【Python】用Py-ART实现速度退模糊

原创已于 2023-03-02 14:34:47 修改 · 1.9k 阅读

12 ·

CC 4.0 BY-SA版权

本文为博主原创文章，未经博主允许不得转载。

文章标签：

#python

于 2023-02-28 21:30:00 首次发布

Python 专栏收录该内容

5 篇文章

订阅专栏

本文介绍了Py-ART库在处理雷达数据时的两种速度退模糊技术——dealias_region_based和dealias_fourdd。通过示例代码展示了如何应用这些方法来提高雷达数据的准确性，其中dealias_region_based基于区域进行退模糊，而dealias_fourdd利用四维连续性进行速度校正。

0 介绍

Py-ART 被大气辐射测量 (ARM) 气候研究机构用于处理来自许多降水和云雷达的数据，但其设计目的是让雷达和大气界的其他人可以使用它来检查、处理和分析来自多种天气雷达的数据。

下面为使用Py-ART提供的两种速度退模糊的方法（dealias_region_based和dealias_fourdd）的实施个例。

1 实现

# First import needed modules.
import cartopy.crs as ccrs
import matplotlib.pyplot as plt
import numpy as np
import pyart

# Read the radar data.
radar = pyart.io.read('xxx')


fig = plt.figure(figsize=[8,16])
ax = plt.subplot(2,1,1,projection=ccrs.PlateCarree())
display = pyart.graph.RadarMapDisplay(radar)
display.plot_ppi_map('reflectivity', sweep=2, resolution='50m',
                     vmin=0, vmax=60,cmap='gist_ncar',
                     projection=ccrs.PlateCarree())

ax2 = plt.subplot(2,1,2,projection=ccrs.PlateCarree())
display = pyart.graph.RadarMapDisplay(radar)
display.plot_ppi_map('velocity', sweep=2, resolution='50m',
                     vmin=-15, vmax=15,
                     projection=ccrs.PlateCarree(), cmap='bwr')
# plt.show()
plt.savefig('region_plot_ppi_map.jpg')

1.1 dealias_region_based

radar.instrument_parameters['nyquist_velocity']['data']
vel_texture = pyart.retrieve.calculate_velocity_texture(radar, vel_field='velocity', wind_size=3, nyq=16.524973)
radar.add_field('velocity_texture', vel_texture, replace_existing=True)

# Set up the gatefilter to be based on the velocity texture.
gatefilter = pyart.filters.GateFilter(radar)
gatefilter.exclude_above('velocity_texture', 3)


nyq = radar.instrument_parameters['nyquist_velocity']['data'][0]
velocity_dealiased = pyart.correct.dealias_region_based(radar, vel_field='velocity', nyquist_vel=nyq,
                                                        centered=True, gatefilter=gatefilter)
radar.add_field('corrected_velocity', velocity_dealiased, replace_existing=True)


# Plot the new velocities, which now look much more realistic.
fig = plt.figure(figsize=[8, 8])
display = pyart.graph.RadarMapDisplay(radar)
display.plot_ppi_map('corrected_velocity', sweep=2, resolution='50m',
                     vmin=-30, vmax=30,
                     projection=ccrs.PlateCarree(), cmap='bwr',
                     gatefilter=gatefilter)
# plt.show()
plt.savefig('region_plot_ppi_map_corrected_velocity.jpg')

1.2 dealias_fourdd

原理：利用径向速度在空间和时间四维的连续性来实现速度退模糊。详情见参考4。

radar.instrument_parameters['nyquist_velocity']['data']

vel_texture = pyart.retrieve.calculate_velocity_texture(radar, vel_field='velocity', wind_size=3, nyq=16.524973)
radar.add_field('velocity_texture', vel_texture, replace_existing=True)

sond_name = "xxx"
dt,profile = pyart.io.read_arm_sonde_vap(sond_name,radar=radar)

gatefilter = pyart.filters.GateFilter(radar)
gatefilter.exclude_transition()
gatefilter.exclude_invalid('velocity')
gatefilter.exclude_invalid('reflectivity')
gatefilter.exclude_outside('reflectivity',0,80)
gatefilter.exclude_above('velocity_texture', 3)

velocity_dealiased = pyart.correct.dealias_fourdd(radar, vel_field='velocity',
                            sonde_profile=profile,gatefilter=gatefilter)
radar.add_field('corrected_velocity', velocity_dealiased, replace_existing=True)


# Plot the new velocities, which now look much more realistic.
fig = plt.figure(figsize=[8, 8])
display = pyart.graph.RadarMapDisplay(radar)
display.plot_ppi_map('corrected_velocity', sweep=2, resolution='50m',
                     vmin=-41, vmax=41,cmap='bwr',
                     projection=ccrs.PlateCarree(),
                     gatefilter=gatefilter)
# plt.show()
plt.savefig('fourdd_plot_ppi_map_corrected_velocity.jpg')