【python由站点数据插值到网格数据方法对比】

1、前言

• 气象海洋中空间数据类型有站点数据、格点数据。
• 站点数据空间分布不连续，不利于进行时空分析；有时需要将站点数据插值到网格中。
• 本文基于一个散点数据对比了griddata、IDW、krige、rbf的插值结果。
• interp2d插值，地理数据中的分辨率转换

2、结果对比

2.1 原始散点站位图

• 原始散点站位如下：
  

2.2 griddata插值

from scipy.interpolate import griddata
lat = df_filter['Lat']
lon = df_filter['Lon']
VIS_Min = df_filter['VIS_Min']
olon = np.linspace(lon.min(), lon.max(), 40)
olat = np.linspace(lat.min(), lat.max(), 40)
olon, olat = np.meshgrid(olon, olat)
VIS_Min_grd=griddata((lon,lat),VIS_Min,(olon, olat),method='linear')
plt.figure(figsize=(8,6))
plt.pcolormesh(olon, olat, VIS_Min_grd)
plt.scatter(lon, lat, s=50, c=VIS_Min, cmap='viridis',edgecolors='k')

• 可视化代码

import os
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.colors as mcolors
import matplotlib as mpl

import cartopy.crs as ccrs
import cartopy.io.shapereader as shpreader
from mpl_toolkits.axes_grid1.inset_locator import inset_axes
from matplotlib.colors import LinearSegmentedColormap

import regionmask
import geopandas as gpd
import warnings

warnings.filterwarnings("ignore")
import matplotlib.ticker as ticker
extents = [115.6, 117.2, 39.7, 40.8]
crs = ccrs.PlateCarree()
fig = plt.figure(figsize=(12, 6))
ax = fig.add_subplot(111, projection=crs)
ax.set_extent(extents, crs)

prov_reader = shpreader.Reader(r'.\bou2_4p.shp')
nineline_reader = shpreader.Reader(r'.\九段线.shp')

ax.add_geometries(prov_reader.geometries(), crs, edgecolor='gray', facecolor='none', lw=1)
ax.add_geometries(nineline_reader.geometries(), crs, edgecolor='gray', facecolor='none', lw=1)

ax.set_xticks(np.arange(extents[0], extents[1] + 0.3, 0.3))
ax.set_yticks(np.arange(extents[2], extents[3] + 0.2, 0.2))
plt.tick_params(axis='bot