In [1]:
from IPython.display import Image
Image(url='https://vesg.ipsl.upmc.fr/thredds/fileServer/IPSLFS/jservon/CliMAF_Notebooks_html/CliMAF-logo-small.png')
Out[1]:
A science-oriented framework to ease the analysis of climate model simulations¶
WP5 ANR Convergence¶
Development team: Stéphane Sénési (CNRM-GAME), Gaëlle Rigoudy (CNRM-GAME), Jérôme Servonnat (LSCE-IPSL), Ludivine Vignon (CNRM-GAME), Laurent Franchisteguy (CNRM-GAME), Patrick Brockmann (LSCE-IPSL)¶
Beta-testing: Olivier Marti (LSCE-IPSL), Marie-Pierre Moine (CERFACS), Emilia Sanchez-Gomez (CERFACS)¶
¶
contact: climaf@meteo.fr¶
users list: climaf-users@meteo.fr¶
Working with CliMAF on CORDEX ensembles¶
In this notebook we will see how to make a multi-model ensemble with CORDEX. Compared with CMIP5 and CMIP6, CORDEX datasets are defined not only by one model but by a GCM/RCM couple (driving_model/model).
- We will see how to make the most of CliMAF to work on those CORDEX ensembles.
- and we will see how to access the land/sea mask
¶
First, import climaf¶
In [2]:
from climaf.api import *
And set verbosity ('critical' -> minimum ; 'debug' -> maximum)¶
In [3]:
clog('critical') # min verbosity = critical < warning < info < debug = max verbosity
1. Make a CORDEX ensemble¶
The specificity of CORDEX is that it works as GCM/RCM couples. We propose here a way to build a multi-model CORDEX ensemble
In [4]:
from natsort import natsorted
# -- Let's make a dictionary to specify the general request:
req_dict = dict(project = 'CORDEX',
CORDEX_domain = 'EUR-11',
frequency = 'daily',
experiment = 'historical',
period = '2000',
variable = 'tas'
)
# -- Do the initial request to retrieve all the possible results:
req = ds(model = '*',
driving_model = '*',
realization = '*',
**req_dict)
# -- We will now:
# - make a loop on all the available 'driving_model' (GCMs)
# - for each driving_model, we check the model (RCMs) available
# - and for each GCM/RCM couple, we check the available realizations
# - store all the matching datasets in myens_dict
# - and create the ensemble with cens
myens_dict = dict()
for GCM in req.explore('choices')['driving_model']:
dumreq = ds(model='*',
realization = '*',
driving_model = GCM,
**req_dict)
RCMs = dumreq.explore('choices')['model']
if not isinstance(RCMs, list):
RCMs = [RCMs]
for RCM in RCMs:
dum2req = ds(model=RCM,
driving_model = GCM,
realization = '*',
**req_dict)
print 'GCM = ',GCM,' RCM = ',RCM
res_choices = dum2req.explore('choices')
if res_choices:
# -- If multiple realizations are available, we take the first one
if isinstance(res_choices['realization'], list):
dum2req.kvp['realization'] = natsorted(dum2req.explore('choices')['realization'])[0]
myens_dict[GCM+'_'+RCM] = dum2req.explore('resolve')
#
my_CORDEX_ens = cens(myens_dict)
In [5]:
clog('critical')
In [6]:
summary(my_CORDEX_ens)
2. Access the mask and land/area fraction sftlf: r0i0p0 and fx¶
In [7]:
from climaf.api import *
In [8]:
req_dict = dict(project = 'CORDEX',
CORDEX_domain = 'EUR-11',
frequency = 'fx',
experiment = '*',
period = 'fx',
realization = 'r0i0p0',
variable = 'sftlf'
)
# -- Do the initial request to retrieve all the possible results:
req = ds(model = '*',
driving_model = '*',
**req_dict)
req.explore('choices')
Out[8]:
And of course, you can apply all the pretreatments you want to the members of your ensemble before getting your list of files.¶
For the available operators, we invite you to go deeper in CliMAF documentation and other CliMAF notebooks.