Santa Cruz Mountains map (also, annotating is slightly hard)

[kratsg]

Here's the code that generates a nice map of the Santa Cruz Mountains (and adds a point identifying the Santa Cruz Institute for Particle Physics)

• a workaround to color the label a different color
• slight issues with figuring out how to annotate things correctly using actual lat/long coordinates

from ridge_map import RidgeMap
import matplotlib
import matplotlib.pyplot as plt
import numpy as np

bgcolor = np.array([65,74,76])/255.

scipp = (-122.060510, 36.998776)
rm = RidgeMap((-122.087116,36.945365,-121.999226,37.023250))
scipp_coords = ((scipp[0] - rm.longs[0])/(rm.longs[1] - rm.longs[0]),(scipp[1] - rm.lats[0])/(rm.lats[1] - rm.lats[0]))

values = rm.get_elevation_data(num_lines=150)
ridges = rm.plot_map(values=rm.preprocess(values=values, lake_flatness=1, water_ntile=0, vertical_ratio=240),
            label='Santa Cruz\nMountains',
            label_x=0.75,
            label_y=0.025,
            label_size=90,
            kind='elevation',
            background_color=bgcolor,
            line_color = plt.get_cmap('cool'),
            size_scale=50)

[c for c in ridges.get_children() if isinstance(c, matplotlib.text.Text) and 'Santa Cruz' in c._text][0].set_color('white')
ridges.text(scipp_coords[0]+0.005, scipp_coords[1]+0.005, 'SCIPP', size=40, color="white", transform=ridges.transAxes, verticalalignment="bottom", zorder=len(values)+10)
ridges.add_patch(plt.Circle(scipp_coords, 0.0025, color="white", fill=True, transform=ridges.transAxes, zorder=len(values)+10))
ridges.get_figure().savefig('santa_cruz.pdf', bbox_inches='tight')

[ColCarroll]

Beautiful! I'll add this to the examples section if that's ok!

I'll look at what I can do for those difficulties. At the very least, adding a RidgeMap.lat_long_to_data method seems useful.

[kratsg]

Beautiful! I'll add this to the examples section if that's ok!

Fine with me!

I'll look at what I can do for those difficulties. At the very least, adding a RidgeMap.lat_long_to_data method seems useful.

Yeah. I mean, there's already a ridges.annotate method, however I couldn't figure out the right combination of coordinates to make things work and annotate as expected, e.g. the following did not work for me

ridges.annotate('SCIPP', scipp)
ridges.annotate('SCIPP', scipp_coords)
ridges.annotate('SCIPP', scipp, transform=ridges.transAxes)
ridges.annotate('SCIPP', scipp_coords, transform=ridges.transAxes)
ridges.annotate('SCIPP', scipp, transform=ridges.transData)
ridges.annotate('SCIPP', scipp_coords, transform=ridges.transData)
ridges.annotate('SCIPP', ridges.transAxes.transform(scipp))
ridges.annotate('SCIPP', ridges.transAxes.transform(scipp_coords))
ridges.annotate('SCIPP', ridges.transAxes.inverted().transform(scipp))
ridges.annotate('SCIPP', ridges.transAxes.inverted().transform(scipp_coords))
ridges.annotate('SCIPP', ridges.transData.transform(scipp))
ridges.annotate('SCIPP', ridges.transData.transform(scipp_coords))
ridges.annotate('SCIPP', ridges.transData.inverted().transform(scipp))
ridges.annotate('SCIPP', ridges.transData.inverted().transform(scipp_coords))

although I admit I'm not very well-versed in MPL transforms, but it seems like the axes used in this package are somewhat not trivial/straightforward to build.

[ColCarroll]

Closing as we added the beautiful example, but would be happy to merge improved annotating.

[cranmer]

I was doing this for a gift. One issue is that it seems the lat/long position is for zero-altitude. So to get the point right one needs the vertical off-set as well and how that's projected into the frame.

[cranmer]

Also, to match the font:

import matplotlib.font_manager as fm
from ridge_map import FontManager

prop = fm.FontProperties(fname= rm.font.get_file())

and then add fontproperties=prop to the ridges.text call

ridges.text(scipp_coords[0]+0.005, scipp_coords[1]+0.005, 'SCIPP', size=40, color="white", transform=ridges.transAxes, verticalalignment="bottom", zorder=len(values)+10,fontproperties=prop)