Develop wave_height_period Figure Module¶
Development of functions for nowcast.figures.wwatch3.wave_height_period web site figure module.
In [1]:
from pathlib import Path
from types import SimpleNamespace
import matplotlib.dates
import matplotlib.pyplot as plt
import requests
import xarray
from salishsea_tools.places import PLACES
from nowcast.figures import shared
import nowcast.figures.website_theme
In [2]:
%matplotlib inline
_get_wwatch3_fields() Function¶
In [3]:
def _get_wwatch3_fields(dataset_url):
## TODO: This is a work-around because neither netCDF4 nor xarray are able
## to load the dataset directly from the URL due to an OpenDAP issue
dataset_id = dataset_url.rsplit('/', 1)[1].split('.', 1)[0]
wwatch3_fields_file = Path('/tmp').joinpath(dataset_id).with_suffix('.nc')
with wwatch3_fields_file.open('wb') as f:
resp = requests.get(f'{dataset_url}')
f.write(resp.content)
try:
wwatch3_fields = xarray.open_dataset(wwatch3_fields_file)
except OSError:
raise ValueError(f'WaveWatch3 fields dataset not found')
return wwatch3_fields
_prep_plot_data() Function¶
In [129]:
def _prep_plot_data(buoy, wwatch3_dataset_url):
wwatch3_fields = _get_wwatch3_fields(wwatch3_dataset_url)
wwatch3 = xarray.Dataset({
'wave_height': wwatch3_fields.hs.sel(
longitude=PLACES[buoy]['lon lat'][0]+360,
latitude=PLACES[buoy]['lon lat'][1],
method='nearest',
),
'peak_freq': wwatch3_fields.fp.sel(
longitude=PLACES[buoy]['lon lat'][0]+360,
latitude=PLACES[buoy]['lon lat'][1],
method='nearest',
),
})
wwatch3_period = slice(
str(wwatch3_fields.time.values[0]), str(wwatch3_fields.time.values[-1])
)
obs = get_ndbc_buoy(buoy)
obs = xarray.Dataset({
'wave_height': obs.loc[wwatch3_period, 'WVHT [m]'],
'dominant_period': obs.loc[wwatch3_period, 'DPD [sec]'],
})
# Change dataset times to Pacific time zone
shared.localize_time(wwatch3)
shared.localize_time(obs)
return SimpleNamespace(
wwatch3=wwatch3,
obs=obs,
)
_prep_fig_axes() Function¶
In [5]:
def _prep_fig_axes(figsize, theme):
fig, (ax_sig_height, ax_peak_freq) = plt.subplots(
2, 1, figsize=figsize, facecolor=theme.COLOURS['figure']['facecolor']
)
fig.autofmt_xdate()
return fig, (ax_sig_height, ax_peak_freq)
_plot_wave_height_time_series() Function¶
In [125]:
def _plot_wave_height_time_series(ax, place, plot_data, theme):
plot_data.obs.wave_height.plot(
ax=ax,
linewidth=2,
label='ECCC Observed',
color=theme.COLOURS['time series']['obs wave height'],
)
plot_data.wwatch3.wave_height.plot(
ax=ax,
linewidth=2,
label='WaveWatch3',
color=theme.COLOURS['time series']['wave height'],
)
legend = ax.legend(prop=theme.FONTS['legend label small'])
legend.set_title('Legend', prop=theme.FONTS['legend title small'])
_wave_height_time_series_labels(ax, place, plot_data, theme)
_wave_height_time_series_labels() Function¶
In [126]:
def _wave_height_time_series_labels(ax, place, plot_data, theme):
ax.set_title(
f'Significant Wave Height at {place}',
fontproperties=theme.FONTS['axes title'],
color=theme.COLOURS['text']['axes title']
)
ax.set_ylabel(
'Significant Wave Height [m]',
fontproperties=theme.FONTS['axis'],
color=theme.COLOURS['text']['axis']
)
ax.grid(axis='both')
theme.set_axis_colors(ax)
_plot_dominant_period_time_series() Function¶
In [127]:
def _plot_dominant_period_time_series(ax, place, plot_data, theme):
plot_data.obs.dominant_period.plot(
ax=ax,
linewidth=2,
label='ECCC Observed',
color=theme.COLOURS['time series']['obs wave period'],
)
(1/plot_data.wwatch3.peak_freq).plot(
ax=ax,
linewidth=2,
label='WaveWatch3',
color=theme.COLOURS['time series']['wave period'],
)
legend = ax.legend(prop=theme.FONTS['legend label small'])
legend.set_title('Legend', prop=theme.FONTS['legend title small'])
_dominant_period_time_series_labels(ax, place, plot_data, theme)
_dominant_period_time_series_labels() Function¶
In [121]:
def _dominant_period_time_series_labels(ax, place, plot_data, theme):
ax.set_title(
f'Dominant Wave Period at {place}',
fontproperties=theme.FONTS['axes title'],
color=theme.COLOURS['text']['axes title']
)
ax.set_xlabel(
f'Time [{plot_data.wwatch3.attrs["tz_name"]}]',
fontproperties=theme.FONTS['axis'],
color=theme.COLOURS['text']['axis']
)
ax.xaxis.set_major_formatter(matplotlib.dates.DateFormatter('%d%b %H:%M'))
ax.set_ylabel(
'Dominant Wave Period [s]',
fontproperties=theme.FONTS['axis'],
color=theme.COLOURS['text']['axis']
)
ax.grid(axis='both')
theme.set_axis_colors(ax)
make_figure() Function¶
This is is the function that will be called by the nowcast.workers.make_plots worker to return a matplotlib.figure.Figure object.
In [114]:
def make_figure(
buoy,
wwatch3_dataset_url,
figsize=(16, 9),
theme=nowcast.figures.website_theme,
):
plot_data = _prep_plot_data(buoy, wwatch3_dataset_url)
fig, (ax_sig_height, ax_peak_freq) = _prep_fig_axes(figsize, theme)
_plot_wave_height_time_series(ax_sig_height, buoy, plot_data, theme)
_plot_dominant_period_time_series(ax_peak_freq, buoy, plot_data, theme)
return fig
Render the Figure¶
The %%timeit cell magic lets us keep an eye on how log the figure takes to process.
Setting -n1 -r1 prevents it from processing the figure more than once
as it might try to do to generate better statistics.
In [130]:
%%timeit -n1 -r1
from importlib import reload
from nowcast.figures import website_theme
from salishsea_tools import places
reload(website_theme)
reload(places)
wwatch3_dataset_url = (
'https://salishsea.eos.ubc.ca/erddap/griddap/ubcSSf2DWaveFields30mV17-02.nc?hs,fp'
)
fig = make_figure('Halibut Bank', wwatch3_dataset_url)
fig = make_figure('Sentry Shoal', wwatch3_dataset_url)
2.79 s ± 0 ns per loop (mean ± std. dev. of 1 run, 1 loop each)
In [ ]: