-
Notifications
You must be signed in to change notification settings - Fork 1
/
extract_ESACCI.py
208 lines (187 loc) · 8.13 KB
/
extract_ESACCI.py
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
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Fri May 21 11:38:44 2021
uses the raw ESA Cloud_cci L3U nc files to make npz files that are more useful here
@author: arndt
"""
import netCDF4 as nc4
import numpy as np
import sys
from datetime import datetime
import os
import glob
import multiprocessing as mlp
import traceback
coordinates = ['lat', 'lon']
properties = ["lwp", "iwp","cer_liq", "cer_ice","cot","ctp","cth", "ctt","cee","stemp_cloudy"
,"cfc"]
properties_daily_asc = ["cwp_asc", "cer_asc", "cot_asc", "ctp_asc", "cth_asc",
"ctt_asc", "cot_desc", "stemp_asc", "cph_asc"]
properties_daily_desc = ["cwp_desc", "cer_desc", "cot_desc", "ctp_desc", "cth_desc",
"ctt_desc", "cot_asc", "stemp_desc", "cph_desc"]
def read_daily(nc_file):
"""Extracts the ascending and descinding orbits and rescales them and puts them in the order the RF expects them to be in
Args:
nc_file (string): path of file to load
Returns:
list of two np.ndarray: array for each orbit. each contains the 2D fields of all properties,
spatiotemporal locations and radiative fluxes and cloud cover
"""
data = nc4.Dataset(nc_file, "r", format="NETCDF4")
fname = os.path.basename(nc_file)
#cloud cover is in a different file for some reason
cltds = nc4.Dataset(os.path.join(os.environ["SCR"],
"ESACCI/MASK",fname.replace("PRODUCTS","MASKTYPE")),"r")
clt_asc = cltds.variables["cmask_asc"]
clt_desc = cltds.variables["cmask_desc"]
del cltds
#radiative fluxes are also in another file
radds = nc4.Dataset(os.path.join(os.environ["SCR"],"ESACCI/RAD",
fname.replace("CLD","RAD")),"r")
toa_asc = np.vstack([radds.variables[ x][...] for x in ["toa_lwup_asc","toa_swup_asc","toa_swdn_asc",
"toa_swup_clr_asc", "toa_lwup_clr_asc" ]])
toa_desc = np.vstack([radds.variables[x][...] for x in ["toa_lwup_desc","toa_swup_desc","toa_swdn_desc",
"toa_swup_clr_desc","toa_lwup_clr_desc"] ])
del radds
#other physical variables
f_properties_asc = np.vstack([data.variables[name][:] for name in properties_daily_asc])
#replace nonsense with missing
f_properties_asc = np.where(f_properties_asc<0, np.nan, f_properties_asc)
#rescale height to meters
f_properties_asc[4]*=1000
#make 2D fiels for the locations
lat = data.variables["lat"][:]
lon = data.variables["lon"][:]
timepoint = data.variables["time"][:]
latlat,lonlon = np.meshgrid(lat,lon)
latlat=latlat.transpose()[np.newaxis,...]
lonlon=lonlon.transpose()[np.newaxis,...]
timepoint = np.ones(latlat.shape)*timepoint
assert np.all(np.unique(latlat)==np.unique(lat))
assert np.all(np.unique(lonlon)==np.unique(lon))
f_loc = np.vstack((latlat,lonlon,timepoint))
#phase descision
isice = f_properties_asc[-1]==2
isliquid = f_properties_asc[-1]==1
#phase water path
lwp=np.copy(f_properties_asc[0])
iwp=np.copy(f_properties_asc[0])
lwp[isice]=0
iwp[isliquid]=0
lwp[~(isice+isliquid)]/=2
iwp[~(isice+isliquid)]/=2
test = ~np.isnan(f_properties_asc[0])
#make sure ice and water add up to total everywhere
assert np.all(lwp[test]+iwp[test]==f_properties_asc[0][test])
#phase effective radius
ceri=np.copy(f_properties_asc[1])
cerl=np.copy(f_properties_asc[1])
ceri[isliquid]=0
cerl[isice]=0
ceri[~(isice+isliquid)]/=2
cerl[~(isice+isliquid)]/=2
test = ~np.isnan(f_properties_asc[1])
assert np.all(ceri[test]+cerl[test]==f_properties_asc[1][test])
#put everything in the expected order
out1=[f_properties_asc[0],lwp,iwp, cerl,ceri]
out1.extend(f_properties_asc[2:-1])
out1.extend(toa_asc)
out1.extend(clt_asc)
out1.extend(f_loc)
out1=np.stack(out1)
#same thing for the descending orbits
f_properties_desc = np.vstack([data.variables[name][:] for name in properties_daily_desc])
f_properties_desc = np.where(f_properties_desc<0, np.nan, f_properties_desc)
f_properties_desc[4]*=1000
#phase descision
isice = f_properties_desc[-1]==2
isliquid = f_properties_desc[-1]==1
#phase water path
lwp=np.copy(f_properties_desc[0])
iwp=np.copy(f_properties_desc[0])
lwp[isice]=0
iwp[isliquid]=0
lwp[~(isice+isliquid)]/=2
iwp[~(isice+isliquid)]/=2
test = ~np.isnan(f_properties_desc[0])
assert np.all(lwp[test]+iwp[test]==f_properties_desc[0][test])
#phase effective radius
ceri=np.copy(f_properties_desc[1])
cerl=np.copy(f_properties_desc[1])
ceri[isliquid]=0
cerl[isice]=0
ceri[~(isice+isliquid)]/=2
cerl[~(isice+isliquid)]/=2
test = ~np.isnan(f_properties_desc[1])
assert np.all(ceri[test]+cerl[test]==f_properties_desc[1][test])
out2=[f_properties_desc[0],lwp,iwp, cerl,ceri]
out2.extend(f_properties_desc[2:-1])
out2.extend(toa_desc)
out2.extend(clt_desc)
timepoint = data.variables["time"][:]
timepoint = np.ones(latlat.shape)*timepoint
f_loc = np.vstack((latlat,lonlon,timepoint))
out2.extend(f_loc)
out2=np.stack(out2)
return [out1, out2]
if __name__=="__main__":
work = os.environ["WORK"]
scratch = os.environ["SCR"]
folder = os.path.join(scratch, "ESACCI/CLD",)
outfolder = os.path.join(folder,"../npz_daily")
if not os.path.exists(outfolder):
os.makedirs(outfolder)
files = glob.glob(os.path.join(folder, "*.nc"))
with open("experiment_log.txt", "a+") as of:
print("extract_ESACCI.py("+str(datetime.today())+") : "+sys.argv[1], file=of)
assert len(files)>0, folder
def convert(file):
if "L3U" in file:
read = read_daily
time="daily"
else:
raise ValueError("file makes no sense?")
if time=="daily":
if not os.path.exists(os.path.join(outfolder,
"1_"+os.path.basename(file).replace(".nc", ".npz"))):
#only processes if it doesnt already exist
print("{} doesnt exist".format(os.path.join(outfolder,
"1_"+os.path.basename(file).replace(".nc", ".npz"))))
p=read(file)
assert len(p[0].shape)>2, p[0].shape
np.savez_compressed(os.path.join(outfolder, "0_"+os.path.basename(file).replace(".nc",".npz")),
props= p[0][:-3].data.astype(np.float32),
locs = p[0][-3:].data.astype(np.float32))
np.savez_compressed(os.path.join(outfolder, "1_"+os.path.basename(file).replace(".nc",".npz")
), props=p[1][:-3].data.astype(np.float32),
locs = p[1][-3:].data.astype(np.float32))
os.remove(file)
#sanity check and removal of input files
fname = os.path.basename(file)
if os.path.exists(os.path.join(outfolder,
"1_"+os.path.basename(file).replace(".nc", ".npz"))):
try:
p=np.load(os.path.join(outfolder,
"1_"+os.path.basename(file).replace(".nc", ".npz")))["props"]
assert np.all(p.shape==(17,3600,7200)),"shape not good"
except Exception:
traceback.print_exc()
return
try:
os.remove(os.path.join(os.environ["SCR"],
"ESACCI/MASK",fname.replace("PRODUCTS","MASKTYPE")))
except Exception:
traceback.print_exc()
try:
os.remove(os.path.join(os.environ["SCR"],
"ESACCI/RAD",fname.replace("CLD","RAD")))
except Exception:
traceback.print_exc()
try:
os.remove(file)
except Exception:
traceback.print_exc()
p=mlp.Pool(20)
print(len(files))
p.map(convert, files)