"""Climate data pre-processing"""
# Built ins
import logging
from packaging.version import Version
import warnings
# External libs
import cftime
import numpy as np
import netCDF4
import xarray as xr
# Locals
from oggm import cfg
from oggm import utils
from oggm import entity_task
from oggm.exceptions import InvalidParamsError
# Module logger
log = logging.getLogger(__name__)
[docs]@entity_task(log, writes=['gcm_data'])
def process_gcm_data(gdir, filesuffix='', prcp=None, temp=None,
year_range=('1961', '1990'), scale_stddev=True,
time_unit=None, calendar=None, source='',
apply_bias_correction=True):
""" Applies the anomaly method to GCM climate data
This function can be applied to any GCM data, if it is provided in a
suitable :py:class:`xarray.DataArray`. See Parameter description for
format details.
For CESM-LME a specific function :py:func:`tasks.process_cesm_data` is
available which does the preprocessing of the data and subsequently calls
this function.
Parameters
----------
gdir : :py:class:`oggm.GlacierDirectory`
where to write the data
filesuffix : str
append a suffix to the filename (useful for ensemble experiments).
prcp : :py:class:`xarray.DataArray`
| monthly total precipitation [mm month-1]
| Coordinates:
| lat float64
| lon float64
| time: cftime object
temp : :py:class:`xarray.DataArray`
| monthly temperature [K]
| Coordinates:
| lat float64
| lon float64
| time cftime object
year_range : tuple of str
the year range for which you want to compute the anomalies. Default
is `('1961', '1990')`
scale_stddev : bool
whether or not to scale the temperature standard deviation as well
time_unit : str
The unit conversion for NetCDF files. It must be adapted to the
length of the time series. The default is to choose
it ourselves based on the starting year.
For example: 'days since 0850-01-01 00:00:00'
calendar : str
If you use an exotic calendar (e.g. 'noleap')
source : str
For metadata: the source of the climate data
apply_bias_correction : boolean
if a bias-correction should be applied. Default is True, only set it to False
if the GCM has already been externally bias-corrected to the applied
observational calibration dataset (true for ISIMIP 3b that is bias-corrected
to W5E5). !!! We assume that temp is in Kelvin and convert to CELSIUS !!!
"""
# Standard sanity checks
months = temp['time.month']
if months[0] != 1:
raise ValueError('We expect the files to start in January!')
if months[-1] != 12:
raise ValueError('We expect the files to end in December!')
if (np.abs(temp['lon']) > 180) or (np.abs(prcp['lon']) > 180):
raise ValueError('We expect the longitude coordinates to be within '
'[-180, 180].')
assert len(prcp) // 12 == len(prcp) / 12, "Somehow we didn't get full years"
assert len(temp) // 12 == len(temp) / 12, "Somehow we didn't get full years"
# Get the reference data to apply the anomaly to
fpath = gdir.get_filepath('climate_historical')
with xr.open_dataset(fpath) as ds_ref:
ds_ref = ds_ref.sel(time=slice(*year_range))
if apply_bias_correction:
# compute monthly anomalies
# of temp
if scale_stddev:
# This is a bit more arithmetic
ts_tmp_sel = temp.sel(time=slice(*year_range))
if len(ts_tmp_sel) // 12 != len(ts_tmp_sel) / 12:
raise InvalidParamsError('year_range cannot contain the first'
'or last calendar year in the series')
if ((len(ts_tmp_sel) // 12) % 2) == 1:
raise InvalidParamsError('We need an even number of years '
'for this to work')
ts_tmp_std = ts_tmp_sel.groupby('time.month').std(dim='time')
std_fac = ds_ref.temp.groupby('time.month').std(dim='time') / ts_tmp_std
std_fac = np.tile(std_fac.data, len(temp) // 12)
# We need an even number of years for this to work
win_size = len(ts_tmp_sel) + 1
def roll_func(x, axis=None):
x = x[:, ::12]
n = len(x[0, :]) // 2
xm = np.nanmean(x, axis=axis)
return xm + (x[:, n] - xm) * std_fac
temp = temp.rolling(time=win_size, center=True,
min_periods=1).reduce(roll_func)
ts_tmp_sel = temp.sel(time=slice(*year_range))
if len(ts_tmp_sel.time) != len(ds_ref.time):
raise InvalidParamsError('The reference climate period and the '
'GCM period after window selection do '
'not match.')
ts_tmp_avg = ts_tmp_sel.groupby('time.month').mean(dim='time')
ts_tmp = temp.groupby('time.month') - ts_tmp_avg
# of precip -- scaled anomalies
ts_pre_avg = prcp.sel(time=slice(*year_range))
ts_pre_avg = ts_pre_avg.groupby('time.month').mean(dim='time')
ts_pre_ano = prcp.groupby('time.month') - ts_pre_avg
# scaled anomalies is the default. Standard anomalies above
# are used later for where ts_pre_avg == 0
ts_pre = prcp.groupby('time.month') / ts_pre_avg
# for temp
loc_tmp = ds_ref.temp.groupby('time.month').mean()
ts_tmp = ts_tmp.groupby('time.month') + loc_tmp
# for prcp
loc_pre = ds_ref.prcp.groupby('time.month').mean()
# scaled anomalies
ts_pre = ts_pre.groupby('time.month') * loc_pre
# standard anomalies
ts_pre_ano = ts_pre_ano.groupby('time.month') + loc_pre
# Correct infinite values with standard anomalies
ts_pre.values = np.where(np.isfinite(ts_pre.values),
ts_pre.values,
ts_pre_ano.values)
# The previous step might create negative values (unlikely). Clip them
ts_pre.values = utils.clip_min(ts_pre.values, 0)
assert np.all(np.isfinite(ts_pre.values))
assert np.all(np.isfinite(ts_tmp.values))
else:
# do no bias correction at all
# (!!! only if GCM is already externally bias corrected)
ts_tmp = temp - 273.15 # convert K to Celsius
ts_pre = prcp # mm month-1
source = source + '_no_OGGM_bias_correction'
gdir.write_monthly_climate_file(temp.time.values,
ts_pre.values, ts_tmp.values,
float(ds_ref.ref_hgt),
prcp.lon.values, prcp.lat.values,
time_unit=time_unit,
calendar=calendar,
file_name='gcm_data',
source=source,
filesuffix=filesuffix)
@entity_task(log, writes=['gcm_data'])
def process_monthly_isimip_data(gdir, output_filesuffix='',
member='mri-esm2-0_r1i1p1f1',
ssp='ssp126',
year_range=('1979', '2014'),
apply_bias_correction=False,
testing=False,
y0=None, y1=None,
**kwargs):
"""Read, process and store the isimip3b gcm data for this glacier.
It stores the data in a format that can be used by the OGGM mass balance
model and in the glacier directory.
Currently, this function is built for the ISIMIP3b
simulations that are on the OGGM servers.
Parameters
----------
output_filesuffix : str
append a suffix to the filename (useful for ensemble experiments).
If it is not set, we create a filesuffix with applied ensemble and ssp
member : str
ensemble member gcm that you want to process
ssp : str
ssp scenario to process (only 'ssp126', 'ssp370' or 'ssp585' are available)
year_range : tuple of str
the year range for which the anomalies are computed
(passed to process_gcm_gdata). Default for ISIMIP3b `('1979', '2014')
apply_bias_correction : bool
whether the bias correction is applied (default is False) or not. As
we use already internally bias-corrected GCMs, it is default set
to False!
testing : boolean
Default is False. If testing is set to True,
the smaller test ISIMIP3b gcm files are downloaded
instead (only useful for pytest)
y0 : int
start year of the ISIMIP3b data processing.
Default is None which processes the entire timeseries.
Set this to the beginning of your bias correction/
projection period minus half of bc period
to make processing faster.
y1 : int
end year of the CMIP data processing.
Set this to the end of your projection period
plus half of bc period. Default is None to process
the entire time series, same as y0.
**kwargs: any kwarg to be passed to ref:`process_gcm_data`
"""
if output_filesuffix == '':
# recognize the gcm climate file for later
output_filesuffix = '_monthly_ISIMIP3b_{}_{}'.format(member, ssp)
# Glacier location
glon = gdir.cenlon
glat = gdir.cenlat
if y0 is not None:
assert y0 < 2014, 'y0 has to be below 2014'
if y1 is not None:
assert y1 > 2014, 'y0 has to be above 2014 at the moment'
if testing:
gcm_server = 'https://cluster.klima.uni-bremen.de/~oggm/test_climate/'
else:
gcm_server = 'https://cluster.klima.uni-bremen.de/~oggm/'
path = f'{gcm_server}/cmip6/isimip3b/flat/2023.2/monthly/'
add = '_global_monthly_flat_glaciers.nc'
fpath_spec = path + '{}_w5e5_'.format(member) + '{ssp}_{var}' + add
fpath_temp = fpath_spec.format(var='tasAdjust', ssp=ssp)
fpath_temp_h = fpath_spec.format(var='tasAdjust', ssp='historical')
fpath_precip = fpath_spec.format(var='prAdjust', ssp=ssp)
fpath_precip_h = fpath_spec.format(var='prAdjust', ssp='historical')
with utils.get_lock():
fpath_temp = utils.file_downloader(fpath_temp)
fpath_temp_h = utils.file_downloader(fpath_temp_h)
fpath_precip = utils.file_downloader(fpath_precip)
fpath_precip_h = utils.file_downloader(fpath_precip_h)
# Read the GCM files
with xr.open_dataset(fpath_temp_h, use_cftime=True) as tempds_hist, \
xr.open_dataset(fpath_temp, use_cftime=True) as tempds_gcm:
# make processing faster
if y0 is not None:
tempds_hist = tempds_hist.sel(time=slice(str(y0), None))
if y1 is not None:
tempds_gcm = tempds_gcm.sel(time=slice(None, str(y1)))
# Check longitude conventions
if tempds_gcm.longitude.min() >= 0 and glon <= 0:
glon += 360
assert tempds_gcm.attrs['experiment'] == ssp
# Take the closest to the glacier
# Should we consider GCM interpolation?
# try:
# computing all the distances and choose the nearest gridpoint
c = ((tempds_gcm.longitude - glon) ** 2 +
(tempds_gcm.latitude - glat) ** 2)
# first select gridpoint, then merge, should be faster!!!
temp_a_gcm = tempds_gcm.isel(points=np.argmin(c.data))
temp_a_hist = tempds_hist.isel(points=np.argmin(c.data))
# merge historical with gcm together
# TODO: change to drop_conflicts when xarray version v0.17.0 can
# be used with salem
temp_a = xr.merge([temp_a_gcm, temp_a_hist],
combine_attrs='override')
temp = temp_a.tasAdjust
temp['lon'] = temp_a.longitude
temp['lat'] = temp_a.latitude
temp.lon.values = temp.lon if temp.lon <= 180 else temp.lon - 360
with xr.open_dataset(fpath_precip_h, use_cftime=True) as precipds_hist, \
xr.open_dataset(fpath_precip, use_cftime=True) as precipds_gcm:
# make processing faster
if y0 is not None:
precipds_hist = precipds_hist.sel(time=slice(str(y0), None))
if y1 is not None:
precipds_gcm = precipds_gcm.sel(time=slice(None, str(y1)))
c = ((precipds_gcm.longitude - glon) ** 2 +
(precipds_gcm.latitude - glat) ** 2)
precip_a_gcm = precipds_gcm.isel(points=np.argmin(c.data))
precip_a_hist = precipds_hist.isel(points=np.argmin(c.data))
precip_a = xr.merge([precip_a_gcm, precip_a_hist],
combine_attrs='override')
precip = precip_a.prAdjust
precip['lon'] = precip_a.longitude
precip['lat'] = precip_a.latitude
# Back to [-180, 180] for OGGM
precip.lon.values = precip.lon if precip.lon <= 180 else precip.lon - 360
# Convert kg m-2 s-1 to mm mth-1 => 1 kg m-2 = 1 mm !!!
assert 'kg m-2 s-1' in precip.units, 'Precip units not understood'
ny, r = divmod(len(temp), 12)
assert r == 0
dimo = [cfg.DAYS_IN_MONTH[m - 1] for m in temp['time.month']]
precip = precip * dimo * (60 * 60 * 24)
process_gcm_data(gdir, filesuffix=output_filesuffix, prcp=precip, temp=temp,
year_range=year_range, source=output_filesuffix,
apply_bias_correction=apply_bias_correction,
**kwargs)
[docs]@entity_task(log, writes=['gcm_data'])
def process_cesm_data(gdir, filesuffix='', fpath_temp=None, fpath_precc=None,
fpath_precl=None, **kwargs):
"""Processes and writes CESM climate data for this glacier.
This function is made for interpolating the Community
Earth System Model Last Millennium Ensemble (CESM-LME) climate simulations,
from Otto-Bliesner et al. (2016), to the high-resolution CL2 climatologies
(provided with OGGM) and writes everything to a NetCDF file.
Parameters
----------
gdir : :py:class:`oggm.GlacierDirectory`
where to write the data
filesuffix : str
append a suffix to the filename (useful for ensemble experiments).
fpath_temp : str
path to the temp file (default: cfg.PATHS['cesm_temp_file'])
fpath_precc : str
path to the precc file (default: cfg.PATHS['cesm_precc_file'])
fpath_precl : str
path to the precl file (default: cfg.PATHS['cesm_precl_file'])
**kwargs: any kwarg to be passed to ref:`process_gcm_data`
"""
# CESM temperature and precipitation data
if fpath_temp is None:
if not ('cesm_temp_file' in cfg.PATHS):
raise ValueError("Need to set cfg.PATHS['cesm_temp_file']")
fpath_temp = cfg.PATHS['cesm_temp_file']
if fpath_precc is None:
if not ('cesm_precc_file' in cfg.PATHS):
raise ValueError("Need to set cfg.PATHS['cesm_precc_file']")
fpath_precc = cfg.PATHS['cesm_precc_file']
if fpath_precl is None:
if not ('cesm_precl_file' in cfg.PATHS):
raise ValueError("Need to set cfg.PATHS['cesm_precl_file']")
fpath_precl = cfg.PATHS['cesm_precl_file']
# read the files
if Version(xr.__version__) < Version('0.11'):
raise ImportError('This task needs xarray v0.11 or newer to run.')
tempds = xr.open_dataset(fpath_temp)
precpcds = xr.open_dataset(fpath_precc)
preclpds = xr.open_dataset(fpath_precl)
# Get the time right - i.e. from time bounds
# Fix for https://github.com/pydata/xarray/issues/2565
with utils.ncDataset(fpath_temp, mode='r') as nc:
time_unit = nc.variables['time'].units
calendar = nc.variables['time'].calendar
try:
# xarray v0.11
time = netCDF4.num2date(tempds.time_bnds[:, 0], time_unit,
calendar=calendar)
except TypeError:
# xarray > v0.11
time = tempds.time_bnds[:, 0].values
# select for location
lon = gdir.cenlon
lat = gdir.cenlat
# CESM files are in 0-360
if lon <= 0:
lon += 360
# take the closest
# Should we consider GCM interpolation?
temp = tempds.TREFHT.sel(lat=lat, lon=lon, method='nearest')
prcp = (precpcds.PRECC.sel(lat=lat, lon=lon, method='nearest') +
preclpds.PRECL.sel(lat=lat, lon=lon, method='nearest'))
temp['time'] = time
prcp['time'] = time
temp.lon.values = temp.lon if temp.lon <= 180 else temp.lon - 360
prcp.lon.values = prcp.lon if prcp.lon <= 180 else prcp.lon - 360
# Convert m s-1 to mm mth-1
if time[0].month != 1:
raise ValueError('We expect the files to start in January!')
ny, r = divmod(len(time), 12)
assert r == 0
ndays = np.tile(cfg.DAYS_IN_MONTH, ny)
prcp = prcp * ndays * (60 * 60 * 24 * 1000)
tempds.close()
precpcds.close()
preclpds.close()
# Here:
# - time_unit='days since 0850-01-01 00:00:00'
# - calendar='noleap'
process_gcm_data(gdir, filesuffix=filesuffix, prcp=prcp, temp=temp,
time_unit=time_unit, calendar=calendar, **kwargs)
[docs]@entity_task(log, writes=['gcm_data'])
def process_cmip_data(gdir, filesuffix='', fpath_temp=None,
fpath_precip=None, y0=None, y1=None,
**kwargs):
"""Read, process and store the CMIP5 and CMIP6 climate data for this glacier.
It stores the data in a format that can be used by the OGGM mass balance
model and in the glacier directory.
Currently, this function is built for the CMIP5 and CMIP6 projection
simulations that are on the OGGM servers.
Parameters
----------
filesuffix : str
append a suffix to the filename (useful for ensemble experiments).
fpath_temp : str
path to the temp file
fpath_precip : str
path to the precip file
y0 : int
start year of the CMIP data processing.
Default is None which processes the entire timeseries.
Set this to the beginning of your bias correction/
projection period minus half of bc period
to make process_cmip_data faster.
y1 : int
end year of the CMIP data processing.
Set this to the end of your projection period
plus half of bc period. Default is None to process
the entire time series, same as y0.
**kwargs: any kwarg to be passed to ref:`process_gcm_data`
"""
# Glacier location
glon = gdir.cenlon
glat = gdir.cenlat
if y0 is not None:
y0 = str(y0)
if y1 is not None:
y1 = str(y1)
# Read the GCM files
with xr.open_dataset(fpath_temp, use_cftime=True) as tempds, \
xr.open_dataset(fpath_precip, use_cftime=True) as precipds:
# only process and save the gcm data selected --> saves some time!
if (y0 is not None) or (y1 is not None):
tempds = tempds.sel(time=slice(y0, y1))
precipds = precipds.sel(time=slice(y0, y1))
# Check longitude conventions
if tempds.lon.min() >= 0 and glon <= 0:
glon += 360
# Take the closest to the glacier
# Should we consider GCM interpolation?
try:
# if gcms are not flattened, do:
# this is the default, so try this first
temp = tempds.tas.sel(lat=glat, lon=glon, method='nearest')
precip = precipds.pr.sel(lat=glat, lon=glon, method='nearest')
except:
# are the gcms flattened? if yes,
# compute all the distances and choose the
# nearest gridpoint
c_tempds = ((tempds.lon - glon) ** 2 +
(tempds.lat - glat) ** 2)
c_precipds = ((precipds.lon - glon) ** 2 +
(precipds.lat - glat) ** 2)
temp_0 = tempds.isel(points=np.argmin(c_tempds.data))
precip_0 = precipds.isel(points=np.argmin(c_precipds.data))
temp = temp_0.tas
temp['lon'] = temp_0.lon
temp['lat'] = temp_0.lat
precip = precip_0.pr
precip['lon'] = precip_0.lon
precip['lat'] = precip_0.lat
# Back to [-180, 180] for OGGM
temp.lon.values = temp.lon if temp.lon <= 180 else temp.lon - 360
precip.lon.values = precip.lon if precip.lon <= 180 else precip.lon - 360
# Convert kg m-2 s-1 to mm mth-1 => 1 kg m-2 = 1 mm !!!
assert 'kg m-2 s-1' in precip.units, 'Precip units not understood'
ny, r = divmod(len(temp), 12)
assert r == 0
dimo = [cfg.DAYS_IN_MONTH[m - 1] for m in temp['time.month']]
precip = precip * dimo * (60 * 60 * 24)
process_gcm_data(gdir, filesuffix=filesuffix, prcp=precip, temp=temp,
source=filesuffix, **kwargs)
@entity_task(log, writes=['gcm_data'])
def process_lmr_data(gdir, fpath_temp=None, fpath_precip=None,
year_range=('1951', '1980'), filesuffix='', **kwargs):
"""Read, process and store the Last Millennium Reanalysis (LMR) data for this glacier.
LMR data: https://atmos.washington.edu/~hakim/lmr/LMRv2/
LMR data is annualised in anomaly format relative to 1951-1980. We
create synthetic timeseries from the reference data.
It stores the data in a format that can be used by the OGGM mass balance
model and in the glacier directory.
Parameters
----------
fpath_temp : str
path to the temp file (default: LMR v2.1 from server above)
fpath_precip : str
path to the precip file (default: LMR v2.1 from server above)
year_range : tuple of str
the year range for which you want to compute the anomalies. Default
for LMR is `('1951', '1980')`
filesuffix : str
append a suffix to the filename (useful for ensemble experiments).
**kwargs: any kwarg to be passed to ref:`process_gcm_data`
"""
# Get the path of GCM temperature & precipitation data
base_url = 'https://atmos.washington.edu/%7Ehakim/lmr/LMRv2/'
if fpath_temp is None:
with utils.get_lock():
fpath_temp = utils.file_downloader(base_url + 'air_MCruns_ensemble_mean_LMRv2.1.nc')
if fpath_precip is None:
with utils.get_lock():
fpath_precip = utils.file_downloader(
base_url + 'prate_MCruns_ensemble_mean_LMRv2.1.nc')
# Glacier location
glon = gdir.cenlon
glat = gdir.cenlat
# Read the GCM files
with xr.open_dataset(fpath_temp, use_cftime=True) as tempds, \
xr.open_dataset(fpath_precip, use_cftime=True) as precipds:
# Check longitude conventions
if tempds.lon.min() >= 0 and glon <= 0:
glon += 360
# Take the closest to the glacier
# Should we consider GCM interpolation?
temp = tempds.air.sel(lat=glat, lon=glon, method='nearest')
precip = precipds.prate.sel(lat=glat, lon=glon, method='nearest')
# Currently we just take the mean of the ensemble, although
# this is probably not advised. The GCM climate will correct
# anyways
temp = temp.mean(dim='MCrun')
precip = precip.mean(dim='MCrun')
# Precip unit is kg/m^2/s we convert to mm month since we apply the anomaly after
precip = precip * 30.5 * (60 * 60 * 24)
# Back to [-180, 180] for OGGM
temp.lon.values = temp.lon if temp.lon <= 180 else temp.lon - 360
precip.lon.values = precip.lon if precip.lon <= 180 else precip.lon - 360
# OK now we have to turn these annual timeseries in monthly data
# We take the ref climate
fpath = gdir.get_filepath('climate_historical')
with xr.open_dataset(fpath) as ds_ref:
ds_ref = ds_ref.sel(time=slice(*year_range))
loc_tmp = ds_ref.temp.groupby('time.month').mean()
loc_pre = ds_ref.prcp.groupby('time.month').mean()
# Make time coord
t = np.cumsum([31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31] * len(temp))
t = cftime.num2date(np.append([0], t[:-1]), 'days since 0000-01-01 00:00:00',
calendar='noleap')
temp = xr.DataArray((loc_tmp.data + temp.data[:, np.newaxis]).flatten(),
coords={'time': t, 'lon': temp.lon, 'lat': temp.lat},
dims=('time',))
# For precip the std dev is very small - lets keep it as is for now but
# this is a bit ridiculous. We clip to zero here to be sure
precip = utils.clip_min((loc_pre.data + precip.data[:, np.newaxis]).flatten(), 0)
precip = xr.DataArray(precip, dims=('time',),
coords={'time': t, 'lon': temp.lon, 'lat': temp.lat})
process_gcm_data(gdir, filesuffix=filesuffix, prcp=precip, temp=temp,
year_range=year_range, calendar='noleap',
source='lmr', **kwargs)
