-
Notifications
You must be signed in to change notification settings - Fork 6
/
Copy pathukmo-midas.py
314 lines (259 loc) · 11.6 KB
/
ukmo-midas.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
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
# -*- coding: utf8 -*-
# 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/>.
import sys
import os
import io
import copy
import tarfile
import gzip
import shutil
import datetime
import multiprocessing as mp
import requests
import xarray as xr
import pandas as pd
import dask
from dask.diagnostics import ProgressBar
import zarr
import numpy as np
import shapely.geometry
import geopandas as gpd
START_YEAR = 1979
END_YEAR = 2018
BASE_DIR = '/home/lunet/gylc4/geodata/MIDAS/'
BASE_FILE = 'midas_rainhrly_{y}01-{y}12.txt'
STATIONS_DIR = os.path.join(BASE_DIR, 'stations')
STATIONS_LIST = os.path.join(BASE_DIR, 'src_id_list.csv')
PRECIP_FILE = '/home/lunet/gylc4/geodata/MIDAS/midas_precip_{s}-{e}.zarr'
SELECT_FILE = '../data/MIDAS/midas_{s}-{e}_precip_select.zarr'.format(s=START_YEAR, e=END_YEAR)
PAIR_FILE = '../data/MIDAS/midas_{s}-{e}_precip_pairs.nc'
# https://www.metoffice.gov.uk/public/weather/climate-extremes/#?tab=climateExtremes
UK_HOURLY_MAX = 92.0
UK_DAILY_MAX = 279.0
# http://artefacts.ceda.ac.uk/badc_datadocs/ukmo-midas/RH_Table.html
HEADER = ['end_time', 'id', 'id_type', 'ob_hour_count', 'version_num', 'met_domain_name',
'src_id', 'rec_st_ind', 'prcp_amt', 'prcp_dur', 'prcp_amt_q', 'prcp_dur_q',
'meto_stmp_time', 'midas_stmp_etime', 'prcp_amt_j'
]
IDX_COL_NAMES = ["id", "src_id", "src_name", "latitude", "longitude"]
DTYPES = {#'end_time':str,
'id':np.uint32, 'id_type':str, 'ob_hour_count':np.uint16,
'version_num':np.uint16, 'met_domain_name':str,
'src_id':np.uint32, 'rec_st_ind':np.uint16,
'prcp_amt':float, 'prcp_dur':float,
'prcp_amt_q':str, 'prcp_dur_q':str,
'meto_stmp_time':str, 'midas_stmp_etime':str, 'prcp_amt_j':str
}
CHUNKS = {'end_time': -1, 'station': 30}
DTYPE = 'float32'
FLOAT_ENCODING = {'dtype': DTYPE, 'compressor': zarr.Blosc(cname='lz4', clevel=9)}
INT_ENCODING = {'dtype': 'uint32', 'compressor': zarr.Blosc(cname='lz4', clevel=9)}
STR_ENCODING = {'dtype': 'U', 'compressor': zarr.Blosc(cname='lz4', clevel=9)}
GEN_ENCODING = {'compressor': zarr.Blosc(cname='lz4', clevel=9)}
KEEP_VARS = ['id', 'ob_hour_count', 'prcp_amt',
'prcp_dur', 'rec_st_ind', 'src_id', 'version_num']
ENCODING = {'id': INT_ENCODING,
# 'id_type': STR_ENCODING,
'ob_hour_count': INT_ENCODING,
'version_num': INT_ENCODING,
# 'met_domain_name': STR_ENCODING,
'src_id': INT_ENCODING,
'rec_st_ind': INT_ENCODING,
'prcp_amt': FLOAT_ENCODING,
'prcp_dur': FLOAT_ENCODING,
# 'prcp_amt_q': STR_ENCODING,
# 'prcp_dur_q': STR_ENCODING,
# 'meto_stmp_time': STR_ENCODING,
# 'midas_stmp_etime': STR_ENCODING,
# 'prcp_amt_j': STR_ENCODING
}
def read_stations(data_dir, start_year, end_year):
"""Read station data from text file.
Return a xarray DataArray.
"""
prefix = 'midas_rainhrly_'
suffix = '.txt'
df_list = []
for year in range(start_year, end_year+1):
filename = BASE_FILE.format(y=year)
print(filename)
filepath = os.path.join(data_dir, filename)
df = pd.read_csv(filepath, sep=',', names=HEADER, na_values=' ',
dtype=DTYPES, engine='c',
index_col=False, parse_dates=['end_time'])
df_list.append(df)
# One dataframe per station
df_all = pd.concat(df_list)
gb = df_all.groupby('src_id')
df_stations = [gb.get_group(x) for x in gb.groups]
# Create a xarray DataArray
# print(df_stations_list.head())
ds_stations = []
idx = [[] for i in range(len(IDX_COL_NAMES))]
for df in df_stations:
# delete duplicated index
df.set_index('end_time', drop=True, inplace=True)
df_dropped = df[~df.index.duplicated(keep='first')]
ds = df_dropped.to_xarray()
ds = ds.expand_dims('station')
src_id = df_dropped['src_id'].unique()[0] # station unique identifier
ds.coords['station'] = [src_id]
ds.drop('src_id')
# ds.coords['src_name'] = xr.DataArray([src_name], dims='station')
# ds.coords['latitude'] = xr.DataArray([lat], dims='station')
# ds.coords['longitude'] = xr.DataArray([lon], dims='station')
ds_stations.append(ds)
ds_concat = xr.concat(ds_stations, dim='station')
return ds_concat
def add_stations_metadata(ds, src_id):
# Read metadata
df_stations_list = pd.read_csv(src_id, index_col=0, sep='|')
df_stations_list = df_stations_list[~df_stations_list.index.duplicated(keep='first')]
# select metadata and keep as xarray
ds_srcid = df_stations_list[['SRC_NAME', 'HIGH_PRCN_LAT', 'HIGH_PRCN_LON']].sort_index().to_xarray()
# Keep only metadata of station present in the precip dataset
ds_list = []
for station in ds['station']:
try:
station_infos = ds_srcid.sel(SRC_ID=station).drop('SRC_ID')
except KeyError:
src_name = xr.DataArray([''], coords={'station': [station]}, dims='station')
lat = xr.DataArray([np.nan], coords={'station': [station]}, dims='station')
lon = xr.DataArray([np.nan], coords={'station': [station]}, dims='station')
station_infos = xr.Dataset({'SRC_NAME': src_name, 'HIGH_PRCN_LAT': lat, 'HIGH_PRCN_LON': lon})
ds_list.append(station_infos)
srcid_sel = xr.concat(ds_list, dim='station')
# Add coordinates to precip dataset
ds.coords['src_name'] = srcid_sel['SRC_NAME']
ds.coords['latitude'] = srcid_sel['HIGH_PRCN_LAT']
ds.coords['longitude'] = srcid_sel['HIGH_PRCN_LON']
return ds
def select_stations(ds, ymin, ymax):
# 90% of all hourly records in 365 days
min_records = int(365*24*.9)
print('## Min record per year:', min_records)
# Keep only the years of interest
ds_short = ds.sel(end_time=slice(str(ymin), str(ymax)))
# Keep only station with coordinates
ds_sel = np.logical_and(np.isfinite(ds_short['latitude']), np.isfinite(ds_short['longitude']))
ds_with_coords = ds_short.where(ds_sel, drop=True)
# Keep only the quality controlled data
ds_cc = ds_with_coords.where(~ds_short['qc0'] & ~ds_short['qc1'] & ~ds_short['qc3']).drop(['qc0', 'qc1', 'qc3'])
# Number of values per year
prcp_values_per_years = ds_cc['prcp_amt'].groupby('end_time.year').count(dim='end_time')
min_years = int(len(prcp_values_per_years['year']) * .9)
print('## Min complete year per station:', min_years)
flag_full_year = xr.where(prcp_values_per_years >= min_records, True, False)
# number of full years per station
num_of_full_years = flag_full_year.sum(dim='year')
# ds_cc['full_years'] = num_of_full_years#.transpose(*ds_cc['prcp_amt'].dims)
# Keep only stations with enough full years
precip_full_years = ds_cc.where(num_of_full_years >= min_years, drop=True)
# Drop years without enough data
dropped_years = precip_full_years.groupby('end_time.year').where(flag_full_year).drop('year')
# Check that no partial year remains
year_count_after_drop = dropped_years['prcp_amt'].groupby('end_time.year').count(dim='end_time')
flag_full_year = xr.where(year_count_after_drop > min_records, True, False)
# num_of_full_years = flag_full_year.sum(dim='year') # number of full years per station
# print(num_of_full_years.load())
arr_year_count_after_drop = year_count_after_drop.values
pos_year_count_after_drop = arr_year_count_after_drop[arr_year_count_after_drop > 0]
assert np.all(pos_year_count_after_drop > min_records)
return dropped_years
def qc0(ds):
"""Flag data that are not hourly, and flagged as not reliable
"""
ds['qc0'] = np.logical_or(np.not_equal(ds['version_num'], 1),
ds['ob_hour_count']!=1)
def qc1(ds):
"""Exceed UK hour record by 20%
"""
ds['qc1'] = np.isfinite(ds['prcp_amt'].where(ds['prcp_amt'] >= UK_HOURLY_MAX * 1.2))
# print(ds['qc1'].sum().load())
# print(ds['prcp_amt'].min().load())
def qc3(ds):
"""Exceed UK 24h record by 20%
"""
# Aggregate on 24h
ds['prcp_daily'] = ds['prcp_amt'].rolling(end_time=24, min_periods=max(int(24*.9), 1)).sum(dim='end_time', skipna=True)
ds['qc3'] = xr.where(ds['prcp_daily'] >= UK_DAILY_MAX * 1.2, True, False)
def qc4(ds):
"""Hourly totals at 0900 hours exceeding 2 times the mean daily rainfall for that month
and preceded by 23h without rainfall
"""
daily_mean_per_month = ds['prcp_daily'].groupby('end_time.month').mean(dim='end_time', skipna=True)
print(daily_mean_per_month)
prcp_0900 = ds['prcp_amt'].sel(end_time=datetime.time(9))
month = prcp_0900['end_time.month']
print(prcp_0900)
over_thres = xr.where(prcp_0900 >= 2*daily_mean_per_month[month], True, False)
print(over_thres.load())
def qc5(ds):
"""Three consecutive measurements at 0900 or 1200 hours following 23h without rainfall
"""
pass
def qc7(ds):
"""Duplicate rainfall in consecutive hours that
exceed two times the mean daily rainfall for that month.
"""
pass
def quality_assessment(ds):
qc0(ds)
qc1(ds)
qc3(ds)
# qc4(ds)
# print(ds['qc0'].sum().load())
# print(ds['qc1'].sum().load())
# print(ds['qc3'].sum().load())
return ds
def to_gdf(ds):
"""return names as a geoDataframe
"""
names = ds.src_name.load().drop('src_name')
df = names.to_dataframe().dropna()
df['geometry'] = [shapely.geometry.Point(lon, lat)
for lon, lat in zip(df['longitude'], df['latitude'])]
df['name'] = df['src_name'].str.decode("utf-8")
del df['src_name']
gdf = gpd.GeoDataFrame(df, geometry='geometry')
return gdf
def main():
# with ProgressBar():
# ds = read_stations(STATIONS_DIR, START_YEAR, END_YEAR).chunk(CHUNKS)[KEEP_VARS]
# print(ds)
# ds.to_zarr(PRECIP_FILE.format(s=START_YEAR, e=END_YEAR), mode='w', encoding=ENCODING)
# Join
# ds_list = []
# for y_start in [1979, 1989, 1999, 2009]:
# ds = xr.open_zarr(PRECIP_FILE.format(s=y_start, e=y_start+9)).drop(['longitude', 'latitude', 'src_name'])
# # print(ds['latitude'].load())
# ds_list.append(ds)
# ds_all = xr.concat(ds_list, dim='end_time', coords='minimal').chunk(CHUNKS)
# Add station metadata and save
# ds_precip = add_stations_metadata(ds_all, STATIONS_LIST)
# ds_all.to_zarr(PRECIP_FILE.format(s=1979, e=2018), mode='w', encoding=ENCODING)
# ds = xr.open_zarr(PRECIP_FILE.format(s=START_YEAR, e=END_YEAR))
# ds = quality_assessment(ds)
# ds_sel = select_stations(ds, START_YEAR, END_YEAR).load()
# print(ds_sel)
# encoding = {'full_years': INT_ENCODING, **ENCODING}
# ds_sel[KEEP_VARS].to_zarr(SELECT_FILE, mode='w', encoding=ENCODING)
ds_sel = xr.open_zarr(SELECT_FILE)
print(ds_sel)
gdf = to_gdf(ds_sel)
out_path = os.path.join('../data/MIDAS', "midas.gpkg")
gdf.to_file(out_path, driver="GPKG")
# print(ds_sel.max().load())
# ds_pairs = station_pairs(ds_sel)
# ds_pairs.load().to_netcdf(PAIR_FILE, mode='w')
if __name__ == "__main__":
sys.exit(main())