如何使用xarray，numpy和Basemap从我的数据集中掩盖海洋区域？

Question

我有一个netCDF文件“ VWC_95percentile_by_seasons.nc”。它包含三个维度seasons（DJF，JJA，MAM，SON），latitude和longitude，变量名称为VWC 95TH PERCENTILE。

变量打印如下：

<xarray.DataArray 'VWC 95TH PERCENTILE' (season: 4, latitude: 106, longitude: 193)>
dask.array<shape=(4, 106, 193), dtype=float64, chunksize=(4, 106, 193)>
Coordinates:
  * latitude   (latitude) float32 -39.2 -39.149525 ... -33.950478 -33.9
  * longitude  (longitude) float32 140.8 140.84792 140.89584 ... 149.95209 150.0
  * season     (season) object 'DJF' 'JJA' 'MAM' 'SON'

我将需要每个季节在一个图上绘制一个图。每个地块都需要从海洋区域屏蔽掉。

我的代码是：

import os
import time, datetime
import xarray as xr
import numpy as np
import pandas as pd
from glob import glob
import matplotlib.pyplot as plt
from mpl_toolkits.basemap import Basemap, maskoceans

shapefile1=r'{PATH}\shp\regions'

ds_path = r'{PATH}\output\VWC_95percentile_by_seasons.nc'
variable_name = 'VWC 95TH PERCENTILE'
lable = 'VWC'
title = 'VWC 95TH PERCENTILE'


print("### LOAD DATA")
ds = xr.open_mfdataset(ds_path)
da = ds[variable_name]
print(da)

print("### PLOTTING")

# define the colormap
cmap = plt.cm.jet
# extract all colors from the .jet map
cmaplist = [cmap(i) for i in range(cmap.N)]
# create the new map
cmap = cmap.from_list('Custom cmap', cmaplist, cmap.N)

# colour levels
levels=[0, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50]

# Plot
p = da.plot(levels=10, cmap=cmap, x='longitude', y='latitude', col='season', cbar_kwargs={'label': lable})        # p is an instance of <class 'xarray.plot.facetgrid.FacetGrid'>
plt.suptitle(title)

for i, ax in enumerate(p.axes.flatten()):

    ax.set_xlabel('')
    ax.set_ylabel('')
    map = Basemap(llcrnrlat=-39.2,urcrnrlat=-33.9,llcrnrlon=140.8,urcrnrlon=150.0,resolution='i',ax=ax)
    map.readshapefile(shapefile1, 'regions', linewidth=0.5)    # draw disctict boundaries
plt.show()

当前绘制：

如何使用xarray，numpy和Basemap遮盖所有海洋区域？