如何将Matplotlib Basemap的maskoceans（）应用于多边形补丁

Question

我想制作新西兰的情节，每个地区根据一些数据进行颜色编码。然而，我用来制作图像的形状文件延伸到土地边缘之外并进入海洋。这意味着当我为多边形着色时，它们最终也会对海洋的部分进行着色。不是想要的回应！

我使用的代码是：

from mpl_toolkits.basemap import Basemap
from matplotlib.patches import Polygon

plt.figure(figsize=(15,15))
lonmax = 180
lonmin = 165
latmax = -33
latmin = -48
map = Basemap(llcrnrlon=lonmin,llcrnrlat=latmin,urcrnrlon=lonmax,urcrnrlat=latmax, resolution = 'i')

map.drawmapboundary(fill_color='white')
map.fillcontinents(color='white',lake_color='white')
map.drawcoastlines()

map.readshapefile('../data/raw/statsnzregional-council-2016-generalised-version-SHP/regional-council-2016-generalised-version', 'regional_council')

ax = plt.gca() # get current axes instance
cm = matplotlib.cm.get_cmap('viridis')
norm = matplotlib.colors.Normalize(vmin=0.05, vmax=0.35)

for info, shape in zip(map.regional_council_info, map.regional_council):
    poly = Polygon(shape, facecolor=cm(norm(region_percent[info['REGC2016_N']])),edgecolor='k')
    ax.add_patch(poly)

plt.show()

产生以下图像。这个图像非常接近我想要的但是我希望着色停在陆地边界而不是着色海洋。

我已经研究过Basemap的maskoceans（）并且相信这可能是解决这个问题的最好方法，但是我不明白如何将它应用到我的特定情况（例如，如何访问lat，lons？什么是在这种情况下数据数组？）

或者有没有办法让新西兰的地图边界成为一个硬边界，所以只打印内部与多边形补丁的重叠？

Answer 1

你需要一些多边形来掩盖多余的区域。 在这里获取掩码文件（nz_mask_w.shp，nz_mask_e.shp）： https://github.com/swatchai/cartopy_asean_proj/tree/master/shape_files 这就是代码：

import matplotlib
import matplotlib.pyplot as plt
from mpl_toolkits.basemap import Basemap
from matplotlib.patches import Polygon
from matplotlib.collections import PatchCollection
import shapefile  # used to read my shapefiles

fig = plt.figure(figsize=(15,15))
lonmax = 179.95
lonmin = 165
latmax = -33
latmin = -48
map = Basemap(llcrnrlon=lonmin, \
              llcrnrlat=latmin, \
              urcrnrlon=lonmax, \
              urcrnrlat=latmax, \
              resolution = 'i')

# this is map theme (change to your need)
map.readshapefile('data/statsnzregional/regional-council-2016-generalised-version', \
                  name='regional_council')

ax = plt.gca()  # get current axes instance

#cm = matplotlib.cm.get_cmap('viridis')
#norm = matplotlib.colors.Normalize(vmin=0.05, vmax=0.35)

for info, shape in zip(map.regional_council_info, map.regional_council):
    poly = Polygon(shape, \
                facecolor=cm(norm(int(info['REGC2016']))/100.), \
                edgecolor='k', \
                zorder=1)  
    # original:- facecolor=cm(norm(info['REGC2016']))
    ax.add_patch(poly)


# mask out the excess areas (use files in data sub folder)
sf = shapefile.Reader("data/nz_mask_w")
ss = sf.shapes()
poly1 = Polygon(ss[0].points)

sf = shapefile.Reader("data/nz_mask_e")
ss = sf.shapes()
poly2 = Polygon(ss[0].points)

ax.add_collection( PatchCollection([poly1,poly2], \
                                   zorder=12, \
                                   facecolor='lightblue', \
                                   edgecolor='lightblue' ) )

map.drawcoastlines(color='blue', linewidth=0.3)
plt.show()