Python Cartopy地图，国家/地区以外的剪贴区域（多边形）

Question

我用来自NaturalEarth的国家边界创建了Stamen地形图。 现在，我想从国家/地区边界之外删除所有数据（在这种情况下为地形）。 我该怎么办？

我在瑞士内外都可以看到地形的示例：

from cartopy.io import shapereader
import cartopy.io.img_tiles as cimgt
import cartopy.crs as ccrs
import geopandas
import matplotlib.pyplot as plt


resolution = '10m'
category = 'cultural'
name = 'admin_0_countries'

shpfilename = shapereader.natural_earth(resolution, category, name)

df = geopandas.read_file(shpfilename)

poly = [df.loc[df['ADMIN'] == 'Switzerland']['geometry'].values[0]]

stamen_terrain = cimgt.Stamen('terrain-background')

fig = plt.figure(figsize=(8,6))

ax = fig.add_subplot(1, 1, 1, projection=stamen_terrain.crs)
ax.add_geometries(poly, crs=ccrs.PlateCarree(), facecolor='none', edgecolor='r')
exts = [poly[0].bounds[0], poly[0].bounds[2], poly[0].bounds[1], poly[0].bounds[3]]
ax.set_extent(exts, crs=ccrs.Geodetic())

ax.add_image(stamen_terrain, 8)
fig.tight_layout()
plt.show()

如何仅显示边界内的地形，而其余部分则显示为白色/透明？

试图朝这个方向努力，但到目前为止没有成功： https://scitools.org.uk/cartopy/docs/latest/gallery/logo.html

Answer 1

您需要一个遮罩以隐藏图像中不需要的部分。这是一个可运行的代码，演示了获取所需图形的所有步骤。

from shapely.geometry import Polygon
from cartopy.io import shapereader
import cartopy.io.img_tiles as cimgt
import cartopy.crs as ccrs
import geopandas
import matplotlib.pyplot as plt

def rect_from_bound(xmin, xmax, ymin, ymax):
    """Returns list of (x,y)'s for a rectangle"""
    xs = [xmax, xmin, xmin, xmax, xmax]
    ys = [ymax, ymax, ymin, ymin, ymax]
    return [(x, y) for x, y in zip(xs, ys)]

# request data for use by geopandas
resolution = '10m'
category = 'cultural'
name = 'admin_0_countries'

shpfilename = shapereader.natural_earth(resolution, category, name)
df = geopandas.read_file(shpfilename)

# get geometry of a country
poly = [df.loc[df['ADMIN'] == 'Switzerland']['geometry'].values[0]]

stamen_terrain = cimgt.Stamen('terrain-background')

# projections that involved
st_proj = stamen_terrain.crs  #projection used by Stamen images
ll_proj = ccrs.PlateCarree()  #CRS for raw long/lat

# create fig and axes using intended projection
fig = plt.figure(figsize=(8,6))
ax = fig.add_subplot(1, 1, 1, projection=st_proj)
ax.add_geometries(poly, crs=ll_proj, facecolor='none', edgecolor='black')

pad1 = .1  #padding, degrees unit
exts = [poly[0].bounds[0] - pad1, poly[0].bounds[2] + pad1, poly[0].bounds[1] - pad1, poly[0].bounds[3] + pad1];
ax.set_extent(exts, crs=ll_proj)

# make a mask polygon by polygon's difference operation
# base polygon is a rectangle, another polygon is simplified switzerland
msk = Polygon(rect_from_bound(*exts)).difference( poly[0].simplify(0.01) )
msk_stm  = st_proj.project_geometry (msk, ll_proj)  # project geometry to the projection used by stamen

# get and plot Stamen images
ax.add_image(stamen_terrain, 8) # this requests image, and plot

# plot the mask using semi-transparency (alpha=0.65) on the masked-out portion
ax.add_geometries( msk_stm, st_proj, zorder=12, facecolor='white', edgecolor='none', alpha=0.65)

ax.gridlines(draw_labels=True)

plt.show()

结果图：

和面具：

编辑1

上面的代码适用于单个国家/地区。如果多个连续的国家/地区是我们的新目标，则需要将所有国家/地区和dissolve选择到一个单一的几何中。只需修改几行代码。

示例：新的目标国家/地区：['挪威'，'瑞典'，'芬兰']

要替换的代码行：

poly = [df.loc[df['ADMIN'] == 'Switzerland']['geometry'].values[0]]

要替换的新代码行：

scan3 = df[ df['ADMIN'].isin(['Norway','Sweden', 'Finland']) ]
scan3_dissolved = scan3.dissolve(by='LEVEL')
poly = [scan3_dissolved['geometry'].values[0]]

示例输出图：