Skip to content

Commit a834a59

Browse files
BHFockBHFock
committed
Add Exercise_02
1 parent f0afbbd commit a834a59

File tree

4 files changed

+885
-0
lines changed

4 files changed

+885
-0
lines changed

‎notebooks/.notebook_shadow_copies/Exercise_02.md‎

Lines changed: 107 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,107 @@
1+
---
2+
jupyter:
3+
jupytext:
4+
text_representation:
5+
extension: .md
6+
format_name: markdown
7+
format_version: '1.3'
8+
jupytext_version: 1.14.4
9+
kernelspec:
10+
display_name: Python 3 (ipykernel)
11+
language: python
12+
name: python3
13+
---
14+
15+
<!-- #region tags=[] -->
16+
# LFRic Iris data manipulation and visualisation practical
17+
18+
Let's apply what we've learned about data processing and visualisation of LFRic data in Iris and PyVista with the two exercises. \
19+
20+
The aim is to use the prompts to write the code yourself, but we have also provided a separate notebook containing the answers if you are stuck \
21+
22+
All the information needed to write the code for this practical can be found in the notebooks in the first part of this practical \
23+
24+
Note: this is delivered in Jupyter labs, but sometime the PyVista and GeoVista plotting is laggy in labs. If you prefer you can run in ipython.
25+
26+
27+
<!-- #endregion -->
28+
29+
## Exercise 2 - Regrid UM data to LFRic and plot using PyVista
30+
31+
32+
Now you can do a similar exercise compared to the previous Exercise 1, but regrid UM data onto a LFRic mesh and plot the data using PyVista
33+
34+
35+
**Step 1** To begin, we need to import the neccesary packages that we will need for this exercise.
36+
37+
```python
38+
%matplotlib inline
39+
import pyvista as pv
40+
import geovista as gv
41+
import geovista.theme
42+
import iris.quickplot as qplt
43+
import iris
44+
from geovista import GeoPlotter
45+
from esmf_regrid.experimental.unstructured_scheme import MeshToGridESMFRegridder, GridToMeshESMFRegridder
46+
from iris.experimental.ugrid.load import PARSE_UGRID_ON_LOAD
47+
pv.rcParams["use_ipyvtk"] = True
48+
iris.FUTURE.datum_support = True # avoids some warnings
49+
```
50+
51+
The pv_conversions script contains two functions which convert LFRic cubes to pyvista objects. Load these two functions:
52+
53+
```python
54+
from pv_conversions import pv_from_lfric_cube
55+
from pv_conversions import pv_from_um_cube
56+
```
57+
58+
**Step 2** Lets chose a different diagnostic, 'surface_temperature' and load both the UM data, as well LFRic data to use as reference grid.
59+
60+
```python
61+
data_path =
62+
lfric_path = data_path + '20210324T0000Z_lf_ugrid.nc'
63+
um_path = data_path + '20210324T0000Z_um_latlon.nc'
64+
65+
with PARSE_UGRID_ON_LOAD.context():
66+
lfric_rho = iris.load_cube(lfric_path, 'surface_air_pressure')
67+
68+
um_temp = iris.load_cube(um_path, 'surface_temperature')
69+
um_temp_t0 = um_temp[0]
70+
```
71+
72+
**Step 3** Initialise the regridder 'GridToMeshESMFRegridder' \
73+
Then use the regridder to regrid the new UM cube created earlier. Print your result and notice the mesh characterisics.
74+
75+
```python
76+
77+
```
78+
79+
**Step 4** Plot the regridded UM data with PyVista. \
80+
(hint: before you can do this you will need to convert you mesh to polydata using pv_from_lfric_cube)
81+
82+
```python
83+
84+
```
85+
86+
**Step 5** Plot the native UM data with PyVista. \
87+
(hint: before you can do this you will need to convert you mesh to polydata using pv_from_um_cube)
88+
89+
```python
90+
91+
```
92+
93+
**Step 6** Now we can plot this data side by side \
94+
(hints: start by using plotter = GeoPlotter(shape=(1,2)), then create your subplots, add your coastlines, add a base layer, and add you mesh) \
95+
note: PyVista and GeoVista can be slow in jupyter labs, but try and move the plots, look at the poles - you might notice the polar sigularity problem of the lat-lon grid.
96+
97+
```python
98+
99+
```
100+
101+
**Step 7** In notebook Section 3, we see how to use the plotter.camera_position = viewpoint functionality. Try this out with the surface temperature data.
102+
103+
```python
104+
105+
```
106+
107+
**Finished!?** These two exercises were just the beginning. If you have time try adding some cells below and extract a zonal mean, or try to select a region of data.

‎notebooks/.notebook_shadow_copies/Exercise_02_answers.md‎

Lines changed: 138 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,138 @@
1+
---
2+
jupyter:
3+
jupytext:
4+
text_representation:
5+
extension: .md
6+
format_name: markdown
7+
format_version: '1.3'
8+
jupytext_version: 1.14.4
9+
kernelspec:
10+
display_name: Python 3 (ipykernel)
11+
language: python
12+
name: python3
13+
---
14+
15+
<!-- #region tags=[] -->
16+
# LFRic Iris data manipulation and visualisation practical
17+
18+
Let's apply what we've learned about data processing and visualisation of LFRic data in Iris and PyVista with the two exercises. \
19+
20+
The aim is to use the prompts to write the code yourself, but we have also provided a separate notebook containing the answers if you are stuck \
21+
22+
All the information needed to write the code for this practical can be found in the notebooks in the first part of this practical \
23+
24+
Note: this is delivered in Jupyter labs, but sometime the PyVista and GeoVista plotting is laggy in labs. If you prefer you can run in ipython.
25+
26+
27+
<!-- #endregion -->
28+
29+
## Exercise 2 - Regrid UM data to LFRic and plot using PyVista
30+
31+
32+
Now you can do a similar exercise compared to the previous Exercise 1, but regrid UM data onto a LFRic mesh and plot the data using PyVista
33+
34+
35+
**Step 1** To begin, we need to import the neccesary packages that we will need for this exercise.
36+
37+
```python
38+
%matplotlib inline
39+
import pyvista as pv
40+
import geovista as gv
41+
import geovista.theme
42+
import iris.quickplot as qplt
43+
import iris
44+
from geovista import GeoPlotter
45+
from esmf_regrid.experimental.unstructured_scheme import MeshToGridESMFRegridder, GridToMeshESMFRegridder
46+
from iris.experimental.ugrid.load import PARSE_UGRID_ON_LOAD
47+
pv.rcParams["use_ipyvtk"] = True
48+
iris.FUTURE.datum_support = True # avoids some warnings
49+
```
50+
51+
The pv_conversions script contains two functions which convert LFRic cubes to pyvista objects. Load these two functions:
52+
53+
```python
54+
from pv_conversions import pv_from_lfric_cube
55+
from pv_conversions import pv_from_um_cube
56+
```
57+
58+
**Step 2** Lets chose a different diagnostic, 'surface_temperature' and load both the UM data, as well LFRic data to use as reference grid.
59+
60+
```python
61+
# Define the location of the data and file names
62+
data_path = ''
63+
lfric_path = data_path + '20210324T0000Z_lf_ugrid.nc'
64+
um_path = data_path + '20210324T0000Z_um_latlon.nc'
65+
66+
with PARSE_UGRID_ON_LOAD.context():
67+
lfric_rho = iris.load_cube(lfric_path, 'surface_air_pressure')
68+
69+
um_temp = iris.load_cube(um_path, 'surface_temperature')
70+
um_temp_t0 = um_temp[0]
71+
```
72+
73+
**Step 3** Initialise the regridder 'GridToMeshESMFRegridder' \
74+
Then use the regridder to regrid the new UM cube created earlier. Print your result and notice the mesh characterisics.
75+
76+
```python
77+
g2m_regridder = GridToMeshESMFRegridder(um_temp_t0, lfric_rho)
78+
regridded_um = g2m_regridder(um_temp_t0)
79+
print(regridded_um)
80+
```
81+
82+
**Step 4** Plot the regridded UM data with PyVista. \
83+
(hint: before you can do this you will need to convert you mesh to polydata using pv_from_lfric_cube)
84+
85+
```python
86+
pv = pv_from_lfric_cube(regridded_um)
87+
pv.plot()
88+
```
89+
90+
**Step 5** Plot the native UM data with PyVista. \
91+
(hint: before you can do this you will need to convert you mesh to polydata using pv_from_um_cube)
92+
93+
```python
94+
um_pv = pv_from_um_cube(um_temp_t0)
95+
um_pv.plot()
96+
```
97+
98+
**Step 6** Now we can plot this data side by side \
99+
(hints: start by using plotter = GeoPlotter(shape=(1,2)), then create your subplots, add your coastlines, add a base layer, and add you mesh) \
100+
note: PyVista and GeoVista can be slow in jupyter labs, but try and move the plots, look at the poles - you might notice the polar sigularity problem of the lat-lon grid.
101+
102+
```python
103+
plotter = GeoPlotter(shape=(1, 2))
104+
105+
plotter.subplot(0, 0)
106+
plotter.add_coastlines()
107+
plotter.add_base_layer(texture=gv.natural_earth_1())
108+
plotter.add_mesh(pv, show_edges=True)
109+
110+
plotter.subplot(0, 1)
111+
plotter.add_coastlines()
112+
#plotter.add_base_layer(texture=gv.natural_earth_1())
113+
plotter.add_mesh(um_pv, show_edges=True)
114+
115+
# Make left+right move together
116+
plotter.link_views()
117+
118+
# Try to rationalise the global view a little.
119+
plotter.view_xz()
120+
121+
plotter.show()
122+
```
123+
124+
**Step 7** In notebook Section 3, we see how to use the plotter.camera_position = viewpoint functionality. Try this out with the surface temperature data.
125+
126+
```python
127+
viewpoint = [
128+
(0.9550352379408845, 0.9378277371075855, 0.9637172962958191),
129+
(0.0, 0.0, 0.0),
130+
(-0.3202752464164225, -0.5004192729867466, 0.8043657860428399)
131+
]
132+
133+
plotter.camera_position = viewpoint
134+
135+
plotter.show()
136+
```
137+
138+
**Finished!?** These two exercises were just the beginning. If you have time try adding some cells below and extract a zonal mean, or try to select a region of data.

0 commit comments

Comments
 (0)