Note

Click here to download the full example code

QPF (Hong Kong)

This example demonstrates how to perform operational deterministic QPF up to three hours from raingauge and radar data, using data from Hong Kong.

Definitions

import os
import numpy as np
import pandas as pd
from pyresample import utils
import xarray
import cartopy.feature as cfeature
import cartopy.crs as ccrs
import matplotlib
import imageio
import matplotlib.pyplot as plt
from matplotlib.colors import BoundaryNorm

from swirlspy.rad.irisref import read_irisref
from swirlspy.qpe.utils import dbz2rr, rr2rf, locate_file, timestamps_ending
from swirlspy.qpe.utils import multiple_acc
from swirlspy.obs.rain import Rain
from swirlspy.qpf.rover import rover
from swirlspy.qpf.sla import sla
from swirlspy.core.resample import grid_resample

plt.switch_backend('agg')

THIS_DIR = os.getcwd()
os.chdir(THIS_DIR)

start_time = pd.Timestamp.now()

Traceback (most recent call last):
  File "/tmp/build/docs/swirlspy/swirlspy/examples/qpf_hk.py", line 30, in <module>
    from swirlspy.qpf.rover import rover
  File "/tmp/build/docs/swirlspy/swirlspy/qpf/rover.py", line 6, in <module>
    from rover.rover import rover as rover_api
ImportError: libopencv_core.so.3.4: cannot open shared object file: No such file or directory

Initialising

This section demonstrates extracting radar reflectivity data.

Step 1: Define a basetime.

# Supply basetime
basetime = pd.Timestamp('201902190800')

Step 2: Using basetime, generate timestamps of desired radar files timestamps_ending() and locate files using locate_file().

# Obtain radar files
dir = THIS_DIR + '/../tests/samples/iris/ppi'
located_files = []
radar_ts = timestamps_ending(
    basetime,
    duration=pd.Timedelta(minutes=60)
)

for timestamp in radar_ts:
    located_files.append(locate_file(dir, timestamp))

Step 3: Read data from radar files into xarray.DataArray using read_irisref().

reflectivity_list = []  # stores reflec from read_iris()
for filename in located_files:
    reflec = read_irisref(
        filename
    )
    reflectivity_list.append(reflec)

Step 4: Define the target grid as a pyresample AreaDefinition.

# Defining target grid
area_id = "hk1980_250km"
description = ("A 250 m resolution rectangular grid "
               "centred at HKO and extending to 250 km "
               "in each direction in HK1980 easting/northing coordinates")
proj_id = 'hk1980'
projection = ('+proj=tmerc +lat_0=22.31213333333334 '
              '+lon_0=114.1785555555556 +k=1 +x_0=836694.05 '
              '+y_0=819069.8 +ellps=intl +towgs84=-162.619,-276.959,'
              '-161.764,0.067753,-2.24365,-1.15883,-1.09425 +units=m '
              '+no_defs')
x_size = 500
y_size = 500
area_extent = (587000, 569000, 1087000, 1069000)
area_def_tgt = utils.get_area_def(
    area_id, description, proj_id, projection, x_size, y_size, area_extent
)

Step 5: Reproject the radar data from read_irisref() from Centered Azimuthal (source) projection to HK 1980 (target) projection.

# Extracting the AreaDefinition of the source projection
area_def_src = reflectivity_list[0].attrs['area_def']

# Reprojecting
reproj_reflectivity_list = []
for reflec in reflectivity_list:
    reproj_reflec = grid_resample(
        reflec, area_def_src, area_def_tgt,
        coord_label=['easting', 'northing']
    )
    reproj_reflectivity_list.append(reproj_reflec)

Step 6: Assigning reflectivity xarrays at the last two timestamps to variables for use during ROVER QPF.

xarray1 = reproj_reflectivity_list[-2]
xarray2 = reproj_reflectivity_list[-1]

initialising_time = pd.Timestamp.now()

Running ROVER and Semi-Lagrangian Advection

  1. Concatenate two reflectivity xarrays along time dimension.

  2. Run ROVER, with the concatenated xarray as the input.

  3. Perform Semi-Lagrangian Advection using the motion fields from rover.

# Combining the two reflectivity DataArrays
# the order of the coordinate keys is now ['y', 'x', 'time']
# as opposed to ['time', 'x', 'y']
reflec_concat = xarray.concat([xarray1, xarray2], dim='time')

# Rover
motion_u, motion_v = rover(
    reflec_concat
)

rover_time = pd.Timestamp.now()

# Semi Lagrangian Advection
reflectivity = sla(
    reflec_concat, motion_u, motion_v, steps=30
)

sla_time = pd.Timestamp.now()

Concatenating observed and forecasted reflectivities

  1. Add forecasted reflectivity to reproj_reflectivity_list.

  2. Concatenate observed and forecasted reflectivity xarray.DataArrays along the time dimension.

reproj_reflectivity_list.append(reflectivity[1:, ...])
reflectivity = xarray.concat(reproj_reflectivity_list, dim='time')

concat_time = pd.Timestamp.now()

Generating radar reflectivity maps

  1. Define the colour scale and the format of the reflectivity plots.

  2. Initialise the figure and the cartopy GeoAxes. Use the to_cartopy_crs() method of the AreaDefinition to obtain the projection system of the axes.

  3. Add GSHHS coastlines using the add_feature() method of cartopy GeoAxes.

  4. Plot data using xarray.plot() and make desired adjustments to labels.

  5. Plot quiver using the quiver() method of cartopy GeoAxes.

In this example, only hourly images will be plotted.

# Defining colour scale and format
levels = [
    -32768,
    10, 15, 20, 24, 28, 32,
    34, 38, 41, 44, 47, 50,
    53, 56, 58, 60, 62
]
cmap = matplotlib.colors.ListedColormap([
    '#FFFFFF', '#08C5F5', '#0091F3', '#3898FF', '#008243', '#00A433',
    '#00D100', '#01F508', '#77FF00', '#E0D100', '#FFDC01', '#EEB200',
    '#F08100', '#F00101', '#E20200', '#B40466', '#ED02F0'
])

norm = BoundaryNorm(levels, ncolors=cmap.N, clip=True)

# generate timesteps for plotting
timesteps = [basetime+pd.Timedelta(minutes=60*i-6) for i in range(4)]

# Defining quiver parameters
qx = motion_u.coords['easting'].values[::5]
qy = motion_u.coords['northing'].values[::5]
qu = motion_u.values[::5, ::5]
qv = motion_v.values[::5, ::5]

# Defining coastlines
hires = cfeature.GSHHSFeature(
            scale='h',
            levels=[1],
            edgecolor='black',
            facecolor='none'
        )

# Plotting
images = []
for time in timesteps:
    # Figure
    plt.figure(figsize=(28, 21))
    plt.axis("equal")
    # Axes
    crs = area_def_tgt.to_cartopy_crs()
    ax = plt.axes(projection=crs)
    # Coastlines
    ax.add_feature(hires)

    # Plot reflectivity data
    quadmesh = reflectivity.sel(time=time).plot(
        cmap=cmap,
        norm=norm,
        extend='neither'
        )

    # Customising labels
    cbar = quadmesh.colorbar
    cbar.ax.set_ylabel(
        reflectivity.attrs['long_name']+'['+reflectivity.attrs['units']+']',
        fontsize=28
    )
    cbar.ax.tick_params(labelsize=24)
    ax.xaxis.set_visible(True)
    ax.yaxis.set_visible(True)
    ax.xaxis.set_tick_params(labelsize=24)
    ax.yaxis.set_tick_params(labelsize=24)
    ax.xaxis.label.set_size(28)
    ax.yaxis.label.set_size(28)

    # Quiver
    ax.quiver(
        qx, qy, qu, qv, pivot='mid', regrid_shape=20
    )

    # Title
    plt.title(
        "Quantitative Precipitation Forecast with Reflectivity Fields\n"
        f"Base time: {basetime}H\n"
        f"Valid time {pd.Timestamp(time)}H",
        fontsize=32
    )
    savepath = os.path.join(
        THIS_DIR,
        ("../tests/outputs/rover-output-map-"
         f"{time.strftime('%Y%m%d%H%M')}.png")
    )
    plt.savefig(savepath)
    images.append(imageio.imread(savepath))

imageio.mimsave(
    THIS_DIR +
    "/../tests/outputs/rover-output-map.gif",
    images, duration=1
)

radar_image_time = pd.Timestamp.now()

Accumulating hourly rainfall for 3-hour forecast

Hourly accumulated rainfall is calculated every 30 minutes, the first endtime is the basetime i.e. T+0min.

  1. Convert reflectivity in dBZ to rainrates in mm/h with dbz2rr().

  2. Changing time coordinates of xarray from start time to endtime.

  3. Convert rainrates to rainfalls in 6 mins with rr2rf().

  4. Accumulate hourly rainfall every 30 minutes using multiple_acc().

# Convert reflectivity to rainrates
rainrates = dbz2rr(reflectivity, a=58.53, b=1.56)

# Converting the coordinates of xarray from start to endtime
rainrates_endtime = rainrates.copy()
rainrates_endtime.coords['time'] = \
    [
        pd.Timestamp(t) + pd.Timedelta(minutes=6)
        for t in rainrates_endtime.coords['time'].values
    ]

# Convert rainrates to accumulated rainfalls every 6 minutes with rr2rf().
rainfalls = rr2rf(rainrates_endtime)

# Accumulate hourly rainfall every 30 minutes
acc_rf = multiple_acc(
    rainfalls, basetime, basetime+pd.Timedelta(hours=3)
)

acc_time = pd.Timestamp.now()

Plotting rainfall maps

  1. Define the colour scheme.

  2. Obtaining the crs of the projection system using to_cartopy_crs() method of AreaDefinition.

  3. Defining the zoom or area extent. Tuple order is (x0, x1, y0, y1) as opposed to pyresample (x0, y0, x1, y1).

  4. Initialise figure.

  5. Initialise cartopy GeoAxes.

  6. Adding area extent defined in (3) to axes.

  7. Adding coastlines (this example uses GSHHS)

  8. Plot xarray using xarray.plot() and make desired adjustments to labels.

  9. Adding title (if required).

# Defining the colour scheme
levels = [
    0, 0.5, 2, 5, 10, 20,
    30, 40, 50, 70, 100, 150,
    200, 300, 400, 500, 600, 700
]

cmap = matplotlib.colors.ListedColormap([
    '#ffffff', '#9bf7f7', '#00ffff', '#00d5cc', '#00bd3d', '#2fd646',
    '#9de843', '#ffdd41', '#ffac33', '#ff621e', '#d23211', '#9d0063',
    '#e300ae', '#ff00ce', '#ff57da', '#ff8de6', '#ffe4fd'
])

norm = BoundaryNorm(levels, ncolors=cmap.N, clip=True)

# Defining projection
crs = area_def_tgt.to_cartopy_crs()
# Defining zoom extent
r = 64000
proj_site = xarray1.proj_site
zoom = (
    proj_site[0]-r, proj_site[0]+r, proj_site[1]-r, proj_site[1]+r
)

images = []
for time in acc_rf.coords['time'].values:
    # Define figure
    plt.figure(figsize=(28, 21))
    plt.axis('equal')
    # Plotting axes
    ax = plt.axes(projection=crs)
    # setting extent
    ax.set_extent(zoom, crs=crs)
    # Adding coastlines
    ax.add_feature(hires)

    # Plot data
    quadmesh = acc_rf.sel(time=time).plot(
        cmap=cmap,
        norm=norm,
        extend='neither'
        )

    # Customizing labels
    # Customising labels
    cbar = quadmesh.colorbar
    cbar.ax.set_ylabel(
        acc_rf.attrs['long_name']+'['+acc_rf.attrs['units']+']',
        fontsize=28
    )
    cbar.ax.tick_params(labelsize=24)
    ax.xaxis.set_visible(True)
    ax.yaxis.set_visible(True)
    ax.xaxis.set_tick_params(labelsize=24)
    ax.yaxis.set_tick_params(labelsize=24)
    ax.xaxis.label.set_size(28)
    ax.yaxis.label.set_size(28)

    # Add title
    t_minus = pd.Timestamp(time) - \
        basetime-pd.Timedelta(minutes=60)
    t_minus = round(t_minus.total_seconds()) // 60
    t_plus = pd.Timestamp(time) - basetime
    t_plus = round(t_plus.total_seconds()) // 60
    plt.title(
        "Quantitative Precipitation Forecast with Hourly"
        " Accumulated Rainfall.\n"
        f"Basetime: {basetime}    Valid time: {pd.Timestamp(time)}\n"
        "Start time: t"
        f"{t_minus:+} minutes   End time: t{t_plus:+} minutes",
        fontsize=32
    )
    savepath = THIS_DIR + \
        ("/../tests/outputs/forecast-" +
         f"{pd.Timestamp(time).strftime('%Y%m%d%H%M')}.png")
    plt.savefig(savepath)
    images.append(imageio.imread(savepath))

imageio.mimsave(
    THIS_DIR +
    "/../tests/outputs/forecast.gif",
    images, duration=1
)

rf_image_time = pd.Timestamp.now()

Extract the rainfall values at a specified location

In this example, the rainfall values at the location is assumed to be the same as the nearest gridpoint.

  1. Read information regarding the radar stations into a pandas.DataFrame.

  2. Extract the rainfall values at the nearest gridpoint to location for given times (in this example, 30 minute intervals).

  3. Store rainfall values over time in a pandas.DataFrame.

  4. Plot the time series of rainfall at different stations.

# Getting radar station coordinates
df = pd.read_csv(
    os.path.join(THIS_DIR, "../tests/samples/hk_raingauge.csv"),
    usecols=[0, 1, 2, 3, 4]
)

# Extract rainfall values at gridpoint closest to the
# location specified for given timesteps and storing it
# in pandas.DataFrame.

rf_time = []
for time in acc_rf.coords['time'].values:
    rf = []
    for index, row in df.iterrows():
        rf.append(acc_rf.sel(
            time=time, northing=row[1],
            easting=row[2],
            method='nearest'
        ).values)
    rf_time.append(rf)

rf_time = np.array(rf_time)

station_rf = pd.DataFrame(
    data=rf_time,
    columns=df.iloc[:, 0],
    index=pd.Index(
        acc_rf.coords['time'].values,
        name='time'
    )
)
print(station_rf)

# Plotting time series graph
ax = station_rf.plot(title="Time Series of Hourly Accumulated Rainfall",
                     grid=True)
ax.set_ylabel("Hourly Accumulated Rainfall (mm)")
plt.savefig(THIS_DIR+"/../tests/outputs/qpf_time_series.png")

extract_time = pd.Timestamp.now()

Checking run time of each component

print(f"Start time: {start_time}")
print(f"Initialising time: {initialising_time}")
print(f"Rover time: {rover_time}")
print(f"SLA time: {sla_time}")
print(f"Concatenating time: {concat_time}")
print(f"Plotting radar image time: {radar_image_time}")
print(f"Accumulating rainfall time: {acc_time}")
print(f"Plotting rainfall map time: {rf_image_time}")
print(f"Extracting and plotting time series time: {extract_time}")

print(f"Time to initialise: {initialising_time-start_time}")
print(f"Time to run rover: {rover_time-initialising_time}")
print(f"Time to perform SLA: {sla_time-rover_time}")
print(f"Time to concatenate xarrays: {concat_time - sla_time}")
print(f"Time to plot radar image: {radar_image_time - concat_time}")
print(f"Time to accumulate rainfall: {acc_time - radar_image_time}")
print(f"Time to plot rainfall maps: {rf_image_time-acc_time}")
print(f"Time to extract and plot time series: {extract_time-rf_image_time}")

Total running time of the script: ( 0 minutes 0.003 seconds)

Gallery generated by Sphinx-Gallery