Basemap, projection='geos', controlling mark_inset() location

Question

I'm trying to make a plot using basemap with an inset zoom. I'm having trouble with the inset box from mark_inset() to appear anywhere besides the lower left hand corner. It works for projection='cyl' but not for projection='geos' which is what I want.

from mpl_toolkits.basemap import Basemap 
from mpl_toolkits.axes_grid1.inset_locator import zoomed_inset_axes
from mpl_toolkits.axes_grid1.inset_locator import mark_inset
import matplotlib.pyplot as plt

fig = plt.figure()
ax = fig.add_subplot(111)

map1 = Basemap(projection='geos', lat_0=0, lon_0=0)
map1.drawmapboundary()
map1.drawcoastlines()

axins = zoomed_inset_axes(ax, 7, loc=3)
axins.set_xlim(-12, 5)
axins.set_ylim(50, 60)

map2 = Basemap(projection='geos', lon_0=0, llcrnrlon=-12, llcrnrlat=50, urcrnrlon=5, urcrnrlat=60[enter image description here][1])
map2.drawcoastlines()

mark_inset(ax, axins, loc1=2, loc2=4, fc="none", ec="0.5") 

Answer 1

Bumping this...can confirm it's an issue.

I haven't tried every projection, but of the dozen I have tried, 'cyl' is the only one where mark_inset works. All others locate it at lower left.

Answer 2

I encountered the same problem. This is because Basemap works internally with projected coordinates - except for the cyl projection; read here :

Calling a Basemap class instance with the arguments lon, lat will convert lon/lat (in degrees) to x/y map projection coordinates (in meters). [...]

For cylindrical equidistant projection (cyl), this does nothing (ie x,y == lon,lat).

To make the marking work properly, you need to first extract the inset map limits, convert them to geographic coordinates, and finally re-project them again onto the main map.

I wrote a small replacement function for mark_inset that additionally needs the Basemap objects m & m2 of both maps (main and inset):

def mark_inset(ax, ax2, m, m2, loc1=(1, 2), loc2=(3, 4), **kwargs):
    """
    Patched mark_inset to work with Basemap.
    Reason: Basemap converts Geographic (lon/lat) to Map Projection (x/y) coordinates

    Additionally: set connector locations separately for both axes:
        loc1 & loc2: tuple defining start and end-locations of connector 1 & 2
    """

    # Doesn't work for Basemap
#    rect = TransformedBbox(inset_axes.viewLim, parent_axes.transData)

#    axzoom_geoLims = np.array(m2(*ax2.viewLim._points.T, inverse=True))
    axzoom_geoLims = m2(ax2.get_xlim(), ax2.get_ylim(), inverse=True)
    rect = TransformedBbox(Bbox(np.array(m(*axzoom_geoLims)).T), ax.transData)

    pp = BboxPatch(rect, fill=False, **kwargs)
    ax.add_patch(pp)

    p1 = BboxConnector(ax2.bbox, rect, loc1=loc1[0], loc2=loc1[1], **kwargs)
    ax2.add_patch(p1)
    p1.set_clip_on(False)
    p2 = BboxConnector(ax2.bbox, rect, loc1=loc2[0], loc2=loc2[1], **kwargs)
    ax2.add_patch(p2)
    p2.set_clip_on(False)

    return pp, p1, p2

Note: the function is based on another answer that allows to set different start and end locations for the connectors. Hence, you can do that with this function as well.

Note2: the ax2 object doesn't need to be necessarily an inset object - it can be any axis object (eg, also a subplot). Therefore, a better name for the function would be mark_geo_zoom . ;-)