Skip to content

Commit ea26738

Browse files
author
Joe Hamman
authored
update rasm example ipython notebook and rst (#982)
1 parent 584e703 commit ea26738

File tree

3 files changed

+428
-320
lines changed

3 files changed

+428
-320
lines changed

‎doc/examples/monthly-means.rst

Lines changed: 58 additions & 39 deletions
Original file line numberDiff line numberDiff line change
@@ -3,33 +3,40 @@
33
Calculating Seasonal Averages from Timeseries of Monthly Means
44
==============================================================
55

6-
Author: `Joe Hamman <http://uw-hydro.github.io/current_member/joe_hamman/>`_
6+
Author: `Joe Hamman <https://github.com/jhamman/>`__
77

8-
The data for this example can be found in the `xray-data <https://github.com/xray/xray-data>`_ repository. This example is also available in an IPython Notebook that is available `here <https://github.com/xray/xray/tree/master/examples/xray_seasonal_means.ipynb>`_.
8+
The data used for this example can be found in the
9+
`xarray-data <https://github.com/pydata/xarray-data>`__ repository.
910

1011
Suppose we have a netCDF or xray Dataset of monthly mean data and we
1112
want to calculate the seasonal average. To do this properly, we need to
1213
calculate the weighted average considering that each month has a
1314
different number of days.
1415

16+
Suppose we have a netCDF or ``xarray.Dataset`` of monthly mean data and
17+
we want to calculate the seasonal average. To do this properly, we need
18+
to calculate the weighted average considering that each month has a
19+
different number of days.
20+
1521
.. code:: python
1622
1723
%matplotlib inline
1824
import numpy as np
1925
import pandas as pd
20-
import xray
26+
import xarray as xr
2127
from netCDF4 import num2date
2228
import matplotlib.pyplot as plt
2329
2430
print("numpy version : ", np.__version__)
25-
print("pandas version : ", pd.version.version)
26-
print("xray version : ", xray.version.version)
31+
print("pandas version : ", pd.__version__)
32+
print("xarray version : ", xr.__version__)
33+
2734
2835
.. parsed-literal::
2936
30-
numpy version : 1.9.2
31-
pandas version : 0.16.2
32-
xray version : 0.5.1
37+
numpy version : 1.11.1
38+
pandas version : 0.18.1
39+
xarray version : 0.8.2
3340
3441
3542
Some calendar information so we can support any netCDF calendar.
@@ -90,21 +97,22 @@ Open the ``Dataset``
9097

9198
.. code:: python
9299
93-
monthly_mean_file = 'RASM_example_data.nc'
94-
ds = xray.open_dataset(monthly_mean_file, decode_coords=False)
100+
ds = xr.tutorial.load_dataset('rasm')
95101
print(ds)
96102
97103
98104
.. parsed-literal::
99105
100-
<xray.Dataset>
106+
<xarray.Dataset>
101107
Dimensions: (time: 36, x: 275, y: 205)
102108
Coordinates:
103109
* time (time) datetime64[ns] 1980-09-16T12:00:00 1980-10-17 ...
104-
* x (x) int64 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ...
105110
* y (y) int64 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ...
111+
* x (x) int64 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ...
106112
Data variables:
107113
Tair (time, y, x) float64 nan nan nan nan nan nan nan nan nan nan ...
114+
yc (y, x) float64 16.53 16.78 17.02 17.27 17.51 17.76 18.0 18.25 ...
115+
xc (y, x) float64 189.2 189.4 189.6 189.7 189.9 190.1 190.2 190.4 ...
108116
Attributes:
109117
title: /workspace/jhamman/processed/R1002RBRxaaa01a/lnd/temp/R1002RBRxaaa01a.vic.ha.1979-09-01.nc
110118
institution: U.W.
@@ -132,9 +140,8 @@ allong the time dimension.
132140
.. code:: python
133141
134142
# Make a DataArray with the number of days in each month, size = len(time)
135-
month_length = xray.DataArray(get_dpm(ds.time.to_index(),
136-
calendar='noleap'),
137-
coords=[ds.time], name='month_length')
143+
month_length = xr.DataArray(get_dpm(ds.time.to_index(), calendar='noleap'),
144+
coords=[ds.time], name='month_length')
138145
139146
# Calculate the weights by grouping by 'time.season'.
140147
# Conversion to float type ('astype(float)') only necessary for Python 2.x
@@ -153,14 +160,16 @@ allong the time dimension.
153160
154161
.. parsed-literal::
155162
156-
<xray.Dataset>
163+
<xarray.Dataset>
157164
Dimensions: (season: 4, x: 275, y: 205)
158165
Coordinates:
159166
* y (y) int64 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ...
160167
* x (x) int64 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ...
161168
* season (season) object 'DJF' 'JJA' 'MAM' 'SON'
162169
Data variables:
163170
Tair (season, y, x) float64 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 ...
171+
xc (season, y, x) float64 189.2 189.4 189.6 189.7 189.9 190.1 ...
172+
yc (season, y, x) float64 16.53 16.78 17.02 17.27 17.51 17.76 18.0 ...
164173
165174
166175
.. code:: python
@@ -172,30 +181,31 @@ allong the time dimension.
172181
.. code:: python
173182
174183
# Quick plot to show the results
175-
is_null = np.isnan(ds_unweighted['Tair'][0].values)
184+
notnull = pd.notnull(ds_unweighted['Tair'][0])
176185
177186
fig, axes = plt.subplots(nrows=4, ncols=3, figsize=(14,12))
178187
for i, season in enumerate(('DJF', 'MAM', 'JJA', 'SON')):
179-
plt.sca(axes[i, 0])
180-
plt.pcolormesh(np.ma.masked_where(is_null, ds_weighted['Tair'].sel(season=season).values),
181-
vmin=-30, vmax=30, cmap='Spectral_r')
182-
plt.colorbar(extend='both')
183-
184-
plt.sca(axes[i, 1])
185-
plt.pcolormesh(np.ma.masked_where(is_null, ds_unweighted['Tair'].sel(season=season).values),
186-
vmin=-30, vmax=30, cmap='Spectral_r')
187-
plt.colorbar(extend='both')
188-
189-
plt.sca(axes[i, 2])
190-
plt.pcolormesh(np.ma.masked_where(is_null, ds_diff['Tair'].sel(season=season).values),
191-
vmin=-0.1, vmax=.1, cmap='RdBu_r')
192-
plt.colorbar(extend='both')
193-
for j in range(3):
194-
axes[i, j].axes.get_xaxis().set_ticklabels([])
195-
axes[i, j].axes.get_yaxis().set_ticklabels([])
196-
axes[i, j].axes.axis('tight')
188+
ds_weighted['Tair'].sel(season=season).where(notnull).plot.pcolormesh(
189+
ax=axes[i, 0], vmin=-30, vmax=30, cmap='Spectral_r',
190+
add_colorbar=True, extend='both')
191+
192+
ds_unweighted['Tair'].sel(season=season).where(notnull).plot.pcolormesh(
193+
ax=axes[i, 1], vmin=-30, vmax=30, cmap='Spectral_r',
194+
add_colorbar=True, extend='both')
195+
196+
ds_diff['Tair'].sel(season=season).where(notnull).plot.pcolormesh(
197+
ax=axes[i, 2], vmin=-0.1, vmax=.1, cmap='RdBu_r',
198+
add_colorbar=True, extend='both')
197199
198200
axes[i, 0].set_ylabel(season)
201+
axes[i, 1].set_ylabel('')
202+
axes[i, 2].set_ylabel('')
203+
204+
for ax in axes.flat:
205+
ax.axes.get_xaxis().set_ticklabels([])
206+
ax.axes.get_yaxis().set_ticklabels([])
207+
ax.axes.axis('tight')
208+
ax.set_xlabel('')
199209
200210
axes[0, 0].set_title('Weighted by DPM')
201211
axes[0, 1].set_title('Equal Weighting')
@@ -206,6 +216,15 @@ allong the time dimension.
206216
fig.suptitle('Seasonal Surface Air Temperature', fontsize=16, y=1.02)
207217
208218
219+
220+
221+
.. parsed-literal::
222+
223+
<matplotlib.text.Text at 0x117c18048>
224+
225+
226+
227+
209228
.. image:: monthly_means_output.png
210229

211230

@@ -214,12 +233,12 @@ allong the time dimension.
214233
# Wrap it into a simple function
215234
def season_mean(ds, calendar='standard'):
216235
# Make a DataArray of season/year groups
217-
year_season = xray.DataArray(ds.time.to_index().to_period(freq='Q-NOV').to_timestamp(how='E'),
218-
coords=[ds.time], name='year_season')
236+
year_season = xr.DataArray(ds.time.to_index().to_period(freq='Q-NOV').to_timestamp(how='E'),
237+
coords=[ds.time], name='year_season')
219238
220239
# Make a DataArray with the number of days in each month, size = len(time)
221-
month_length = xray.DataArray(get_dpm(ds.time.to_index(), calendar=calendar),
222-
coords=[ds.time], name='month_length')
240+
month_length = xr.DataArray(get_dpm(ds.time.to_index(), calendar=calendar),
241+
coords=[ds.time], name='month_length')
223242
# Calculate the weights by grouping by 'time.season'
224243
weights = month_length.groupby('time.season') / month_length.groupby('time.season').sum()
225244

0 commit comments

Comments
 (0)