def process_radlw_file(self):
''' Create longwave radiation file '''
nframes = self.nframes / self.nframes_per_day # daily file
radlw_tmp = np.empty((self.ny, self.nx))
radlw_out = np.empty((nframes, self.ny, self.nx))
# open input file
fid_radlw = ioncdf.opennc(self.file_radlw)
# read coordinates and time
lon = ioncdf.readnc(fid_radlw, 'lon')
lat = ioncdf.readnc(fid_radlw, 'lat')
time = ioncdf.readnc(fid_radlw, self.name_time_radlw)
# open radlw factor
fid_factor = ioncdf.opennc(self.longwave_factor)
radlw_factor = ioncdf.readnc(fid_factor, 'radlw')
ioncdf.closenc(fid_factor)
# flip upside down for ROMS
if self.target_model == 'ROMS':
radlw_factor = radlw_factor[::-1, :]
# run the computation
correct_heat_budget = 1.0088 # attempt to close heat budget
for kt in np.arange(0, nframes):
radlw_tmp[:, :] = ioncdf.readnc_oneframe(fid_radlw,
self.name_radlw, kt)
radlw_out[kt, :, :] = (radlw_tmp[:, :] * radlw_factor[:, :] *
correct_heat_budget) * self.lsm[:, :]
# output file informations
if self.target_model == 'ROMS':
my_dict = {'varname':'lwrad_down','time_dim':'lrf_time','time_var':'lrf_time','long name':'Downwelling longwave radiation',\
'units':'W.m-2','fileout':self.output_dir + 'radlw_' + self.dataset + '_' + str(self.year) + '_ROMS.nc'}
elif self.target_model == 'NEMO':
my_dict = {'varname':'radlw','time_dim':'time','time_var':'time','long name':'Downwelling longwave radiation',\
'units':'W.m-2','fileout':self.output_dir + 'radlw_' + self.dataset + '_' + str(self.year) + '.nc'}
my_dict[
'description'] = 'DFS 5.2 (MEOM/LGGE) contact : [email protected]'
my_dict['spval'] = self.spval
my_dict['reftime'] = self.reftime
my_dict['time_valid_min'] = time.min()
my_dict['time_valid_max'] = time.max()
my_dict['var_valid_min'] = radlw_out.min()
my_dict['var_valid_max'] = radlw_out.max()
# close input file and write output
ioncdf.closenc(fid_radlw)
ioncdf.write_ncfile(lon, lat, time, radlw_out, my_dict)
# clear arrays
radlw_tmp = None
radlw_out = None
return None
def process_v10_file(self):
''' Create meridional wind file '''
v10_tmp = np.empty((self.ny, self.nx))
v10_out = np.empty((self.nframes, self.ny, self.nx))
# open input file
fid_v10 = ioncdf.opennc(self.file_v10)
# read coordinates and time
lon = ioncdf.readnc(fid_v10, 'lon')
lat = ioncdf.readnc(fid_v10, 'lat')
time = ioncdf.readnc(fid_v10, self.name_time_v10)
# open background velocity
fid_bgd = ioncdf.opennc(self.v10_background)
v10_background = ioncdf.readnc(fid_bgd, 'v10')
ioncdf.closenc(fid_bgd)
# flip upside down for ROMS
if self.target_model == 'ROMS':
v10_background = v10_background[::-1, :]
# run the computation
for kt in np.arange(0, self.nframes):
v10_tmp[:, :] = ioncdf.readnc_oneframe(fid_v10, self.name_v10, kt)
v10_out[kt, :, :] = (v10_tmp[:, :] +
v10_background[:, :]) * self.lsm[:, :]
# output file informations
if self.target_model == 'ROMS':
my_dict = {'varname':'Vwind','time_dim':'wind_time','time_var':'wind_time','long name':'Meridional wind speed at 10m',\
'units':'m/s','fileout':self.output_dir + 'v10_' + self.dataset + '_' + str(self.year) + '_ROMS.nc'}
elif self.target_model == 'NEMO':
my_dict = {'varname':'v10','time_dim':'time','time_var':'time','long name':'Meridional wind speed at 10m',\
'units':'m/s','fileout':self.output_dir + 'v10_' + self.dataset + '_' + str(self.year) + '.nc'}
my_dict[
'description'] = 'DFS 5.2 (MEOM/LGGE) contact : [email protected]'
my_dict['spval'] = self.spval
my_dict['reftime'] = self.reftime
my_dict['var_valid_min'] = v10_out.min()
my_dict['var_valid_max'] = v10_out.max()
my_dict['time_valid_min'] = time.min()
my_dict['time_valid_max'] = time.max()
# close input file and write output
ioncdf.closenc(fid_v10)
ioncdf.write_ncfile(lon, lat, time, v10_out, my_dict)
# clear arrays
v10_tmp = None
v10_out = None
return None
def process_precip_file(self):
''' Create precip file '''
nframes = self.nframes / self.nframes_per_day # daily file
precip_tmp = np.empty((self.ny, self.nx))
precip_out = np.empty((nframes, self.ny, self.nx))
# open input file
fid_precip = ioncdf.opennc(self.file_precip)
# read coordinates and time
lon = ioncdf.readnc(fid_precip, 'lon')
lat = ioncdf.readnc(fid_precip, 'lat')
time = ioncdf.readnc(fid_precip, self.name_time_precip)
# copy
for kt in np.arange(0, nframes):
precip_tmp[:, :] = ioncdf.readnc_oneframe(fid_precip,
self.name_precip, kt)
precip_out[kt, :, :] = (precip_tmp[:, :]) * self.lsm[:, :]
# output file informations
if self.target_model == 'ROMS':
my_dict = {'varname':'rain','time_dim':'rain_time','time_var':'rain_time','long name':'Total Precipitation',\
'units':'kg.m-2.s-1','fileout':self.output_dir + 'precip_' + self.dataset + '_' + str(self.year) + '_ROMS.nc'}
elif self.target_model == 'NEMO':
my_dict = {'varname':'precip','time_dim':'time','time_var':'time','long name':'Total Precipitation',\
'units':'kg.m-2.s-1','fileout':self.output_dir + 'precip_' + self.dataset + '_' + str(self.year) + '.nc'}
my_dict[
'description'] = 'DFS 5.2 (MEOM/LGGE) contact : [email protected]'
my_dict['spval'] = self.spval
my_dict['reftime'] = self.reftime
my_dict['time_valid_min'] = time.min()
my_dict['time_valid_max'] = time.max()
my_dict['var_valid_min'] = precip_out.min()
my_dict['var_valid_max'] = precip_out.max()
# close input file and write output
ioncdf.closenc(fid_precip)
ioncdf.write_ncfile(lon, lat, time, precip_out, my_dict)
# clear arrays
precip_tmp = None
precip_out = None
return None
def process_msl_file(self):
''' Create msl file '''
msl_tmp = np.empty((self.ny, self.nx))
msl_out = np.empty((self.nframes, self.ny, self.nx))
# open input file
fid_msl = ioncdf.opennc(self.file_msl)
# read coordinates and time
lon = ioncdf.readnc(fid_msl, 'lon')
lat = ioncdf.readnc(fid_msl, 'lat')
time = ioncdf.readnc(fid_msl, self.name_time_msl)
# copy
for kt in np.arange(0, self.nframes):
msl_tmp[:, :] = ioncdf.readnc_oneframe(fid_msl, self.name_msl, kt)
msl_out[kt, :, :] = (msl_tmp[:, :]) * self.lsm[:, :]
# output file informations
if self.target_model == 'ROMS':
my_dict = {'varname':'Pair','time_dim':'pair_time','time_var':'pair_time','long name':'Mean Sea Level Pressure',\
'units':'Pa','fileout':self.output_dir + 'msl_' + self.dataset + '_' + str(self.year) + '_ROMS.nc'}
elif self.target_model == 'NEMO':
my_dict = {'varname':'msl','time_dim':'time','time_var':'time','long name':'Mean Sea Level Pressure',\
'units':'Pa','fileout':self.output_dir + 'msl_' + self.dataset + '_' + str(self.year) + '.nc'}
my_dict[
'description'] = 'DFS 5.2 (MEOM/LGGE) contact : [email protected]'
my_dict['spval'] = self.spval
my_dict['reftime'] = self.reftime
my_dict['time_valid_min'] = time.min()
my_dict['time_valid_max'] = time.max()
my_dict['var_valid_min'] = msl_out.min()
my_dict['var_valid_max'] = msl_out.max()
# close input file and write output
ioncdf.closenc(fid_msl)
ioncdf.write_ncfile(lon, lat, time, msl_out, my_dict)
# clear arrays
msl_tmp = None
msl_out = None
return None
def process_tcc_file(self):
''' Create tcc file '''
tcc_tmp = np.empty((self.ny, self.nx))
tcc_out = np.empty((self.nframes, self.ny, self.nx))
# open input file
fid_tcc = ioncdf.opennc(self.file_tcc)
# read coordinates and time
lon = ioncdf.readnc(fid_tcc, 'lon')
lat = ioncdf.readnc(fid_tcc, 'lat')
time = ioncdf.readnc(fid_tcc, self.name_time_tcc)
# copy
for kt in np.arange(0, self.nframes):
tcc_tmp[:, :] = ioncdf.readnc_oneframe(fid_tcc, self.name_tcc, kt)
tcc_out[kt, :, :] = (tcc_tmp[:, :]) * self.lsm[:, :]
# output file informations
if self.target_model == 'ROMS':
my_dict = {'varname':'Cloud','time_dim':'cloud_time','time_var':'cloud_time','long name':'Total Cloud Cover',\
'units':'','fileout':self.output_dir + 'tcc_' + self.dataset + '_' + str(self.year) + '_ROMS.nc'}
elif self.target_model == 'NEMO':
my_dict = {'varname':'tcc','time_dim':'time','time_var':'time','long name':'Total Cloud Cover',\
'units':'','fileout':self.output_dir + 'tcc_' + self.dataset + '_' + str(self.year) + '.nc'}
my_dict[
'description'] = 'DFS 5.2 (MEOM/LGGE) contact : [email protected]'
my_dict['spval'] = self.spval
my_dict['reftime'] = self.reftime
my_dict['time_valid_min'] = time.min()
my_dict['time_valid_max'] = time.max()
my_dict['var_valid_min'] = tcc_out.min()
my_dict['var_valid_max'] = tcc_out.max()
# close input file and write output
ioncdf.closenc(fid_tcc)
ioncdf.write_ncfile(lon, lat, time, tcc_out, my_dict)
# clear arrays
tcc_tmp = None
tcc_out = None
return None
def process_q2_file(self):
''' Create q2 file '''
t2_old = np.empty((self.ny, self.nx))
t2_new = np.empty((self.ny, self.nx))
msl = np.empty((self.ny, self.nx))
q2_tmp = np.empty((self.ny, self.nx))
q2_out = np.empty((self.nframes, self.ny, self.nx))
# open input file
fid_q2 = ioncdf.opennc(self.file_q2)
# read coordinates and time
lon = ioncdf.readnc(fid_q2, 'lon')
lat = ioncdf.readnc(fid_q2, 'lat')
time = ioncdf.readnc(fid_q2, self.name_time_q2)
# We need original t2 and msl from ERAinterim
fid_t2old = ioncdf.opennc(self.file_t2)
fid_msl = ioncdf.opennc(self.file_msl)
fid_t2new = ioncdf.opennc(self.file_t2_new)
# run the computation
for kt in np.arange(0, self.nframes):
# read all the fields
# ROMS has t2 in degC, others in Kelvin - qsat_from_t2_and_msl expects Celcius
if self.target_model == 'ROMS':
t2_old[:, :] = ioncdf.readnc_oneframe(fid_t2old, self.name_t2,
kt)
t2_new[:, :] = ioncdf.readnc_oneframe(fid_t2new, self.name_t2,
kt)
else:
t2_old[:, :] = ioncdf.readnc_oneframe(fid_t2old, self.name_t2,
kt) - 273.15
t2_new[:, :] = ioncdf.readnc_oneframe(fid_t2new, self.name_t2,
kt) - 273.15
msl[:, :] = ioncdf.readnc_oneframe(fid_msl, self.name_msl, kt)
q2_tmp[:, :] = ioncdf.readnc_oneframe(fid_q2, self.name_q2, kt)
# compute humidity at saturation
q_sat_new = humidity_toolbox.qsat_from_t2_and_msl(t2_new, msl)
q_sat_old = humidity_toolbox.qsat_from_t2_and_msl(t2_old, msl)
# compute new specific humidity
q2_tmp[:, :] = (q2_tmp[:, :] * q_sat_new / q_sat_old)
q2_tmp[np.where(self.lsm[:, :] == 0)] = 0.
q2_out[kt, :, :] = q2_tmp[:, :]
# output file informations
if self.target_model == 'ROMS':
my_dict = {'varname':'Qair','time_dim':'qair_time','time_var':'qair_time','long name':'Specific Humidity',\
'units':'kg/kg','fileout':self.output_dir + 'q2_' + self.dataset + '_' + str(self.year) + '_ROMS.nc'}
elif self.target_model == 'NEMO':
my_dict = {'varname':'q2','time_dim':'time','time_var':'time','long name':'Specific Humidity',\
'units':'kg/kg','fileout':self.output_dir + 'q2_' + self.dataset + '_' + str(self.year) + '.nc'}
my_dict[
'description'] = 'DFS 5.2 (MEOM/LGGE) contact : [email protected]'
my_dict['spval'] = self.spval
my_dict['reftime'] = self.reftime
my_dict['time_valid_min'] = time.min()
my_dict['time_valid_max'] = time.max()
my_dict['var_valid_min'] = q2_out.min()
my_dict['var_valid_max'] = q2_out.max()
# close input file and write output
ioncdf.closenc(fid_q2)
ioncdf.write_ncfile(lon, lat, time, q2_out, my_dict)
# clear arrays
q2_tmp = None
q2_out = None
return None
def process_t2_file(self):
''' Create Air temperature file '''
t2_tmp = np.empty((self.ny, self.nx))
t2_out = np.empty((self.nframes, self.ny, self.nx))
# open input file
fid_t2 = ioncdf.opennc(self.file_t2)
# read coordinates and time
lon = ioncdf.readnc(fid_t2, 'lon')
lat = ioncdf.readnc(fid_t2, 'lat')
time = ioncdf.readnc(fid_t2, self.name_time_t2)
#------- Antarctic correction --------
# make sure that we use latitude from south to north
if self.target_model == 'ROMS':
lat_s2n = lat
else:
lat_s2n = lat[::-1]
value_south = -2.0 # we remove 2 degrees C
lattrans1 = -60 # start linear transition to correction at 60S
lattrans2 = -75 # correction is at full value at 75S
jtrans1 = (np.abs(lat_s2n - lattrans1)).argmin()
jtrans2 = (np.abs(lat_s2n - lattrans2)).argmin()
correction_south = np.zeros((self.ny, self.nx))
correction_south[0:jtrans2, :] = value_south
for jj in np.arange(jtrans2, jtrans1):
correction_south[jj, :] = value_south * float(
jj - jtrans1) / float(jtrans2 - jtrans1)
if not self.target_model == 'ROMS':
correction_south = correction_south[::-1, :]
#------- Arctic correction --------
correction_north = np.zeros((self.ny, self.nx))
fid_off = ioncdf.opennc(self.t2_poles_offset)
monthly_offset_poles = ioncdf.readnc(fid_off, 'Tair')
ioncdf.closenc(fid_off)
fid_ifra = ioncdf.opennc(self.ice_nsidc)
monthly_ice_fraction = ioncdf.readnc(fid_ifra, 'ifrac')
ioncdf.closenc(fid_ifra)
# poles correction north of 70 N
lattrans3 = 65 # correction poles transition starts at 65, full at 70N
jtrans3 = (np.abs(lat_s2n - lattrans3)).argmin()
val_oce = -0.7
# run the computation
for kt in np.arange(0, self.nframes):
t2_tmp[:, :] = ioncdf.readnc_oneframe(fid_t2, self.name_t2, kt)
#----- arctic correction ------
# interp monthly offset to daily value
this_day = 1 + (kt / self.nframes_per_day)
this_day_weights = self.time_interp_weights(this_day)
this_day_offset = self.interp_data_to_day(monthly_offset_poles,
this_day_weights)
this_day_ifrac = self.interp_data_to_day(monthly_ice_fraction,
this_day_weights)
# compute artic correction
correction_north[jtrans3:,:] = this_day_offset[jtrans3:,:]*this_day_ifrac[jtrans3:,:] + \
val_oce*(1. -this_day_ifrac[jtrans3:,:])
if not self.target_model == 'ROMS':
correction_north[:, :] = correction_north[::-1, :]
#------ add northern and southern hemisphere correction ------
t2_out[kt, :, :] = (t2_tmp[:, :] + correction_north[:, :] +
correction_south[:, :]) * self.lsm[:, :]
# output file informations
if self.target_model == 'ROMS':
my_dict = {'varname':'Tair','time_dim':'tair_time','time_var':'tair_time','long name':'Air Temperature at 2m',\
'units':'degC','fileout':self.output_dir + 't2_' + self.dataset + '_' + str(self.year) + '_ROMS.nc'}
elif self.target_model == 'NEMO':
my_dict = {'varname':'t2','time_dim':'time','time_var':'time','long name':'Air Temperature at 2m',\
'units':'K','fileout':self.output_dir + 't2_' + self.dataset + '_' + str(self.year) + '.nc'}
my_dict[
'description'] = 'DFS 5.2 (MEOM/LGGE) contact : [email protected]'
my_dict['spval'] = self.spval
my_dict['reftime'] = self.reftime
my_dict['var_valid_min'] = t2_out.min()
my_dict['var_valid_max'] = t2_out.max()
my_dict['time_valid_min'] = time.min()
my_dict['time_valid_max'] = time.max()
# close input file and write output
ioncdf.closenc(fid_t2)
ioncdf.write_ncfile(lon, lat, time, t2_out, my_dict)
# clear arrays
t2_tmp = None
t2_out = None
# save this for q2
self.file_t2_new = my_dict['fileout']
return None
def __init__(self, dict_input, dict_datafiles):
# constants
self.rho_w = 1000.
self.nsec_per_day = 86400.
self.reftime = dt.datetime(1900, 1, 1, 0, 0)
self.spval = 1.0e+15
self.dataset = 'DFS5.2'
# read inputs
self.dict_input = dict_input
for key in dict_input:
exec('self.' + key + '=dict_input[key]')
for key in dict_datafiles:
exec('self.' + key + '=dict_datafiles[key]')
# set time related stuff
if calendar.isleap(self.year):
self.ndays = 366
self.days_in_month = np.array(
[31., 29., 31., 30., 31., 30., 31., 31., 30., 31., 30., 31.])
else:
self.ndays = 365
self.days_in_month = np.array(
[31., 28., 31., 30., 31., 30., 31., 31., 30., 31., 30., 31.])
# forcing set freq
if self.freq == '3h':
self.nframes_per_day = 8
elif self.freq == '6h':
self.nframes_per_day = 4
self.nframes = self.ndays * self.nframes_per_day
print self.year, 'has', self.nframes, 'frames'
# variable names
if self.target_model == 'ROMS':
self.name_t2 = 'Tair'
self.name_time_t2 = 'tair_time'
self.name_q2 = 'Qair'
self.name_time_q2 = 'qair_time'
self.name_u10 = 'Uwind'
self.name_time_u10 = 'wind_time'
self.name_v10 = 'Vwind'
self.name_time_v10 = 'wind_time'
self.name_radsw = 'swrad'
self.name_time_radsw = 'srf_time'
self.name_radlw = 'lwrad_down'
self.name_time_radlw = 'lrf_time'
self.name_precip = 'rain'
self.name_time_precip = 'rain_time'
self.name_snow = 'rain'
self.name_time_snow = 'rain_time'
self.name_msl = 'Pair'
self.name_time_msl = 'pair_time'
else:
self.name_t2 = 't2'
self.name_time_t2 = 'time'
self.name_q2 = 'q2'
self.name_time_q2 = 'time'
self.name_u10 = 'u10'
self.name_time_u10 = 'time'
self.name_v10 = 'v10'
self.name_time_v10 = 'time'
self.name_radsw = 'radsw'
self.name_time_radsw = 'time'
self.name_radlw = 'radlw'
self.name_time_radlw = 'time'
self.name_precip = 'precip'
self.name_time_precip = 'time'
self.name_snow = 'snow'
self.name_time_snow = 'time'
self.name_msl = 'msl'
self.name_time_msl = 'time'
self.name_tcc = 'tcc'
self.name_time_tcc = 'time'
fid_lsm = ioncdf.opennc(self.mask)
self.lsm = ioncdf.readnc(fid_lsm, 'lsm')
ioncdf.closenc(fid_lsm)
return None