observationoperator.py 11.9 KB
Newer Older
brunner's avatar
brunner committed
1
2
3
4
#!/usr/bin/env python
# model.py

import logging
brunner's avatar
brunner committed
5
6
7
import os
import sys
import subprocess
brunner's avatar
brunner committed
8
import da.cosmo.io4 as io
brunner's avatar
brunner committed
9
10
11
12
13
import numpy as np
from netCDF4 import Dataset
from datetime import datetime, timedelta
from dateutil import rrule
from cdo import *
brunner's avatar
brunner committed
14
from . import site_height
brunner's avatar
brunner committed
15
16
from itertools import repeat
from multiprocessing import Pool
brunner's avatar
brunner committed
17
from da.tools.general import to_datetime
brunner's avatar
brunner committed
18
19
20

identifier = 'ObservationOperator'
version = '10'
brunner's avatar
brunner committed
21

brunner's avatar
brunner committed
22
23
cdo = Cdo()

brunner's avatar
brunner committed
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
################### Begin Class ObservationOperator ###################
class ObservationOperator(object):
    def __init__(self, dacycle=None):
        self.ID = identifier
        self.version = version
        self.restart_filelist = []
        self.output_filelist = []
        self.outputdir = None # Needed for opening the samples.nc files created 

        logging.info('Observation Operator object initialized: %s' % self.ID)

        if dacycle != None:
            self.dacycle = dacycle
        else:
            self.dacycle = {}

    
    def get_initial_data(self):
        """ This method places all initial data needed by an ObservationOperator in the proper folder for the model """

    def setup(self,dacycle):
        """ Perform all steps necessary to start the observation operator through a simple Run() call """

        self.dacycle = dacycle
        self.outputdir = dacycle['dir.output']

    def prepare_run(self):
        """ Prepare the running of the actual forecast model, for example compile code """

	# Define the name of the file that will contain the modeled output of each observation

        self.simulated_file = os.path.join(self.outputdir, 'samples_simulated.%s.nc' % self.dacycle['time.sample.stamp'])
        self.forecast_nmembers = int(self.dacycle['da.optimizer.nmembers'])

brunner's avatar
brunner committed
58
    def run(self,lag,dacycle):
brunner's avatar
brunner committed
59

brunner's avatar
brunner committed
60
        absolute_start_time = str((to_datetime(dacycle['abs.time.start'])).strftime('%Y-%m-%d'))
brunner's avatar
brunner committed
61
62
63
        starth = abs((to_datetime(dacycle['abs.time.start'])-dacycle['time.start']).days)*24
        endh = abs((to_datetime(dacycle['abs.time.start'])-dacycle['time.finish']).days)*24

brunner's avatar
brunner committed
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
        f = io.CT_CDF(self.simulated_file, method='create')
        logging.debug('Creating new simulated observation file in ObservationOperator (%s)' % self.simulated_file)
	
        dimid = f.createDimension('obs_num', size=None)
        dimid = ('obs_num',)
        savedict = io.std_savedict.copy() 
        savedict['name'] = "obs_num"
        savedict['dtype'] = "int"
        savedict['long_name'] = "Unique_Dataset_observation_index_number"
        savedict['units'] = ""
        savedict['dims'] = dimid
        savedict['comment'] = "Unique index number within this dataset ranging from 0 to UNLIMITED."
        f.add_data(savedict,nsets=0)

        dimmember = f.createDimension('nmembers', size=self.forecast_nmembers)
        dimmember = ('nmembers',)
        savedict = io.std_savedict.copy() 
        savedict['name'] = "flask"
        savedict['dtype'] = "float"
brunner's avatar
brunner committed
83
84
        savedict['long_name'] = "mole_fraction_of_trace_gas_in_dry_air"
        savedict['units'] = "ppm"
brunner's avatar
brunner committed
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
        savedict['dims'] = dimid + dimmember
        savedict['comment'] = "Simulated model value created by COSMO"
        f.add_data(savedict,nsets=0)

	# Open file with x,y,z,t of model samples that need to be sampled

        f_in = io.ct_read(self.dacycle['ObsOperator.inputfile'],method='read') 

	# Get simulated values and ID

        ids = f_in.get_variable('obs_num')
        obs = f_in.get_variable('observed')
        mdm = f_in.get_variable('modeldatamismatch')

        f_in.close()

        shape = (self.forecast_nmembers,mdm.size)
        model_data=np.empty(shape=shape)   # 3x7

#        self.obspack_dir = dacycle.dasystem['obspack.input.dir']

#        infile = os.path.join(self.obspack_dir, 'summary', '%s_dataset_summary.txt' % (self.obspack_id,))
#        infile = "/store/empa/em05/parsenov/obspack/summary/obspack_co2_1_GLOBALVIEWplus_v3.2_2017-11-02_dataset_summary.txt"
#        f = open(infile, 'r')
#        lines = f.readlines()
#        f.close()
#        ncfilelist = []

#        for line in lines:
#            if not line.startswith('# dataset:'): continue

#            items = line.split(':')
#            ncfile = items[1].strip()

#            ncfilelist += [ncfile]

#        for ncfile in ncfilelist:
#            infile = os.path.join(ncfile + '.nc')

brunner's avatar
brunner committed
124
125
 # UNCOMMENT FROM HERE

brunner's avatar
brunner committed
126
        for dt in rrule.rrule(rrule.HOURLY, dtstart=dacycle['time.start']+timedelta(hours=24*lag*int(dacycle['time.cycle'])), until=dacycle['time.start']+timedelta(hours=(lag+1)*24*int(dacycle['time.cycle']))):
brunner's avatar
brunner committed
127
            logging.info('Multiplying emissions with parameters for lag %d, date %s' % (lag, dt.strftime('%Y%m%d%H')))
brunner's avatar
brunner committed
128
129
            for ens in range(0,self.forecast_nmembers):
                ens = str(ens).zfill(3)
brunner's avatar
brunner committed
130
131
                cdo.setunit("'kg m-2 s-1' -expr,GPP_"+ens+"_F=CO2_GPP_F*parametermap -merge "+os.path.join(dacycle['da.bio.input'], 'gpp_%s.nc' % dt.strftime('%Y%m%d%H')), input = os.path.join(dacycle['restartmap.dir'],"parameters_gpp_lag"+str(lag)+"."+ens+".nc"), output = os.path.join(dacycle['da.bio.input'], 'ensemble', "gpp_"+ens+"_%s.nc" % dt.strftime('%Y%m%d%H')))
                cdo.setunit("'kg m-2 s-1' -expr,RESP_"+ens+"_F=CO2_RESP_F*parametermap -merge "+os.path.join(dacycle['da.bio.input'], 'ra_%s.nc' % dt.strftime('%Y%m%d%H')), input = os.path.join(dacycle['restartmap.dir'],"parameters_resp_lag"+str(lag)+"."+ens+".nc"), output = os.path.join(dacycle['da.bio.input'], 'ensemble', "ra_"+ens+"_%s.nc" % dt.strftime('%Y%m%d%H')))
brunner's avatar
brunner committed
132
133
134
            cdo.merge(input = os.path.join(dacycle['da.bio.input'], 'ensemble', "gpp_???_%s.nc" % dt.strftime('%Y%m%d%H')), output = os.path.join(dacycle['da.bio.input'], 'ensemble', "gpp_%s.nc" % dt.strftime('%Y%m%d%H')))
            cdo.merge(input = os.path.join(dacycle['da.bio.input'], 'ensemble', "ra_???_%s.nc" % dt.strftime('%Y%m%d%H')), output = os.path.join(dacycle['da.bio.input'], 'ensemble', "ra_%s.nc" % dt.strftime('%Y%m%d%H')))

brunner's avatar
brunner committed
135
        os.chdir(dacycle['da.obsoperator.home'])
brunner's avatar
brunner committed
136

brunner's avatar
brunner committed
137
138
        if os.path.exists("/scratch/snx3000/parsenov/ctdas/"+absolute_start_time+"_"+str(starth+lag*168)+"_"+str(endh+lag*168)+"/cosmo/output/"):
            os.rename("/scratch/snx3000/parsenov/ctdas/"+absolute_start_time+"_"+str(starth+lag*168)+"_"+str(endh+lag*168), "/scratch/snx3000/parsenov/ctdas/non_opt_"+dacycle['time.start'].strftime('%Y%m%d%H')+"_"+str(starth+lag*168)+"_"+str(endh+lag*168))
brunner's avatar
brunner committed
139

brunner's avatar
brunner committed
140
        os.system('python run_chain.py ctdas '+absolute_start_time+' '+str(starth+lag*168)+' '+str(endh+lag*168)+' -j meteo icbc emissions biofluxes int2lm post_int2lm cosmo')
brunner's avatar
brunner committed
141
        os.chdir(dacycle['dir.da_run'])
brunner's avatar
brunner committed
142
143
144
145
146

        args = [
            (dacycle, hstart, hstop, self.forecast_nmembers)
            for dacycle, (hstart, hstop), self.forecast_nmembers
            in zip(repeat(dacycle),
brunner's avatar
brunner committed
147
                [(starth+168*lag,endh+168*lag-1)],
brunner's avatar
brunner committed
148
149
150
                repeat(self.forecast_nmembers))
        ]

brunner's avatar
brunner committed
151
152
        with Pool(3) as pool:
            pool.starmap(self.extract_model_data, args)
brunner's avatar
brunner committed
153
154

        for i in range(0,self.forecast_nmembers):
brunner's avatar
brunner committed
155
            idx = str(i).zfill(3)
brunner's avatar
brunner committed
156
            cosmo_file = os.path.join('/store/empa/em05/parsenov/cosmo_data/model_'+idx+'_%s.nc' % dacycle['time.sample.stamp'])
brunner's avatar
brunner committed
157
            ifile = Dataset(cosmo_file, mode='r')
brunner's avatar
brunner committed
158
            model_data[i,:] = (np.squeeze(ifile.variables['CO2'][:])*29./44.01)*1E6   # in ppm
brunner's avatar
brunner committed
159
160
161
162
163
164
165
166
167
168
169
            ifile.close()

        for j,data in enumerate(zip(ids,obs,mdm)):
            f.variables['obs_num'][j] = data[0]		
            f.variables['flask'][j,:] = model_data[:,j]
        f.close()
#### WARNING ACHTUNG PAZNJA POZOR VNEMANIE data[2] is model data mismatch (=1000) by default in tools/io4.py!!! pavle


        logging.info('ObservationOperator finished successfully, output file written (%s)' % self.simulated_file)

brunner's avatar
brunner committed
170
    def run_forecast_model(self, lag, dacycle):
brunner's avatar
brunner committed
171
        self.prepare_run()
brunner's avatar
brunner committed
172
        self.run(lag, dacycle)
brunner's avatar
brunner committed
173

brunner's avatar
brunner committed
174
    def extract_model_data(self,dacycle,hstart,hstop,ensnum):
brunner's avatar
brunner committed
175

brunner's avatar
brunner committed
176
        self.dacycle = dacycle
brunner's avatar
brunner committed
177
        time_stamp = str((dacycle['time.start']+timedelta(hours=hstart)).strftime('%Y%m%d%H'))+'_'+str((dacycle['time.start']+timedelta(hours=hstop)).strftime('%Y%m%d%H'))
brunner's avatar
brunner committed
178
        abs_start_time = str((to_datetime(dacycle['abs.time.start'])).strftime('%Y%m%d%H'))
brunner's avatar
brunner committed
179

brunner's avatar
brunner committed
180
181
        cosmo_out = "/scratch/snx3000/parsenov/ctdas/"+abs_start_time+"_"+str(hstart)+"_"+str(hstop+1)+"/cosmo/output/"
        hhl_cosmo_out = "/scratch/snx3000/parsenov/ctdas/"+abs_start_time+"_0_168/cosmo/output/"
brunner's avatar
brunner committed
182
        cosmo_save = "/store/empa/em05/parsenov/cosmo_data/"
brunner's avatar
brunner committed
183
        hhl_fn = hhl_cosmo_out+'lffd'+abs_start_time+'c.nc'
brunner's avatar
brunner committed
184
185
186
187
        cdo.selname("HHL", input = hhl_fn, output = cosmo_out+"hhl.nc")

        for ens in range(0,ensnum):
            ens = str(ens).zfill(3)
brunner's avatar
brunner committed
188
            files2cat=[]
brunner's avatar
brunner committed
189
            for dt in rrule.rrule(rrule.HOURLY, dtstart=to_datetime(dacycle['abs.time.start'])+timedelta(hours=hstart), until=to_datetime(dacycle['abs.time.start'])+timedelta(hours=hstop)):
brunner's avatar
brunner committed
190
                dt=dt.strftime('%Y%m%d%H')
brunner's avatar
brunner committed
191
                logging.info('Extracting output for ens %s, time %s' % (str(ens),str(dt)))
brunner's avatar
brunner committed
192
193
                co2_in_fn = cosmo_out+'lffd'+dt+'.nc'
                co2_out_fn = cosmo_out+'CO2_'+ens+'_'+dt+'.nc'
brunner's avatar
brunner committed
194
                cdo.expr("'CO2=(CO2_BG-GPP_"+ens+"+RESP_"+ens+"+CO2_A_CH+CO2_A)/(1.-QV)'", input = "-selname,QV,CO2_BG,GPP_"+ens+",RESP_"+ens+",CO2_A_CH,CO2_A "+co2_in_fn, output = co2_out_fn)
brunner's avatar
brunner committed
195
196
197
198
                files2cat.append(co2_out_fn)

            cdo.cat(input = files2cat, output = cosmo_out+"CO2_"+ens+"_"+time_stamp+".nc")

brunner's avatar
brunner committed
199
200
            cdo.remapnn("lon=7.99_lat=46.54,", input = cosmo_out+"CO2_"+ens+"_"+time_stamp+".nc", output = cosmo_out+"CO2_jfj_"+ens+"_"+time_stamp+".nc")
            cdo.remapnn("lon=7.99_lat=46.54,", input = cosmo_out+"hhl.nc", output = cosmo_out+"hhl_jfj.nc")
brunner's avatar
brunner committed
201

brunner's avatar
brunner committed
202
203
            cdo.remapnn("lon=8.40_lat=47.48,", input = cosmo_out+"CO2_"+ens+"_"+time_stamp+".nc", output = cosmo_out+"CO2_lhw_"+ens+"_"+time_stamp+".nc")
            cdo.remapnn("lon=8.40_lat=47.48,", input = cosmo_out+"hhl.nc", output = cosmo_out+"hhl_lhw.nc")
brunner's avatar
brunner committed
204

brunner's avatar
brunner committed
205
206
            cdo.remapnn("lon=8.18_lat=47.19,", input = cosmo_out+"CO2_"+ens+"_"+time_stamp+".nc", output = cosmo_out+"CO2_brm_"+ens+"_"+time_stamp+".nc")
            cdo.remapnn("lon=8.18_lat=47.19,", input = cosmo_out+"hhl.nc", output = cosmo_out+"hhl_brm.nc")
brunner's avatar
brunner committed
207

brunner's avatar
brunner committed
208
209
            cdo.remapnn("lon=7.92_lat=47.92,", input = cosmo_out+"CO2_"+ens+"_"+time_stamp+".nc", output = cosmo_out+"CO2_ssl_"+ens+"_"+time_stamp+".nc")
            cdo.remapnn("lon=7.92_lat=47.92,", input = cosmo_out+"hhl.nc", output = cosmo_out+"hhl_ssl.nc")
brunner's avatar
brunner committed
210

brunner's avatar
brunner committed
211
            sites = ("lhw","brm","jfj","ssl")
brunner's avatar
brunner committed
212
213
214
215
216
217
218
219
220
            for s,ss in enumerate(sites):
                site_height.main(cosmo_out, str(ens), ss, time_stamp)

            cdo.intlevel("860", input = cosmo_out+"CO2_60lev_"+ens+"_lhw_"+time_stamp+".nc", output = cosmo_out+"/modelled_"+ens+"_lhw_"+time_stamp+".nc")
            cdo.intlevel("797", input = cosmo_out+"CO2_60lev_"+ens+"_brm_"+time_stamp+".nc", output = cosmo_out+"modelled_"+ens+"_brm_"+time_stamp+".nc")
            cdo.intlevel("3580", input = cosmo_out+"CO2_60lev_"+ens+"_jfj_"+time_stamp+".nc", output = cosmo_out+"modelled_"+ens+"_jfj_"+time_stamp+".nc")
            cdo.intlevel("1205", input = cosmo_out+"CO2_60lev_"+ens+"_ssl_"+time_stamp+".nc", output = cosmo_out+"modelled_"+ens+"_ssl_"+time_stamp+".nc")

            cdo.cat(input = cosmo_out+"modelled_"+ens+"_brm_"+time_stamp+".nc "+cosmo_out+"modelled_"+ens+"_jfj_"+time_stamp+".nc "+cosmo_out+"modelled_"+ens+"_lhw_"+time_stamp+".nc "+cosmo_out+"modelled_"+ens+"_ssl_"+time_stamp+".nc ", output = cosmo_save+"model_"+ens+"_"+time_stamp+".nc")
brunner's avatar
brunner committed
221
222
223
224
225
226
227
228
229
230
231
232
233
234


################### End Class ObservationOperator ###################

class RandomizerObservationOperator(ObservationOperator):
    """ This class holds methods and variables that are needed to use a random number generated as substitute
        for a true observation operator. It takes observations and returns values for each obs, with a specified 
        amount of white noise added 
    """



if __name__ == "__main__":
    pass