|
1
|
|
|
# Copyright (c) 2008-2015 MetPy Developers. |
|
2
|
|
|
# Distributed under the terms of the BSD 3-Clause License. |
|
3
|
|
|
# SPDX-License-Identifier: BSD-3-Clause |
|
4
|
|
|
"""Contains calculation of kinematic parameters (e.g. divergence or vorticity).""" |
|
5
|
|
|
from __future__ import division |
|
6
|
|
|
|
|
7
|
|
|
import functools |
|
8
|
|
|
import warnings |
|
9
|
|
|
|
|
10
|
|
|
import numpy as np |
|
11
|
|
|
from pyproj import Geod |
|
12
|
|
|
|
|
13
|
|
|
from ..calc.tools import get_layer |
|
14
|
|
|
from ..cbook import is_string_like, iterable |
|
15
|
|
|
from ..constants import Cp_d, g |
|
16
|
|
|
from ..package_tools import Exporter |
|
17
|
|
|
from ..units import atleast_2d, check_units, concatenate, units |
|
18
|
|
|
|
|
19
|
|
|
exporter = Exporter(globals()) |
|
20
|
|
|
|
|
21
|
|
|
|
|
22
|
|
|
def _gradient(f, *args, **kwargs): |
|
23
|
|
|
"""Wrap :func:`numpy.gradient` to handle units.""" |
|
24
|
|
|
if len(args) < f.ndim: |
|
25
|
|
|
args = list(args) |
|
26
|
|
|
args.extend([units.Quantity(1., 'dimensionless')] * (f.ndim - len(args))) |
|
27
|
|
|
grad = np.gradient(f, *(a.magnitude for a in args), **kwargs) |
|
28
|
|
|
if f.ndim == 1: |
|
29
|
|
|
return units.Quantity(grad, f.units / args[0].units) |
|
30
|
|
|
return [units.Quantity(g, f.units / dx.units) for dx, g in zip(args, grad)] |
|
31
|
|
|
|
|
32
|
|
|
|
|
33
|
|
|
def _stack(arrs): |
|
34
|
|
|
return concatenate([a[np.newaxis] for a in arrs], axis=0) |
|
35
|
|
|
|
|
36
|
|
|
|
|
37
|
|
|
def _get_gradients(u, v, dx, dy): |
|
38
|
|
|
"""Return derivatives for components to simplify calculating convergence and vorticity.""" |
|
39
|
|
|
dudy, dudx = _gradient(u, dy, dx) |
|
40
|
|
|
dvdy, dvdx = _gradient(v, dy, dx) |
|
41
|
|
|
return dudx, dudy, dvdx, dvdy |
|
42
|
|
|
|
|
43
|
|
|
|
|
44
|
|
|
def _is_x_first_dim(dim_order): |
|
45
|
|
|
"""Determine whether x is the first dimension based on the value of dim_order.""" |
|
46
|
|
|
if dim_order is None: |
|
47
|
|
|
warnings.warn('dim_order is using the default setting (currently "xy"). This will ' |
|
48
|
|
|
'change to "yx" in the next version. It is recommended that you ' |
|
49
|
|
|
'specify the appropriate ordering ("xy", "yx") for your data by ' |
|
50
|
|
|
'passing the `dim_order` argument to the calculation.', FutureWarning) |
|
51
|
|
|
dim_order = 'xy' |
|
52
|
|
|
return dim_order == 'xy' |
|
53
|
|
|
|
|
54
|
|
|
|
|
55
|
|
|
def _check_and_flip(arr): |
|
56
|
|
|
"""Transpose array or list of arrays if they are 2D.""" |
|
57
|
|
|
if hasattr(arr, 'ndim'): |
|
58
|
|
|
if arr.ndim >= 2: |
|
59
|
|
|
return arr.T |
|
60
|
|
|
else: |
|
61
|
|
|
return arr |
|
62
|
|
|
elif not is_string_like(arr) and iterable(arr): |
|
63
|
|
|
return tuple(_check_and_flip(a) for a in arr) |
|
64
|
|
|
else: |
|
65
|
|
|
return arr |
|
66
|
|
|
|
|
67
|
|
|
|
|
68
|
|
|
def ensure_yx_order(func): |
|
69
|
|
|
"""Wrap a function to ensure all array arguments are y, x ordered, based on kwarg.""" |
|
70
|
|
|
@functools.wraps(func) |
|
71
|
|
|
def wrapper(*args, **kwargs): |
|
72
|
|
|
# Check what order we're given |
|
73
|
|
|
dim_order = kwargs.pop('dim_order', None) |
|
74
|
|
|
x_first = _is_x_first_dim(dim_order) |
|
75
|
|
|
|
|
76
|
|
|
# If x is the first dimension, flip (transpose) every array within the function args. |
|
77
|
|
|
if x_first: |
|
78
|
|
|
args = tuple(_check_and_flip(arr) for arr in args) |
|
79
|
|
|
for k, v in kwargs: |
|
80
|
|
|
kwargs[k] = _check_and_flip(v) |
|
81
|
|
|
|
|
82
|
|
|
ret = func(*args, **kwargs) |
|
83
|
|
|
|
|
84
|
|
|
# If we flipped on the way in, need to flip on the way out so that output array(s) |
|
85
|
|
|
# match the dimension order of the original input. |
|
86
|
|
|
if x_first: |
|
87
|
|
|
return _check_and_flip(ret) |
|
88
|
|
|
else: |
|
89
|
|
|
return ret |
|
90
|
|
|
|
|
91
|
|
|
# Inject a docstring for the dim_order argument into the function's docstring. |
|
92
|
|
|
dim_order_doc = """ |
|
93
|
|
|
dim_order : str or ``None``, optional |
|
94
|
|
|
The ordering of dimensions in passed in arrays. Can be one of ``None``, ``'xy'``, |
|
95
|
|
|
or ``'yx'``. ``'xy'`` indicates that the dimension corresponding to x is the leading |
|
96
|
|
|
dimension, followed by y. ``'yx'`` indicates that x is the last dimension, preceded |
|
97
|
|
|
by y. ``None`` indicates that the default ordering should be assumed, |
|
98
|
|
|
which will change in version 0.6 from 'xy' to 'yx'. Can only be passed as a keyword |
|
99
|
|
|
argument, i.e. func(..., dim_order='xy').""" |
|
100
|
|
|
|
|
101
|
|
|
# Find the first blank line after the start of the parameters section |
|
102
|
|
|
params = wrapper.__doc__.find('Parameters') |
|
103
|
|
|
blank = wrapper.__doc__.find('\n\n', params) |
|
104
|
|
|
wrapper.__doc__ = wrapper.__doc__[:blank] + dim_order_doc + wrapper.__doc__[blank:] |
|
105
|
|
|
|
|
106
|
|
|
return wrapper |
|
107
|
|
|
|
|
108
|
|
|
|
|
109
|
|
|
@exporter.export |
|
110
|
|
|
@ensure_yx_order |
|
111
|
|
|
def v_vorticity(u, v, dx, dy): |
|
112
|
|
|
r"""Calculate the vertical vorticity of the horizontal wind. |
|
113
|
|
|
|
|
114
|
|
|
The grid must have a constant spacing in each direction. |
|
115
|
|
|
|
|
116
|
|
|
Parameters |
|
117
|
|
|
---------- |
|
118
|
|
|
u : (M, N) ndarray |
|
119
|
|
|
x component of the wind |
|
120
|
|
|
v : (M, N) ndarray |
|
121
|
|
|
y component of the wind |
|
122
|
|
|
dx : float |
|
123
|
|
|
The grid spacing in the x-direction |
|
124
|
|
|
dy : float |
|
125
|
|
|
The grid spacing in the y-direction |
|
126
|
|
|
|
|
127
|
|
|
Returns |
|
128
|
|
|
------- |
|
129
|
|
|
(M, N) ndarray |
|
130
|
|
|
vertical vorticity |
|
131
|
|
|
|
|
132
|
|
|
See Also |
|
133
|
|
|
-------- |
|
134
|
|
|
h_convergence, convergence_vorticity |
|
135
|
|
|
|
|
136
|
|
|
""" |
|
137
|
|
|
_, dudy, dvdx, _ = _get_gradients(u, v, dx, dy) |
|
138
|
|
|
return dvdx - dudy |
|
139
|
|
|
|
|
140
|
|
|
|
|
141
|
|
|
@exporter.export |
|
142
|
|
|
@ensure_yx_order |
|
143
|
|
|
def h_convergence(u, v, dx, dy): |
|
144
|
|
|
r"""Calculate the horizontal convergence of the horizontal wind. |
|
145
|
|
|
|
|
146
|
|
|
The grid must have a constant spacing in each direction. |
|
147
|
|
|
|
|
148
|
|
|
Parameters |
|
149
|
|
|
---------- |
|
150
|
|
|
u : (M, N) ndarray |
|
151
|
|
|
x component of the wind |
|
152
|
|
|
v : (M, N) ndarray |
|
153
|
|
|
y component of the wind |
|
154
|
|
|
dx : float |
|
155
|
|
|
The grid spacing in the x-direction |
|
156
|
|
|
dy : float |
|
157
|
|
|
The grid spacing in the y-direction |
|
158
|
|
|
|
|
159
|
|
|
Returns |
|
160
|
|
|
------- |
|
161
|
|
|
(M, N) ndarray |
|
162
|
|
|
The horizontal convergence |
|
163
|
|
|
|
|
164
|
|
|
See Also |
|
165
|
|
|
-------- |
|
166
|
|
|
v_vorticity, convergence_vorticity |
|
167
|
|
|
|
|
168
|
|
|
""" |
|
169
|
|
|
dudx, _, _, dvdy = _get_gradients(u, v, dx, dy) |
|
170
|
|
|
return dudx + dvdy |
|
171
|
|
|
|
|
172
|
|
|
|
|
173
|
|
|
@exporter.export |
|
174
|
|
|
@ensure_yx_order |
|
175
|
|
|
def convergence_vorticity(u, v, dx, dy): |
|
176
|
|
|
r"""Calculate the horizontal convergence and vertical vorticity of the horizontal wind. |
|
177
|
|
|
|
|
178
|
|
|
The grid must have a constant spacing in each direction. |
|
179
|
|
|
|
|
180
|
|
|
Parameters |
|
181
|
|
|
---------- |
|
182
|
|
|
u : (M, N) ndarray |
|
183
|
|
|
x component of the wind |
|
184
|
|
|
v : (M, N) ndarray |
|
185
|
|
|
y component of the wind |
|
186
|
|
|
dx : float |
|
187
|
|
|
The grid spacing in the x-direction |
|
188
|
|
|
dy : float |
|
189
|
|
|
The grid spacing in the y-direction |
|
190
|
|
|
|
|
191
|
|
|
Returns |
|
192
|
|
|
------- |
|
193
|
|
|
convergence, vorticity : tuple of (M, N) ndarrays |
|
194
|
|
|
The horizontal convergence and vertical vorticity, respectively |
|
195
|
|
|
|
|
196
|
|
|
See Also |
|
197
|
|
|
-------- |
|
198
|
|
|
v_vorticity, h_convergence |
|
199
|
|
|
|
|
200
|
|
|
Notes |
|
201
|
|
|
----- |
|
202
|
|
|
This is a convenience function that will do less work than calculating |
|
203
|
|
|
the horizontal convergence and vertical vorticity separately. |
|
204
|
|
|
|
|
205
|
|
|
""" |
|
206
|
|
|
dudx, dudy, dvdx, dvdy = _get_gradients(u, v, dx, dy) |
|
207
|
|
|
return dudx + dvdy, dvdx - dudy |
|
208
|
|
|
|
|
209
|
|
|
|
|
210
|
|
|
@exporter.export |
|
211
|
|
|
@ensure_yx_order |
|
212
|
|
|
def shearing_deformation(u, v, dx, dy): |
|
213
|
|
|
r"""Calculate the shearing deformation of the horizontal wind. |
|
214
|
|
|
|
|
215
|
|
|
The grid must have a constant spacing in each direction. |
|
216
|
|
|
|
|
217
|
|
|
Parameters |
|
218
|
|
|
---------- |
|
219
|
|
|
u : (M, N) ndarray |
|
220
|
|
|
x component of the wind |
|
221
|
|
|
v : (M, N) ndarray |
|
222
|
|
|
y component of the wind |
|
223
|
|
|
dx : float |
|
224
|
|
|
The grid spacing in the x-direction |
|
225
|
|
|
dy : float |
|
226
|
|
|
The grid spacing in the y-direction |
|
227
|
|
|
|
|
228
|
|
|
Returns |
|
229
|
|
|
------- |
|
230
|
|
|
(M, N) ndarray |
|
231
|
|
|
Shearing Deformation |
|
232
|
|
|
|
|
233
|
|
|
See Also |
|
234
|
|
|
-------- |
|
235
|
|
|
stretching_convergence, shearing_stretching_deformation |
|
236
|
|
|
|
|
237
|
|
|
""" |
|
238
|
|
|
_, dudy, dvdx, _ = _get_gradients(u, v, dx, dy) |
|
239
|
|
|
return dvdx + dudy |
|
240
|
|
|
|
|
241
|
|
|
|
|
242
|
|
|
@exporter.export |
|
243
|
|
|
@ensure_yx_order |
|
244
|
|
|
def stretching_deformation(u, v, dx, dy): |
|
245
|
|
|
r"""Calculate the stretching deformation of the horizontal wind. |
|
246
|
|
|
|
|
247
|
|
|
The grid must have a constant spacing in each direction. |
|
248
|
|
|
|
|
249
|
|
|
Parameters |
|
250
|
|
|
---------- |
|
251
|
|
|
u : (M, N) ndarray |
|
252
|
|
|
x component of the wind |
|
253
|
|
|
v : (M, N) ndarray |
|
254
|
|
|
y component of the wind |
|
255
|
|
|
dx : float |
|
256
|
|
|
The grid spacing in the x-direction |
|
257
|
|
|
dy : float |
|
258
|
|
|
The grid spacing in the y-direction |
|
259
|
|
|
|
|
260
|
|
|
Returns |
|
261
|
|
|
------- |
|
262
|
|
|
(M, N) ndarray |
|
263
|
|
|
Stretching Deformation |
|
264
|
|
|
|
|
265
|
|
|
See Also |
|
266
|
|
|
-------- |
|
267
|
|
|
shearing_deformation, shearing_stretching_deformation |
|
268
|
|
|
|
|
269
|
|
|
""" |
|
270
|
|
|
dudx, _, _, dvdy = _get_gradients(u, v, dx, dy) |
|
271
|
|
|
return dudx - dvdy |
|
272
|
|
|
|
|
273
|
|
|
|
|
274
|
|
|
@exporter.export |
|
275
|
|
|
@ensure_yx_order |
|
276
|
|
|
def shearing_stretching_deformation(u, v, dx, dy): |
|
277
|
|
|
r"""Calculate the horizontal shearing and stretching deformation of the horizontal wind. |
|
278
|
|
|
|
|
279
|
|
|
The grid must have a constant spacing in each direction. |
|
280
|
|
|
|
|
281
|
|
|
Parameters |
|
282
|
|
|
---------- |
|
283
|
|
|
u : (M, N) ndarray |
|
284
|
|
|
x component of the wind |
|
285
|
|
|
v : (M, N) ndarray |
|
286
|
|
|
y component of the wind |
|
287
|
|
|
dx : float |
|
288
|
|
|
The grid spacing in the x-direction |
|
289
|
|
|
dy : float |
|
290
|
|
|
The grid spacing in the y-direction |
|
291
|
|
|
|
|
292
|
|
|
Returns |
|
293
|
|
|
------- |
|
294
|
|
|
shearing, strectching : tuple of (M, N) ndarrays |
|
295
|
|
|
The horizontal shearing and stretching deformation, respectively |
|
296
|
|
|
|
|
297
|
|
|
See Also |
|
298
|
|
|
-------- |
|
299
|
|
|
shearing_deformation, stretching_deformation |
|
300
|
|
|
|
|
301
|
|
|
Notes |
|
302
|
|
|
----- |
|
303
|
|
|
This is a convenience function that will do less work than calculating |
|
304
|
|
|
the shearing and streching deformation terms separately. |
|
305
|
|
|
|
|
306
|
|
|
""" |
|
307
|
|
|
dudx, dudy, dvdx, dvdy = _get_gradients(u, v, dx, dy) |
|
308
|
|
|
return dvdx + dudy, dudx - dvdy |
|
309
|
|
|
|
|
310
|
|
|
|
|
311
|
|
|
@exporter.export |
|
312
|
|
|
@ensure_yx_order |
|
313
|
|
|
def total_deformation(u, v, dx, dy): |
|
314
|
|
|
r"""Calculate the horizontal total deformation of the horizontal wind. |
|
315
|
|
|
|
|
316
|
|
|
The grid must have a constant spacing in each direction. |
|
317
|
|
|
|
|
318
|
|
|
Parameters |
|
319
|
|
|
---------- |
|
320
|
|
|
u : (M, N) ndarray |
|
321
|
|
|
x component of the wind |
|
322
|
|
|
v : (M, N) ndarray |
|
323
|
|
|
y component of the wind |
|
324
|
|
|
dx : float |
|
325
|
|
|
The grid spacing in the x-direction |
|
326
|
|
|
dy : float |
|
327
|
|
|
The grid spacing in the y-direction |
|
328
|
|
|
|
|
329
|
|
|
Returns |
|
330
|
|
|
------- |
|
331
|
|
|
(M, N) ndarray |
|
332
|
|
|
Total Deformation |
|
333
|
|
|
|
|
334
|
|
|
See Also |
|
335
|
|
|
-------- |
|
336
|
|
|
shearing_deformation, stretching_deformation, shearing_stretching_deformation |
|
337
|
|
|
|
|
338
|
|
|
Notes |
|
339
|
|
|
----- |
|
340
|
|
|
This is a convenience function that will do less work than calculating |
|
341
|
|
|
the shearing and streching deformation terms separately and calculating the |
|
342
|
|
|
total deformation "by hand". |
|
343
|
|
|
|
|
344
|
|
|
""" |
|
345
|
|
|
dudx, dudy, dvdx, dvdy = _get_gradients(u, v, dx, dy) |
|
346
|
|
|
return np.sqrt((dvdx + dudy)**2 + (dudx - dvdy)**2) |
|
347
|
|
|
|
|
348
|
|
|
|
|
349
|
|
|
@exporter.export |
|
350
|
|
|
@ensure_yx_order |
|
351
|
|
|
def advection(scalar, wind, deltas): |
|
352
|
|
|
r"""Calculate the advection of a scalar field by the wind. |
|
353
|
|
|
|
|
354
|
|
|
The order of the dimensions of the arrays must match the order in which |
|
355
|
|
|
the wind components are given. For example, if the winds are given [u, v], |
|
356
|
|
|
then the scalar and wind arrays must be indexed as x,y (which puts x as the |
|
357
|
|
|
rows, not columns). |
|
358
|
|
|
|
|
359
|
|
|
Parameters |
|
360
|
|
|
---------- |
|
361
|
|
|
scalar : N-dimensional array |
|
362
|
|
|
Array (with N-dimensions) with the quantity to be advected. |
|
363
|
|
|
wind : sequence of arrays |
|
364
|
|
|
Length N sequence of N-dimensional arrays. Represents the flow, |
|
365
|
|
|
with a component of the wind in each dimension. For example, for |
|
366
|
|
|
horizontal advection, this could be a list: [u, v], where u and v |
|
367
|
|
|
are each a 2-dimensional array. |
|
368
|
|
|
deltas : sequence |
|
369
|
|
|
A (length N) sequence containing the grid spacing in each dimension. |
|
370
|
|
|
|
|
371
|
|
|
Returns |
|
372
|
|
|
------- |
|
373
|
|
|
N-dimensional array |
|
374
|
|
|
An N-dimensional array containing the advection at all grid points. |
|
375
|
|
|
|
|
376
|
|
|
""" |
|
377
|
|
|
# This allows passing in a list of wind components or an array. |
|
378
|
|
|
wind = _stack(wind) |
|
379
|
|
|
|
|
380
|
|
|
# If we have more than one component, we need to reverse the order along the first |
|
381
|
|
|
# dimension so that the wind components line up with the |
|
382
|
|
|
# order of the gradients from the ..., y, x ordered array. |
|
383
|
|
|
if wind.ndim > scalar.ndim: |
|
384
|
|
|
wind = wind[::-1] |
|
385
|
|
|
|
|
386
|
|
|
# Gradient returns a list of derivatives along each dimension. We convert |
|
387
|
|
|
# this to an array with dimension as the first index. Reverse the deltas to line up |
|
388
|
|
|
# with the order of the dimensions. |
|
389
|
|
|
grad = _stack(_gradient(scalar, *deltas[::-1])) |
|
390
|
|
|
|
|
391
|
|
|
# Make them be at least 2D (handling the 1D case) so that we can do the |
|
392
|
|
|
# multiply and sum below |
|
393
|
|
|
grad, wind = atleast_2d(grad, wind) |
|
394
|
|
|
|
|
395
|
|
|
return (-grad * wind).sum(axis=0) |
|
396
|
|
|
|
|
397
|
|
|
|
|
398
|
|
|
@exporter.export |
|
399
|
|
|
@ensure_yx_order |
|
400
|
|
|
def frontogenesis(thta, u, v, dx, dy, dim_order='yx'): |
|
401
|
|
|
r"""Calculate the 2D kinematic frontogenesis of a temperature field. |
|
402
|
|
|
|
|
403
|
|
|
The implementation is a form of the Petterssen Frontogenesis and uses the formula |
|
404
|
|
|
outlined in [Bluestein1993]_ pg.248-253. |
|
405
|
|
|
|
|
406
|
|
|
.. math:: F=\frac{1}{2}\left|\nabla \theta\right|[D cos(2\beta)-\delta] |
|
407
|
|
|
|
|
408
|
|
|
* :math:`F` is 2D kinematic frontogenesis |
|
409
|
|
|
* :math:`\theta` is potential temperature |
|
410
|
|
|
* :math:`D` is the total deformation |
|
411
|
|
|
* :math:`\beta` is the angle between the axis of dilitation and the isentropes |
|
412
|
|
|
* :math:`\delta` is the divergence |
|
413
|
|
|
|
|
414
|
|
|
Notes |
|
415
|
|
|
----- |
|
416
|
|
|
Assumes dim_order='yx', unless otherwise specified. |
|
417
|
|
|
|
|
418
|
|
|
Parameters |
|
419
|
|
|
---------- |
|
420
|
|
|
thta : (M, N) ndarray |
|
421
|
|
|
Potential temperature |
|
422
|
|
|
u : (M, N) ndarray |
|
423
|
|
|
x component of the wind |
|
424
|
|
|
v : (M, N) ndarray |
|
425
|
|
|
y component of the wind |
|
426
|
|
|
dx : float |
|
427
|
|
|
The grid spacing in the x-direction |
|
428
|
|
|
dy : float |
|
429
|
|
|
The grid spacing in the y-direction |
|
430
|
|
|
|
|
431
|
|
|
Returns |
|
432
|
|
|
------- |
|
433
|
|
|
(M, N) ndarray |
|
434
|
|
|
2D Frotogenesis in [temperature units]/m/s |
|
435
|
|
|
|
|
436
|
|
|
|
|
437
|
|
|
Conversion factor to go from [temperature units]/m/s to [tempature units/100km/3h] |
|
438
|
|
|
:math:`1.08e4*1.e5` |
|
439
|
|
|
|
|
440
|
|
|
""" |
|
441
|
|
|
# Get gradients of potential temperature in both x and y |
|
442
|
|
|
grad = _gradient(thta, dy, dx) |
|
443
|
|
|
ddy_thta, ddx_thta = grad[-2:] # Throw away unused gradient components |
|
444
|
|
|
|
|
445
|
|
|
# Compute the magnitude of the potential temperature gradient |
|
446
|
|
|
mag_thta = np.sqrt(ddx_thta**2 + ddy_thta**2) |
|
447
|
|
|
|
|
448
|
|
|
# Get the shearing, stretching, and total deformation of the wind field |
|
449
|
|
|
shrd, strd = shearing_stretching_deformation(u, v, dx, dy, dim_order=dim_order) |
|
450
|
|
|
tdef = total_deformation(u, v, dx, dy, dim_order=dim_order) |
|
451
|
|
|
|
|
452
|
|
|
# Get the divergence of the wind field |
|
453
|
|
|
div = h_convergence(u, v, dx, dy, dim_order=dim_order) |
|
454
|
|
|
|
|
455
|
|
|
# Compute the angle (beta) between the wind field and the gradient of potential temperature |
|
456
|
|
|
psi = 0.5 * np.arctan2(shrd, strd) |
|
457
|
|
|
beta = np.arcsin((-ddx_thta * np.cos(psi) - ddy_thta * np.sin(psi)) / mag_thta) |
|
458
|
|
|
|
|
459
|
|
|
return 0.5 * mag_thta * (tdef * np.cos(2 * beta) - div) |
|
460
|
|
|
|
|
461
|
|
|
|
|
462
|
|
|
@exporter.export |
|
463
|
|
|
@ensure_yx_order |
|
464
|
|
|
def geostrophic_wind(heights, f, dx, dy): |
|
465
|
|
|
r"""Calculate the geostrophic wind given from the heights or geopotential. |
|
466
|
|
|
|
|
467
|
|
|
Parameters |
|
468
|
|
|
---------- |
|
469
|
|
|
heights : (M, N) ndarray |
|
470
|
|
|
The height field, with either leading dimensions of (x, y) or trailing dimensions |
|
471
|
|
|
of (y, x), depending on the value of ``dim_order``. |
|
472
|
|
|
f : array_like |
|
473
|
|
|
The coriolis parameter. This can be a scalar to be applied |
|
474
|
|
|
everywhere or an array of values. |
|
475
|
|
|
dx : scalar |
|
476
|
|
|
The grid spacing in the x-direction |
|
477
|
|
|
dy : scalar |
|
478
|
|
|
The grid spacing in the y-direction |
|
479
|
|
|
|
|
480
|
|
|
Returns |
|
481
|
|
|
------- |
|
482
|
|
|
A 2-item tuple of arrays |
|
483
|
|
|
A tuple of the u-component and v-component of the geostrophic wind. |
|
484
|
|
|
|
|
485
|
|
|
""" |
|
486
|
|
|
if heights.dimensionality['[length]'] == 2.0: |
|
487
|
|
|
norm_factor = 1. / f |
|
488
|
|
|
else: |
|
489
|
|
|
norm_factor = g / f |
|
490
|
|
|
|
|
491
|
|
|
# If heights has more than 2 dimensions, we need to pass in some dummy |
|
492
|
|
|
# grid deltas so that we can still use np.gradient. It may be better to |
|
493
|
|
|
# to loop in this case, but that remains to be done. |
|
494
|
|
|
deltas = [dy, dx] |
|
495
|
|
|
if heights.ndim > 2: |
|
496
|
|
|
deltas = [units.Quantity(1., units.m)] * (heights.ndim - 2) + deltas |
|
497
|
|
|
|
|
498
|
|
|
grad = _gradient(heights, *deltas) |
|
499
|
|
|
dy, dx = grad[-2:] # Throw away unused gradient components |
|
500
|
|
|
return -norm_factor * dy, norm_factor * dx |
|
501
|
|
|
|
|
502
|
|
|
|
|
503
|
|
|
@exporter.export |
|
504
|
|
|
@check_units('[length]', '[temperature]') |
|
505
|
|
|
def montgomery_streamfunction(height, temperature): |
|
506
|
|
|
r"""Compute the Montgomery Streamfunction on isentropic surfaces. |
|
507
|
|
|
|
|
508
|
|
|
The Montgomery Streamfunction is the streamfunction of the geostrophic wind on an |
|
509
|
|
|
isentropic surface. This quantity is proportional to the geostrophic wind in isentropic |
|
510
|
|
|
coordinates, and its gradient can be interpreted similarly to the pressure gradient in |
|
511
|
|
|
isobaric coordinates. |
|
512
|
|
|
|
|
513
|
|
|
Parameters |
|
514
|
|
|
---------- |
|
515
|
|
|
height : `pint.Quantity` |
|
516
|
|
|
Array of geopotential height of isentropic surfaces |
|
517
|
|
|
temperature : `pint.Quantity` |
|
518
|
|
|
Array of temperature on isentropic surfaces |
|
519
|
|
|
|
|
520
|
|
|
Returns |
|
521
|
|
|
------- |
|
522
|
|
|
stream_func : `pint.Quantity` |
|
523
|
|
|
|
|
524
|
|
|
Notes |
|
525
|
|
|
----- |
|
526
|
|
|
The formula used is that from [Lackmann2011]_ p. 69 for T in Kelvin and height in meters: |
|
527
|
|
|
.. math:: sf = gZ + C_pT |
|
528
|
|
|
* :math:`sf` is Montgomery Streamfunction |
|
529
|
|
|
* :math:`g` is avg. gravitational acceleration on Earth |
|
530
|
|
|
* :math:`Z` is geopotential height of the isentropic surface |
|
531
|
|
|
* :math:`C_p` is specific heat at constant pressure for dry air |
|
532
|
|
|
* :math:`T` is temperature of the isentropic surface |
|
533
|
|
|
|
|
534
|
|
|
See Also |
|
535
|
|
|
-------- |
|
536
|
|
|
get_isentropic_pressure |
|
537
|
|
|
|
|
538
|
|
|
""" |
|
539
|
|
|
return (g * height) + (Cp_d * temperature) |
|
540
|
|
|
|
|
541
|
|
|
|
|
542
|
|
|
@exporter.export |
|
543
|
|
|
@check_units('[speed]', '[speed]', '[pressure]', '[length]', |
|
544
|
|
|
'[length]', '[length]', '[speed]', '[speed]') |
|
545
|
|
|
def storm_relative_helicity(u, v, p, hgt, top, bottom=0 * units('meter'), |
|
546
|
|
|
storm_u=0 * units('m/s'), storm_v=0 * units('m/s')): |
|
547
|
|
|
# Partially adapted from similar SharpPy code |
|
548
|
|
|
r"""Calculate storm relative helicity. |
|
549
|
|
|
|
|
550
|
|
|
Needs u and v wind components, heights and pressures, |
|
551
|
|
|
and top and bottom of SRH layer. An optional storm |
|
552
|
|
|
motion vector can be specified. SRH is calculated using the |
|
553
|
|
|
equation specified on p. 230-231 in the Markowski and Richardson |
|
554
|
|
|
meso textbook [Markowski2010]. |
|
555
|
|
|
|
|
556
|
|
|
.. math:: \int\limits_0^d (\bar v - c) \cdot \bar\omega_{h} \,dz |
|
557
|
|
|
|
|
558
|
|
|
This is applied to the data from a hodograph with the following summation: |
|
559
|
|
|
|
|
560
|
|
|
.. math:: \sum_{n = 1}^{N-1} [(u_{n+1} - c_{x})(v_{n} - c_{y}) - |
|
561
|
|
|
(u_{n} - c_{x})(v_{n+1} - c_{y})] |
|
562
|
|
|
|
|
563
|
|
|
Parameters |
|
564
|
|
|
---------- |
|
565
|
|
|
u : array-like |
|
566
|
|
|
The u components of winds, same length as hgts |
|
567
|
|
|
v : array-like |
|
568
|
|
|
The u components of winds, same length as hgts |
|
569
|
|
|
p : array-like |
|
570
|
|
|
Pressure in hPa, same length as hgts |
|
571
|
|
|
hgt : array-like |
|
572
|
|
|
The heights associatd with the data, provided in meters above mean |
|
573
|
|
|
sea level and converted into meters AGL. |
|
574
|
|
|
top : number |
|
575
|
|
|
The height of the top of the desired layer for SRH. |
|
576
|
|
|
bottom : number |
|
577
|
|
|
The height at the bottom of the SRH layer. Default is sfc (None). |
|
578
|
|
|
storm_u : number |
|
579
|
|
|
u component of storm motion |
|
580
|
|
|
storm_v : number |
|
581
|
|
|
v component of storm motion |
|
582
|
|
|
|
|
583
|
|
|
Returns |
|
584
|
|
|
------- |
|
585
|
|
|
number |
|
586
|
|
|
p_srh : positive storm-relative helicity |
|
587
|
|
|
number |
|
588
|
|
|
n_srh : negative storm-relative helicity |
|
589
|
|
|
number |
|
590
|
|
|
t_srh : total storm-relative helicity |
|
591
|
|
|
|
|
592
|
|
|
""" |
|
593
|
|
|
# Converting to m/s to make sure output is in m^2/s^2 |
|
594
|
|
|
u = u.to('meters/second') |
|
595
|
|
|
v = v.to('meters/second') |
|
596
|
|
|
storm_u = storm_u.to('meters/second') |
|
597
|
|
|
storm_v = storm_v.to('meters/second') |
|
598
|
|
|
|
|
599
|
|
|
w_int = get_layer(p, u, v, heights=hgt, bottom=bottom, depth=top - bottom) |
|
600
|
|
|
|
|
601
|
|
|
sru = w_int[1] - storm_u |
|
602
|
|
|
srv = w_int[2] - storm_v |
|
603
|
|
|
|
|
604
|
|
|
int_layers = sru[1:] * srv[:-1] - sru[:-1] * srv[1:] |
|
605
|
|
|
|
|
606
|
|
|
p_srh = int_layers[int_layers.magnitude > 0.].sum() |
|
607
|
|
|
n_srh = int_layers[int_layers.magnitude < 0.].sum() |
|
608
|
|
|
t_srh = p_srh + n_srh |
|
609
|
|
|
|
|
610
|
|
|
return p_srh, n_srh, t_srh |
|
611
|
|
|
|
|
612
|
|
|
|
|
613
|
|
|
@exporter.export |
|
614
|
|
|
def calc_dx_dy(longitude, latitude, shape='sphere', radius=6370997.): |
|
615
|
|
|
r"""Calculate the distance between grid points that are in a latitude/longitude format. |
|
616
|
|
|
|
|
617
|
|
|
Calculate the distance between grid points when the grid spacing is defined by |
|
618
|
|
|
delta lat/lon rather than delta x/y |
|
619
|
|
|
|
|
620
|
|
|
Parameters |
|
621
|
|
|
---------- |
|
622
|
|
|
longitude : array_like |
|
623
|
|
|
array of longitudes defining the grid |
|
624
|
|
|
latitude : array_like |
|
625
|
|
|
array of latitudes defining the grid |
|
626
|
|
|
shape : optional |
|
627
|
|
|
the shape of the model globe |
|
628
|
|
|
common shapes: 'sphere', 'WGS84', 'GRS80' |
|
629
|
|
|
radius : float, optional |
|
630
|
|
|
the radius of the earth |
|
631
|
|
|
|
|
632
|
|
|
Returns |
|
633
|
|
|
------- |
|
634
|
|
|
dx, dy: 2D arrays of distances between grid points |
|
635
|
|
|
in the x and y direction with units of meters |
|
636
|
|
|
|
|
637
|
|
|
Notes |
|
638
|
|
|
----- |
|
639
|
|
|
Accepts, 1D or 2D arrays for latitude and longitude |
|
640
|
|
|
Assumes [Y, X] for 2D arrays |
|
641
|
|
|
|
|
642
|
|
|
""" |
|
643
|
|
|
if radius != 6370997.: |
|
644
|
|
|
g = Geod(a=radius, b=radius) |
|
645
|
|
|
else: |
|
646
|
|
|
g = Geod(ellps=shape) |
|
647
|
|
|
|
|
648
|
|
|
if latitude.ndim == 1: |
|
649
|
|
|
longitude, latitude = np.meshgrid(longitude, latitude) |
|
650
|
|
|
|
|
651
|
|
|
dy = np.zeros(latitude.shape) |
|
652
|
|
|
dx = np.zeros(longitude.shape) |
|
653
|
|
|
|
|
654
|
|
|
for i in range(longitude.shape[1]): |
|
655
|
|
|
for j in range(latitude.shape[0] - 1): |
|
656
|
|
|
_, _, dy[j, i] = g.inv(longitude[j, i], latitude[j, i], |
|
657
|
|
|
longitude[j + 1, i], latitude[j + 1, i]) |
|
658
|
|
|
dy[j + 1, :] = dy[j, :] |
|
659
|
|
|
|
|
660
|
|
|
for i in range(longitude.shape[1] - 1): |
|
661
|
|
|
for j in range(latitude.shape[0]): |
|
662
|
|
|
_, _, dx[j, i] = g.inv(longitude[j, i], latitude[j, i], |
|
663
|
|
|
longitude[j, i + 1], latitude[j, i + 1]) |
|
664
|
|
|
dx[:, i + 1] = dx[:, i] |
|
665
|
|
|
|
|
666
|
|
|
xdiff_sign = np.sign(longitude[0, 1] - longitude[0, 0]) |
|
667
|
|
|
ydiff_sign = np.sign(latitude[1, 0] - latitude[0, 0]) |
|
668
|
|
|
return xdiff_sign * dx * units.meter, ydiff_sign * dy * units.meter |
|
669
|
|
|
|
Unidata /
MetPy
