1

I am trying to display a contourf plot as you can see below in the code snippet on a Folium map.

I can see the filled contour plot just fine. Also, I added a color bar with exact same colors using branca at this line:

bmap     = branca.colormap.LinearColormap(colorl, vmin=levs[0], 
vmax=levs[-1]).to_step(len(levs),index=levs)
geojsonf = geojsoncontour.contourf_to_geojson(
contourf=pcontf,
min_angle_deg=3.0,
ndigits=5,
stroke_width=1,
fill_opacity=0.9)

As you can see in the output image, colors don't match.

enter image description here

I suspect opacity I use for the contour plot might play a role here but changing opacity does not make it better.

I also tried making Circle markers (not shown here) with the same colors but still no luck. I cannot get pyplot colors to match.

Any suggestion is greatly appreciated. Also is there a better way to accomplish the same task? I basically have a 2D NumPy array with values ranging from -50 to 50on a reprojected lat-lon grid. I need to be able to show the shaded contours and associated values in the bar.

fig = plt.figure(figsize=[10, 15], dpi=None)  
ax = fig.subplots()
jet =plt.get_cmap('jet') 
clevs= np.array(levs)
cnorm = plt.Normalize(vmin=levs[0],vmax=levs[-1])
clevels = [levs[0]] + list(0.5*(clevs[1:]+clevs[:-1])) + [levs[-1]]
colors=jet(cnorm(clevels))
colorsm = color.ListedColormap(colors)

pcontf = ax.contourf(lons,lats,data,levels=levs,cmap=colorsm)

mapa = folium.Map([np.mean(lats), np.mean(lons)], zoom_start=10,tiles='Stamen Terrain')

colorl = []
for i,val in enumerate(colors):
    carr= colors[i-1]
    ccol = (carr[1],carr[2],carr[3])
    colorl.insert(i,ccol) 
bmap     = branca.colormap.LinearColormap(colorl, vmin=levs[0], 
vmax=levs[-1]).to_step(len(levs),index=levs)
geojsonf = geojsoncontour.contourf_to_geojson(
contourf=pcontf,
min_angle_deg=3.0,
ndigits=5,
stroke_width=1,
fill_opacity=0.9)

folium.GeoJson(
geojsonf,
style_function=lambda x: {
    'color':     x['properties']['stroke'],
    'weight':    x['properties']['stroke-width'],
    'fillColor': x['properties']['fill'],
    'opacity':   0.9,
}).add_to(mapa)
bmap.add_to(mapa)
Sabito stands with Ukraine
  • 4,271
  • 8
  • 34
  • 56
Nihat Cubukcu
  • 11
  • 1
  • what is "branca"? – Paul H Aug 17 '20 at 15:25
  • Paul H https://pypi.org/project/branca/ is handy for things like creating a colormap on Folium. There could be other ways but I found branca the most straightforward. I could not find any other way to add colorbar on Folium. – Nihat Cubukcu Aug 18 '20 at 02:32

1 Answers1

0

I believe you have to recreate the Matplotlib colormap in Folium first. This is how I did it (in the example my values range from 0 to 310, m is the Folium map):

  1. Creating a colormap with Matplotlib:
    import matplotlib as mpl
    import branca.colormap as cm
    import numpy as np
    cmap = mpl.cm.get_cmap(name='rainbow',lut=310)

  2. Creating a list with appropriate step size:
    index_list = range(0,310,10).tolist()
    cmap_list = cmap(index_list).tolist()

  3. Creating a Folium colormap identical to the Matplotlib colormap:
    cmap_foliump = cm.LinearColormap(
    cmap_list, vmin=0, vmax=310, index=index_list, caption='my colormap')

  4. Adding to Folium map:
    m.add_child(cmap_folium)

Make sure index_list is comprised of true integers. If you create it with numpy.arange() just add .astype(int) prior to tolist()

Reno Filla
  • 3
  • 3