geodedetic_image_overlay.md

Geodedetic image overlay

import numpy as np


def sample_data(shape=(73, 145)):
    nlats, nlons = shape
    lats = np.linspace(-np.pi / 2, np.pi / 2, nlats)
    lons = np.linspace(0, 2 * np.pi, nlons)
    lons, lats = np.meshgrid(lons, lats)
    wave = 0.75 * (np.sin(2 * lats) ** 8) * np.cos(4 * lons)
    mean = 0.5 * np.cos(2 * lats) * ((np.sin(2 * lats)) ** 2 + 2)

    lats = np.rad2deg(lats)
    lons = np.rad2deg(lons)
    data = wave + mean

    return lons, lats, data


lon, lat, data = sample_data(shape=(73, 145))
lon -= 180

%matplotlib inline

import matplotlib

cm = matplotlib.colormaps["cubehelix"]

normed_data = (data - data.min()) / (data.max() - data.min())
colored_data = cm(normed_data)

Bad

import folium


m = folium.Map(location=[lat.mean(), lon.mean()], zoom_start=1)

folium.raster_layers.ImageOverlay(
    image=colored_data,
    bounds=[[lat.min(), lon.min()], [lat.max(), lon.max()]],
    opacity=0.25,
).add_to(m)

m

Good

m = folium.Map(location=[lat.mean(), lon.mean()], zoom_start=1)

folium.raster_layers.ImageOverlay(
    image=colored_data,
    bounds=[[lat.min(), lon.min()], [lat.max(), lon.max()]],
    mercator_project=True,
    opacity=0.25,
).add_to(m)


m

Same as above but with cartopy

import cartopy.crs as ccrs
from cartopy.img_transform import warp_array

source_extent = [lon.min(), lon.max(), lat.min(), lat.max()]

new_data = warp_array(
    colored_data,
    target_proj=ccrs.GOOGLE_MERCATOR,
    source_proj=ccrs.PlateCarree(),
    target_res=data.shape,
    source_extent=source_extent,
    target_extent=None,
    mask_extrapolated=False,
)


m = folium.Map(location=[lat.mean(), lon.mean()], zoom_start=1)

folium.raster_layers.ImageOverlay(
    image=new_data[0],
    bounds=[[lat.min(), lon.min()], [lat.max(), lon.max()]],
    opacity=0.25,
).add_to(m)

m

TODO: Try rasterio. Rasterio can warp images and arrays.

Compare to original

From https://scitools.org.uk/cartopy/docs/v0.16/gallery/waves.html