1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 | #cython: linetrace=True
#distutils: define_macros=CYTHON_TRACE_NOGIL=1
#cython: language_level=3
"""
AMSIMP Backend Class. For information about this class is described below.
Copyright (C) 2020 AMSIMP
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <https://www.gnu.org/licenses/>.
"""
# -----------------------------------------------------------------------------------------#
# Importing Dependencies
import os, sys
from datetime import datetime
import socket
from amsimp.download cimport Download
from amsimp.download import Download
import numpy as np
from astropy import constants as constant
from astropy import units
from astropy.units.quantity import Quantity
cimport numpy as np
from cpython cimport bool
import matplotlib.pyplot as plt
from matplotlib import ticker
import cartopy.crs as ccrs
from cartopy.util import add_cyclic_point
import iris
from iris.coords import DimCoord
from iris.cube import Cube
from iris.coord_systems import GeogCS
# -----------------------------------------------------------------------------------------#
cdef class Backend(Download):
"""
AMSIMP Backend Class - This is the base class for AMSIMP, as such, all
other classes within AMSIMP inherit this class, either directly or
indirectly.
For methods to be included in this class they must meet one of the following
criteria:
(1) they are considered essential components of the software.
(2) the output of these methods are generated solely from the methods
classified as (1).
(3) these methods import data from the Global Data Assimilation System.
This data is retrieved from the AMSIMP data repository, which is updated
every hour.
(4) they don't classify nicely into any other class.
(5) they offer a visualisation of any of the methods found in this class.
Below is a list of the methods included within this class, with a short
description of their intended purpose and a bracketed number signifying
which of the above criteria they meet. Please see the relevant class methods
for more information. Please note that the unit information with AMSIMP
is provided by the unit module within astropy.
latitude_lines ~ generates a NumPy array of latitude lines (1). The unit
of measurement is degrees.
longitude_lines ~ generates a NumPy array of longitude lines (1). The unit
of measurement is degrees.
pressure_surfaces ~ generates a NumPy array of constant pressure
surfaces (1). The unit of measurement is hectopascals. This is the
isobaric coordinate system.
coriolis_parameter ~ generates a NumPy arrray of the Coriolis parameter
at various latitudes of the Earth's surface (2). The unit of measurement
is radians per second.
geopotential_height ~ outputs a NumPy array of geopotential height (3). The
unit of measurement is metres.
relative_humidity ~ outputs a NumPy array of relative humidity (3). The unit
of measurement is percentage.
temperature ~ outputs a NumPy array of temperature (3). The unit of
measurement is Kelvin.
remove_all_files ~ this method deletes any file created by the software,
whilst downloading data from the AMSIMP Initial Atmospheric Conditions
Data Repository (3). It does not, however, delete any forecast generated by,
the, simulate, method. Please note that you may specifcy whether you would like
to save such files throught the utilisation of the parameter, remove_files,
when the class is initialised.
pressure_thickness ~ outputs a NumPy array of atmospheric pressure
thickness (4). The unit of measurement is metres.
troposphere_boundaryline ~ generates a NumPy array of the
troposphere - stratosphere boundary line (4). The unit of measurement is
metres.
longitude_contourf ~ generates a contour plot for a desired atmospheric
process, with the axes being latitude, and longitude (5).
altitude_contourf ~ generates a contour plot for a desired atmospheric
process, with the axes being latitude, and altitude (5).
thickness_contourf ~ generates a contour plot for pressure thickness,
with the axes being latitude and longitude. This plot is then layed
on top of a EckertIII global projection (5).
"""
# Define units of measurement for AMSIMP.
units = units
def __cinit__(
self,
int delta_latitude=5,
int delta_longitude=5,
forecast_length=72,
delta_t=2,
bool ai=True,
data_size=150,
epochs=200,
input_date=None,
bool input_data=False,
psurfaces=None,
lat=None,
lon=None,
height=None,
temp=None,
rh=None,
u=None,
v=None,
dict constants={
"sidereal_day": (23 + (56 / 60)) * 3600,
"angular_rotation_rate": ((2 * np.pi) / ((23 + (56 / 60)) * 3600)),
"planet_radius": constant.R_earth.value,
"planet_mass": constant.M_earth.value,
"specific_heat_capacity_psurface": 718,
"gravitational_acceleration": 9.80665,
"planet": "Earth"
}
):
"""
The parameters, delta_latitude and delta_longitude, defines the
horizontal resolution between grid points within the software. The
parameter values must be between 1 and 10 degrees if you are utilising
the default initial conditions included within the software. Defaults
to a value of 5 degrees.
The parameter, input_date, allows the end-user to specify the initial
date at which the forecast is generated from. Please note that
currently you cannot specify dates before Janurary 2020. Please also
note that the latest data available to the software is typically
from three days ago. This is due to the rate at which the NOAA
updates the data on the Global Data Assimilation System website. This
feature is intended for situations where the default initial conditions
included within the software are being used.
The parameter, input_data, is a boolean value, and when set to True,
it will allow the end-user to provide their own initial conditions
to the software. Please note that this feature has not been fully
tested as of this moment, and such is experimental for the time
being.
The parameter, ai, is a boolean value, and when set to True, a
recurrent neural network will be trained on the amount of days
specified by the parameter, data_size. A prediction by the
recurrent neural network will be generated, and will be
combined with a traditional physical model of the atmosphere.
It can also be used independently.
"""
# Make the aforementioned variables available else where in the class.
self.delta_latitude = delta_latitude
self.delta_longitude = delta_longitude
self.input_date = input_date
self.input_data = input_data
# The date at which the initial conditions was gathered (i.e. how
# recent the data is).
if self.input_date == None:
data_date_file = self.download(
"https://github.com/amsimp/initial-conditions/raw/master/date.npy",
bar=False,
)
data_date = np.load(data_date_file)
os.remove(data_date_file)
date = datetime(
int(data_date[2]),
int(data_date[1]),
int(data_date[0]),
int(data_date[3]),
)
self.date = date
else:
# Ensure input_date is of the type "datetime.datetime".
if type(self.input_date) != datetime:
raise Exception(
"input_date must of the type 'datetime.datetime'. The value of input_date was: {}".format(
self.input_date
)
)
date = self.input_date
self.date = date
# Ensure input_data is a boolean value.
if not isinstance(self.input_data, bool):
raise Exception(
"input_data must be a boolean value. The value of input_data was: {}".format(
self.input_data
)
)
# Function to ensure that the input data is 3 dimensional.
def dimension(input_variable):
if np.ndim(input_variable) != 3:
raise Exception(
"All input data variables (height, rh, temp, u, v) must be a 3 dimensional."
)
# Function to check if a function is a NumPy array, or a list.
def type_check(input_variable):
if type(input_variable) == Quantity:
pass
elif type(input_variable) == np.ndarray:
pass
elif type(input_variable) == list:
pass
else:
raise Exception(
"All input data variables must be either NumPy arrays, or lists."
)
# Function to check if grid input variables are 1 dimensional.
def dimension_grid(input_variable):
if np.ndim(input_variable) != 1:
raise Exception(
"All grid input variables (psurfaces, lat, lon) must be a 1 dimensional."
)
# Function to convert input lists into NumPy arrays.
def list_to_numpy(input_variable):
if type(input_variable) == list:
input_variable = np.asarray(input_variable)
return input_variable
if self.input_data:
# Ensure input data variables are either NumPy arrays, or lists.
type_check(psurfaces)
type_check(lat)
type_check(lon)
type_check(height)
type_check(rh)
type_check(temp)
type_check(u)
type_check(v)
# Check if grid input variables are 1 dimensional.
dimension_grid(psurfaces)
dimension_grid(lat)
dimension_grid(lon)
# Check if input data is 3 dimensional.
dimension(height)
dimension(rh)
dimension(temp)
dimension(u)
dimension(v)
# Convert input data lists to NumPy arrays.
list_to_numpy(psurfaces)
list_to_numpy(lat)
list_to_numpy(lon)
list_to_numpy(height)
list_to_numpy(rh)
list_to_numpy(temp)
list_to_numpy(u)
list_to_numpy(v)
# Add units to input variables.
# psurfaces variable.
if type(psurfaces) != Quantity:
psurfaces = psurfaces * units.hPa
# lat variable.
if type(lat) != Quantity:
lat = lat * units.deg
# lon variable.
if type(height) != Quantity:
lon = lon * units.deg
# height variable.
if type(height) != Quantity:
height = height * units.m
# rh variable.
if type(rh) != Quantity:
rh = rh * units.percent
# temp variable.
if type(temp) != Quantity:
temp = temp * units.K
# u variable.
if type(u) != Quantity:
u = u * (units.m / units.s)
# v variable.
if type(v) != Quantity:
v = v * (units.m / units.s)
# Ensure self.delta_latitude is between 1 and 10 degrees.
if not input_data:
if self.delta_latitude > 10 or self.delta_latitude < 1:
raise Exception(
"delta_latitude must be a positive integer between 1 and 10. The value of delta_latitude was: {}".format(
self.delta_latitude
)
)
# Ensure self.delta_longitude is between 1 and 10 degrees.
if self.delta_longitude > 10 or self.delta_longitude < 1:
raise Exception(
"delta_longitude must be a positive integer between 1 and 10. The value of delta_longitude was: {}".format(
self.delta_longitude
)
)
folder = "https://github.com/amsimp/initial-conditions/raw/master/initial_conditions/"
# Define date.
year = self.date.year
month = self.date.month
day = self.date.day
hour = self.date.hour
# Adds zero before single digit numbers.
if day < 10:
day = "0" + str(day)
if month < 10:
month = "0" + str(month)
if hour < 10:
hour = "0" + str(hour)
# Converts integers to strings.
day = str(day)
month = str(month)
year = str(year)
hour = str(hour)
folder += (
year + "/" + month + "/" + day + "/" + hour + "/"
)
# The url to the NumPy pressure surfaces file stored on the AMSIMP
# Initial Conditions Data repository.
url = folder + "initial_conditions.nc"
# Download the NumPy file and store the NumPy array into a variable.
try:
cube = iris.load("initial_conditions.nc")
except OSError:
fname = self.download(url)
cube = iris.load(fname)
height = cube.extract("geopotential_height")[0]
temp = cube.extract("air_temperature")[0]
rh = cube.extract("relative_humidity")[0]
u = cube.extract("x_wind")[0]
v = cube.extract("y_wind")[0]
# Constants dictionary.
# Sidereal day.
if type(constants["sidereal_day"]) != Quantity:
constants["sidereal_day"] = constants["sidereal_day"] * units.s
# Angular rotation rate.
if type(constants["angular_rotation_rate"]) != Quantity:
constants["angular_rotation_rate"] = constants["angular_rotation_rate"] * (
units.rad / units.s
)
# Planet radius.
if type(constants["planet_radius"]) != Quantity:
constants["planet_radius"] = constants["planet_radius"] * units.m
# Planet mass.
if type(constants["planet_mass"]) != Quantity:
constants["planet_mass"] = constants["planet_mass"] * units.kg
# Specific heat capacity on a constant pressure surface.
if type(constants["specific_heat_capacity_psurface"]) != Quantity:
constants["specific_heat_capacity_psurface"] = constants[
"specific_heat_capacity_psurface"
] * (units.J / (units.kg * units.K))
# Gravitational acceleration.
if type(constants["gravitational_acceleration"]) != Quantity:
constants["gravitational_acceleration"] = constants["gravitational_acceleration"] * (
units.m / (units.s ** 2)
)
# Planet
planet = constants["planet"].lower()
planet = planet.replace(" ", "")
planet = planet.capitalize()
# Gas constant
R = 287.05799596 * (units.J / (units.kg * units.K))
constants["gas_constant"] = R
# Universal gravitational constant.
G = constant.G
G = G.value * G.unit
constants["universal_gravitational_constant"] = G
# Make the input data variables available else where in the class.
self.psurfaces = psurfaces
self.lat = lat
self.lon = lon
self.input_height = height
self.input_rh = rh
self.input_temp = temp
self.input_u = u
self.input_v = v
self.constants = constants
# Predefined Constants.
# Angular rotation rate of the planet.
self.sidereal_day = constants["sidereal_day"]
self.Upomega = constants["angular_rotation_rate"]
# Mean radius of the planet.
self.a = self.constants["planet_radius"]
# Mass of the planet.
self.M = self.constants["planet_mass"]
# The specific heat capacity on a constant pressure surface.
self.c_p = self.constants["specific_heat_capacity_psurface"]
# Gravitational acceleration.
self.g = self.constants["gravitational_acceleration"]
# Planet
self.planet = self.constants["planet"]
# Gas Constant.
self.R = constants["gas_constant"]
# Universal gravitational constant.
self.G = constants["universal_gravitational_constant"]
# Function to check for an internet connection.
def is_connected():
try:
host = socket.gethostbyname("www.github.com")
s = socket.create_connection((host, 80), 2)
s.close()
return True
except OSError:
pass
return False
# Check for an internet connection.
if not input_data:
if not is_connected():
raise Exception(
"You must connect to the internet in order to utilise AMSIMP."
+ " Apologies for any inconvenience caused."
)
# RNN.
self.epochs = epochs
self.data_size = data_size
# Ensure epochs is an integer value.
if not isinstance(self.epochs, int):
raise Exception(
"epochs must be a integer value. The value of epochs was: {}".format(
self.ai
)
)
# Ensure epochs is a natural number.
if not self.epochs > 0:
raise Exception(
"epochs must be a integer value. The value of epochs was: {}".format(
self.ai
)
)
# Ensure data_size is an integer value.
if not isinstance(self.data_size, int):
raise Exception(
"data_size must be a integer value. The value of data_size was: {}".format(
self.ai
)
)
# Ensure data_size is a natural number and is greater than 14.
if not self.data_size > 14:
raise Exception(
"data_size must be a integer value. The value of data_size was: {}".format(
self.ai
)
)
# -----------------------------------------------------------------------------------------#
cpdef np.ndarray latitude_lines(self):
"""
Generates a NumPy array of latitude lines.
"""
cdef float i
if not self.input_data:
latitude_lines = [
i
for i in np.arange(
-89, 90, self.delta_latitude
)
]
# Convert list to NumPy array and add the unit of measurement.
latitude_lines = np.asarray(latitude_lines) * units.deg
else:
latitude_lines = self.lat
return latitude_lines
cpdef np.ndarray longitude_lines(self):
"""
Generates a NumPy array of longitude lines.
"""
cdef float i
if not self.input_data:
longitude_lines = [
i
for i in np.arange(
0, 360, self.delta_longitude
)
]
# Convert list to NumPy array and add the unit of measurement.
longitude_lines = np.asarray(longitude_lines) * units.deg
else:
longitude_lines = self.lon
return longitude_lines
cpdef np.ndarray pressure_surfaces(self, dim_3d=False):
"""
Generates a NumPy array of the constant pressure surfaces. This
is the isobaric coordinate system.
"""
if not self.input_data:
pressure_surfaces = self.input_rh.coords('pressure')[0].points
pressure_surfaces = pressure_surfaces[::-1] * self.units.Pa
pressure_surfaces = pressure_surfaces.to(self.units.hPa)
pressure_surfaces = pressure_surfaces.value
else:
pressure_surfaces = self.psurfaces.value
# Convert Pressure Array into 3D Array.
if dim_3d:
list_pressure = []
for p in pressure_surfaces:
p = np.zeros((
len(self.latitude_lines()), len(self.longitude_lines())
)) + p
p = p.tolist()
list_pressure.append(p)
pressure_surfaces = np.asarray(list_pressure)
pressure_surfaces *= self.units.hPa
return pressure_surfaces
# -----------------------------------------------------------------------------------------#
cpdef np.ndarray gradient_longitude(self, parameter=None):
"""
Explain here.
"""
# Determine gradient with respect to longitude.
cdef np.ndarray longitude = np.radians(self.longitude_lines().value)
# Define variable types.
cdef np.ndarray parameter_central, parameter_boundaries
cdef np.ndarray parameter_lgmt, parameter_rgmt
cdef np.ndarray lon_lgmt, lon_rgmt, lon_boundaries
# Compute gradient.
# Central points.
parameter_central = np.gradient(
parameter, longitude, axis=2
)
# Boundaries.
# Define atmospheric parameter boundaries.
parameter_lgmt = parameter[:, :, -3:].value
parameter_rgmt = parameter[:, :, :3].value
# Define longitude boundaries.
lon_lgmt = longitude[-3:]
lon_rgmt = longitude[:3]
# Concatenate atmospheric parameter boundaries.
parameter_boundaries = np.concatenate(
(parameter_lgmt, parameter_rgmt), axis=2
) * parameter.unit
# Concatenate longitude boundaries.
lon_boundaries = np.concatenate((lon_lgmt, lon_rgmt))
# Compute gradient at boundaries.
parameter_boundaries = np.gradient(
parameter_boundaries, lon_boundaries, axis=2
)
# Redefine parameter as computed gradient.
parameter = parameter_central
parameter[:, :, -1] = parameter_boundaries[:, :, 2]
parameter[:, :, 0] = parameter_boundaries[:, :, 3]
# Make 1d latitude array into a 3d latitude array.
cdef np.ndarray latitude = np.radians(self.latitude_lines())
latitude = self.make_3dimensional_array(
parameter=latitude, axis=1
)
# Handle longitudinal variation with respect to latitude.
gradient_longitude = (
(1 / (self.a * np.cos(latitude.value))) * parameter
)
return gradient_longitude
cpdef np.ndarray gradient_latitude(self, parameter=None):
"""
Explain here.
"""
# Define variable.
cdef np.ndarray latitude = np.radians(self.latitude_lines().value)
# Determine gradient with respect to latitude.
gradient_latitude = np.gradient(
parameter.value, self.a.value * latitude, axis=1
) * (parameter.unit / units.m)
return gradient_latitude
cpdef np.ndarray gradient_pressure(self, parameter=None):
"""
Explain here.
"""
# Define variable.
cdef np.ndarray pressure = self.pressure_surfaces()
# Determine gradient with respect to pressure.
gradient_pressure = np.gradient(
parameter.value, pressure.value, axis=0
) * (parameter.unit / units.hPa)
return gradient_pressure
cpdef np.ndarray make_3dimensional_array(self, parameter=None, axis=1):
"""
Explain here.
"""
if axis == 0:
a = self.latitude_lines().value
b = self.longitude_lines().value
elif axis == 1:
a = self.pressure_surfaces().value
b = self.longitude_lines().value
elif axis == 2:
a = self.pressure_surfaces().value
b = self.latitude_lines().value
list_parameter = []
for param in parameter:
unit = param.unit
param = param.value + np.zeros((len(a), len(b)))
param = param.tolist()
list_parameter.append(param)
parameter = np.asarray(list_parameter) * unit
if axis == 1:
parameter = np.transpose(parameter, (1, 0, 2))
elif axis == 2:
parameter = np.transpose(parameter, (1, 2, 0))
return parameter
cpdef dict retrieve_constants(self):
"""
Explain here.
"""
return self.constants
# -----------------------------------------------------------------------------------------
cpdef np.ndarray coriolis_parameter(self):
"""
Generates a NumPy arrray of the Coriolis parameter at various latitudes
of the Earth's surface.
The Coriolis parameter is defined as two times the angular rotation of
the Earth by the sin of the latitude you are interested in.
Equation:
f_0 = 2 \* Upomega * sin(\phi)
"""
coriolis_parameter = (
2 * self.Upomega * np.sin(np.radians(self.latitude_lines()))
)
return coriolis_parameter
# -----------------------------------------------------------------------------------------#
cpdef np.ndarray geopotential_height(self):
"""
This method imports geopotential height data from a NetCDF file, which is
located in the AMSIMP Initial Conditions Data Repository on GitHub.
The data stored within this repo is updated every six hours by amsimp-bot.
Following which, it outputs a NumPy array in the shape of
(len(pressure_surfaces), len(latitude_lines), len(longitude_lines)).
Geopotential Height is the height above sea level of a pressure level.
For example, if a station reports that the 500 hPa height at its
location is 5600 m, it means that the level of the atmosphere over
that station at which the atmospheric pressure is 500 hPa is 5600
meters above sea level.
I strongly recommend storing the output into a variable in order to
prevent the need to repeatly self.download the file. For more information,
visit https://github.com/amsimp/initial-conditions.
"""
if not self.input_data:
# Input data.
height = self.input_height
pressure = self.pressure_surfaces().to(self.units.Pa)
# Grid.
grid_points = [
('pressure', pressure.value),
('latitude', self.latitude_lines().value),
('longitude', self.longitude_lines().value),
]
# Interpolation
height = height.interpolate(
grid_points, iris.analysis.Linear()
)
# Get data.
geopotential_height = height.data
geopotential_height = np.asarray(geopotential_height.tolist())
geopotential_height *= self.units.m
else:
geopotential_height = self.input_height
return geopotential_height
cpdef np.ndarray relative_humidity(self):
"""
This method imports relative humidity data from a NetCDF file, which is
located in the AMSIMP Initial Conditions Data Repository on GitHub.
The data stored within this repo is updated every six hours by amsimp-bot.
Following which, it outputs a NumPy array in the shape of
(len(pressure_surfaces), len(latitude_lines), len(longitude_lines)).
Relative Humidity is the amount of water vapour present in air expressed
as a percentage of the amount needed for saturation at the same temperature.
I strongly recommend storing the output into a variable in order to
prevent the need to repeatly self.download the file. For more information,
visit https://github.com/amsimp/initial-conditions.
"""
if not self.input_data:
# Input data.
rh = self.input_rh
pressure = self.pressure_surfaces().to(self.units.Pa)
# Grid.
grid_points = [
('pressure', pressure.value),
('latitude', self.latitude_lines().value),
('longitude', self.longitude_lines().value),
]
# Interpolation
rh = rh.interpolate(
grid_points, iris.analysis.Linear()
)
# Get data.
relative_humidity = rh.data
relative_humidity = np.asarray(relative_humidity.tolist())
relative_humidity *= self.units.percent
else:
relative_humidity = self.input_rh
return relative_humidity
cpdef np.ndarray temperature(self):
"""
This method imports temperature data from a NetCDF file, which is
located in the AMSIMP Initial Conditions Data Repository on GitHub.
The data stored within this repo is updated every six hours by amsimp-bot.
Following which, it outputs a NumPy array in the shape of
(len(pressure_surfaces), len(latitude_lines), len(longitude_lines)).
Temperature is defined as the mean kinetic energy density of molecular
motion.
I strongly recommend storing the output into a variable in order to
prevent the need to repeatly self.download the file. For more information,
visit https://github.com/amsimp/initial-conditions.
"""
if not self.input_data:
# Input data.
temp = self.input_temp
pressure = self.pressure_surfaces().to(self.units.Pa)
# Grid.
grid_points = [
('pressure', pressure.value),
('latitude', self.latitude_lines().value),
('longitude', self.longitude_lines().value),
]
# Interpolation
temp = temp.interpolate(
grid_points, iris.analysis.Linear()
)
# Get data.
temperature = temp.data
temperature = np.asarray(temperature.tolist())
temperature *= self.units.K
else:
temperature = self.input_temp
return temperature
cpdef exit(self):
"""
This method deletes any file created by the software,
whilst downloading data from the AMSIMP Initial Atmospheric
Conditions Data Repository. It does not, however, delete any
forecast generated and saved by the simualte method.
This command will also cause the Python interpreter to close
and exit.
"""
# Initial atmospheric conditions file.
try:
os.remove("initial_conditions.nc")
except OSError:
pass
# Close Python.
sys.exit()
# -----------------------------------------------------------------------------------------#
cpdef np.ndarray pressure_thickness(self, p1=1000, p2=500):
"""
Generates a NumPy array of pressure thickness between two defined
constant pressure surfaces.
Pressure thickness is defined as the distance between two
pressure surfaces. Pressure thickness is determined through the
hypsometric equation.
Equation:
h = z_2 - z_1
"""
# Ensure p1 is greater than p2.
if p1 < p2:
raise Exception("Please note that p1 must be greater than p2.")
# Find the nearest constant pressure surface to p1 and p2 in pressure.
cdef int indx_p1 = (np.abs(self.pressure_surfaces().value - p1)).argmin()
cdef int indx_p2 = (np.abs(self.pressure_surfaces().value - p2)).argmin()
pressure_thickness = (
self.geopotential_height()[indx_p2] - self.geopotential_height()[indx_p1]
)
return pressure_thickness
cpdef np.ndarray troposphere_boundaryline(self):
"""
Generates a NumPy array of the troposphere - stratosphere
boundary line in the shape (len(longitude_lines), len(latitude_lines).
This is calculated by looking at the vertical temperature profile in
the method, amsimp.Backend.temperature().
Explain here.
"""
cdef np.ndarray temperature = self.temperature().value
grad_temperature = np.gradient(temperature)
cdef np.ndarray gradient_temperature = grad_temperature[0]
gradient_temperature = np.transpose(gradient_temperature, (2, 1, 0))
cdef np.ndarray psurface = self.pressure_surfaces().value
cdef list trop_strat_line = []
cdef list lat_trop_strat_line
cdef np.ndarray temp, t
cdef int n
cdef float t_n, p
for temp in gradient_temperature:
lat_trop_strat_line = []
for t in temp:
n = 0
while n < len(t):
t_n = t[n]
if t[n] >= 0:
p = psurface[n]
if p < 400:
lat_trop_strat_line.append(p)
n = len(t)
n += 1
trop_strat_line.append(lat_trop_strat_line)
troposphere_boundaryline = np.asarray(trop_strat_line)
troposphere_boundaryline = np.transpose(troposphere_boundaryline, (1, 0))
troposphere_boundaryline *= units.hPa
return troposphere_boundaryline
# -----------------------------------------------------------------------------------------#
def longitude_contourf(self, which="air_temperature", psurface=1000):
"""
Plots a desired atmospheric process on a contour plot, with the axes
being latitude and longitude. This plot is then layed on top of a
EckertIII global projection.
For the raw data, please see the other methods found in this class.
If you would like a particular atmospheric process to
be added to this method, either create an issue on
GitHub or send an email to support@amsimp.com.
"""
if self.planet == "Earth":
# Ensure, which, is a string.
if not isinstance(which, str):
raise Exception(
"which must be a string of the name of the atmospheric parameter of interest."
)
# Index of the nearest pressure surface in amsimp.Backend.pressure_surfaces()
indx_psurface = (np.abs(self.pressure_surfaces().value - psurface)).argmin()
# Defines the axes, and the data.
latitude, longitude = np.meshgrid(
self.latitude_lines().value,
self.longitude_lines().value
)
if which == "temperature" or which == "air_temperature":
data = self.temperature()[indx_psurface, :, :]
data_type = "Air Temperature"
unit = " (K)"
elif which == "geopotential_height" or which == "height":
data = self.geopotential_height()[indx_psurface, :, :]
data_type = "Geopotential Height"
unit = " (m)"
elif which == "density" or which == "atmospheric_density":
data = self.density()[indx_psurface, :, :]
data_type = "Atmospheric Density"
unit = " ($\\frac{kg}{m^3}$)"
elif which == "humidity" or which == "relative_humidity":
data = self.relative_humidity()[indx_psurface, :, :]
data_type = "Relative Humidity"
unit = " (%)"
elif which == "virtual_temperature":
data = self.virtual_temperature()[indx_psurface, :, :]
data_type = "Virtual Temperature"
unit = " (K)"
elif which == "vapor_pressure":
data = self.vapor_pressure()[indx_psurface, :, :]
data_type = "Vapor Pressure"
unit = " (hPa)"
elif which == "potential_temperature":
data = self.potential_temperature()[indx_psurface, :, :]
data_type = "Potential Temperature"
unit = " (K)"
elif which == "precipitable_water" or which == "precipitable_water_vapor":
data = self.precipitable_water()
data_type = "Precipitable Water Vapor"
unit = " (mm)"
elif which == "zonal_wind":
data = self.wind()[0][indx_psurface, :, :]
data_type = "Zonal Wind"
unit = " ($\\frac{m}{s}$)"
elif which == "meridional_wind":
data = self.wind()[1][indx_psurface, :, :]
data_type = "Meridional Wind"
unit = " ($\\frac{m}{s}$)"
elif which == "wind":
data = self.wind()
data = np.sqrt(data[0]**2 + data[1]**2)[indx_psurface, :, :]
data_type = "Wind"
unit = " ($\\frac{m}{s}$)"
elif which == "mixing_ratio":
data = self.mixing_ratio()[indx_psurface, :, :]
data_type = "Mixing Ratio"
unit = " (Dimensionless)"
else:
raise Exception(
"Invalid keyword. which must be a string of an atmospheric parameter included with AMSIMP."
)
if psurface < self.pressure_surfaces()[-1].value or psurface > self.pressure_surfaces()[0].value:
raise Exception(
"psurface must be a real number within the isobaric boundaries. The value of psurface was: {}".format(
psurface
)
)
# EckertIII projection details.
ax = plt.axes(projection=ccrs.EckertIII())
ax.set_global()
ax.coastlines()
ax.gridlines()
# Contourf plotting.
minimum = data.min()
maximum = data.max()
levels = np.linspace(minimum, maximum, 21)
data, lon = add_cyclic_point(data, coord=self.longitude_lines().value)
data, lat = add_cyclic_point(
np.transpose(data), coord=self.latitude_lines().value
)
data = np.transpose(data)
contour = plt.contourf(
lon,
lat,
data,
transform=ccrs.PlateCarree(),
levels=levels,
)
# Add SALT.
plt.xlabel("Latitude ($\phi$)")
plt.ylabel("Longitude ($\lambda$)")
if not self.input_data:
title = (
data_type + " ("
+ str(self.date.year) + '-' + str(self.date.month) + '-'
+ str(self.date.day) + " " + str(self.date.hour)
+ ":00 h)"
)
else:
title = data_type
if which != "surface_temperature" and which != "precipitable_water" and which != "precipitable_water_vapor":
title = (
title + " (" + "Pressure Surface = " + str(self.pressure_surfaces()[indx_psurface]) +")"
)
plt.title(title)
# Colorbar creation.
colorbar = plt.colorbar()
tick_locator = ticker.MaxNLocator(nbins=15)
colorbar.locator = tick_locator
colorbar.update_ticks()
colorbar.set_label(
data_type + unit
)
plt.show()
plt.close()
else:
raise NotImplementedError(
"Visualisations for planetary bodies other than Earth is not currently implemented."
)
def psurface_contourf(self, which="air_temperature", central_long=352.3079):
"""
Plots a desired atmospheric process on a contour plot,
with the axes being latitude, and pressure surfaces.
For the raw data, please see the other methods found in
this class.
If you would like a particular atmospheric process to
be added to this method, either create an issue on
GitHub or send an email to support@amsimp.com.
"""
# Ensure, which, is a string.
if not isinstance(which, str):
raise Exception(
"which must be a string of the name of the atmospheric parameter of interest."
)
# Ensure central_long is between 0 and 359.
if central_long < 0 or central_long > 359:
raise Exception(
"central_long must be a real number between 0 and 359. The value of central_long was: {}".format(
central_long
)
)
# Index of the nearest central_long in amsimp.Backend.longtitude_lines()
indx_long = (np.abs(self.longitude_lines().value - central_long)).argmin()
# Defines the axes, and the data.
latitude, pressure_surfaces = np.meshgrid(
self.latitude_lines().value, self.pressure_surfaces()
)
if which == "temperature" or which == "air_temperature":
data = self.temperature()[:, :, indx_long]
data_type = "Air Temperature"
unit = " (K)"
elif which == "density" or which == "atmospheric_density":
data = self.density()[:, :, indx_long]
data_type = "Atmospheric Density"
unit = " ($\\frac{kg}{m^3}$)"
elif which == "humidity" or which == "relative_humidity":
data = self.relative_humidity()[:, :, indx_long]
data_type = "Relative Humidity"
unit = " (%)"
elif which == "virtual_temperature":
data = self.virtual_temperature()[:, :, indx_long]
data_type = "Virtual Temperature"
unit = " (K)"
elif which == "vapor_pressure":
data = self.vapor_pressure()[:, :, indx_long]
data_type = "Vapor Pressure"
unit = " (hPa)"
elif which == "potential_temperature":
data = self.potential_temperature()[:, :, indx_long]
data_type = "Potential Temperature"
unit = " (K)"
elif which == "zonal_wind":
data = self.wind()[0][:, :, indx_long]
data_type = "Zonal Wind"
unit = " ($\\frac{m}{s}$)"
elif which == "meridional_wind":
data = self.wind()[1][:, :, indx_long]
data_type = "Meridional Wind"
unit = " ($\\frac{m}{s}$)"
elif which == "mixing_ratio":
data = self.mixing_ratio()[:, :, indx_long]
data_type = "Mixing Ratio"
unit = " (Dimensionless)"
else:
raise Exception(
"Invalid keyword. which must be a string of an atmospheric parameter included with AMSIMP."
)
# Contourf plotting.
minimum = data.min()
maximum = data.max()
levels = np.linspace(minimum, maximum, 21)
plt.contourf(
latitude,
pressure_surfaces,
data,
levels=levels,
)
# Add SALT.
plt.xlabel("Latitude ($\phi$)")
plt.ylabel("Pressure (hPa)")
plt.yscale('log')
plt.gca().invert_yaxis()
if not self.input_data:
plt.title(
data_type + " ("
+ str(self.date.year) + '-' + str(self.date.month) + '-'
+ str(self.date.day) + " " + str(self.date.hour)
+ ":00 h, Longitude = "
+ str(np.round(self.longitude_lines()[indx_long], 2)) + ")"
)
else:
plt.title(
data_type + " ("
+ "Longitude = "
+ str(np.round(self.longitude_lines()[indx_long], 2)) + ")"
)
# Colorbar creation.
colorbar = plt.colorbar()
tick_locator = ticker.MaxNLocator(nbins=15)
colorbar.locator = tick_locator
colorbar.update_ticks()
colorbar.set_label(
data_type + unit
)
# Average boundary line between the troposphere and the stratosphere.
troposphere_boundaryline = self.troposphere_boundaryline()
avg_tropstratline = np.mean(troposphere_boundaryline) + np.zeros(
len(troposphere_boundaryline)
)
# Plot average boundary line on the contour plot.
plt.plot(
latitude[1],
avg_tropstratline,
color="black",
linestyle="dashed",
label="Troposphere - Stratosphere Boundary Line",
)
plt.legend(loc=0)
plt.show()
plt.close()
def thickness_contourf(self, p1=1000, p2=500):
"""
Plots pressure thickness on a contour plot, with the axes being
latitude and longitude. This plot is then layed on top of a EckertIII
global projection.
For the raw data, please use the amsimp.Backend.pressure_thickness()
method.
"""
if self.planet == "Earth":
# Defines the axes, and the data.
latitude, longitude = self.latitude_lines().value, self.longitude_lines().value
pressure_thickness = self.pressure_thickness(p1=p1, p2=p2)
# EckertIII projection details.
ax = plt.axes(projection=ccrs.EckertIII())
ax.set_global()
ax.coastlines()
ax.gridlines()
# Contourf plotting.
minimum = pressure_thickness.min()
maximum = pressure_thickness.max()
levels = np.linspace(minimum, maximum, 21)
pressure_thickness, longitude = add_cyclic_point(
pressure_thickness, coord=longitude
)
pressure_thickness, latitude = add_cyclic_point(
np.transpose(pressure_thickness), coord=latitude
)
pressure_thickness = np.transpose(pressure_thickness)
contour = plt.contourf(
longitude,
latitude,
pressure_thickness,
transform=ccrs.PlateCarree(),
levels=levels,
)
# Index of the rain / snow line
indx_snowline = (np.abs(levels.value - 5400)).argmin()
contour.collections[indx_snowline].set_color('black')
contour.collections[indx_snowline].set_linewidth(1)
# Add SALT.
plt.xlabel("Latitude ($\phi$)")
plt.ylabel("Longitude ($\lambda$)")
if not self.input_data:
plt.title("Pressure Thickness ("
+ str(self.date.year) + '-' + str(self.date.month) + '-'
+ str(self.date.day) + " " + str(self.date.hour) + ":00 h" + ")"
)
else:
plt.title("Pressure Thickness")
# Colorbar creation.
colorbar = plt.colorbar()
tick_locator = ticker.MaxNLocator(nbins=15)
colorbar.locator = tick_locator
colorbar.update_ticks()
colorbar.set_label(
"Pressure Thickness (" + str(p1) + " hPa - " + str(p2) + " hPa) (m)"
)
# Footnote
if p1 == 1000 and p2 == 500:
plt.figtext(
0.99,
0.01,
"Rain / Snow Line is marked by the black line (5,400 m).",
horizontalalignment="right",
)
plt.show()
plt.close()
else:
raise NotImplementedError(
"Visualisations for planetary bodies other than Earth is not currently implemented."
)
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 | #cython: linetrace=True
#distutils: define_macros=CYTHON_TRACE_NOGIL=1
#cython: language_level=3
"""
AMSIMP Wind Class. For information about this class is described below.
Copyright (C) 2020 AMSIMP
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <https://www.gnu.org/licenses/>.
"""
# -----------------------------------------------------------------------------------------#
# Importing Dependencies
import os
import iris
import cartopy.crs as ccrs
from cartopy.util import add_cyclic_point
import matplotlib.pyplot as plt
from matplotlib import ticker
import numpy as np
from amsimp.moist cimport Moist
from amsimp.moist import Moist
cimport numpy as np
# -----------------------------------------------------------------------------------------#
cdef class Wind(Moist):
"""
AMSIMP Wind Class - This class is concerned with calculating numerical
values for wind in the troposphere and the stratosphere.
Below is a list of the methods included within this class, with a short
description of their intended purpose. Please see the relevant class methods
for more information.
wind ~ generates a NumPy array of the components of wind. The unit of
measurement is metres per second.
static_stability ~ generates a NumPy array of the the static
stability within a vertical profile.
verical_motion ~ generates a NumPy array of the vertical motion
with the atmosphere, by integrating the isobaric version of the
mass continunity equation.
wind_contourf ~ generates a vertical profile of geostrophic wind,
and outputs this as a contour plot.
globe ~ generates a geostrophic wind contour plot, adds wind vectors to
that said plot, and overlays both on a Nearside Projection of the Earth.
"""
cpdef np.ndarray static_stability(self):
"""
Generates a NumPy array of the the static stability within
a vertical profile.
Static stability is defined as the stability of the
atmosphere in hydrostatic equilibrium with respect to vertical
displacements.
Equation:
sigma = - frac{R T}{p theta} frac{partial theta}{partial p}
"""
cdef np.ndarray theta = self.potential_temperature(moist=False)
cdef np.ndarray temperature = self.temperature()
cdef np.ndarray pressure = self.pressure_surfaces(dim_3d=True)
static_stability = - self.R * (
temperature / (pressure * theta)
) * self.gradient_p(
parameter=theta
)
return static_stability
cpdef tuple wind(self):
"""
This method imports wind data from a GRIB file, which is
located in the AMSIMP Initial Conditions Data Repository on GitHub.
The data stored within this repo is updated every six hours by amsimp-bot.
Following which, it outputs a NumPy array in the shape of
(len(pressure_surfaces), len(latitude_lines), len(longitude_lines)).
Wind is the flow of gases on a large scale. Wind consists of
the bulk movement of air.
I strongly recommend storing the output into a variable in order to
prevent the need to repeatly download the file. For more information,
visit https://github.com/amsimp/initial-conditions.
For geostrophic wind, please see the method, wind.
"""
if not self.input_data:
# Input data.
# Zonal.
u = self.input_u
# Meridional.
v = self.input_v
pressure = self.pressure_surfaces().to(self.units.Pa)
# Grid.
grid_points = [
('pressure', pressure.value),
('latitude', self.latitude_lines().value),
('longitude', self.longitude_lines().value),
]
# Interpolation
# Zonal.
u = u.interpolate(
grid_points, iris.analysis.Linear()
)
# Meridional.
v = v.interpolate(
grid_points, iris.analysis.Linear()
)
# Get data.
# Zonal.
u = u.data
u = np.asarray(u.tolist())
# Meridional.
v = v.data
v = np.asarray(v.tolist())
u *= self.units.m / self.units.s
v *= self.units.m / self.units.s
else:
# Zonal.
u = self.input_u
# Meridional.
v = self.input_v
return u, v
cpdef np.ndarray vertical_motion(self):
"""
Generates a NumPy array of the vertical motion within
the atmosphere, by integrating the isobaric version of
the mass continunity equation.
Typical large-scale vertical motions in the atmosphere
are of the order of 0.01 – 0.1 m s−1.
Equation:
frac{\partial u}{\partial x} + frac{\partial v}{\partial y} = - frac{\partial \omega}{\partial p}
"""
# Define variables.
cdef np.ndarray latitude = np.radians(self.latitude_lines()).value
cdef np.ndarray longitude = np.radians(self.longitude_lines()).value
cdef np.ndarray pressure = self.pressure_surfaces(dim_3d=True)
cdef np.ndarray u, v
u, v = self.wind()
# Determine the LHS pf the equation by calculating the derivatives.
# Change in meridional wind with respect to latitude.
v_dy = self.gradient_latitude(parameter=v)
# Change in zonal wind with respect to longitude.
u_dx = self.gradient_longitude(parameter=u)
LHS = u_dx + v_dy
LHS *= -1
# Integrate the continunity equation.
cdef list vertical_motion_list = []
cdef int n = 0
cdef int len_pressure = len(pressure)
cdef omega, omega_unit
while n < len_pressure:
p1 = n
p2 = n+2
y = LHS[p1:p2, :, :]
p = pressure[p1:p2, :, :]
omega = np.trapz(
y=y, x=p, axis=0
)
omega_unit = omega.unit
vertical_motion_list.append(omega.value)
n += 1
# Convert list to NumPy array.
vertical_motion = np.asarray(vertical_motion_list)
# Add units.
vertical_motion *= omega_unit
# Ensure the shape is correct.
if np.shape(vertical_motion) != np.shape(u):
raise Exception("Unable to determine vertical motion"
+ " at this time. Please contact the developer for futher"
+ " assistance.")
return vertical_motion
# -----------------------------------------------------------------------------------------#
def globe(self, central_lat=53.1424, central_long=352.3079, psurface=1000, which_wind=0):
"""
This particular method adds wind vectors to a contour plot. It also
overlays both of these elements onto a Nearside Perspective projection
of the Earth.
By default, the perspective view is looking directly down at the city
of Dublin in the country of Ireland. If which_wind is set to 0, it will
plot the non-geostophic wind data, otherwise, if it is set to 1, it will
plot the geostrophic wind data.
Note:
The NumPy method, seterr, is used to suppress a weird RunTime warning
error.
"""
if self.planet == "Earth":
# Ensure central_lat is between -90 and 90.
if central_lat < -90 or central_lat > 90:
raise Exception(
"central_lat must be a real number between -90 and 90. The value of central_lat was: {}".format(
central_lat
)
)
# Ensure central_long is between 0 and 359.
if central_long < 0 or central_long > 359:
raise Exception(
"central_long must be a real number between 0 and 359. The value of central_long was: {}".format(
central_long
)
)
# Ensure psurface is between 1000 and 1 hPa above sea level.
if psurface < 1 or psurface > 1000:
raise Exception(
"psurface must be a real number between 1 and 1,000. The value of psurface was: {}".format(
psurface
)
)
# Index of the nearest alt in amsimp.Backend.altitude_level()
indx_psurface = (np.abs(self.pressure_surfaces().value - psurface)).argmin()
# Ignore NumPy errors.
np.seterr(all="ignore")
# Define the axes, and the data.
latitude = self.latitude_lines().value
longitude = self.longitude_lines().value
wind = self.wind()
u = wind[0][indx_psurface, :, :]
v = wind[1][indx_psurface, :, :]
title = "Wind"
u_norm = u / np.sqrt(u ** 2 + v ** 2)
v_norm = v / np.sqrt(u ** 2 + v ** 2)
wind = np.sqrt(u ** 2 + v ** 2)
ax = plt.axes(
projection=ccrs.NearsidePerspective(
central_longitude=central_long, central_latitude=central_lat
)
)
# Add latitudinal and longitudinal grid lines, as well as,
# coastlines to the globe.
ax.set_global()
ax.coastlines()
ax.gridlines()
# Contour plotting.
minimum = wind.min()
maximum = wind.max()
levels = np.linspace(minimum, maximum, 21)
wind, lon = add_cyclic_point(wind, coord=longitude)
wind, lat = add_cyclic_point(np.transpose(wind), coord=latitude)
wind = np.transpose(wind)
contourf = plt.contourf(
lon,
lat,
wind,
transform=ccrs.PlateCarree(),
levels=levels,
)
# Reduce density of wind vectors.
skip_lat = latitude.shape[0] / 23
skip_lat = int(np.round(skip_lat))
skip_lon = longitude.shape[0] / 23
skip_lon = int(np.round(skip_lat))
# Add geostrophic wind vectors.
plt.quiver(
longitude[::skip_lon],
latitude[::skip_lat],
u_norm.value[::skip_lat, ::skip_lon],
v_norm.value[::skip_lat, ::skip_lon],
transform=ccrs.PlateCarree(),
)
# Add SALT.
plt.xlabel("Latitude ($\phi$)")
plt.ylabel("Longitude ($\lambda$)")
if not self.input_data:
plt.title(title + " ("
+ str(self.date.year) + '-' + str(self.date.month) + '-'
+ str(self.date.day) + " " + str(self.date.hour)
+ ":00 h, Pressure Surface = "
+ str(self.pressure_surfaces()[indx_psurface]) +")"
)
else:
plt.title(title + " ("
+ "Pressure Surface = "
+ str(self.pressure_surfaces()[indx_psurface]) +")"
)
# Add colorbar.
colorbar = plt.colorbar(contourf)
tick_locator = ticker.MaxNLocator(nbins=15)
colorbar.locator = tick_locator
colorbar.update_ticks()
colorbar.set_label("Velocity ($\\frac{m}{s}$)")
# Footnote
if which_wind == 1:
plt.figtext(
0.99,
0.01,
"Note: Geostrophic balance does not hold near the equator.",
horizontalalignment="right",
)
plt.show()
plt.close()
else:
raise NotImplementedError(
"Visualisations for planetary bodies other than Earth is not currently implemented."
)
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 | #cython: linetrace=True
#distutils: define_macros=CYTHON_TRACE_NOGIL=1
#cython: language_level=3
"""
AMSIMP Moist Thermodynamics Class. For information about this class is described below.
Copyright (C) 2020 AMSIMP
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <https://www.gnu.org/licenses/>.
"""
# -----------------------------------------------------------------------------------------#
# Importing Dependencies
import numpy as np
from scipy.integrate import quad
import cartopy.crs as ccrs
from cartopy.util import add_cyclic_point
import matplotlib.pyplot as plt
from matplotlib import ticker
from amsimp.backend cimport Backend
from amsimp.backend import Backend
cimport numpy as np
from cpython cimport bool
# -----------------------------------------------------------------------------------------#
cdef class Moist(Backend):
"""
AMSIMP Moist Thermodynamics Class - This class is concerned with incorpating
moisture into the atmospheric model. This is done through the utilisation of
Mosit Thermodynamics. Atmospheric thermodynamics is the study of heat-to-work
transformations (and their reverse) that take place in the earth's atmosphere
and manifest as weather or climate. Atmospheric thermodynamics use the laws
of classical thermodynamics, to describe and explain such phenomena as
the properties of moist air
vapor_pressure ~ generates a NumPy array of saturated vapor pressure.
virtual_temperature ~ generates a NumPy array of the virtual temperature.
The unit of measurement is Kelvin.
density ~ generates a NumPy array of the atmospheric density. The unit of
measurement is kilograms per cubic metric.
exner_function ~ generates a NumPy array of the Exner function (4). This
method has no unit of measurement, i.e. it is dimensionless.
mixing_ratio ~ generates a NumPy array of the mixing ratio (water vapor).
This method has no unit of measurement, i.e. it is dimensionless.
potential_temperature ~ outputs a NumPy array of potential temperature.
The unit of measurement is Kelvin.
precipitable_water ~ generates a NumPy array of precipitable water
vapor. The unit of measurement is millimetres.
"""
cpdef np.ndarray vapor_pressure(self):
"""
Generates a NumPy array of vapor pressure.
Vapor pressure, in meteorology, is the partial pressure of water vapor.
The partial pressure of water vapor is the pressure that water vapor
contributes to the total atmospheric pressure.
Equation (Saturated Vapor Pressure):
e_s = 6.112 \* \exp((17.67 * T) / (T + 243.15))
Equation (Vapor Pressure):
e = frac{r_h \* e_s}{100}
"""
# Convert temperature in Kelvin to degrees centigrade.
cdef np.ndarray temperature = self.temperature().value - 273.15
# Saturated water vapor pressure
sat_vapor_pressure = 0.61121 * np.exp(
(
18.678 - (temperature / 234.5)
) * (temperature / (257.14 + temperature)
)
)
# Add units of measurement.
sat_vapor_pressure *= self.units.kPa
sat_vapor_pressure = sat_vapor_pressure.to(self.units.hPa)
# Vapor pressure, accounting for relative humidity.
vapor_pressure = self.relative_humidity().value * sat_vapor_pressure
vapor_pressure /= 100
return vapor_pressure
cpdef np.ndarray virtual_temperature(self):
"""
Generates a NumPy array of the virtual temperature.
The virtual temperature is the temperature at which dry air
would have the same density as the moist air, at a given
pressure. In other words, two air samples with the same
virtual temperature have the same density, regardless
of their actual temperature or relative humidity.
Equation:
T_v = frac{T}{1 - frac{0.378 e}{p}}
"""
virtual_temperature = self.temperature() / (
1 - (
self.vapor_pressure() / self.pressure_surfaces(dim_3d=True)
) * (1 - 0.622)
)
return virtual_temperature
# -----------------------------------------------------------------------------------------#
cpdef np.ndarray density(self):
"""
Generates a NumPy array of atmospheric pressure utilising the Ideal
Gas Law.
The atmospheric density is the mass of the atmosphere per unit
volume. The ideal gas equation is the equation of state for the
atmosphere, and is defined as an equation relating temperature,
pressure, and specific volume of a system in theromodynamic
equilibrium.
Equation:
rho = frac{p}{R * T_v}
"""
cdef np.ndarray pressure_surfaces = self.pressure_surfaces(dim_3d=True)
density = pressure_surfaces / (self.virtual_temperature() * self.R)
# Switch to the appropriate SI units.
density = density.si
return density
cpdef np.ndarray exner_function(self):
"""
Generates a NumPy array of the exner function.
The Exner function can be viewed as non-dimensionalized pressure.
Equation:
\Pi = frac{T}{theta}
"""
exner_function = self.virtual_temperature() / self.potential_temperature()
return exner_function
cpdef np.ndarray mixing_ratio(self):
""""
Generates a NumPy array of the mixing ratio.
The mixing ratio is the ratio of the mass of a variable atmospheric
constituent to the mass of dry air. In this particular case, it refers
to water vapor.
Equation:
q = frac{0.622 * e}{p - e}
"""
mixing_ratio = (
0.622 * self.vapor_pressure()
) / (
self.pressure_surfaces(dim_3d=True) - self.vapor_pressure()
)
return mixing_ratio
cpdef np.ndarray potential_temperature(self, moist=False):
"""
Generates a NumPy array of potential temperature.
The potential temperature of a parcel of fluid at pressure P is the
temperature that the parcel would attain if adiabatically brought
to a standard reference pressure.
Equation:
theta = T \* (frac{P}{P_0}) ** (-R / c_p)
"""
# Ensure moist is a boolean value.
if not isinstance(moist, bool):
raise Exception(
"moist must be a boolean value. The value of moist was: {}".format(
moist
)
)
# Determine whether to calculate wet-bulb potential temperature, or
# potential temperature.
cdef np.ndarray temperature
if moist:
temperature = self.virtual_temperature().value
else:
temperature = self.temperature().value
cdef np.ndarray pressure_surfaces = self.pressure_surfaces(dim_3d=True).value
cdef float R = self.R.value
cdef float c_p = self.c_p.value
cdef list list_potentialtemperature = []
cdef int n = 0
cdef int len_psurfaces = len(pressure_surfaces)
while n < len_psurfaces:
theta = temperature[n] * (
(pressure_surfaces[n] / pressure_surfaces[0]) ** (-R / c_p)
)
list_potentialtemperature.append(theta)
n += 1
potential_temperature = np.asarray(list_potentialtemperature)
potential_temperature *= self.units.K
return potential_temperature
# -----------------------------------------------------------------------------------------#
def __mixing_ratio(self, pressure, vapor_pressure):
"""
This method is solely utilised for integration in the
amsimp.Water.precipitable_water() method.
"""
y = (0.622 * vapor_pressure) / (pressure - vapor_pressure)
return y
cpdef np.ndarray precipitable_water(self, sum_pwv=True):
"""
Generates a NumPy array of saturated precipitable water vapor.
Precipitable water is the total atmospheric water vapor contained in a
vertical column of unit cross-sectional area extending between any two
specified levels. For a contour plot of this data, please use the
amsimp.Water.contourf() method.
Equation:
PW = frac{1}{rho \* g} \* \int r dp
"""
# Defining some variables.
cdef np.ndarray pressure = self.pressure_surfaces(dim_3d=True).to(self.units.Pa).value
pressure = np.transpose(pressure, (2, 1, 0))
cdef np.ndarray vapor_pressure = self.vapor_pressure().to(self.units.Pa)
vapor_pressure = np.transpose(vapor_pressure.value, (2, 1, 0))
cdef g = self.g
cdef rho_w = 997 * (self.units.kg / self.units.m ** 3)
# Integrate the mixing ratio with respect to pressure between the
# pressure boundaries of p1, and p2.
cdef list list_precipitablewater = []
cdef list pwv_lat, pwv_alt
cdef tuple integration_term
cdef float p1, p2, e, Pwv_intergrationterm
cdef P_wv
cdef int len_pressurelong = len(pressure)
cdef int len_pressurelat = len(pressure[0])
cdef int len_pressurealt = len(pressure[0][0]) - 1
cdef int i, n, k
i = 0
while i < len_pressurelong:
pwv_lat = []
n = 0
while n < len_pressurelat:
pwv_alt = []
k = 0
while k < len_pressurealt:
p1 = pressure[i][n][k]
p2 = pressure[i][n][k + 1]
e = (vapor_pressure[i][n][k] + vapor_pressure[i][n][k + 1]) / 2
integration_term = quad(self.__mixing_ratio, p1, p2, args=(e,))
Pwv_intergrationterm = integration_term[0]
pwv_alt.append(Pwv_intergrationterm)
k += 1
if sum_pwv:
P_wv = np.sum(pwv_alt)
else:
P_wv = pwv_alt
pwv_lat.append(P_wv)
n += 1
list_precipitablewater.append(pwv_lat)
i += 1
precipitable_water = np.asarray(list_precipitablewater) * self.units.Pa
if np.shape(precipitable_water) == (len_pressurelong, len_pressurelat):
precipitable_water = np.transpose(precipitable_water)
else:
precipitable_water = np.transpose(precipitable_water, (2, 1, 0))
# Multiplication term by integration.
precipitable_water *= -1 / (rho_w * g)
precipitable_water = precipitable_water.to(self.units.mm)
return precipitable_water
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 | #cython: linetrace=True
#distutils: define_macros=CYTHON_TRACE_NOGIL=1
#cython: language_level=3
"""
AMSIMP Recurrent Neural Network and Dynamics Class. For information about
this class is described below.
Copyright (C) 2020 AMSIMP
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <https://www.gnu.org/licenses/>.
"""
# -----------------------------------------------------------------------------------------#
# Importing Dependencies
from datetime import timedelta, datetime
import os, sys
from sklearn.preprocessing import MinMaxScaler
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import LSTM, Dense, RepeatVector, TimeDistributed
from tensorflow.keras.layers import Bidirectional
from tensorflow.keras.optimizers import Adam
import tensorflow as tf
import matplotlib.pyplot as plt
from matplotlib import style, ticker, gridspec
import numpy as np
from astropy import constants as constant
import cartopy.crs as ccrs
from cartopy.util import add_cyclic_point
from cpython cimport bool
from amsimp.wind cimport Wind
from amsimp.wind import Wind
from astropy.units.quantity import Quantity
import iris
from iris.coords import DimCoord, AuxCoord
from iris.cube import Cube, CubeList
from iris import save
from progress.bar import IncrementalBar
# -----------------------------------------------------------------------------------------#
cdef class RNN(Wind):
"""
AMSIMP Recurrent Neural Network Class - This class is concerned with
determining the evolution of the state of the atmosphere through
the utilisation of a recurrent neural network. Weather forecasting
has traditionally been done by physical models of the atmosphere,
which are unstable to perturbations, and thus are inaccurate for
large periods of time. Since machine learning techniques are more
robust to perturbations, it would be logical to combine a neural network
with a physical model. Weather forecasting is a sequential data problem,
therefore, a recurrent neural network is the most suitable option for
this task. For this particular version of the software, the
recurrent neural network will be trained solely on temperature and
relative humidity data. Geopotential height will be added in a future release.
A seed is set in order to ensure reproducibility.
Below is a list of the methods included within this class, with a short
description of their intended purpose. Please see the relevant class methods
for more information:
download_historical_data ~ generates a NumPy array of the previous state
of the atmosphere. The number of days of data which is generated can
be specified through the utilisation of the parameter, data_size, when
the class is initialised.
standardise_data ~ generates a NumPy array which standardises the data
generated by the download_historical_data method.
preprocess_data ~ generates the training dataset for the recurrent
neural network.
model_prediction ~ generates a NumPy array of evolution of the
future state of the atmosphere. This is the prediction generated
by the recurrent neural network by training on the generated and
preprocessed training data.
"""
# Feature Scaling
temp_sc = MinMaxScaler(feature_range=(0,1))
rh_sc = MinMaxScaler(feature_range=(0,1))
height_sc = MinMaxScaler(feature_range=(0,1))
# Ensure reproducibility.
tf.random.set_seed(13)
def download_historical_data(self):
"""
Generates a NumPy array of the previous state of the atmosphere.
The number of days of data which is generated can be specified
through the utilisation of the parameter, data_size, when the
class is initialised.
This data is retrieved from the historical dataset contained
on the AMSIMP Initial Atmospheric Conditions Repository.
"""
# Folder containing historical data on GitHub.
folder = "https://github.com/amsimp/initial-conditions/raw/master/initial_conditions/"
# Data lists.
# Temperature.
cdef list T_list = []
# Geopotential Height.
cdef list geo_list = []
# Relative Humidity.
cdef list rh_list = []
# Define variable types.
cdef np.ndarray T, height, rh
# Beginning date of the dataset.
cdef date = self.date
date = date + timedelta(days=-self.data_size)
# Download progresss.
max_bar = self.data_size * 4
bar = IncrementalBar('Downloading Historical Data', max=max_bar)
# Check if the parameter, input_data, is set to True.
if self.input_data:
raise NotImplementedError(
"Currently, it is not possible to train the neural network on user defined datasets."
)
# Remove initial conditions file.
try:
os.remove("initial_conditions.nc")
except OSError:
pass
for i in range(max_bar):
# Configure the Wind class, so, that it aligns with the
# paramaters defined by the user.
config = Wind(
delta_latitude=self.delta_latitude,
delta_longitude=self.delta_longitude,
input_date=date,
)
# Redefine NumPy arrays.
# Geopotential Height.
height = config.geopotential_height().value
height = height.flatten()
# Temperature.
T = config.temperature().value
T = T.flatten()
# Relative Humidity.
rh = config.relative_humidity().value
rh = rh.flatten()
# Append to list.
# Geopotential Height.
geo_list.append(height)
# Temperature.
T_list.append(T)
# Relative Humidity.
rh_list.append(rh)
# Add six hours to date to get the next dataset.
date = date + timedelta(hours=+6)
# Remove download file.
os.remove("initial_conditions.nc")
# Increment progress bar.
bar.next()
# Convert lists to NumPy arrays.
# Geopotential Height.
geopotential_height = np.asarray(geo_list)
# Temperature.
temperature = np.asarray(T_list)
# Relative Humidity.
relative_humidity = np.asarray(rh_list)
# Finish bar.
bar.finish()
# Output.s
output = (temperature, relative_humidity, geopotential_height)
return output
def standardise_data(self):
"""
Generates a NumPy array which standardises the data
generated by the download_historical_data method. It is important
to scale features before training a neural network.
Normalisation is a common way of doing this scaling by
subtracting the mean and dividing by the standard deviation
of each feature. In order for the most optimal performance,
the method ``MinMaxScaler" from the library, scikit-learn,
is utilised within the software.
"""
# Define atmospheric parameters.
historical_data = self.download_historical_data()
temperature = historical_data[0]
relative_humidity = historical_data[1]
geopotential_height = historical_data[2]
# Standardise the data.
# Temperature.
temperature = self.temp_sc.fit_transform(temperature)
# Relative Humidity.
relative_humidity = self.rh_sc.fit_transform(relative_humidity)
# Geopotential Height.
geopotential_height = self.height_sc.fit_transform(geopotential_height)
# Output.
output = (temperature, relative_humidity, geopotential_height)
return output
def preprocess_data(
self, dataset, past_history, future_target
):
"""
Generates the training dataset for the recurrent neural network.
The data set in question is updated every six hours by the National Oceanic
and Atmospheric Administration. This means for a single day,
there will be four observations. The goal for the software will be to,
first predict the relevant atmospheric parameter in seven days time
given the last thirty days of data and combine this RNN prediction with the
physical model prediction in an attempt to make a more accurate prediction
overall. In order to make such predictions, it is necessary to create a
window of the last 120 (30 x 4) observations to train the model.
"""
X, y = list(), list()
for i in range(len(dataset)):
# Find the end.
end_ix = i + past_history
out_end_ix = end_ix + future_target
# Determine if we are beyond the dataset.
if out_end_ix > len(dataset):
break
# Gather the input and output components.
seq_x, seq_y = dataset[i:end_ix, :], dataset[end_ix:out_end_ix, :]
# Append to list.
X.append(seq_x)
y.append(seq_y)
return np.asarray(X), np.asarray(y)
def model_prediction(self):
"""
Generates a NumPy array of evolution of the future state of the atmosphere.
This is the prediction generated by the recurrent neural network by training
on the generated and preprocessed training data. Please see the other
methods within this class for preprocessing information.
"""
# Dataset.
dataset = self.standardise_data()
# Temperature.
temperature = dataset[0]
# Relative Humidity.
relative_humidity = dataset[1]
# Geopotential Height.
geopotential_height = dataset[2]
# The network is shown data from the last 15 days.
past_history = 20 * 4
# The network predicts the next 7 days worth of steps.
future_target = 7 * 4
# The dataset is preprocessed.
# Temperature.
x_temp, y_temp = self.preprocess_data(
temperature, past_history, future_target
)
# Relative Humidity.
x_rh, y_rh = self.preprocess_data(
relative_humidity, past_history, future_target
)
# Geopotential Height.
x_height, y_height = self.preprocess_data(
geopotential_height, past_history, future_target
)
# Number of features.
features = x_temp.shape[2]
# Create, and train models.
# Temperature model.
# Optimiser.
opt_temp = Adam(lr=1e-6, decay=1e-10, clipvalue=0.6)
# Create.
temp_model = Sequential()
temp_model.add(
LSTM(
400, activation='tanh', input_shape=(past_history, features)
)
)
temp_model.add(RepeatVector(future_target))
temp_model.add(LSTM(400, activation='tanh', return_sequences=True))
temp_model.add(LSTM(400, activation='relu', return_sequences=True))
temp_model.add(LSTM(400, activation='relu', return_sequences=True))
temp_model.add(TimeDistributed(Dense(features)))
temp_model.compile(optimizer=opt_temp, loss='mse', metrics=['mean_absolute_error'])
# Train.
temp_model.fit(
x_temp, y_temp, epochs=self.epochs, batch_size=20
)
# Relative Humidity model.
# Optimiser.
opt_rh = Adam(lr=1e-6, decay=1e-10, clipvalue=0.6)
# Create.
rh_model = Sequential()
rh_model.add(
LSTM(
400, activation='tanh', input_shape=(past_history, features)
)
)
rh_model.add(RepeatVector(future_target))
rh_model.add(LSTM(400, activation='tanh', return_sequences=True))
rh_model.add(LSTM(400, activation='relu', return_sequences=True))
rh_model.add(LSTM(400, activation='relu', return_sequences=True))
rh_model.add(TimeDistributed(Dense(features)))
rh_model.compile(optimizer=opt_rh, loss='mse', metrics=['mean_absolute_error'])
# Train.
rh_model.fit(
x_rh, y_rh, epochs=self.epochs, batch_size=20
)
# Geopotential height model.
# Optimiser.
opt_height = Adam(lr=1e-6, decay=1e-10, clipvalue=0.6)
# Create.
height_model = Sequential()
height_model.add(
LSTM(
400, activation='tanh', input_shape=(past_history, features)
)
)
height_model.add(RepeatVector(future_target))
height_model.add(LSTM(400, activation='tanh', return_sequences=True))
height_model.add(LSTM(400, activation='relu', return_sequences=True))
height_model.add(LSTM(400, activation='relu', return_sequences=True))
height_model.add(TimeDistributed(Dense(features)))
height_model.compile(optimizer=opt_height, loss='mse', metrics=['mean_absolute_error'])
# Train.
height_model.fit(
x_height, y_height, epochs=self.epochs, batch_size=20
)
# Prediction.
# Set up inputs.
predict_temp_input = temperature[-past_history:, :]
predict_temp_input = predict_temp_input.reshape(
1, predict_temp_input.shape[0], predict_temp_input.shape[1]
)
predict_rh_input = relative_humidity[-past_history:, :]
predict_rh_input = predict_rh_input.reshape(
1, predict_rh_input.shape[0], predict_rh_input.shape[1]
)
predict_height_input = geopotential_height[-past_history:, :]
predict_height_input = predict_height_input.reshape(
1, predict_height_input.shape[0], predict_height_input.shape[1]
)
# Make predictions.
predict_temp = temp_model.predict(predict_temp_input)
predict_temp = predict_temp[0]
predict_rh = rh_model.predict(predict_rh_input)
predict_rh = predict_rh[0]
predict_height = rh_model.predict(predict_height_input)
predict_height = predict_height[0]
# Invert normalisation.
predict_temp = self.temp_sc.inverse_transform(predict_temp)
predict_rh = self.rh_sc.inverse_transform(predict_rh)
predict_height = self.rh_sc.inverse_transform(predict_height)
# Reshape into 3d arrays.
# Dimensions.
len_time = len(predict_temp)
len_pressure = len(self.pressure_surfaces())
len_lat = len(self.latitude_lines())
len_lon = len(self.longitude_lines())
# Reshape.
predict_temp = predict_temp.reshape(len_time, len_pressure, len_lat, len_lon)
predict_rh = predict_rh.reshape(len_time, len_pressure, len_lat, len_lon)
predict_height = predict_height.reshape(len_time, len_pressure, len_lat, len_lon)
return predict_temp, predict_rh, predict_height
cdef class Dynamics(RNN):
"""
AMSIMP Dynamics Class - Also, known as Motus Aeris @ AMSIMP. This class
generates a simulation of tropospheric and stratsopheric dynamics.
Predictions are made by numerically solving the isobaric version of the
Primitive Equations (they are coupled set of nonlinear PDEs). The initial conditions
are defined in the class methods of Water, Moist, and Backend. For more information
on the initial conditions, please see those classes.
Below is a list of the methods included within this class, with a short
description of their intended purpose. Please see the relevant class methods
for more information.
forecast_period ~ generates the period of time over which the forecast will
be generated for. Please also note that this method also outputs the change in time
used. This is used in the prognostic equations.
simulate ~ generates a simulation of tropospheric and stratsopheric dynamics.
visualise ~ please explain here.
"""
def __cinit__(
self,
int delta_latitude=5,
int delta_longitude=5,
forecast_length=72,
delta_t=2,
bool ai=True,
data_size=150,
epochs=200,
input_date=None,
bool input_data=False,
psurfaces=None,
lat=None,
lon=None,
height=None,
temp=None,
rh=None,
u=None,
v=None,
dict constants={
"sidereal_day": (23 + (56 / 60)) * 3600,
"angular_rotation_rate": ((2 * np.pi) / ((23 + (56 / 60)) * 3600)),
"planet_radius": constant.R_earth.value,
"planet_mass": constant.M_earth.value,
"specific_heat_capacity_psurface": 718,
"gravitational_acceleration": 9.80665,
"planet": "Earth"
}
):
"""
The parameter, forecast_length, defines the length of the
simulation (defined in hours). Defaults to a value of 72.
The parameter, delta_t, defines the change with respect to time
(in minutes) defined in the simulation. Defaults to a value of 2.
For more information, please refer to amsimp.Backend.__cinit__()
method.
"""
# Add units to forecast length variable.
if type(forecast_length) != Quantity:
forecast_length = forecast_length * self.units.h
# Add units to delta_t variable.
if type(delta_t) != Quantity:
delta_t = delta_t * self.units.min
# Declare class variables.
super().__init__(delta_latitude)
super().__init__(delta_longitude)
self.forecast_length = forecast_length
self.delta_t = delta_t
self.ai = ai
super().__init__(input_date)
super().__init__(input_data)
super().__init__(height)
super().__init__(temp)
super().__init__(rh)
super().__init__(u)
super().__init__(v)
super().__init__(constants)
# Ensure self.forecast_length is greater than, or equal to 1.
if self.forecast_length.value <= 0:
raise Exception(
"forecast_length must be a positive number greater than, or equal to 1. The value of forecast_length was: {}".format(
self.forecast_length
)
)
# Ensure ai is a boolean value.
if not isinstance(self.ai, bool):
raise Exception(
"ai must be a boolean value. The value of ai was: {}".format(
self.ai
)
)
cpdef tuple forecast_period(self):
"""
Generates the period of time over which the forecast will be generated
for.
Explain more.
"""
segments = int((self.forecast_length / self.delta_t).si.value) + 1
# Define the forecast period.
forecast_period = np.linspace(
0, self.forecast_length.value, segments
) * self.forecast_length.unit
# Convert to seconds.
delta_t = self.delta_t.to(self.units.s)
return forecast_period, delta_t
def __perturbations_errorcheck(self, input_perturbation):
"""
Explain here.
"""
if input_perturbation != None:
# Check if first index of tuple is a function.
if not callable(input_perturbation):
raise Exception(
"perturbations must be callable functions."
)
def __interpolation_cube(self, input_cube, grid_points):
"""
Explain here.
"""
output = input_cube.interpolate(grid_points, iris.analysis.Linear())
return output
cpdef simulate(
self,
bool save_file=False,
perturbations_temperature=None,
perturbations_zonalwind=None,
perturbations_meridionalwind=None,
perturbations_mixingratio=None
):
"""
Generates a simulation of tropospheric and stratsopheric dynamics.
Predictions are made by numerically solving the Primitive
Equations. Depending on the parameter specifed in the initialisation
of the class, a recurrent neural network may be incorpated in
the output.
The Primitive Equations are a set of nonlinear partial differential
equations that are used to approximate global atmospheric flow and
are used in most atmospheric models. They consist of three main sets
of balance equations: a continuity equation, conservation of
momentum, and a thermal energy equation.
The Lax–Friedrichs method is used to numerically solve the Primitive
Equations within the software. It is a numerical method for the
solution of hyperbolic partial differential equations based on
finite differences. The method can be described as the
FTCS (forward in time, centered in space) scheme with a numerical
dissipation term of 1/2. One can view the Lax–Friedrichs method as an
alternative to Godunov's scheme, where one avoids solving a Riemann
problem at each cell interface
The parameter, save_file, may be utilised to save the output of
this class. The output will be saved as a NetCDF file. These
files can be opened by using the Iris library, which can
be downloaded via Anaconda Cloud.
"""
# Error checking.
np.seterr(all='raise')
# Ensure save_file is a boolean value.
if not isinstance(save_file, bool):
raise Exception(
"save_file must be a boolean value. The value of save_file was: {}".format(
save_file
)
)
# Perturbations error checking.
# Air temperature.
self.__perturbations_errorcheck(perturbations_temperature)
# Zonal wind.
self.__perturbations_errorcheck(perturbations_zonalwind)
# Meridional wind.
self.__perturbations_errorcheck(perturbations_meridionalwind)
# Mixing Ratio.
self.__perturbations_errorcheck(perturbations_mixingratio)
# Define variables that do not vary with respect to time.
cdef np.ndarray latitude = self.latitude_lines().value
cdef np.ndarray longitude = self.longitude_lines().value
cdef np.ndarray pressure = self.pressure_surfaces()
cdef np.ndarray pressure_3d = self.pressure_surfaces(dim_3d=True)
cdef np.ndarray interpolate_pressure = np.insert(
pressure, 0, pressure[0] + np.abs(pressure[0] - pressure[1])
)
cdef time = self.forecast_period()[0]
# The Coriolis parameter at various latitudes of the Earth's surface,
# under various approximations.
# No approximation.
cdef np.ndarray f = self.coriolis_parameter().value / self.units.s
f = self.make_3dimensional_array(parameter=f, axis=1)
# Define the change with respect to time.
cdef delta_t = self.forecast_period()[1]
# Forecast length.
cdef forecast_length = self.forecast_length.to(self.units.s)
cdef int t = 0
# Define initial conditions.
# Gravitational Acceleration.
cdef g = self.g
# Geopotential Height.
cdef np.ndarray height = self.geopotential_height()
# Wind.
cdef tuple wind = self.wind()
# Zonal Wind.
cdef np.ndarray u = wind[0]
# Meridional Wind.
cdef np.ndarray v = wind[1]
# Vertical Motion.
cdef np.ndarray omega = self.vertical_motion()
# Temperature.
# Air Temperature.
cdef np.ndarray T = self.temperature()
# Virtual Temperature.
cdef np.ndarray T_v = self.virtual_temperature()
# Relative Humidity.
cdef np.ndarray rh = self.relative_humidity()
# Mixing Ratio.
cdef np.ndarray q = self.mixing_ratio()
# Precipitable Water.
cdef np.ndarray pwv = self.precipitable_water()
# Prediction from Recurrent Neural Network.
cdef np.ndarray prediction_ai_temp, prediction_ai_height, prediction_ai_rh
cdef int iterator_ai
if self.ai:
prediction_ai = self.model_prediction()
prediction_ai_height = prediction_ai[2] * self.units.m
prediction_ai_temp = prediction_ai[0] * self.units.K
prediction_ai_rh = prediction_ai[1] * self.units.percent
iterator_ai = 0
# Define extra variable types.
# Numy Arrays.
cdef np.ndarray T_n, q_n, u_n, v_n, height_n, A, B, C, D, E
cdef np.ndarray RHS, e, T_c, sat_vapor_pressure, geopotential_height
cdef np.ndarray mean_u, mean_v, mean_T, mean_q, temperature
cdef np.ndarray virtual_temperature, zonal_wind, meridional_wind
cdef np.ndarray static_stability, relative_humidity, mixing_ratio
cdef np.ndarray precipitable_water, Tv_layers, z_0, z_1, z_2, h
cdef np.ndarray Tv_layer, log_pressure_layer
# Booleans.
cdef bool break_loop
# Ints / Floats.
cdef int len_p = len(pressure)
cdef int i
# Tuples.
cdef tuple shape, shape_2d
# Create a bar to determine progress.
max_bar = len(time)
bar = IncrementalBar('Progress', max=max_bar)
# Start progress bar.
bar.next()
# Define NumPy array for ouputs.
shape = (len(time), len(pressure), len(latitude), len(longitude))
shape_2d = (len(time), len(latitude), len(longitude))
# Geopotential Height.
geopotential_height = np.zeros(shape) * height.unit
geopotential_height[0, :, :, :] = height
# Temperature.
# Air Temperature.
temperature = np.zeros(shape) * T.unit
temperature[0, :, :, :] = T
T_n = T.copy()
# Virtual Temperature.
virtual_temperature = np.zeros(shape) * T_v.unit
virtual_temperature[0, :, :, :] = T_v
# Geostrophic Wind.
# Zonal Wind.
zonal_wind = np.zeros(shape) * u.unit
zonal_wind[0, :, :, :] = u
u_n = u.copy()
# Meridional Wind.
meridional_wind = np.zeros(shape) * v.unit
meridional_wind[0, :, :, :] = v
v_n = v.copy()
# Vertical Motion.
vertical_motion = np.zeros(shape) * omega.unit
vertical_motion[0, :, :, :] = omega
# Relative Humidity.
relative_humidity = np.zeros(shape) * rh.unit
relative_humidity[0, :, :, :] = rh
# Mixing Ratio.
mixing_ratio = np.zeros(shape) * q.unit
mixing_ratio[0, :, :, :] = q
q_n = q.copy()
# Precipitable Water.
precipitable_water = np.zeros(shape_2d) * pwv.unit
precipitable_water[0, :, :] = pwv
# Define the initial state for the perturbation functions.
config = Wind(
delta_latitude=self.delta_latitude,
delta_longitude=self.delta_longitude,
input_data=True,
psurfaces=self.pressure_surfaces(),
lat=self.latitude_lines(),
lon=self.longitude_lines(),
height=height,
temp=T,
rh=rh,
u=u,
v=v,
constants=self.constants
)
# Define the coordinates for the cubes.
# Latitude.
lat = DimCoord(
latitude,
standard_name='latitude',
units='degrees'
)
# Longitude
lon = DimCoord(
longitude,
standard_name='longitude',
units='degrees'
)
# Pressure Surfaces.
p = DimCoord(
pressure,
long_name='pressure',
units='hPa'
)
# Time.
forecast_period = DimCoord(
time,
standard_name='forecast_period',
units='hours'
)
# Forecast reference time.
ref_time = AuxCoord(
self.date.strftime("%Y-%m-%d %H:%M:%S"),
standard_name='forecast_reference_time'
)
# Determine geopotential height of lowest constant pressure surface for
# the integration of the Hydrostatic Equation.
# Define grid to interpolate onto.
grid_points = [
('pressure', interpolate_pressure.value),
('latitude', latitude),
('longitude', longitude),
]
# Define cube.
z = Cube(height.value,
standard_name='geopotential_height',
units='m',
dim_coords_and_dims=[
(p, 0), (lat, 1), (lon, 2)
],
)
# Interpolation of geopotential height based on new grid.
z = z.interpolate(grid_points, iris.analysis.Linear())
z_0 = np.asarray(z.data.tolist())[0] * height.unit
cdef int nt = 1
try:
while t < forecast_length.value:
# Wind
# Zonal Wind.
# Determine each term in the zonal momentum equation.
A = self.gradient_longitude(parameter=-u*u)
B = self.gradient_latitude(parameter=-v*u)
C = self.gradient_pressure(parameter=-omega*u)
D = self.gradient_longitude(parameter=-g*height)
E = f * v
# Add any perturbations of zonal wind defined by the user.
if perturbations_zonalwind != None:
perturbations = perturbations_zonalwind(config)
else:
perturbations = np.zeros(u.value.shape) * E.unit
# Sum the RHS terms and multiple by the time step.
RHS = (A + B + C + D + E + perturbations) * delta_t
mean_u = (
(
u[2:, 1:-1, 1:-1] + u[:-2, 1:-1, 1:-1]
) + (
u[1:-1, 2:, 1:-1] + u[1:-1, :-2, 1:-1]
) + (
u[1:-1, 1:-1, 2:] + u[1:-1, 1:-1, :-2]
)
) / 6
u_n[1:-1, 1:-1, 1:-1] = mean_u + RHS[1:-1, 1:-1, 1:-1]
# Meridional Wind.
# Determine each term in the meridional momentum equation.
A = self.gradient_longitude(parameter=-u*v)
B = self.gradient_latitude(parameter=-v*v)
C = self.gradient_pressure(parameter=-omega*v)
D = self.gradient_latitude(parameter=-g*height)
E = -f * u
# Add any perturbations of meridional wind defined by the user.
if perturbations_meridionalwind != None:
perturbations = perturbations_meridionalwind(config)
else:
perturbations = np.zeros(v.value.shape) * E.unit
# Sum the RHS terms and multiple by the time step.
RHS = (A + B + C + D + E + perturbations) * delta_t
mean_v = (
(
v[2:, 1:-1, 1:-1] + v[:-2, 1:-1, 1:-1]
) + (
v[1:-1, 2:, 1:-1] + v[1:-1, :-2, 1:-1]
) + (
v[1:-1, 1:-1, 2:] + v[1:-1, 1:-1, :-2]
)
) / 6
v_n[1:-1, 1:-1, 1:-1] = mean_v + RHS[1:-1, 1:-1, 1:-1]
# Temperature.
# Air Temperature.
# Thermodynamic Equation.
A = self.gradient_longitude(parameter=-u*T)
B = self.gradient_latitude(parameter=-v*T)
C = self.gradient_pressure(parameter=-omega*T)
D = omega * ((self.R * T) / (pressure_3d * self.c_p))
# Add any perturbations of air temperature defined by the user.
if perturbations_temperature != None:
perturbations = perturbations_temperature(config)
else:
perturbations = np.zeros(T.value.shape) * D.unit
# Sum the RHS terms and multiple by the time step.
RHS = (A + B + C + D + perturbations) * delta_t
mean_T = (
(
T[2:, 1:-1, 1:-1] + T[:-2, 1:-1, 1:-1]
) + (
T[1:-1, 2:, 1:-1] + T[1:-1, :-2, 1:-1]
) + (
T[1:-1, 1:-1, 2:] + T[1:-1, 1:-1, :-2]
)
) / 6
T_n[1:-1, 1:-1, 1:-1] = mean_T + RHS[1:-1, 1:-1, 1:-1]
# Mixing ratio
# The scalar gradients of the mixing ratio.
dq_dx = self.gradient_longitude(parameter=q)
dq_dy = self.gradient_latitude(parameter=q)
dq_dp = self.gradient_pressure(parameter=q)
# Advect mixing ratio via wind.
A = self.gradient_longitude(parameter=-u*q)
B = self.gradient_latitude(parameter=-v*q)
C = self.gradient_pressure(parameter=-omega*q)
# Add any perturbations of mixing ratio defined by the user.
if perturbations_mixingratio != None:
perturbations = perturbations_mixingratio(config)
else:
perturbations = np.zeros(q.value.shape) * C.unit
# Sum the RHS terms and multiple by the time step.
RHS = (A + B + C + perturbations) * delta_t
mean_q = (
(
q[2:, 1:-1, 1:-1] + q[:-2, 1:-1, 1:-1]
) + (
q[1:-1, 2:, 1:-1] + q[1:-1, :-2, 1:-1]
) + (
q[1:-1, 1:-1, 2:] + q[1:-1, 1:-1, :-2]
)
) / 6
q_n[1:-1, 1:-1, 1:-1] = mean_q + RHS[1:-1, 1:-1, 1:-1]
# Vapor pressure.
e = pressure_3d * q_n / (0.622 + q_n)
# Convert temperature in Kelvin to degrees centigrade.
T_c = T_n.value - 273.15
# Saturated vapor pressure.
sat_vapor_pressure = 0.61121 * np.exp(
(
18.678 - (T_c / 234.5)
) * (T_c / (257.14 + T_c)
)
)
# Add units of measurement.
sat_vapor_pressure *= self.units.kPa
sat_vapor_pressure = sat_vapor_pressure.to(self.units.hPa)
# Relative Humidity.
rh = (e.value / sat_vapor_pressure.value) * 100
rh[rh > 100] = 100
rh[rh < 0] = 0
rh *= self.units.percent
# Virtual Temperature.
T_v = T_n / (
1 - (
e / pressure_3d
) * (1 - 0.622)
)
# Geopotential Height (Hydrostatic Balance).
# Define variables.
z_1 = z_0
height_n = np.zeros(height.value.shape) * height.unit
# Interpolation of virtual temperature.
# Define cube.
cube = Cube(T_v.value,
standard_name='virtual_temperature',
units='K',
dim_coords_and_dims=[
(p, 0), (lat, 1), (lon, 2)
],
)
# Interpolate.
cube = cube.interpolate(grid_points, iris.analysis.Linear())
# Convert to NumPy array
Tv_layers = np.asarray(cube.data.tolist()) * T_v.unit
for i in range(len_p):
# Virtual temperature layer.
Tv_layer = Tv_layers[i:i+2, :, :]
# Log pressure of layer.
log_pressure_layer = np.log(interpolate_pressure[i:i+2].value)
# Determine thickness of layer.
h = (
-self.R / self.g * np.trapz(
Tv_layer, x=log_pressure_layer, axis=0
)
)
# Determine geopotential height of constant pressure
# surface.
z_2 = z_1 + h
# Redefine bottom pressure surface.
z_1 = z_2
# Add to NumPy array.
height_n[i, :, :] = z_2
# Recurrent Neural Network.
if self.ai:
# Apply predictions every six hours.
if t != 0 and t % 21600 == 0:
# Geopotential Height.
height = (height + prediction_ai_height[iterator_ai]) / 2
# Temperature.
T = (T + prediction_ai_temp[iterator_ai]) / 2
# Relative Humidity.
rh = (rh + prediction_ai_rh[iterator_ai]) / 2
# Iterator.
iterator_ai += 1
# Configure the Wind class, so, that it aligns with the
# paramaters defined by the user.
break_loop = False
while not break_loop:
try:
config = Wind(
delta_latitude=self.delta_latitude,
delta_longitude=self.delta_longitude,
input_data=True,
psurfaces=self.pressure_surfaces(),
lat=self.latitude_lines(),
lon=self.longitude_lines(),
height=height_n,
temp=T_n,
rh=rh,
u=u_n,
v=v_n,
constants=self.constants
)
break_loop = True
except:
pass
# Define other prognostic variables.
# Precipitable Water.
pwv = config.precipitable_water()
# Add predictions to NumPy arrays.
# Geopotential Height.
geopotential_height[nt, :, :, :] = height_n
height = height_n
# Geostrophic Wind.
# Zonal Wind.
zonal_wind[nt, :, :, :] = u_n
u = u_n
# Meridional Wind.
meridional_wind[nt, :, :, :] = v_n
v = v_n
# Temperature.
# Air Temperature.
temperature[nt, :, :, :] = T_n
T = T_n
# Virtual Temperature.
virtual_temperature[nt, :, :, :] = T_v
# Relative Humidity.
relative_humidity[nt, :, :, :] = rh
# Mixing Ratio.
mixing_ratio[nt, :, :, :] = q_n
q = q_n
# Precipitable Water.
precipitable_water[nt, :, :] = pwv
# Add time step.
t += delta_t.value
nt += 1
bar.next()
except KeyboardInterrupt:
pass
# Cubes.
grid_points = [
('forecast_period', time.value),
('pressure', pressure.value),
('latitude', latitude),
('longitude', longitude),
]
grid_points_pwv = [
('forecast_period', time.value),
('latitude', latitude),
('longitude', longitude),
]
# Geopotential Height Cube.
height_cube = Cube(geopotential_height[:, 1:-1, 1:-1, 1:-1],
standard_name='geopotential_height',
units='m',
dim_coords_and_dims=[
(forecast_period, 0), (p[1:-1], 1), (lat[1:-1], 2), (lon[1:-1], 3)
],
attributes={
'source': 'Motus Aeris @ AMSIMP',
}
)
height_cube = self.__interpolation_cube(
input_cube=height_cube, grid_points=grid_points
)
height_cube.add_aux_coord(ref_time)
# Wind Cubes.
# Zonal Wind Cube.
u_cube = Cube(zonal_wind[:, 1:-1, 1:-1, 1:-1],
standard_name='x_wind',
units='m s-1',
dim_coords_and_dims=[
(forecast_period, 0), (p[1:-1], 1), (lat[1:-1], 2), (lon[1:-1], 3)
],
attributes={
'source': 'Motus Aeris @ AMSIMP',
}
)
u_cube = self.__interpolation_cube(
input_cube=u_cube, grid_points=grid_points
)
u_cube.add_aux_coord(ref_time)
# Meridional Wind Cube.
v_cube = Cube(meridional_wind[:, 1:-1, 1:-1, 1:-1],
standard_name='y_wind',
units='m s-1',
dim_coords_and_dims=[
(forecast_period, 0), (p[1:-1], 1), (lat[1:-1], 2), (lon[1:-1], 3)
],
attributes={
'source': 'Motus Aeris @ AMSIMP',
}
)
v_cube = self.__interpolation_cube(
input_cube=v_cube, grid_points=grid_points
)
v_cube.add_aux_coord(ref_time)
# Temperature.
# Air Temperature.
T_cube = Cube(temperature[:, 1:-1, 1:-1, 1:-1],
standard_name='air_temperature',
units='K',
dim_coords_and_dims=[
(forecast_period, 0), (p[1:-1], 1), (lat[1:-1], 2), (lon[1:-1], 3)
],
attributes={
'source': 'Motus Aeris @ AMSIMP',
}
)
T_cube = self.__interpolation_cube(
input_cube=T_cube, grid_points=grid_points
)
T_cube.add_aux_coord(ref_time)
# Virtual Temperature.
Tv_cube = Cube(virtual_temperature[:, 1:-1, 1:-1, 1:-1],
standard_name='virtual_temperature',
units='K',
dim_coords_and_dims=[
(forecast_period, 0), (p[1:-1], 1), (lat[1:-1], 2), (lon[1:-1], 3)
],
attributes={
'source': 'Motus Aeris @ AMSIMP',
}
)
Tv_cube = self.__interpolation_cube(
input_cube=Tv_cube, grid_points=grid_points
)
Tv_cube.add_aux_coord(ref_time)
# Relative Humidity.
rh_cube = Cube(relative_humidity[:, 1:-1, 1:-1, 1:-1],
standard_name='relative_humidity',
units='%',
dim_coords_and_dims=[
(forecast_period, 0), (p[1:-1], 1), (lat[1:-1], 2), (lon[1:-1], 3)
],
attributes={
'source': 'Motus Aeris @ AMSIMP',
}
)
rh_cube = self.__interpolation_cube(
input_cube=rh_cube, grid_points=grid_points
)
rh_cube.add_aux_coord(ref_time)
# Mixing Ratio.
q_cube = Cube(mixing_ratio[:, 1:-1, 1:-1, 1:-1],
long_name='mixing_ratio',
dim_coords_and_dims=[
(forecast_period, 0), (p[1:-1], 1), (lat[1:-1], 2), (lon[1:-1], 3)
],
attributes={
'source': 'Motus Aeris @ AMSIMP',
}
)
q_cube = self.__interpolation_cube(
input_cube=q_cube, grid_points=grid_points
)
q_cube.add_aux_coord(ref_time)
# Precipitable Water.
pwv_cube = Cube(precipitable_water[:, 1:-1, 1:-1],
long_name='precipitable_water',
units='mm',
dim_coords_and_dims=[
(forecast_period, 0), (lat[1:-1], 1), (lon[1:-1], 2)
],
attributes={
'source': 'Motus Aeris @ AMSIMP',
}
)
pwv_cube = self.__interpolation_cube(
input_cube=pwv_cube, grid_points=grid_points_pwv
)
pwv_cube.add_aux_coord(ref_time)
# Create Cube list of output parameters.
output = CubeList([
height_cube,
u_cube,
v_cube,
T_cube,
Tv_cube,
rh_cube,
q_cube,
pwv_cube
])
# If specified, save the forecast in the file format, .nc.
if save_file:
# Establish the file name.
filename = 'motusaeris_amsimp_' + str(self.date.year)
filename += str(self.date.month) + str(self.date.day)
filename += str(self.date.hour) + '.nc'
# Save.
save(output, filename)
# Finish progress bar.
bar.finish()
return output
def visualise(
self,
data=None,
plot=[
"air_temperature",
"geopotential_height",
"precipitable_water",
"relative_humidity"
],
psurface=1000
):
"""
Explain here.
Need to fix.
"""
# Declare variable types.
# NumPy arrays
cdef np.ndarray time, lat, lon, longitude, data1, data2, data3, data4
cdef np.ndarray level1, level2, level3, level4
# Floats.
cdef float min1, min2, min3, min4, max1, max2, max3, max4
if self.planet == "Earth":
# Error checking.
# Ensure a dataset is provided.
if data == None:
raise Exception("The dataset for simulation must be defined.")
# Pressure Surface.
try:
pressure = data[0].coords("pressure")[0].points
except:
pressure = data[1].coords("pressure")[0].points
if psurface < pressure[-1] or psurface > pressure[0]:
raise Exception(
"psurface must be a real number within the isobaric boundaries. The value of psurface was: {}".format(
psurface
)
)
# Index of the nearest pressure surface in amsimp.Backend.pressure_surfaces()
indx_psurface = (np.abs(pressure - psurface)).argmin()
# Define the forecast period.
time = data[0].coords("forecast_period")[0].points
time_unit = str(data[0].coords("forecast_period")[0].units)
# Grid.
lat = data[0].coords("latitude")[0].points
lon = data[0].coords("longitude")[0].points
longitude = lon
# Style of graph.
style.use("fivethirtyeight")
# Define layout.
gs = gridspec.GridSpec(2, 2)
fig = plt.figure(figsize=(18.5, 7.5))
fig.subplots_adjust(hspace=0.340, bottom=0.105, top=0.905)
plt.ion()
# Graph 1.
ax1 = plt.subplot(gs[0, 0], projection=ccrs.EckertIII())
label1 = plot[0]
data1 = data.extract(label1)[0].data
if data1.ndim == 3:
data1 = data1;
else:
data1 = data1[:, indx_psurface, :, :];
data1 = np.asarray(data1.tolist())
unit1 = " (" + str(data.extract(label1)[0].metadata.units) + ")"
label1 = label1.replace("_", " ").title()
# Graph 2.
ax2 = plt.subplot(gs[1, 0], projection=ccrs.EckertIII())
label2 = plot[1]
data2 = data.extract(label2)[0].data
if data2.ndim == 3:
data2 = data2;
else:
data2 = data2[:, indx_psurface, :, :];
data2 = np.asarray(data2.tolist())
unit2 = " (" + str(data.extract(label2)[0].metadata.units) + ")"
label2 = label2.replace("_", " ").title()
# Graph 3.
ax3 = plt.subplot(gs[0, 1], projection=ccrs.EckertIII())
label3 = plot[2]
data3 = data.extract(label3)[0].data
if data3.ndim == 3:
data3 = data3;
else:
data3 = data3[:, indx_psurface, :, :];
data3 = np.asarray(data3.tolist())
unit3 = " (" + str(data.extract(label3)[0].metadata.units) + ")"
label3 = label3.replace("_", " ").title()
# Graph 4.
ax4 = plt.subplot(gs[1, 1], projection=ccrs.EckertIII())
label4 = plot[3]
data4 = data.extract(label4)[0].data[:, indx_psurface, :, :]
if data4.ndim == 3:
data4 = data4;
else:
data4 = data4[:, indx_psurface, :, :];
data4 = np.asarray(data4.tolist())
unit4 = " (" + str(data.extract(label4)[0].metadata.units) + ")"
label4 = label4.replace("_", " ").title()
# Get date.
date = data[0].coords("forecast_reference_time")[0].points[0]
for i in range(len(time)):
# Plot 1.
# EckertIII projection details.
ax1.set_global()
ax1.coastlines()
ax1.gridlines()
# Add SALT to the graph.
ax1.set_xlabel("Longitude ($\lambda$)")
ax1.set_ylabel("Latitude ($\phi$)")
ax1.set_title(label1)
# Contour plot.
cmap1 = plt.get_cmap("hot")
min1 = np.min(data1)
max1 = np.max(data1)
level1 = np.linspace(min1, max1, 21)
data1_plot, lon = add_cyclic_point(data1, coord=longitude)
contour1 = ax1.contourf(
lon,
lat,
data1_plot[i, :, :],
cmap=cmap1,
levels=level1,
transform=ccrs.PlateCarree()
)
# Checks for a colorbar.
if i == 0:
cb1 = fig.colorbar(contour1, ax=ax1)
tick_locator = ticker.MaxNLocator(nbins=10)
cb1.locator = tick_locator
cb1.update_ticks()
cb1.set_label(unit1)
# Plot 2.
cmap2 = plt.get_cmap("jet")
min2 = np.min(data2)
max2 = np.max(data2)
level2 = np.linspace(min2, max2, 21)
data2_plot, lon = add_cyclic_point(data2, coord=longitude)
contour2 = ax2.contourf(
lon,
lat,
data2_plot[i, :, :],
cmap=cmap2,
levels=level2,
transform=ccrs.PlateCarree()
)
# EckertIII projection details.
ax2.set_global()
ax2.coastlines()
ax2.gridlines()
# Checks for a colorbar.
if i == 0:
cb2 = fig.colorbar(contour2, ax=ax2)
tick_locator = ticker.MaxNLocator(nbins=10)
cb2.locator = tick_locator
cb2.update_ticks()
cb2.set_label(unit2)
# Add SALT to the graph.
ax2.set_xlabel("Longitude ($\lambda$)")
ax2.set_ylabel("Latitude ($\phi$)")
ax2.set_title(label2)
# Plot 3.
cmap3 = plt.get_cmap("seismic")
min3 = np.min(data3)
max3 = np.max(data3)
level3 = np.linspace(min3, max3, 21)
data3_plot, lon = add_cyclic_point(data3, coord=longitude)
contour3 = ax3.contourf(
lon,
lat,
data3_plot[i, :, :],
cmap=cmap3,
levels=level3,
transform=ccrs.PlateCarree()
)
# EckertIII projection details.
ax3.set_global()
ax3.coastlines()
ax3.gridlines()
# Checks for a colorbar.
if i == 0:
cb3 = fig.colorbar(contour3, ax=ax3)
tick_locator = ticker.MaxNLocator(nbins=10)
cb3.locator = tick_locator
cb3.update_ticks()
cb3.set_label(unit3)
# Add SALT to the graph.
ax3.set_xlabel("Longitude ($\lambda$)")
ax3.set_ylabel("Latitude ($\phi$)")
ax3.set_title(label3)
# Plot 4.
min4 = np.min(data4)
max4 = np.max(data4)
level4 = np.linspace(min4, max4, 21)
data4_plot, lon = add_cyclic_point(data4, coord=longitude)
contour4 = ax4.contourf(
lon,
lat,
data4_plot[i, :, :],
levels=level4,
transform=ccrs.PlateCarree()
)
# EckertIII projection details.
ax4.set_global()
ax4.coastlines()
ax4.gridlines()
# Checks for a colorbar.
if i == 0:
cb4 = fig.colorbar(contour4, ax=ax4)
tick_locator = ticker.MaxNLocator(nbins=10)
cb4.locator = tick_locator
cb4.update_ticks()
cb4.set_label(unit4)
# Add SALT to the graph.
ax4.set_xlabel("Longitude ($\lambda$)")
ax4.set_ylabel("Latitude ($\phi$)")
ax4.set_title(label4)
# Title
title = (
"Motus Aeris @ AMSIMP (" + date + ", +" + str(np.round(time[i], 2)) + time_unit + ", " + str(pressure[indx_psurface]) + " hPa)"
)
fig.suptitle(title)
# Check if folder needs to created, and if so create one.
try:
os.mkdir('visualisations')
except OSError:
pass
# Save image to folder.
plt.savefig('visualisations/timestep_' + str(i), dpi=300)
# Displaying simualtion.
plt.show()
plt.pause(0.01)
if i < (len(time) - 1):
ax1.clear()
ax2.clear()
ax3.clear()
ax4.clear()
else:
# Compile photos into video using ffmeg.
os.system("ffmpeg -framerate 30 -i visualisations/timestep_%d.png -vf format=yuv420p visualise.mp4")
# Remove images.
for i in range(len(time)):
os.remove("visualisations/timestep_" + str(i) + ".png")
# Remove folder.
try:
os.rmdir('visualisations')
except OSError:
pass
else:
raise NotImplementedError(
"Visualisations for planetary bodies other than Earth is not currently implemented."
)
|