def main(stream=False):
# Get options
args = options()
nw = args.number_of_workers
if not os.path.isdir('PNG/HT'):
print('Creating a directory to place figures (PNG/HT)... \n')
os.system('mkdir -p PNG/HT')
if not os.path.isdir('ncfiles'):
print('Creating a directory to place figures (ncfiles)... \n')
os.system('mkdir ncfiles')
# Read in the yaml file
diag_config_yml = yaml.load(open(args.diag_config_yml_path,'r'), Loader=yaml.Loader)
# Create the case instance
dcase = DiagsCase(diag_config_yml['Case'])
args.case_name = dcase.casename
args.savefigs = True; args.outdir = 'PNG/HT'
RUNDIR = dcase.get_value('RUNDIR')
print('Run directory is:', RUNDIR)
print('Casename is:', dcase.casename)
print('Variables to be processed:', args.variables)
print('Number of workers to be used:', nw)
# set avg dates
avg = diag_config_yml['Avg']
if not args.start_date : args.start_date = avg['start_date']
if not args.end_date : args.end_date = avg['end_date']
# read grid info
grd = MOM6grid(RUNDIR+'/'+dcase.casename+'.mom6.static.nc')
depth = grd.depth_ocean
# remote Nan's, otherwise genBasinMasks won't work
depth[np.isnan(depth)] = 0.0
basin_code = m6toolbox.genBasinMasks(grd.geolon, grd.geolat, depth)
parallel, cluster, client = m6toolbox.request_workers(nw)
print('Reading dataset...')
startTime = datetime.now()
variables = args.variables
def preprocess(ds):
''' Compute montly averages and return the dataset with variables'''
for var in variables:
print('Processing {}'.format(var))
if var not in ds.variables:
print('WARNING: ds does not have variable {}. Creating dataarray with zeros'.format(var))
jm, im = grd.geolat.shape
tm = len(ds.time)
da = xr.DataArray(np.zeros((tm, jm, im)), dims=['time','yq','xh'], \
coords={'yq' : grd.yq, 'xh' : grd.xh, 'time' : ds.time}).rename(var)
ds = xr.merge([ds, da])
#return ds[variables].resample(time="1Y", closed='left', \
# keep_attrs=True).mean(dim='time', keep_attrs=True)
return ds[variables]
if parallel:
ds = xr.open_mfdataset(RUNDIR+'/'+dcase.casename+'.mom6.hm_*.nc', \
parallel=True, data_vars='minimal', chunks={'time': 12},\
coords='minimal', compat='override', preprocess=preprocess)
else:
ds = xr.open_mfdataset(RUNDIR+'/'+dcase.casename+'.mom6.hm_*.nc', \
data_vars='minimal', coords='minimal', compat='override', \
preprocess=preprocess)
print('Time elasped: ', datetime.now() - startTime)
print('Selecting data between {} and {}...'.format(args.start_date, args.end_date))
startTime = datetime.now()
ds_sel = ds.sel(time=slice(args.start_date, args.end_date))
print('Time elasped: ', datetime.now() - startTime)
print('Computing yearly means...')
startTime = datetime.now()
ds_sel = ds_sel.resample(time="1Y", closed='left',keep_attrs=True).mean('time',keep_attrs=True)
print('Time elasped: ', datetime.now() - startTime)
print('Computing time mean...')
startTime = datetime.now()
ds_sel = ds_sel.mean('time').load()
print('Time elasped: ', datetime.now() - startTime)
if parallel:
print('Releasing workers...')
client.close(); cluster.close()
varName = 'T_ady_2d'
print('Saving netCDF files...')
attrs = {'description': 'Time-mean poleward heat transport by components ', 'units': ds[varName].units,
'start_date': args.start_date, 'end_date': args.end_date, 'casename': dcase.casename}
m6toolbox.add_global_attrs(ds_sel,attrs)
ds_sel.to_netcdf('ncfiles/'+dcase.casename+'_heat_transport.nc')
# create a ndarray subclass
class C(np.ndarray): pass
if varName in ds.variables:
tmp = np.ma.masked_invalid(ds_sel[varName].values)
tmp = tmp[:].filled(0.)
advective = tmp.view(C)
advective.units = ds[varName].units
else:
raise Exception('Could not find "T_ady_2d" in file "%s"'%(args.infile+args.monthly))
varName = 'T_diffy_2d'
if varName in ds.variables:
tmp = np.ma.masked_invalid(ds_sel[varName].values)
tmp = tmp[:].filled(0.)
diffusive = tmp.view(C)
diffusive.units = ds[varName].units
else:
diffusive = None
warnings.warn('Diffusive temperature term not found. This will result in an underestimation of the heat transport.')
varName = 'T_lbd_diffy_2d'
if varName in ds.variables:
tmp = np.ma.masked_invalid(ds_sel[varName].values)
tmp = tmp[:].filled(0.)
lbd = tmp.view(C)
#lbd.units = ds[varName].units
else:
lbd = None
warnings.warn('Lateral boundary mixing term not found. This will result in an underestimation of the heat transport.')
plt_heat_transport_model_vs_obs(advective, diffusive, lbd, basin_code, grd, args)
return
def main():
# Get options
args = options()
nw = args.number_of_workers
if not os.path.isdir('PNG/MOC'):
print('Creating a directory to place figures (PNG/MOC)... \n')
os.system('mkdir -p PNG/MOC')
if not os.path.isdir('ncfiles'):
print('Creating a directory to place figures (ncfiles)... \n')
os.system('mkdir ncfiles')
# Read in the yaml file
diag_config_yml = yaml.load(open(args.diag_config_yml_path, 'r'),
Loader=yaml.Loader)
# Create the case instance
dcase = DiagsCase(diag_config_yml['Case'])
args.case_name = dcase.casename
args.savefigs = True
args.outdir = 'PNG/MOC/'
RUNDIR = dcase.get_value('RUNDIR')
print('Run directory is:', RUNDIR)
print('Casename is:', dcase.casename)
print('Number of workers to be used:', nw)
# set avg dates
avg = diag_config_yml['Avg']
if not args.start_date: args.start_date = avg['start_date']
if not args.end_date: args.end_date = avg['end_date']
# read grid info
grd = MOM6grid(RUNDIR + '/' + dcase.casename + '.mom6.static.nc')
depth = grd.depth_ocean
# remote Nan's, otherwise genBasinMasks won't work
depth[np.isnan(depth)] = 0.0
basin_code = m6toolbox.genBasinMasks(grd.geolon, grd.geolat, depth)
parallel, cluster, client = m6toolbox.request_workers(nw)
print('Reading {} dataset...'.format(args.file_name))
startTime = datetime.now()
# load data
def preprocess(ds):
variables = ['vmo', 'vhml', 'vhGM']
for v in variables:
if v not in ds.variables:
ds[v] = xr.zeros_like(ds.vo)
return ds[variables]
if parallel:
ds = xr.open_mfdataset(
RUNDIR + '/' + dcase.casename + args.file_name,
parallel=True,
combine="nested", # concatenate in order of files
concat_dim="time", # concatenate along time
preprocess=preprocess,
).chunk({"time": 12})
else:
ds = xr.open_mfdataset(RUNDIR+'/'+dcase.casename+args.file_name, data_vars='minimal', \
coords='minimal', compat='override', preprocess=preprocess)
print('Time elasped: ', datetime.now() - startTime)
# compute yearly means first since this will be used in the time series
print('Computing yearly means...')
startTime = datetime.now()
ds_yr = ds.resample(time="1Y", closed='left').mean('time')
print('Time elasped: ', datetime.now() - startTime)
print('Selecting data between {} and {}...'.format(args.start_date,
args.end_date))
startTime = datetime.now()
ds_sel = ds_yr.sel(time=slice(args.start_date, args.end_date))
print('Time elasped: ', datetime.now() - startTime)
print('Computing time mean...')
startTime = datetime.now()
ds_mean = ds_sel.mean('time').compute()
print('Time elasped: ', datetime.now() - startTime)
# create a ndarray subclass
class C(np.ndarray):
pass
varName = 'vmo'
conversion_factor = 1.e-9
tmp = np.ma.masked_invalid(ds_mean[varName].values)
tmp = tmp[:].filled(0.)
VHmod = tmp.view(C)
VHmod.units = ds[varName].units
Zmod = m6toolbox.get_z(ds, depth, varName) # same here
if args.case_name != '': case_name = args.case_name
else: case_name = ''
# Global MOC
m6plot.setFigureSize([16, 9], 576, debug=False)
axis = plt.gca()
cmap = plt.get_cmap('dunnePM')
zg = Zmod.min(axis=-1)
psiPlot = MOCpsi(VHmod) * conversion_factor
psiPlot = 0.5 * (psiPlot[0:-1, :] + psiPlot[1::, :])
yyg = grd.geolat_c[:, :].max(axis=-1) + 0 * zg
ci = m6plot.pmCI(0., 40., 5.)
plotPsi(
yyg, zg, psiPlot, ci, 'Global MOC [Sv],' + 'averaged between ' +
args.start_date + ' and ' + args.end_date)
plt.xlabel(r'Latitude [$\degree$N]')
plt.suptitle(case_name)
plt.gca().invert_yaxis()
findExtrema(yyg, zg, psiPlot, max_lat=-30.)
findExtrema(yyg, zg, psiPlot, min_lat=25., min_depth=250.)
findExtrema(yyg, zg, psiPlot, min_depth=2000., mult=-1.)
objOut = args.outdir + str(case_name) + '_MOC_global.png'
plt.savefig(objOut)
if 'zl' in ds:
zl = ds.zl.values
elif 'z_l' in ds:
zl = ds.z_l.values
else:
raise ValueError("Dataset does not have vertical coordinate zl or z_l")
# create dataset to store results
moc = xr.Dataset(data_vars={
'moc': (('zl', 'yq'), psiPlot),
'amoc': (('zl', 'yq'), np.zeros((psiPlot.shape))),
'moc_FFM': (('zl', 'yq'), np.zeros((psiPlot.shape))),
'moc_GM': (('zl', 'yq'), np.zeros((psiPlot.shape))),
'amoc_45': (('time'), np.zeros((ds_yr.time.shape))),
'moc_GM_ACC': (('time'), np.zeros((ds_yr.time.shape))),
'amoc_26': (('time'), np.zeros((ds_yr.time.shape)))
},
coords={
'zl': zl,
'yq': ds.yq,
'time': ds_yr.time
})
attrs = {
'description': 'MOC time-mean sections and time-series',
'units': 'Sv',
'start_date': avg['start_date'],
'end_date': avg['end_date'],
'casename': dcase.casename
}
m6toolbox.add_global_attrs(moc, attrs)
# Atlantic MOC
m6plot.setFigureSize([16, 9], 576, debug=False)
cmap = plt.get_cmap('dunnePM')
m = 0 * basin_code
m[(basin_code == 2) | (basin_code == 4) | (basin_code == 6) |
(basin_code == 7) | (basin_code == 8)] = 1
ci = m6plot.pmCI(0., 22., 2.)
z = (m * Zmod).min(axis=-1)
psiPlot = MOCpsi(VHmod,
vmsk=m * np.roll(m, -1, axis=-2)) * conversion_factor
psiPlot = 0.5 * (psiPlot[0:-1, :] + psiPlot[1::, :])
yy = grd.geolat_c[:, :].max(axis=-1) + 0 * z
plotPsi(
yy, z, psiPlot, ci, 'Atlantic MOC [Sv],' + 'averaged between ' +
args.start_date + ' and ' + args.end_date)
plt.xlabel(r'Latitude [$\degree$N]')
plt.suptitle(case_name)
plt.gca().invert_yaxis()
findExtrema(yy, z, psiPlot, min_lat=26.5, max_lat=27.,
min_depth=250.) # RAPID
findExtrema(yy, z, psiPlot, max_lat=-33.)
findExtrema(yy, z, psiPlot)
findExtrema(yy, z, psiPlot, min_lat=5.)
objOut = args.outdir + str(case_name) + '_MOC_Atlantic.png'
plt.savefig(objOut, format='png')
moc['amoc'].data = psiPlot
print('Plotting AMOC profile at 26N...')
rapid_vertical = xr.open_dataset(
'/glade/work/gmarques/cesm/datasets/RAPID/moc_vertical.nc')
fig, ax = plt.subplots(nrows=1, ncols=1)
ax.plot(rapid_vertical.stream_function_mar.mean('time'),
rapid_vertical.depth,
'k',
label='RAPID')
ax.plot(moc['amoc'].sel(yq=26, method='nearest'), zl, label=case_name)
ax.legend()
plt.gca().invert_yaxis()
plt.grid()
ax.set_xlabel('AMOC @ 26N [Sv]')
ax.set_ylabel('Depth [m]')
objOut = args.outdir + str(case_name) + '_MOC_profile_26N.png'
plt.savefig(objOut, format='png')
print('Computing time series...')
startTime = datetime.now()
# time-series
dtime = ds_yr.time
amoc_26 = np.zeros(len(dtime))
amoc_45 = np.zeros(len(dtime))
moc_GM_ACC = np.zeros(len(dtime))
if args.debug: startTime = datetime.now()
# loop in time
for t in range(len(dtime)):
tmp = np.ma.masked_invalid(ds_yr[varName][t, :].values)
tmp = tmp[:].filled(0.)
# m is still Atlantic ocean
psi = MOCpsi(tmp, vmsk=m * np.roll(m, -1, axis=-2)) * conversion_factor
psi = 0.5 * (psi[0:-1, :] + psi[1::, :])
amoc_26[t] = findExtrema(yy,
z,
psi,
min_lat=26.,
max_lat=27.,
plot=False,
min_depth=250.)
amoc_45[t] = findExtrema(yy,
z,
psi,
min_lat=44.,
max_lat=46.,
plot=False,
min_depth=250.)
tmp_GM = np.ma.masked_invalid(ds_yr['vhGM'][t, :].values)
tmp_GM = tmp_GM[:].filled(0.)
psiGM = MOCpsi(tmp_GM) * conversion_factor
psiGM = 0.5 * (psiGM[0:-1, :] + psiGM[1::, :])
moc_GM_ACC[t] = findExtrema(yyg,
zg,
psiGM,
min_lat=-65.,
max_lat=-30,
mult=-1.,
plot=False)
print('Time elasped: ', datetime.now() - startTime)
# add dataarays to the moc dataset
moc['amoc_26'].data = amoc_26
moc['amoc_45'].data = amoc_45
moc['moc_GM_ACC'].data = moc_GM_ACC
if parallel:
print('Releasing workers ...')
client.close()
cluster.close()
print('Plotting...')
# load AMOC time series data (5th) cycle used in Danabasoglu et al., doi:10.1016/j.ocemod.2015.11.007
path = '/glade/p/cesm/omwg/amoc/COREII_AMOC_papers/papers/COREII.variability/data.original/'
amoc_core_26 = xr.open_dataset(path + 'AMOCts.cyc5.26p5.nc')
# load AMOC from POP JRA-55
amoc_pop_26 = xr.open_dataset(
'/glade/u/home/bryan/MOM6-modeloutputanalysis/'
'AMOC_series_26n.g210.GIAF_JRA.v13.gx1v7.01.nc')
# load RAPID time series
rapid = xr.open_dataset(
'/glade/work/gmarques/cesm/datasets/RAPID/moc_transports.nc').resample(
time="1Y", closed='left', keep_attrs=True).mean('time',
keep_attrs=True)
# plot
fig = plt.figure(figsize=(12, 6))
plt.plot(np.arange(len(moc.time)) + 1958.5,
moc['amoc_26'].values,
color='k',
label=case_name,
lw=2)
# core data
core_mean = amoc_core_26['MOC'].mean(axis=0).data
core_std = amoc_core_26['MOC'].std(axis=0).data
plt.plot(amoc_core_26.time,
core_mean,
'k',
label='CORE II (group mean)',
color='#1B2ACC',
lw=1)
plt.fill_between(amoc_core_26.time,
core_mean - core_std,
core_mean + core_std,
alpha=0.25,
edgecolor='#1B2ACC',
facecolor='#089FFF')
# pop data
plt.plot(np.arange(len(amoc_pop_26.time)) + 1958.5,
amoc_pop_26.AMOC_26n.values,
color='r',
label='POP',
lw=1)
# rapid
plt.plot(np.arange(len(rapid.time)) + 2004.5,
rapid.moc_mar_hc10.values,
color='green',
label='RAPID',
lw=1)
plt.title('AMOC @ 26 $^o$ N', fontsize=16)
plt.ylim(5, 20)
plt.xlim(1948, 1958.5 + len(moc.time))
plt.xlabel('Time [years]', fontsize=16)
plt.ylabel('Sv', fontsize=16)
plt.legend(fontsize=13, ncol=2)
objOut = args.outdir + str(case_name) + '_MOC_26N_time_series.png'
plt.savefig(objOut, format='png')
amoc_core_45 = xr.open_dataset(path + 'AMOCts.cyc5.45.nc')
amoc_pop_45 = xr.open_dataset(
'/glade/u/home/bryan/MOM6-modeloutputanalysis/'
'AMOC_series_45n.g210.GIAF_JRA.v13.gx1v7.01.nc')
# plot
fig = plt.figure(figsize=(12, 6))
plt.plot(np.arange(len(moc.time)) + 1958.5,
moc['amoc_45'],
color='k',
label=case_name,
lw=2)
# core data
core_mean = amoc_core_45['MOC'].mean(axis=0).data
core_std = amoc_core_45['MOC'].std(axis=0).data
plt.plot(amoc_core_45.time,
core_mean,
'k',
label='CORE II (group mean)',
color='#1B2ACC',
lw=2)
plt.fill_between(amoc_core_45.time,
core_mean - core_std,
core_mean + core_std,
alpha=0.25,
edgecolor='#1B2ACC',
facecolor='#089FFF')
# pop data
plt.plot(np.arange(len(amoc_pop_45.time)) + 1958.5,
amoc_pop_45.AMOC_45n.values,
color='r',
label='POP',
lw=1)
plt.title('AMOC @ 45 $^o$ N', fontsize=16)
plt.ylim(5, 20)
plt.xlim(1948, 1958 + len(moc.time))
plt.xlabel('Time [years]', fontsize=16)
plt.ylabel('Sv', fontsize=16)
plt.legend(fontsize=14)
objOut = args.outdir + str(case_name) + '_MOC_45N_time_series.png'
plt.savefig(objOut, format='png')
# Submesoscale-induced Global MOC
class C(np.ndarray):
pass
varName = 'vhml'
conversion_factor = 1.e-9
tmp = np.ma.masked_invalid(ds_mean[varName].values)
tmp = tmp[:].filled(0.)
VHml = tmp.view(C)
VHml.units = ds[varName].units
Zmod = m6toolbox.get_z(ds, depth, varName) # same here
m6plot.setFigureSize([16, 9], 576, debug=False)
axis = plt.gca()
cmap = plt.get_cmap('dunnePM')
z = Zmod.min(axis=-1)
psiPlot = MOCpsi(VHml) * conversion_factor
psiPlot = 0.5 * (psiPlot[0:-1, :] + psiPlot[1::, :])
yy = grd.geolat_c[:, :].max(axis=-1) + 0 * z
ci = m6plot.pmCI(0., 20., 2.)
plotPsi(yy,
z,
psiPlot,
ci,
'Global FFH MOC [Sv],' + 'averaged between ' + args.start_date +
' and ' + args.end_date,
zval=[0., -400., -1000.])
plt.xlabel(r'Latitude [$\degree$N]')
plt.suptitle(case_name)
plt.gca().invert_yaxis()
objOut = args.outdir + str(case_name) + '_FFH_MOC_global.png'
plt.savefig(objOut)
moc['moc_FFM'].data = psiPlot
# GM-induced Global MOC
class C(np.ndarray):
pass
varName = 'vhGM'
conversion_factor = 1.e-9
tmp = np.ma.masked_invalid(ds_mean[varName].values)
tmp = tmp[:].filled(0.)
VHGM = tmp.view(C)
VHGM.units = ds[varName].units
Zmod = m6toolbox.get_z(ds, depth, varName) # same here
m6plot.setFigureSize([16, 9], 576, debug=False)
axis = plt.gca()
cmap = plt.get_cmap('dunnePM')
z = Zmod.min(axis=-1)
psiPlot = MOCpsi(VHGM) * conversion_factor
psiPlot = 0.5 * (psiPlot[0:-1, :] + psiPlot[1::, :])
yy = grd.geolat_c[:, :].max(axis=-1) + 0 * z
ci = m6plot.pmCI(0., 20., 2.)
plotPsi(
yy, z, psiPlot, ci, 'Global GM MOC [Sv],' + 'averaged between ' +
args.start_date + ' and ' + args.end_date)
plt.xlabel(r'Latitude [$\degree$N]')
plt.suptitle(case_name)
plt.gca().invert_yaxis()
findExtrema(yy, z, psiPlot, min_lat=-65., max_lat=-30, mult=-1.)
objOut = args.outdir + str(case_name) + '_GM_MOC_global.png'
plt.savefig(objOut)
moc['moc_GM'].data = psiPlot
print('Saving netCDF files...')
moc.to_netcdf('ncfiles/' + str(case_name) + '_MOC.nc')
return
def driver(args):
nw = args.number_of_workers
if not os.path.isdir('PNG/TS_levels'):
print('Creating a directory to place figures (PNG)... \n')
os.system('mkdir -p PNG/TS_levels')
if not os.path.isdir('ncfiles'):
print('Creating a directory to place netCDF files (ncfiles)... \n')
os.system('mkdir ncfiles')
# Read in the yaml file
diag_config_yml = yaml.load(open(args.diag_config_yml_path, 'r'),
Loader=yaml.Loader)
# Create the case instance
dcase = DiagsCase(diag_config_yml['Case'])
RUNDIR = dcase.get_value('RUNDIR')
args.casename = dcase.casename
print('Run directory is:', RUNDIR)
print('Casename is:', args.casename)
print('Number of workers: ', nw)
# set avg dates
avg = diag_config_yml['Avg']
if not args.start_date: args.start_date = avg['start_date']
if not args.end_date: args.end_date = avg['end_date']
# read grid info
grd = MOM6grid(RUNDIR + '/' + args.casename + '.mom6.static.nc')
grd_xr = MOM6grid(RUNDIR + '/' + args.casename + '.mom6.static.nc',
xrformat=True)
# create masks
depth = grd.depth_ocean
# remote Nan's, otherwise genBasinMasks won't work
depth[np.isnan(depth)] = 0.0
basin_code = genBasinMasks(grd.geolon, grd.geolat, depth, xda=True)
# TODO: improve how obs are selected
if args.obs == 'PHC2':
# load PHC2 data
obs_path = '/glade/p/cesm/omwg/obs_data/phc/'
obs_temp = xr.open_mfdataset(obs_path +
'PHC2_TEMP_tx0.66v1_34lev_ann_avg.nc',
decode_coords=False,
decode_times=False)
obs_salt = xr.open_mfdataset(obs_path +
'PHC2_SALT_tx0.66v1_34lev_ann_avg.nc',
decode_coords=False,
decode_times=False)
elif args.obs == 'WOA18':
# load WOA18 data
obs_path = '/glade/u/home/gmarques/Notebooks/CESM_MOM6/WOA18_remapping/'
obs_temp = xr.open_dataset(
obs_path + 'WOA18_TEMP_tx0.66v1_34lev_ann_avg.nc',
decode_times=False).rename({'theta0': 'TEMP'})
obs_salt = xr.open_dataset(obs_path +
'WOA18_SALT_tx0.66v1_34lev_ann_avg.nc',
decode_times=False).rename({'s_an': 'SALT'})
else:
raise ValueError("The obs selected is not available.")
parallel, cluster, client = request_workers(nw)
print('Reading surface dataset...')
startTime = datetime.now()
variables = ['thetao', 'so', 'time', 'time_bnds']
def preprocess(ds):
''' Compute montly averages and return the dataset with variables'''
return ds[variables] #.resample(time="1Y", closed='left', \
#keep_attrs=True).mean(dim='time', keep_attrs=True)
if parallel:
ds = xr.open_mfdataset(RUNDIR+'/'+dcase.casename+'.mom6.h_*.nc', \
parallel=True, data_vars='minimal', \
coords='minimal', compat='override', preprocess=preprocess)
else:
ds = xr.open_mfdataset(RUNDIR+'/'+dcase.casename+'.mom6.h_*.nc', \
data_vars='minimal', coords='minimal', compat='override', preprocess=preprocess)
print('Time elasped: ', datetime.now() - startTime)
print('Selecting data between {} and {}...'.format(args.start_date,
args.end_date))
startTime = datetime.now()
ds = ds.sel(time=slice(args.start_date, args.end_date))
print('Time elasped: ', datetime.now() - startTime)
print('\n Computing yearly means...')
startTime = datetime.now()
ds = ds.resample(time="1Y", closed='left',
keep_attrs=True).mean('time', keep_attrs=True)
print('Time elasped: ', datetime.now() - startTime)
print('Time averaging...')
startTime = datetime.now()
temp = np.ma.masked_invalid(ds.thetao.mean('time').values)
salt = np.ma.masked_invalid(ds.so.mean('time').values)
print('Time elasped: ', datetime.now() - startTime)
print('Computing stats for different basins...')
startTime = datetime.now()
# construct a 3D area with land values masked
area = np.ma.masked_where(grd.wet == 0, grd.area_t)
tmp = np.repeat(area[np.newaxis, :, :], len(obs_temp.depth), axis=0)
area_mom3D = xr.DataArray(tmp,
dims=('depth', 'yh', 'xh'),
coords={
'depth': obs_temp.depth.values,
'yh': grd.yh,
'xh': grd.xh
})
for k in range(len(area_mom3D.depth)):
area_mom3D[k, :] = grd_xr.area_t.where(
grd_xr.depth_ocean >= area_mom3D.depth[k])
# temp
thetao_mean = ds.thetao.mean('time')
temp_diff = thetao_mean.rename({
'z_l': 'depth'
}).rename('TEMP') - obs_temp['TEMP']
temp_stats = myStats_da(temp_diff, area_mom3D,
basins=basin_code).rename('thetao_bias_stats')
# salt
so_mean = ds.so.mean('time')
salt_diff = so_mean.rename({
'z_l': 'depth'
}).rename('SALT') - obs_salt['SALT']
salt_stats = myStats_da(salt_diff, area_mom3D,
basins=basin_code).rename('so_bias_stats')
# plots
depth = temp_stats.depth.values
basin = temp_stats.basin.values
interfaces = np.zeros(len(depth) + 1)
for k in range(1, len(depth) + 1):
interfaces[k] = interfaces[k -
1] + (2 *
(depth[k - 1] - interfaces[k - 1]))
reg = np.arange(len(temp_stats.basin.values) + 1)
figname = 'PNG/TS_levels/' + str(dcase.casename) + '_'
temp_label = r'Potential temperature [$^o$C]'
salt_label = 'Salinity [psu]'
# minimum
score_plot2(basin,
interfaces,
temp_stats[:, 0, :],
nbins=30,
cmap=plt.cm.viridis,
cmin=temp_stats[:, 0, :].min().values,
units=temp_label,
fname=figname + 'thetao_bias_min.png',
title='Minimun temperature difference (model-{})'.format(
args.obs))
score_plot2(basin,
interfaces,
salt_stats[:, 0, :],
nbins=30,
cmap=plt.cm.viridis,
cmin=salt_stats[:, 0, :].min().values,
units=salt_label,
fname=figname + 'so_bias_min.png',
title='Minimun salinity difference (model-{})'.format(
args.obs))
# maximum
score_plot2(basin,
interfaces,
temp_stats[:, 1, :],
nbins=30,
cmap=plt.cm.viridis,
cmin=temp_stats[:, 1, :].min().values,
units=temp_label,
fname=figname + 'thetao_bias_max.png',
title='Maximum temperature difference (model-{})'.format(
args.obs))
score_plot2(basin,
interfaces,
salt_stats[:, 1, :],
nbins=30,
cmap=plt.cm.viridis,
cmin=salt_stats[:, 1, :].min().values,
units=salt_label,
fname=figname + 'so_bias_max.png',
title='Maximum salinity difference (model-{})'.format(
args.obs))
# mean
score_plot2(basin,
interfaces,
temp_stats[:, 2, :],
nbins=30,
cmap=plt.cm.seismic,
units=temp_label,
fname=figname + 'thetao_bias_mean.png',
title='Mean temperature difference (model-{})'.format(
args.obs))
score_plot2(basin,
interfaces,
salt_stats[:, 2, :],
nbins=30,
cmap=plt.cm.seismic,
units=salt_label,
fname=figname + 'so_bias_mean.png',
title='Mean salinity difference (model-{})'.format(args.obs))
# std
score_plot2(basin,
interfaces,
temp_stats[:, 3, :],
nbins=30,
cmap=plt.cm.viridis,
cmin=1.0E-15,
units=temp_label,
fname=figname + 'thetao_bias_std.png',
title='Std temperature difference (model-{})'.format(args.obs))
score_plot2(basin,
interfaces,
salt_stats[:, 3, :],
nbins=30,
cmap=plt.cm.viridis,
cmin=1.0E-15,
units=salt_label,
fname=figname + 'so_bias_std.png',
title='Std salinity difference (model-{})'.format(args.obs))
# rms
score_plot2(basin,
interfaces,
temp_stats[:, 4, :],
nbins=30,
cmap=plt.cm.viridis,
cmin=1.0E-15,
units=temp_label,
fname=figname + 'thetao_bias_rms.png',
title='Rms temperature difference (model-{})'.format(args.obs))
score_plot2(basin,
interfaces,
salt_stats[:, 4, :],
nbins=30,
cmap=plt.cm.viridis,
cmin=1.0E-15,
units=salt_label,
fname=figname + 'so_bias_rms.png',
title='Rms salinity difference (model-{})'.format(args.obs))
print('Time elasped: ', datetime.now() - startTime)
print('Saving netCDF files...')
startTime = datetime.now()
attrs = {
'description': 'model - obs at depth levels',
'start_date': args.start_date,
'end_date': args.end_date,
'casename': dcase.casename,
'obs': args.obs,
'module': os.path.basename(__file__)
}
# create dataset to store results
add_global_attrs(temp_stats, attrs)
temp_stats.to_netcdf('ncfiles/' + str(args.casename) +
'_thetao_bias_ann_mean_stats.nc')
add_global_attrs(salt_stats, attrs)
salt_stats.to_netcdf('ncfiles/' + str(args.casename) +
'_so_bias_ann_mean_stats.nc')
thetao = xr.DataArray(thetao_mean,
dims=['z_l', 'yh', 'xh'],
coords={
'z_l': ds.z_l,
'yh': grd.yh,
'xh': grd.xh
}).rename('thetao')
temp_bias = np.ma.masked_invalid(thetao.values - obs_temp['TEMP'].values)
ds_thetao = xr.Dataset(data_vars={
'thetao': (('z_l', 'yh', 'xh'), thetao),
'thetao_bias': (('z_l', 'yh', 'xh'), temp_bias)
},
coords={
'z_l': ds.z_l,
'yh': grd.yh,
'xh': grd.xh
})
add_global_attrs(ds_thetao, attrs)
ds_thetao.to_netcdf('ncfiles/' + str(args.casename) +
'_thetao_time_mean.nc')
so = xr.DataArray(ds.so.mean('time'),
dims=['z_l', 'yh', 'xh'],
coords={
'z_l': ds.z_l,
'yh': grd.yh,
'xh': grd.xh
}).rename('so')
salt_bias = np.ma.masked_invalid(so.values - obs_salt['SALT'].values)
ds_so = xr.Dataset(data_vars={
'so': (('z_l', 'yh', 'xh'), so),
'so_bias': (('z_l', 'yh', 'xh'), salt_bias)
},
coords={
'z_l': ds.z_l,
'yh': grd.yh,
'xh': grd.xh
})
add_global_attrs(ds_so, attrs)
ds_so.to_netcdf('ncfiles/' + str(args.casename) + '_so_time_mean.nc')
print('Time elasped: ', datetime.now() - startTime)
if parallel:
print('\n Releasing workers...')
client.close()
cluster.close()
print('Global plots...')
km = len(obs_temp['depth'])
for k in range(km):
if ds['z_l'][k].values < 1200.0:
figname = 'PNG/TS_levels/' + str(dcase.casename) + '_' + str(
ds['z_l'][k].values) + '_'
temp_obs = np.ma.masked_invalid(obs_temp['TEMP'][k, :].values)
xycompare(temp[k, :],
temp_obs,
grd.geolon,
grd.geolat,
area=grd.area_t,
title1='model temperature, depth =' +
str(ds['z_l'][k].values) + 'm',
title2='observed temperature, depth =' +
str(obs_temp['depth'][k].values) + 'm',
suptitle=dcase.casename + ', averaged ' +
str(args.start_date) + ' to ' + str(args.end_date),
extend='both',
dextend='neither',
clim=(-1.9, 30.),
dlim=(-2, 2),
dcolormap=plt.cm.bwr,
save=figname + 'global_temp.png')
salt_obs = np.ma.masked_invalid(obs_salt['SALT'][k, :].values)
xycompare(salt[k, :],
salt_obs,
grd.geolon,
grd.geolat,
area=grd.area_t,
title1='model salinity, depth =' +
str(ds['z_l'][k].values) + 'm',
title2='observed salinity, depth =' +
str(obs_temp['depth'][k].values) + 'm',
suptitle=dcase.casename + ', averaged ' +
str(args.start_date) + ' to ' + str(args.end_date),
extend='both',
dextend='neither',
clim=(30., 39.),
dlim=(-2, 2),
dcolormap=plt.cm.bwr,
save=figname + 'global_salt.png')
print('Antarctic plots...')
for k in range(km):
if (ds['z_l'][k].values < 1200.):
temp_obs = np.ma.masked_invalid(obs_temp['TEMP'][k, :].values)
polarcomparison(temp[k, :],
temp_obs,
grd,
title1='model temperature, depth =' +
str(ds['z_l'][k].values) + 'm',
title2='observed temperature, depth =' +
str(obs_temp['depth'][k].values) + 'm',
extend='both',
dextend='neither',
clim=(-1.9, 10.5),
dlim=(-2, 2),
dcolormap=plt.cm.bwr,
suptitle=dcase.casename + ', averaged ' +
str(args.start_date) + ' to ' + str(args.end_date),
proj='SP',
save=figname + 'antarctic_temp.png')
salt_obs = np.ma.masked_invalid(obs_salt['SALT'][k, :].values)
polarcomparison(salt[k, :],
salt_obs,
grd,
title1='model salinity, depth =' +
str(ds['z_l'][k].values) + 'm',
title2='observed salinity, depth =' +
str(obs_temp['depth'][k].values) + 'm',
extend='both',
dextend='neither',
clim=(33., 35.),
dlim=(-2, 2),
dcolormap=plt.cm.bwr,
suptitle=dcase.casename + ', averaged ' +
str(args.start_date) + ' to ' + str(args.end_date),
proj='SP',
save=figname + 'antarctic_salt.png')
print('Arctic plots...')
for k in range(km):
if (ds['z_l'][k].values < 100.):
temp_obs = np.ma.masked_invalid(obs_temp['TEMP'][k, :].values)
polarcomparison(temp[k, :],
temp_obs,
grd,
title1='model temperature, depth =' +
str(ds['z_l'][k].values) + 'm',
title2='observed temperature, depth =' +
str(obs_temp['depth'][k].values) + 'm',
extend='both',
dextend='neither',
clim=(-1.9, 11.5),
dlim=(-2, 2),
dcolormap=plt.cm.bwr,
suptitle=dcase.casename + ', averaged ' +
str(args.start_date) + ' to ' + str(args.end_date),
proj='NP',
save=figname + 'arctic_temp.png')
salt_obs = np.ma.masked_invalid(obs_salt['SALT'][k, :].values)
polarcomparison(salt[k, :],
salt_obs,
grd,
title1='model salinity, depth =' +
str(ds['z_l'][k].values) + 'm',
title2='observed salinity, depth =' +
str(obs_temp['depth'][k].values) + 'm',
extend='both',
dextend='neither',
clim=(31.5, 35.),
dlim=(-2, 2),
dcolormap=plt.cm.bwr,
suptitle=dcase.casename + ', averaged ' +
str(args.start_date) + ' to ' + str(args.end_date),
proj='NP',
save=figname + 'arctic_salt.png')
return
def horizontal_mean_diff_rms(grd, dcase, basins, args):
'''
Compute horizontal mean difference and rms: model versus observations.
Parameters
----------
grd : OrderedDict
Dictionary with statistics computed using function myStats_da
dcase : case object
Object created using mom6_tools.DiagsCase.
basins : DataArray
Basins mask to apply. Returns horizontal mean RMSE for each basin provided.
Basins must be generated by genBasinMasks.
args : object
Object with command line options.
Returns
-------
Plots horizontal mean difference and rms for different basins.
'''
RUNDIR = dcase.get_value('RUNDIR')
area = grd.area_t.where(grd.wet > 0)
if args.debug: print('RUNDIR:', RUNDIR)
parallel, cluster, client = request_workers(args.number_of_workers)
def preprocess(ds):
if 'thetao' not in ds.variables:
ds["thetao"] = xr.zeros_like(ds.h)
if 'so' not in ds.variables:
ds["so"] = xr.zeros_like(ds.h)
return ds
# read dataset
startTime = datetime.now()
print('Reading dataset...')
ds = xr.open_mfdataset(
RUNDIR + '/' + dcase.casename + '.mom6.h_*.nc',
parallel=True,
combine="nested", # concatenate in order of files
concat_dim="time", # concatenate along time
preprocess=preprocess,
).chunk({"time": 12})
if args.debug:
print(ds)
print('Time elasped: ', datetime.now() - startTime)
print('Selecting data between {} and {}...'.format(args.start_date,
args.end_date))
ds = ds.sel(time=slice(args.start_date, args.end_date))
# Compute climatologies
thetao_model = ds.thetao.resample(time="1Y", closed='left', keep_attrs=True).mean(dim='time', \
keep_attrs=True)
salt_model = ds.so.resample(time="1Y", closed='left', keep_attrs=True).mean(dim='time', \
keep_attrs=True)
# TODO: improve how obs are selected
if args.obs == 'PHC2':
# load PHC2 data
obs_path = '/glade/p/cesm/omwg/obs_data/phc/'
obs_temp = xr.open_dataset(obs_path +
'PHC2_TEMP_tx0.66v1_34lev_ann_avg.nc',
decode_times=False)
obs_salt = xr.open_dataset(obs_path +
'PHC2_SALT_tx0.66v1_34lev_ann_avg.nc',
decode_times=False)
# get theta and salt and rename coordinates to be the same as the model's
thetao_obs = obs_temp.TEMP.rename({
'X': 'xh',
'Y': 'yh',
'depth': 'z_l'
})
salt_obs = obs_salt.SALT.rename({
'X': 'xh',
'Y': 'yh',
'depth': 'z_l'
})
elif args.obs == 'WOA18':
# load WOA18 data
obs_path = '/glade/u/home/gmarques/Notebooks/CESM_MOM6/WOA18_remapping/'
obs_temp = xr.open_dataset(obs_path +
'WOA18_TEMP_tx0.66v1_34lev_ann_avg.nc',
decode_times=False)
obs_salt = xr.open_dataset(obs_path +
'WOA18_SALT_tx0.66v1_34lev_ann_avg.nc',
decode_times=False)
# get theta and salt and rename coordinates to be the same as the model's
thetao_obs = obs_temp.theta0.rename({'depth': 'z_l'})
salt_obs = obs_salt.s_an.rename({'depth': 'z_l'})
else:
raise ValueError("The obs selected is not available.")
# set coordinates to the same as the model's
thetao_obs['xh'] = thetao_model.xh
thetao_obs['yh'] = thetao_model.yh
salt_obs['xh'] = salt_model.xh
salt_obs['yh'] = salt_model.yh
# compute difference
temp_diff = thetao_model - thetao_obs
salt_diff = salt_model - salt_obs
# construct a 3D area with land values masked
area3d = np.repeat(area.values[np.newaxis, :, :],
len(temp_diff.z_l),
axis=0)
mask3d = xr.DataArray(area3d,
dims=(temp_diff.dims[1:4]),
coords={
temp_diff.dims[1]: temp_diff.z_l,
temp_diff.dims[2]: temp_diff.yh,
temp_diff.dims[3]: temp_diff.xh
})
area3d_masked = mask3d.where(temp_diff[0, :] == temp_diff[0, :])
# Horizontal Mean difference (model - obs)
print('\n Computing Horizontal Mean difference for temperature...')
startTime = datetime.now()
temp_bias = HorizontalMeanDiff_da(temp_diff,
weights=area3d_masked,
basins=basins,
debug=args.debug).rename('temp_bias')
print('Time elasped: ', datetime.now() - startTime)
print('\n Computing Horizontal Mean difference for salt...')
startTime = datetime.now()
salt_bias = HorizontalMeanDiff_da(salt_diff,
weights=area3d_masked,
basins=basins,
debug=args.debug).rename('salt_bias')
print('Time elasped: ', datetime.now() - startTime)
# Horizontal Mean rms (model - obs)
print('\n Computing Horizontal Mean rms for temperature...')
startTime = datetime.now()
temp_rms = HorizontalMeanRmse_da(temp_diff,
weights=area3d_masked,
basins=basins,
debug=args.debug).rename('temp_rms')
print('Time elasped: ', datetime.now() - startTime)
print('\n Computing Horizontal Mean rms for salt...')
salt_rms = HorizontalMeanRmse_da(salt_diff,
weights=area3d_masked,
basins=basins,
debug=args.debug).rename('salt_rms')
print('Time elasped: ', datetime.now() - startTime)
if parallel:
print('Releasing workers...')
client.close()
cluster.close()
print('Saving netCDF files...')
attrs = {
'start_date': args.start_date,
'end_date': args.end_date,
'casename': dcase.casename,
'obs': args.obs,
'module': os.path.basename(__file__)
}
add_global_attrs(temp_bias, attrs)
temp_bias.to_netcdf('ncfiles/' + str(dcase.casename) + '_temp_bias.nc')
add_global_attrs(salt_bias, attrs)
salt_bias.to_netcdf('ncfiles/' + str(dcase.casename) + '_salt_bias.nc')
add_global_attrs(temp_rms, attrs)
temp_rms.to_netcdf('ncfiles/' + str(dcase.casename) + '_temp_rms.nc')
add_global_attrs(salt_rms, attrs)
salt_rms.to_netcdf('ncfiles/' + str(dcase.casename) + '_salt_rms.nc')
# temperature
for reg in temp_bias.region:
print('Generating temperature plots for:', str(reg.values))
# remove Nan's
temp_diff_reg = temp_bias.sel(region=reg).dropna('z_l')
temp_rms_reg = temp_rms.sel(region=reg).dropna('z_l')
if temp_diff_reg.z_l.max() <= 1000.0:
splitscale = None
else:
splitscale = [0., -1000., -temp_diff_reg.z_l.max()]
savefig_diff = 'PNG/Horizontal_mean_biases/' + str(
dcase.casename) + '_' + str(reg.values) + '_temp_diff.png'
savefig_rms = 'PNG/Horizontal_mean_biases/' + str(
dcase.casename) + '_' + str(reg.values) + '_temp_rms.png'
ztplot(temp_diff_reg.values,
temp_diff_reg.time.values,
temp_diff_reg.z_l.values * -1,
ignore=np.nan,
splitscale=splitscale,
suptitle=dcase._casename,
contour=True,
title=str(reg.values) +
', Potential Temperature [C], diff (model - obs)',
extend='both',
colormap='dunnePM',
autocenter=True,
tunits='Year',
show=False,
clim=(-3, 3),
save=savefig_diff,
interactive=True)
ztplot(temp_rms_reg.values,
temp_rms_reg.time.values,
temp_rms_reg.z_l.values * -1,
ignore=np.nan,
splitscale=splitscale,
suptitle=dcase._casename,
contour=True,
title=str(reg.values) +
', Potential Temperature [C], rms (model - obs)',
extend='both',
colormap='dunnePM',
autocenter=False,
tunits='Year',
show=False,
clim=(0, 6),
save=savefig_rms,
interactive=True)
plt.close('all')
# salinity
for reg in salt_bias.region:
print('Generating salinity plots for ', str(reg.values))
# remove Nan's
salt_diff_reg = salt_bias.sel(region=reg).dropna('z_l')
salt_rms_reg = salt_rms.sel(region=reg).dropna('z_l')
if salt_diff_reg.z_l.max() <= 1000.0:
splitscale = None
else:
splitscale = [0., -1000., -salt_diff_reg.z_l.max()]
savefig_diff = 'PNG/Horizontal_mean_biases/' + str(
dcase.casename) + '_' + str(reg.values) + '_salt_diff.png'
savefig_rms = 'PNG/Horizontal_mean_biases/' + str(
dcase.casename) + '_' + str(reg.values) + '_salt_rms.png'
ztplot(salt_diff_reg.values,
salt_diff_reg.time.values,
salt_diff_reg.z_l.values * -1,
ignore=np.nan,
splitscale=splitscale,
suptitle=dcase._casename,
contour=True,
title=str(reg.values) + ', Salinity [psu], diff (model - obs)',
extend='both',
colormap='dunnePM',
autocenter=True,
tunits='Year',
show=False,
clim=(-1.5, 1.5),
save=savefig_diff,
interactive=True)
ztplot(salt_rms_reg.values,
salt_rms_reg.time.values,
salt_rms_reg.z_l.values * -1,
ignore=np.nan,
splitscale=splitscale,
suptitle=dcase._casename,
contour=True,
title=str(reg.values) + ', Salinity [psu], rms (model - obs)',
extend='both',
colormap='dunnePM',
autocenter=False,
tunits='Year',
show=False,
clim=(0, 3),
save=savefig_rms,
interactive=True)
plt.close('all')
return
def extract_time_series(fname, variables, grd, dcase, args):
'''
Extract time-series and saves annual means.
Parameters
----------
fname : str
Name of the file to be processed.
variables : str
List of variables to be processed.
grd : OrderedDict
Dictionary with statistics computed using function myStats_da
dcase : case object
Object created using mom6_tools.DiagsCase.
args : object
Object with command line options.
Returns
-------
NetCDF file with annual means.
'''
parallel, cluster, client = request_workers(args.number_of_workers)
RUNDIR = dcase.get_value('RUNDIR')
def preprocess(ds):
''' Compute montly averages and return the dataset with variables'''
return ds[variables].resample(time="1M", closed='left', \
keep_attrs=True).mean(dim='time', keep_attrs=True)
# read forcing files
startTime = datetime.now()
print('Reading dataset...')
if parallel:
ds = xr.open_mfdataset(RUNDIR+'/'+dcase.casename+fname, \
chunks={'time': 365}, parallel=True, data_vars='minimal',
coords='minimal', preprocess=preprocess)
else:
ds = xr.open_mfdataset(RUNDIR + '/' + dcase.casename + fname,
data_vars='minimal',
compat='override',
coords='minimal',
preprocess=preprocess)
print('Time elasped: ', datetime.now() - startTime)
# add attrs and save
attrs = {'description': 'Annual averages of global mean ocean properties.'}
add_global_attrs(ds, attrs)
ds.to_netcdf('ncfiles/' + str(dcase.casename) + '_ann_ave_global_means.nc')
if parallel:
# close processes
print('Releasing workers...\n')
client.close()
cluster.close()
return
def main():
# Get options
args = options()
nw = args.number_of_workers
if not os.path.isdir('ncfiles'):
print('Creating a directory to place figures (ncfiles)... \n')
os.system('mkdir ncfiles')
# Read in the yaml file
diag_config_yml = yaml.load(open(args.diag_config_yml_path,'r'), Loader=yaml.Loader)
# Create the case instance
dcase = DiagsCase(diag_config_yml['Case'])
args.case_name = dcase.casename
RUNDIR = dcase.get_value('RUNDIR')
print('Run directory is:', RUNDIR)
print('Casename is:', dcase.casename)
print('Number of workers to be used:', nw)
# set avg dates
avg = diag_config_yml['Avg']
if not args.start_date : args.start_date = avg['start_date']
if not args.end_date : args.end_date = avg['end_date']
# read grid info
grd = MOM6grid(RUNDIR+'/'+dcase.casename+'.mom6.static.nc')
depth = grd.depth_ocean
# remote Nan's, otherwise genBasinMasks won't work
depth[np.isnan(depth)] = 0.0
basin_code = m6toolbox.genBasinMasks(grd.geolon, grd.geolat, depth)
parallel, cluster, client = m6toolbox.request_workers(nw)
print('Reading {} dataset...'.format(args.file_name))
startTime = datetime.now()
# load data
def preprocess(ds):
variables = ['diftrblo', 'difmxylo' ,'difmxybo', 'diftrelo']
for v in variables:
if v not in ds.variables:
ds[v] = xr.zeros_like(ds.vo)
return ds[variables]
ds = xr.open_mfdataset(RUNDIR+'/'+dcase.casename+args.file_name,
parallel=True,
combine="nested", # concatenate in order of files
concat_dim="time", # concatenate along time
preprocess=preprocess,
).chunk({"time": 12})
print('Time elasped: ', datetime.now() - startTime)
# compute yearly means first
print('Computing yearly means...')
startTime = datetime.now()
ds_yr = ds.resample(time="1Y", closed='left').mean('time')
print('Time elasped: ', datetime.now() - startTime)
print('Selecting data between {} and {}...'.format(args.start_date, args.end_date))
startTime = datetime.now()
ds_sel = ds_yr.sel(time=slice(args.start_date, args.end_date))
print('Time elasped: ', datetime.now() - startTime)
print('Computing time mean...')
startTime = datetime.now()
ds_mean = ds_sel.mean('time').compute()
print('Time elasped: ', datetime.now() - startTime)
attrs = {'description': 'Time-mean mixing coefficients', 'units': 'm^2/s', 'start_date': avg['start_date'],
'end_date': avg['end_date'], 'casename': dcase.casename}
m6toolbox.add_global_attrs(ds_mean,attrs)
print('Saving netCDF files...')
ds_mean.to_netcdf('ncfiles/'+str(args.case_name)+'_avg_mixing_coeffs.nc')
print('Releasing workers ...')
client.close(); cluster.close()
return