风场图

本节介绍如何使用 cedarkit-maps 绘制 850 hPa 风场图。

首先导入需要使用的包。包括：

• 数据结构：numpy, pandas

• 可视化：matplotlib

• cedarkit 工具套件

  • 数据准备：reki

    • 数据查找函数： find_local_file

    • GRIB2文件要素加载函数：load_field_from_file

  • 气象可视化：cedarkit.maps

    • 绘图板：Panel

    • 底图布局：EastAsiaMapDomain

    • 绘图样式：BarbStyle

import pandas as pd

from reki.data_finder import find_local_file
from reki.format.grib.eccodes import load_field_from_file

from cedarkit.maps.style import BarbStyle
from cedarkit.maps.chart import Panel
from cedarkit.maps.domains import EastAsiaMapDomain

设置绘图的数据参数，使用 CMA-GFS 2024 年 4 月 1 日 00 时次 024 时效数据。

system_name = "CMA-GFS"
data_type = "cma_gfs_gmf/grib2/orig"
start_time = pd.to_datetime("2024-04-01 00:00:00")
forecast_time = pd.to_timedelta("24h")

加载数据

获取 GRIB2 数据路径

file_path = find_local_file(
    data_type,
    start_time=start_time,
    forecast_time=forecast_time,
)
file_path

PosixPath('/g3/COMMONDATA/OPER/CEMC/GFS_GMF/Prod-grib/2024040100/ORIG/gmf.gra.2024040100024.grb2')

加载 850 hPa 风场。风场在 ecCodes 的内置要素名为 u 和 v

u_850_field = load_field_from_file(
    file_path,
    parameter="u",
    level_type="pl",
    level=850
)
v_850_field = load_field_from_file(
    file_path,
    parameter="v",
    level_type="pl",
    level=850
)
v_850_field

<xarray.DataArray 'v' (latitude: 1440, longitude: 2880)> Size: 33MB
array([[ 8.80414551,  8.79414551,  8.77414551, ...,  8.83414551,
         8.82414551,  8.81414551],
       [ 7.98414551,  7.97414551,  7.95414551, ...,  8.02414551,
         8.00414551,  7.99414551],
       [ 7.62414551,  7.61414551,  7.59414551, ...,  7.66414551,
         7.65414551,  7.63414551],
       ...,
       [-5.68585449, -5.69585449, -5.70585449, ..., -5.69585449,
        -5.68585449, -5.68585449],
       [-5.25585449, -5.26585449, -5.26585449, ..., -5.27585449,
        -5.25585449, -5.24585449],
       [-4.25585449, -4.25585449, -4.25585449, ..., -4.25585449,
        -4.24585449, -4.24585449]])
Coordinates:
    time        datetime64[ns] 8B 2024-04-01
    step        timedelta64[ns] 8B 1 days
    valid_time  datetime64[ns] 8B 2024-04-02
    pl          float64 8B 850.0
  * latitude    (latitude) float64 12kB 89.94 89.81 89.69 ... -89.81 -89.94
  * longitude   (longitude) float64 23kB 0.0 0.125 0.25 ... 359.6 359.8 359.9
Attributes: (12/17)
    GRIB_edition:             2
    GRIB_centre:              babj
    GRIB_subCentre:           0
    GRIB_tablesVersion:       4
    GRIB_localTablesVersion:  0
    GRIB_dataType:            fc
    ...                       ...
    GRIB_stepType:            instant
    GRIB_stepUnits:           1
    GRIB_stepRange:           24
    GRIB_endStep:             24
    GRIB_count:               189
    long_name:                discipline=0 parmcat=2 parm=3

xarray.DataArray

'v'

• latitude: 1440
• longitude: 2880

• 8.804 8.794 8.774 8.764 8.754 ... -4.256 -4.256 -4.256 -4.246 -4.246

  array([[ 8.80414551,  8.79414551,  8.77414551, ...,  8.83414551,
           8.82414551,  8.81414551],
         [ 7.98414551,  7.97414551,  7.95414551, ...,  8.02414551,
           8.00414551,  7.99414551],
         [ 7.62414551,  7.61414551,  7.59414551, ...,  7.66414551,
           7.65414551,  7.63414551],
         ...,
         [-5.68585449, -5.69585449, -5.70585449, ..., -5.69585449,
          -5.68585449, -5.68585449],
         [-5.25585449, -5.26585449, -5.26585449, ..., -5.27585449,
          -5.25585449, -5.24585449],
         [-4.25585449, -4.25585449, -4.25585449, ..., -4.25585449,
          -4.24585449, -4.24585449]])

• Coordinates: (6)
  • time
    ()
    datetime64[ns]
    2024-04-01

    array('2024-04-01T00:00:00.000000000', dtype='datetime64[ns]')

  • step
    ()
    timedelta64[ns]
    1 days

    array(86400000000000, dtype='timedelta64[ns]')

  • valid_time
    ()
    datetime64[ns]
    2024-04-02

    array('2024-04-02T00:00:00.000000000', dtype='datetime64[ns]')

  • pl
    ()
    float64
    850.0

    array(850.)

  • latitude
    (latitude)
    float64
    89.94 89.81 89.69 ... -89.81 -89.94

    units :

    degrees_north

    standard_name :

    latitude

    long_name :

    latitude

    array([ 89.9375,  89.8125,  89.6875, ..., -89.6875, -89.8125, -89.9375])

  • longitude
    (longitude)
    float64
    0.0 0.125 0.25 ... 359.8 359.9

    units :

    degrees_east

    standard_name :

    longitude

    long_name :

    longitude

    array([0.00000e+00, 1.25000e-01, 2.50000e-01, ..., 3.59625e+02, 3.59750e+02,
           3.59875e+02])

• Indexes: (2)
  • latitude
    PandasIndex

    PandasIndex(Index([ 89.9375,  89.8125,  89.6875,  89.5625,  89.4375,  89.3125,  89.1875,
            89.0625,  88.9375,  88.8125,
           ...
           -88.8125, -88.9375, -89.0625, -89.1875, -89.3125, -89.4375, -89.5625,
           -89.6875, -89.8125, -89.9375],
          dtype='float64', name='latitude', length=1440))

  • longitude
    PandasIndex

    PandasIndex(Index([    0.0,   0.125,    0.25,   0.375,     0.5,   0.625,    0.75,   0.875,
               1.0,   1.125,
           ...
            358.75, 358.875,   359.0, 359.125,  359.25, 359.375,   359.5, 359.625,
            359.75, 359.875],
          dtype='float64', name='longitude', length=2880))

• Attributes: (17)

  GRIB_edition :

  2

  GRIB_centre :

  babj

  GRIB_subCentre :

  0

  GRIB_tablesVersion :

  4

  GRIB_localTablesVersion :

  0

  GRIB_dataType :

  fc

  GRIB_dataDate :

  20240401

  GRIB_dataTime :

  0

  GRIB_validityDate :

  20240402

  GRIB_validityTime :

  0

  GRIB_step :

  24

  GRIB_stepType :

  instant

  GRIB_stepUnits :

  1

  GRIB_stepRange :

  24

  GRIB_endStep :

  24

  GRIB_count :

  189

  long_name :

  discipline=0 parmcat=2 parm=3

配置

定义风场图样式 (BarbStyle)

wind_barb_style = BarbStyle(
    barbcolor="red",
    flagcolor="red",
    linewidth=0.3,
)

绘制

创建中国区域底图布局

domain = EastAsiaMapDomain()

绘图风场图，设置标题。

注：这里将风场稀疏化后再绘图，并只在主区域（第 1 层）绘图，即 layer=[0]

panel = Panel(domain=domain)
panel.plot([[u_850_field[::14, ::14], v_850_field[::14, ::14]]], style=wind_barb_style, layer=[0])

domain.set_title(
    panel=panel,
    graph_name="850 hPa WIND (m/s) windbarb",
    system_name=system_name,
    start_time=start_time,
    forecast_time=forecast_time,
)