elevation (m)')
plt.xlabel('Distance (m)')
plt.show() which results in the plot in attachment. I think it looks okay, but the vectors at the right are quite big (so big they go out the ice sheet outlines) and at the bottom they are quite small. I therefore wanted to try streamlines instead of quiver (which should plot continuous lines of the ice flow instead of vectors), to make it look more aesthetic/realistic:
fig, ax = plt.subplots()
ax.streamplot(realxh[0:nz,0:nx-1], realzh[0:nz,0:nx-1], ui[0:nz,0:nx-1], wi[0:nz,0:nx-1]*(domainsize/(np.max(sur)-np.min(bed))),color='b')
ax.set_title('Ice sheet 2D flow field')
plt.ylabel('Elevation (m)')
plt.xlabel('Distance (m)')
plt.show() But it results in the following error:
Error:Traceback (most recent call last):
File "/Users/yv/PycharmProjects/icesheet/main.py", line 382, in <module>
ax.streamplot(realxh[0:nz,0:nx-1], realzh[0:nz,0:nx-1], ui[0:nz,0:nx-1], wi[0:nz,0:nx-1]*(domainsize/(np.max(sur)-np.min(bed))),color='b')
File "/Users/yv/PycharmProjects/icesheet/.venv/lib/python3.12/site-packages/matplotlib/streamplot.py", line 342, in __init__
raise ValueError("The columns of 'y' must be equal")
ValueError: The columns of 'y' must be equalDoes anyone know how to resolve this? Apparently the shape of my coordinate files are not valid in this case. I have looked online for hours but did not find the answer. It works fine for quiver, but not for streamlines. I put the .mat files of the matrices in attachment (zip) so you can test it out (can be done by from scipy.io import loadmat). You can put nx= 31, nz = 15, and domainsize/(np.max(sur)-np.min(bed)) = 175. The latter scale factor is done to scale the y-axis with the size of the x-axis (the vertical ice thickness and its discretization scheme is lots thinner than the ice sheet horizontal extent).