1

我正在使用 Metpy 制作一张漂亮的降水图。我认为这更像是一个 matplotlib 问题，但我对如何在这个图中添加两个水平颜色条感到困惑。总体思路是在这张地图上显示降雪率和降雨率，这是我通过在一个图形上使用两个等高线填充来完成的。我想显示两个颜色条，因为我有两个单独的降雪率和降雨率。理想情况下，我希望两个颜色条水平并排。我得到的最接近的方法是将它们堆叠在一起，并在它们之间留出很大的空间。这是我草率的代码。

import metpy
from siphon.catalog import TDSCatalog
from datetime import datetime, timedelta
%matplotlib inline
import cartopy.crs as ccrs
import cartopy.feature as cfeature
from metpy.units import units
import metpy.calc as mpcalc
import matplotlib.pyplot as plt
import matplotlib.colors as cls
from xarray.backends import NetCDF4DataStore
import xarray as xr
from scipy.ndimage import gaussian_filter
import numpy as np

# Set up access via NCSS
gfs_catalog = ('http://thredds.ucar.edu/thredds/catalog/grib/NCEP/GFS/'
              'Global_0p5deg/catalog.xml?dataset=grib/NCEP/GFS/Global_0p5deg/Best')
cat = TDSCatalog(gfs_catalog)
ncss = cat.datasets[0].subset()
query3 = ncss.query()
query3.accept('netcdf')
query3.variables('Pressure_reduced_to_MSL_msl', 'Precipitation_rate_surface', 'Snow_depth_surface', 'Categorical_Snow_surface')
now = datetime.utcnow()
query3.time_range(now, now + timedelta(days=4))
query3.lonlat_box(west=-140, east=-60, north=60, south=20)
data3 = ncss.get_data(query3)
ds3 = xr.open_dataset(NetCDF4DataStore(data3))

#parsing data
isSnow_var = ds3.metpy.parse_cf('Categorical_Snow_surface')
precip_var = ds3.metpy.parse_cf('Precipitation_rate_surface')
longitude = precip_var.metpy.x
latitude = precip_var.metpy.y
time_index = 11

#All the Precip Stuff
precip_inch_hour = precip_var[time_index].squeeze() *  141.73228346457 
precip2 = mpcalc.smooth_n_point(precip_inch_hour, 5, 1)

#Converting to 10:1 snow ratio
snow_precip = (precip_inch_hour * isSnow_var[time_index].squeeze()) * 10

# Plot using CartoPy and Matplotlib
mapproj = ccrs.LambertConformal(central_latitude=45., central_longitude=-100.)

# Set projection of data
data_projection = ccrs.PlateCarree()

# Grab data for plotting state boundaries
states_provinces = cfeature.NaturalEarthFeature(
       category='cultural',
       name='admin_1_states_provinces_lakes',
       scale='50m',
       facecolor='none')

# Set extent and plot map lines
fig = plt.figure(1, figsize=(25.,25.))
ax = plt.subplot(111, projection=mapproj)
ax.set_extent([-125., -70, 25., 50.], ccrs.PlateCarree())
ax.coastlines('50m', edgecolor='black', linewidth=0.75)
ax.add_feature(states_provinces, edgecolor='black', linewidth=0.5)

#colormap data
precip_colors = [
   "#bde9bf",  # 0.01 - 0.02 inches 1
   "#adddb0",  # 0.02 - 0.03 inches 2
   "#9ed0a0",  # 0.03 - 0.04 inches 3
   "#8ec491",  # 0.04 - 0.05 inches 4
   "#7fb882",  # 0.05 - 0.06 inches 5
   "#70ac74",  # 0.06 - 0.07 inches 6
   "#60a065",  # 0.07 - 0.08 inches 7
   "#519457",  # 0.08 - 0.09 inches 8
   "#418849",  # 0.09 - 0.10 inches 9
   "#307c3c",  # 0.10 - 0.12 inches 10
   "#1c712e",  # 0.12 - 0.14 inches 11
   "#f7f370",  # 0.14 - 0.16 inches 12
   "#fbdf65",  # 0.16 - 0.18 inches 13
   "#fecb5a",  # 0.18 - 0.2 inches 14
   "#ffb650",  # 0.2 - 0.3 inches 15
   "#ffa146",  # 0.3 - 0.4 inches 16
   "#ff8b3c",   # 0.4 - 0.5 inches 17
   "#ff8b3c"   # 0.5 - 0.6 inches 18
]

precip_colormap = cls.ListedColormap(precip_colors)

#Precip Rate
clev_precip =  np.concatenate((np.arange(0.01, 0.1, .01), np.arange(.1, .2, .02), np.arange(.2, .61, .1)))
norm = cls.BoundaryNorm(clev_precip, 18)
cf = ax.contourf(longitude, latitude, precip2, clev_precip, cmap=precip_colormap, norm=norm, transform=ccrs.PlateCarree())
cb = plt.colorbar(cf, ticks=clev_precip, aspect=65, orientation = 'horizontal', shrink=0.6, pad=0.01)

snow_colors = [
   "#63c9d5",  # 0.1 - 0.2 inches 1
   "#5fb4ca",  # 0.2 - 0.3 inches 2
   "#5a9fc0",  # 0.3 - 0.4 inches 3
   "#558ab5",  # 0.4 - 0.5 inches 4
   "#4e76aa",  # 0.5 - 0.6 inches 5
   "#4763a0",  # 0.6 - 0.7 inches 6
   "#3e4f95",  # 0.7 - 0.8 inches 7
   "#353c8b",  # 0.8 - 0.9 inches 8
   "#292980",  # 0.9 - 1.0 inches 9
   "#493387",  # 1.0 - 1.2 inches 10
   "#643e8e",  # 1.2 - 1.4 inches 11
   "#7c4995",  # 1.4 - 1.6 inches 12
   "#94559c",  # 1.6 - 0.18 inches 13
   "#ab61a3",  # 1.8 - 2 inches 14
   "#c36eaa",  # 2 - 3 inches 15
   "#da7bb0",  # 3 - 4 inches 16
   "#f288b7"   # 4 - 5 inches 17
]

snow_colormap = cls.ListedColormap(snow_colors)

#Snow Rate
clev_snow =  np.concatenate((np.arange(.1, 1, .1), np.arange(1, 2, .2), np.arange(2, 6, 1)))
norm2 = cls.BoundaryNorm(clev_snow, 17)
cf2 = ax.contourf(longitude, latitude, snow_precip, clev_snow, cmap=snow_colormap, norm=norm2, transform=ccrs.PlateCarree())
cb2 = plt.colorbar(cf2, ticks=clev_snow, orientation = 'horizontal', pad=0.01, shrink=0.6, aspect=65)

#Valid Time
vtime = isSnow_var.metpy.time[time_index].values

#Title Info
plt.title('MSLP (hPa) with Highs and Lows, 1000-500 hPa Thickness (m), Rain (in/hr), Snow 10:1 (in/hr)', loc='left')
plt.title(f'VALID: {vtime}', loc='right')

绘图示例

我已经尝试过 subplots 并且我已经尝试过 gridSpec 但我一直遇到不同的障碍，当我尝试这些不同的方法时会遇到不同的错误。主要是我认为这与我对 Metpy、Cartopy 和 Matplotlib 的入门级知识有关。欣赏此问题的任何见解和可能的解决方案。谢谢！

4

1 回答 1

0

首先，感谢您以可重现的方式发布所有代码！

这看起来像是subaxes的解决方案。使用fig.add_axes您可以在您想要的位置精确地创建子轴。

代替您的cb陈述，我使用了以下两个块：

cax1 = fig.add_axes([0.15, 0.25, 0.3, 0.02])
cb = fig.colorbar(cf, cax=cax1, orientation = 'horizontal')

cax2 = fig.add_axes([0.55, 0.25, 0.3, 0.02])
cb2 = fig.colorbar(cf2, cax=cax2, orientation = 'horizontal')

这将创建两个并排的水平颜色条。这四个数字分别是左下角的位置x然后y和x然后y的大小，都是图形大小的比例。我没有在这台机器上使用metpy，所以不得不使用随机数据，但这不应该影响你的结果：

并排颜色条1

注意，当您创建新轴时，任何调用都plt.<function>将默认使用最后创建的轴，在本例中为 cb2。出于这个原因，你应该使用 mantplotlib 首选的面向对象的编程风格，所以创建你的标题ax.set_title而不是plt.title

于 2020-12-29T10:38:30.697 回答