observationoperator.py 12.4 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
from da.cosmo.icbc4ctdas import ct
brunner's avatar
brunner committed
16
17
from itertools import repeat
from multiprocessing import Pool
brunner's avatar
brunner committed
18
from da.tools.general import to_datetime
brunner's avatar
brunner committed
19
20
21

identifier = 'ObservationOperator'
version = '10'
brunner's avatar
brunner committed
22

brunner's avatar
brunner committed
23
24
cdo = Cdo()

brunner's avatar
brunner committed
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
################### 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
59
60
    def run(self,lag,dacycle,statevector):
        members = statevector.ensemble_members[lag]
brunner's avatar
brunner committed
61
        absolute_start_time = str((to_datetime(dacycle['abs.time.start'])).strftime('%Y%m%d%H'))
brunner's avatar
brunner committed
62
63
64
        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
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
        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
84
85
        savedict['long_name'] = "mole_fraction_of_trace_gas_in_dry_air"
        savedict['units'] = "ppm"
brunner's avatar
brunner committed
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
        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
125
126
 # UNCOMMENT FROM HERE

brunner's avatar
brunner committed
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
#        co2_bg = np.empty(self.forecast_nmembers)
#
 #       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']))):
  #          logging.info('Multiplying emissions with parameters for lag %d, date %s' % (lag, dt.strftime('%Y%m%d%H')))
   #         for ens in range(0,self.forecast_nmembers):
    #            dthh = dt.strftime('%H')
     #           co2_bg[ens] = members[ens].param_values[-1]
      #          ens = str(ens).zfill(3)
       #         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')))
         #   logging.info('Background CO2 params are (%s)' % co2_bg)
          #  if dthh=='00':
           #     ct(dt.strftime('%Y%m%d'), co2_bg)
#
 #           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
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
        os.chdir(dacycle['da.obsoperator.home'])

        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))

        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')
        os.chdir(dacycle['dir.da_run'])

        args = [
            (dacycle, hstart, hstop, self.forecast_nmembers)
            for dacycle, (hstart, hstop), self.forecast_nmembers
            in zip(repeat(dacycle),
                [(starth+168*lag,endh+168*lag-1)],
                repeat(self.forecast_nmembers))
        ]

        with Pool(3) as pool:
            pool.starmap(self.extract_model_data, args)

brunner's avatar
brunner committed
163
        for i in range(0,self.forecast_nmembers):
brunner's avatar
brunner committed
164
            idx = str(i).zfill(3)
brunner's avatar
brunner committed
165
            cosmo_file = os.path.join('/store/empa/em05/parsenov/cosmo_data/model_'+idx+'_%s.nc' % dacycle['time.sample.stamp'])
brunner's avatar
brunner committed
166
            ifile = Dataset(cosmo_file, mode='r')
brunner's avatar
brunner committed
167
            model_data[i,:] = (np.squeeze(ifile.variables['CO2'][:])*29./44.01)*1E6   # in ppm
brunner's avatar
brunner committed
168
169
170
171
172
173
174
175
176
177
178
            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
179
    def run_forecast_model(self, lag, dacycle, statevector):
brunner's avatar
brunner committed
180
        self.prepare_run()
brunner's avatar
brunner committed
181
        self.run(lag, dacycle, statevector)
brunner's avatar
brunner committed
182

brunner's avatar
brunner committed
183
    def extract_model_data(self,dacycle,hstart,hstop,ensnum):
brunner's avatar
brunner committed
184

brunner's avatar
brunner committed
185
        self.dacycle = dacycle
brunner's avatar
brunner committed
186
187
     #   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'))
        time_stamp = dacycle['time.sample.stamp']
brunner's avatar
brunner committed
188
        abs_start_time = str((to_datetime(dacycle['abs.time.start'])).strftime('%Y%m%d%H'))
brunner's avatar
brunner committed
189

brunner's avatar
brunner committed
190
191
        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
192
        cosmo_save = "/store/empa/em05/parsenov/cosmo_data/"
brunner's avatar
brunner committed
193
        hhl_fn = hhl_cosmo_out+'lffd'+abs_start_time+'c.nc'
brunner's avatar
brunner committed
194
195
196
197
        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
198
            files2cat=[]
brunner's avatar
brunner committed
199
            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
200
                dt=dt.strftime('%Y%m%d%H')
brunner's avatar
brunner committed
201
                logging.info('Extracting output for ens %s, time %s' % (str(ens),str(dt)))
brunner's avatar
brunner committed
202
203
                co2_in_fn = cosmo_out+'lffd'+dt+'.nc'
                co2_out_fn = cosmo_out+'CO2_'+ens+'_'+dt+'.nc'
brunner's avatar
brunner committed
204
                cdo.expr("'CO2=(BG_"+ens+"-GPP_"+ens+"+RESP_"+ens+"+CO2_A_CH+CO2_A)/(1.-QV)'", input = "-selname,QV,BG_"+ens+",GPP_"+ens+",RESP_"+ens+",CO2_A_CH,CO2_A "+co2_in_fn, output = co2_out_fn)
brunner's avatar
brunner committed
205
206
207
208
                files2cat.append(co2_out_fn)

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

brunner's avatar
brunner committed
209
210
            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
211

brunner's avatar
brunner committed
212
213
            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
214

brunner's avatar
brunner committed
215
216
            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
217

brunner's avatar
brunner committed
218
219
            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
220

brunner's avatar
brunner committed
221
            sites = ("lhw","brm","jfj","ssl")
brunner's avatar
brunner committed
222
223
224
            for s,ss in enumerate(sites):
                site_height.main(cosmo_out, str(ens), ss, time_stamp)

brunner's avatar
brunner committed
225
            cdo.intlevel("860", input = cosmo_out+"CO2_60lev_"+ens+"_lhw_"+time_stamp+".nc", output = cosmo_out+"modelled_"+ens+"_lhw_"+time_stamp+".nc")
brunner's avatar
brunner committed
226
227
228
229
230
            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
231
232
233
234
235
236
237
238
239
240
241
242
243
244


################### 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