Boundary Pass Central Node Model Results Slab¶
Assemble time series from a model results slab for Rich:
- NEMO-3.6 nowcast-green runs
- full water column
- hourly average results at Boundary Pass location of ONC mooring
- hourly average results at ONC central node
- 13-26 Jul-2016
Note: The code in this notebook uses language features that are available only in Python 3.6 and later.
In [4]:
import os
import pathlib
import arrow
import cmocean
import numpy as np
from matplotlib.path import Path
from matplotlib.patches import PathPatch
import matplotlib.pyplot as plt
import xarray as xr
In [5]:
%matplotlib inline
In [6]:
# ONC description of central node location:
# Depth: 297 m
# Latitude: 49.040066666
# Longitude: -123.425825
central_ji = (424, 266)
# ONC description of Boundary Pass mooring location:
# Depth: 230 m
# Latitude: 48.7665
# Longitude: -123.039556
boundary_ji = (343, 289)
Region of Interest¶
In [7]:
bathy = xr.open_dataset('https://salishsea.eos.ubc.ca/erddap/griddap/ubcSSnBathymetry2V16-07')
In [8]:
fig, ax = plt.subplots(1, 1, figsize=(10, 8))
ax.set_aspect(5/4.4)
cmap = cmocean.cm.deep
cmap.set_bad('burlywood')
bathy.bathymetry.isel(gridY=slice(325, 451), gridX=slice(200, 350)).plot(ax=ax, cmap=cmap)
# Mark closest model points to Boundary Pass mooring and VENUS Central node with red stars
ax.plot(boundary_ji[1], boundary_ji[0], marker='*', color='red')
ax.plot(central_ji[1], central_ji[0], marker='*', color='red')
# Mark the x-y region of interest with at lime green rectangle
codes = [Path.MOVETO] + [Path.LINETO]*3 + [Path.CLOSEPOLY]
offsets = {'north': 0, 'south': 5, 'east': 25, 'west': 15}
vertices = [
(central_ji[1] - offsets['west'], boundary_ji[0] - offsets['south']),
(central_ji[1] - offsets['west'], central_ji[0] + offsets['north']),
(boundary_ji[1] + offsets['east'], central_ji[0] + offsets['north']),
(boundary_ji[1] + offsets['east'], boundary_ji[0] - offsets['south']),
(0, 0)]
vertices = np.array(vertices, float)
path = Path(vertices, codes)
pathpatch = PathPatch(path, facecolor='None', edgecolor='lime')
ax.add_patch(pathpatch)
Out[8]:
<matplotlib.patches.PathPatch at 0x7f485af70cc0>
Datasets¶
- Index ranges in the
yandxdirections that define the region of interest above - Results date range
- Time origin for the dataset we are building
- Results archive storage path
- List of dataset paths to process
In [18]:
y_slice = slice(boundary_ji[0] - offsets['south'], central_ji[0] + offsets['north'])
x_slice = slice(central_ji[1] - offsets['west'], boundary_ji[1] + offsets['east'])
start_date = arrow.get('2016-07-13')
end_date = arrow.get('2016-07-13')
time_origin = start_date.format('YYYY-MM-DD HH:mm:ss')
results_archive = pathlib.Path('/results/SalishSea/nowcast-green/')
def build_datasets_list(grid):
datasets = []
for a in arrow.Arrow.range('day', start_date, end_date):
ddmmmyy = a.format('DDMMMYY').lower()
yyyymmdd = a.format('YYYYMMDD')
results_dir = f'{ddmmmyy}'
grid_1h = f'SalishSea_1h_{yyyymmdd}_{yyyymmdd}_grid_{grid}.nc'
datasets.append(os.fspath(results_archive/results_dir/grid_1h))
return datasets
The code below:
opens the list of datasets generated above
indexes into them to isolate the
(y_slice, x_slice)region of interestcreates a new dataset object containing, for each model grid point in the region of interest:
- longitude and latitude
- full depth salinity and temperature profiles
stores the new dataset as a netCDF4 file
Do not run the cell below in this notebook for more than 1 day for testing purposes!
It requires a LOT of memory.
To produce the whole 2016-06-13 through 2016-06-26 dataset:
- save this notebook as a Python module
- use the
environment.yamlfile in this directory to create a conda environment to run the module in onsalish - activate the conda environment and execute the module from the command-line
Temperature and Salinity¶
In [ ]:
creation_time = arrow.now().format('YYYY-MM-DD HH:mm:ss ZZ')
start_yyyymmdd = start_date.format('YYYYMMDD')
end_yyyymmdd = end_date.format('YYYYMMDD')
grid = 'T'
datasets = build_datasets_list(grid)
with xr.open_mfdataset(datasets) as results:
slab = results.isel(y=y_slice, x=x_slice)
ds = xr.Dataset(
data_vars={
'nav_lon': slab.nav_lon,
'nav_lat': slab.nav_lat,
'vosaline': slab.vosaline,
'votemper': slab.votemper,
},
attrs={
'name': f'SalishSea_1h_{start_yyyymmdd}_{end_yyyymmdd}_grid_{grid}',
'title': 'SalishSeaCast Boundary Pass and ONC Central Node Temperature and Salinity Results',
'conventions': 'CF-1.5',
'source_code': 'https://bitbucket.org/salishsea/results/src/tip/BoundaryPassCentralNodeForRich/BoundaryPassCentralNodeResultsTimeSeries.py',
'rendered_notebook': 'https://nbviewer.jupyter.org/urls/bitbucket.org/salishsea/results/raw/tip/BoundaryPassCentralNodeForRich/BoundaryPassCentralNodeResultsSlab.ipynb',
'history': f'{creation_time}: Processed from results files in /results/SalishSea/nowcast-green/',
},
)
ds.vosaline.attrs['standard_name'] = 'sea_water_reference_salinity'
ds.vosaline.attrs['units'] = 'g kg-1'
ds.time_centered.attrs['time_origin'] = f'hours since {time_origin}'
ds.time_counter.attrs['time_origin'] = f'hours since {time_origin}'
ds.to_netcdf(
f'SalishSea_1h_{start_yyyymmdd}_{end_yyyymmdd}_grid_{grid}.nc',
format='netCDF4', engine='netcdf4',
encoding={
'time_counter': {'units': 'hours since 2016-06-13 00:00:00'},
'time_centered': {'units': 'hours since 2016-06-13 00:00:00'},
},
unlimited_dims='time_counter',
)
u Velocity Component¶
In [19]:
creation_time = arrow.now().format('YYYY-MM-DD HH:mm:ss ZZ')
start_yyyymmdd = start_date.format('YYYYMMDD')
end_yyyymmdd = end_date.format('YYYYMMDD')
grid = 'U'
datasets = build_datasets_list(grid)
with xr.open_mfdataset(datasets) as results:
slab = results.isel(y=y_slice, x=x_slice)
ds = xr.Dataset(
data_vars={
'nav_lon': slab.nav_lon,
'nav_lat': slab.nav_lat,
'vozocrtx': slab.vozocrtx,
},
attrs={
'name': f'SalishSea_1h_{start_yyyymmdd}_{end_yyyymmdd}_grid_{grid}',
'title': 'SalishSeaCast Boundary Pass and ONC Central Node {grid.lower()} Velocity Component Results',
'conventions': 'CF-1.5',
'source_code': 'https://bitbucket.org/salishsea/results/src/tip/BoundaryPassCentralNodeForRich/BoundaryPassCentralNodeResultsTimeSeries.py',
'rendered_notebook': 'https://nbviewer.jupyter.org/urls/bitbucket.org/salishsea/results/raw/tip/BoundaryPassCentralNodeForRich/BoundaryPassCentralNodeResultsSlab.ipynb',
'history': f'{creation_time}: Processed from results files in /results/SalishSea/nowcast-green/',
},
)
ds.time_centered.attrs['time_origin'] = f'hours since {time_origin}'
ds.time_counter.attrs['time_origin'] = f'hours since {time_origin}'
ds.to_netcdf(
f'SalishSea_1h_{start_yyyymmdd}_{end_yyyymmdd}_grid_{grid}.nc',
format='netCDF4', engine='netcdf4',
encoding={
'time_counter': {'units': 'hours since 2016-06-13 00:00:00'},
'time_centered': {'units': 'hours since 2016-06-13 00:00:00'},
},
unlimited_dims='time_counter',
)
v Velocity Component¶
In [20]:
creation_time = arrow.now().format('YYYY-MM-DD HH:mm:ss ZZ')
start_yyyymmdd = start_date.format('YYYYMMDD')
end_yyyymmdd = end_date.format('YYYYMMDD')
grid = 'V'
datasets = build_datasets_list(grid)
with xr.open_mfdataset(datasets) as results:
slab = results.isel(y=y_slice, x=x_slice)
ds = xr.Dataset(
data_vars={
'nav_lon': slab.nav_lon,
'nav_lat': slab.nav_lat,
'vomecrty': slab.vomecrty,
},
attrs={
'name': f'SalishSea_1h_{start_yyyymmdd}_{end_yyyymmdd}_grid_{grid}',
'title': f'SalishSeaCast Boundary Pass and ONC Central Node {grid.lower()} Velocity Component Results',
'conventions': 'CF-1.5',
'source_code': 'https://bitbucket.org/salishsea/results/src/tip/BoundaryPassCentralNodeForRich/BoundaryPassCentralNodeResultsTimeSeries.py',
'rendered_notebook': 'https://nbviewer.jupyter.org/urls/bitbucket.org/salishsea/results/raw/tip/BoundaryPassCentralNodeForRich/BoundaryPassCentralNodeResultsSlab.ipynb',
'history': f'{creation_time}: Processed from results files in /results/SalishSea/nowcast-green/',
},
)
ds.time_centered.attrs['time_origin'] = f'hours since {time_origin}'
ds.time_counter.attrs['time_origin'] = f'hours since {time_origin}'
ds.to_netcdf(
f'SalishSea_1h_{start_yyyymmdd}_{end_yyyymmdd}_grid_{grid}.nc',
format='netCDF4', engine='netcdf4',
encoding={
'time_counter': {'units': 'hours since 2016-06-13 00:00:00'},
'time_centered': {'units': 'hours since 2016-06-13 00:00:00'},
},
unlimited_dims='time_counter',
)
w Velocity Component¶
In [ ]:
creation_time = arrow.now().format('YYYY-MM-DD HH:mm:ss ZZ')
start_yyyymmdd = start_date.format('YYYYMMDD')
end_yyyymmdd = end_date.format('YYYYMMDD')
grid = 'W'
datasets = build_datasets_list(grid)
with xr.open_mfdataset(datasets) as results:
slab = results.isel(y=y_slice, x=x_slice)
ds = xr.Dataset(
data_vars={
'nav_lon': slab.nav_lon,
'nav_lat': slab.nav_lat,
'vovecrtz': slab.vovecrtz,
},
attrs={
'name': f'SalishSea_1h_{start_yyyymmdd}_{end_yyyymmdd}_grid_{grid}',
'title': f'SalishSeaCast Boundary Pass and ONC Central Node {grid.lower()} Velocity Component Results',
'conventions': 'CF-1.5',
'source_code': 'https://bitbucket.org/salishsea/results/src/tip/BoundaryPassCentralNodeForRich/BoundaryPassCentralNodeResultsTimeSeries.py',
'rendered_notebook': 'https://nbviewer.jupyter.org/urls/bitbucket.org/salishsea/results/raw/tip/BoundaryPassCentralNodeForRich/BoundaryPassCentralNodeResultsSlab.ipynb',
'history': f'{creation_time}: Processed from results files in /results/SalishSea/nowcast-green/',
},
)
ds.time_centered.attrs['time_origin'] = f'hours since {time_origin}'
ds.time_counter.attrs['time_origin'] = f'hours since {time_origin}'
ds.to_netcdf(
f'SalishSea_1h_{start_yyyymmdd}_{end_yyyymmdd}_grid_{grid}.nc',
format='netCDF4', engine='netcdf4',
encoding={
'time_counter': {'units': 'hours since 2016-06-13 00:00:00'},
'time_centered': {'units': 'hours since 2016-06-13 00:00:00'},
},
unlimited_dims='time_counter',
)
In [ ]: