def climatologize_ncdt(ncdt):
climdt={}
for ckey in ncdt.keys():
dat = ncdt[ckey]
cdat,_ = cutl.climatologize(dat)
climdt[ckey] = cdat
return climdt
def load_LE(fdict,casename='historical',conv=1,season=None,timesel1='1979-01-01,1989-12-31',timesel2='2002-01-01,2012-12-31'):
ledatc = le.load_LEdata(fdict,casename,timesel=timesel1,
rettype='ndarray',conv=conv,ftype='fullts',local=True,verb=False)
ledatp = le.load_LEdata(fdict,casename,timesel=timesel2,
rettype='ndarray',conv=conv,ftype='fullts',local=True,verb=False)
lediff = ledatp-ledatc
print lediff.shape
leclimo,_ = cutl.climatologize(lediff.T)
print leclimo.shape
return leclimo
cansicnc = cansicnc[:,:,0:-1]
cansicnp = cansicnp[:,:,0:-1]
cansicnd = cansicnp - cansicnc
else:
# SST (var names are sic still...)
field = 'gt'
if doBCs:
bcstr='BC'
# use the actual BC files for the simulations
fhadsic = basepath + 'HadISST/hadisst_kemhadctl_128x64_0001_0125_gt.nc'
fhadsicp = basepath + 'HadISST/hadisst_kemhadpert_128x64_0001_0125_gt.nc'
# or GTadjusted_BC_HadISST_2002-2011_0000120100-0125120100_abs10thresh.nc
hadsicc,hadsiccstd = cutl.climatologize(cnc.getNCvar(fhadsic,'GT')) # I think, K?
hadsicp,hadsicpstd = cutl.climatologize(cnc.getNCvar(fhadsicp,'GT'))
# @@ Hurrell files are same as not doBCs
fhurrsic = basepath + 'HURRELL/MODEL_SST.T42_' + timeper + 'climo.nc' #SST, degC, 128x64
fhurrsicp = basepath + 'HURRELL/MODEL_SST.T42_' + timeperp2 + 'climo.nc' #SST, degC, 128x64
else:
fhadsic = basepath + 'HadISST/hadisst1.1_bc_128_64_1870_2013m03_gt_' + timeper + 'climo.nc' #GT, K, 129x64
fhurrsic = basepath + 'HURRELL/MODEL_SST.T42_' + timeper + 'climo.nc' #SST, degC, 128x64
fhadsicp = basepath + 'HadISST/hadisst1.1_bc_128_64_1870_2013m03_gt_' + timeperp + 'climo.nc' #GT, K, 129x64
fhurrsicp = basepath + 'HURRELL/MODEL_SST.T42_' + timeperp2 + 'climo.nc' #SST, degC, 128x64
hadsicc = cnc.getNCvar(fhadsic,'GT') # I think, K?
hadsicp = cnc.getNCvar(fhadsicp,'GT')
for eii in range(1,ensnum+1):
skey = etype + str(eii)
casenamec = bcasenamec + skey
casenamep = bcasenamep + skey
fnamec = basepath + casenamec+ subdir + casenamec + '_' + field + '_001-121_ts.nc'
fnamep = basepath + casenamep+ subdir + casenamep + '_' + field + '_001-121_ts.nc'
# monthly calc
fldc = cnc.getNCvar(fnamec,ncfield,timesel=timesel)*conv
fldp = cnc.getNCvar(fnamep,ncfield,timesel=timesel)*conv
fldd = fldp-fldc
# take the pattern correlation
flddclimo,flddstd = cutl.climatologize(fldd) # climo first (don't need to do for BCs technically)
flddcclimo,flddcstd = cutl.climatologize(flddc) # climo first. baseline diff data
diffdict[skey] = flddclimo
# for each month, compute pattern corr
pc = np.zeros((12))
for mii,mon in enumerate(con.get_mon()):
tmp = np.squeeze(flddclimo[mii,lat>latlim,...])
tmpcmp = np.squeeze(flddcclimo[mii,lat>latlim,...])
pc[mii] = cutl.pattcorr(tmp.flatten()*weights.flatten(),tmpcmp.flatten()*weights.flatten())
pcmeandict[skey] = pc # monthly
# seasonal calc
fldcsea = np.zeros((4,len(lat),len(lon)))
def getNCvar(filename,field,timesel=None,levsel=None,monsel=None,seas=None,calc=None,remlon=1,sqz=True,att=False):
""" gets a variable from netcdf file.
Time is assumed to be the 1st dimension, Lon is assumed to be the last.
If any calculations are requested to be performed on the data, the user
needs to make sure that the requested operations can be performed (b/c
some of the other functions only handle certain # of dims, etc. My bad.)
filename: full path to file
field: NC variable to read in
timesel: comma-delim string of date range in 'YYYY-MM-DD' fmt to select (a la CDO)
levsel: select level in Pa (e.g. 50000 for 500hPa)
monsel: select month from timeseries
seas: seasonally (annually) average or return climatology {climo|ANN|DJF|JJA|NDJ|MAM|SON}
calc: zm (zonal mean),
remlon: removes extra wrap-around longitude for zonal mean.
default is 1, remove it
sqz: squeeze the data if 'getting all data'. Default True.
Trying to avoid situation where need singular dims and squeeze them out
(e.g. MOC variable in CCSM4)
att: if True, include the full netcdf variable (including attributes). Most useful for time.
returns fld
"""
# IF ON MAC: CDO bindings don't work yet, use old function 3/25/2014 #########
plat = platform.system()
if False: #@@@@TESTING 9/20/2016 plat == 'Darwin': # means I'm on my mac
# Call old func
if calc != None:
print '@@ not sure calc will work on mac. calc=' + calc
if levsel!=None:
plev= np.array([100, 200, 300, 500, 700, 1000, 2000, 3000, 5000, 7000, 10000, 12500,
15000, 17500, 20000, 22500, 25000, 30000, 35000, 40000, 45000, 50000,
55000, 60000, 65000, 70000, 75000, 77500, 80000, 82500, 85000, 87500,
90000, 92500, 95000, 97500, 100000])
level=cutl.find_nearest(plev,levsel)
else:
level=None
if timesel == '0002-01-01,0061-12-31':
print 'hard-coded skipping of first year of 61-yr chunk @@'
fld = getNCvar_old(filename,field,seas=seas, monsel=monsel,timechunk=(12,),level=level,calc=calc,sqz=sqz)
else:
# if timesel=='0002-01-01,0121-12-31' then just don't set timechunk because
# files on the mac are already selected to skip first year, and they reside
# in the 'timsel' subdirectory. Check for that?
if timesel=='0002-01-01,0121-12-31':
if 'timsel/' not in filename:
print 'On mac, use files in timsel/ subdirectory! @@ NEEDS TESTING'
fld = getNCvar_old(filename,field,seas=seas,monsel=monsel,level=level,calc=calc,sqz=sqz,timesel=timesel) # doesn't work with all arguments yet @@
return fld
else: # on linux workstation in Vic
ncfile = openNC(filename)
ndims = len(ncfile.dimensions)
ncvar = ncfile.variables[field]
#print ncvar # @@@@@
#### READ VARIABLE FROM NC FILE ########
if timesel == None and calc == None:
if levsel !=None:
if monsel != None:
fld = np.squeeze(cdo.sellevel(levsel,input = cdo.selmon(monsel,input = filename),returnMaArray = field))
else:
fld = np.squeeze(cdo.sellevel(levsel,input = filename, returnMaArray = field))
os.system('rm -rf /tmp/cdoPy*')
else:
if monsel != None:
#print 'timesel==None and calc==None and monsel !=None'
fld = np.squeeze(cdo.selmon(monsel,input = filename, returnMaArray = field))
#print fld.shape
os.system('rm -rf /tmp/cdoPy*')
else: # get everything
if sqz:
#print field + ': squeezing data upon read all' # @@@
# for most situations, this is what we want. @@@@
fld=np.squeeze(ncfile.variables[field][...])
else:
fld = ncfile.variables[field][...]
elif timesel != None and calc == 'zm':
# have to remove the lon before zonal mean, which means have to separate the
# select dates and zm. thus can't use CDO for zm (unless can pass it data instead of a file?)
#fld = np.squeeze(cdo.zonmean( input = cdo.seldate(timesel,input = filename), returnMaArray = field))
print 'assuming T42(63) 64x128 resolution for zonal mean'
if levsel != None:
if monsel != None:
fld = np.squeeze(cdo.seldate(timesel,input = cdo.zonmean( input =
cdo.selindexbox(1,128,1,64,input =
cdo.sellevel(levsel,input =
cdo.selmon(monsel, input = filename)))),
returnMaArray = field))# @@@@
else:
fld = np.squeeze(cdo.seldate(timesel,input = cdo.zonmean( input =
cdo.selindexbox(1,128,1,64,input =
cdo.sellevel(levsel,input = filename))),
returnMaArray = field))
else:
if monsel != None:
fld = np.squeeze(cdo.seldate(timesel,input =
cdo.zonmean( input =
cdo.selindexbox(1,128,1,64,input =
cdo.selmon(monsel,input = filename))),
returnMaArray = field))
else:
fld = np.squeeze(cdo.seldate(timesel,input =
cdo.zonmean(input =
cdo.selindexbox(1,128,1,64,input = filename)),
returnMaArray = field))
os.system('rm -rf /tmp/cdoPy*')
## if remlon:
## # remove extra lon
## fld = np.squeeze(fld[...,0:-1])
## lastdimidx = ndims-1
## fld = np.mean(fld,lastdimidx)
elif timesel != None and calc != None:
if levsel != None and monsel == None:
fld = np.squeeze(cdo.seldate(timesel,input =
cdo.sellevel(levsel,input = filename),
returnMaArray = field))
elif levsel != None and monsel != None:
fld = np.squeeze(
cdo.seldate(timesel,input =
cdo.sellevel(levsel,input =
cdo.selmon(monsel,input = filename)),
returnMaArray = field))
elif levsel == None and monsel != None:
fld = np.squeeze(cdo.seldate(timesel,input =
cdo.selmon(monsel,input = filename),
returnMaArray = field))
else: # levsel and monsel are both None
fld = np.squeeze(cdo.seldate(timesel,input = filename, returnMaArray = field))
os.system('rm -rf /tmp/cdoPy*')
print "only calc='zm' is implemented now. Returning only selected date range/level/month."
elif timesel != None:
if levsel != None and monsel == None:
fld = np.squeeze(cdo.seldate(timesel,input = cdo.sellevel(levsel,input = filename),returnMaArray = field))
elif levsel != None and monsel != None:
fld = np.squeeze(
cdo.seldate(timesel,input =
cdo.sellevel(levsel,input =
cdo.selmon(monsel,input = filename)),
returnMaArray = field))
elif levsel == None and monsel != None:
fld = np.squeeze(cdo.seldate(timesel,input =
cdo.selmon(monsel,input = filename),
returnMaArray = field))
else: # levsel and monsel are both None
fld = np.squeeze(cdo.seldate(timesel,input = filename, returnMaArray = field))
os.system('rm -rf /tmp/cdoPy*')
elif calc == 'zm': # and timesel must be None
print 'assuming T42(63) 64x128 resolution for zonal mean'
if levsel != None and monsel == None:
fld = np.squeeze(cdo.sellevel(levsel,input =
cdo.zonmean(input =
cdo.selindexbox(1,128,1,64,input =
filename)),
returnMaArray = field))
elif levsel != None and monsel != None:
fld = np.squeeze(
cdo.sellevel(levsel,input =
cdo.zonmean(input =
cdo.selindexbox(1,128,1,64,input =
cdo.selmon(monsel,input =
filename))),
returnMaArray = field))
elif levsel == None and monsel != None:
fld = np.squeeze(cdo.zonmean(input =
cdo.selindexbox(1,128,1,64,input =
cdo.selmon(monsel,input =
filename)),
returnMaArray = field))
else: # get all data
fld = np.squeeze(cdo.zonmean(input =
cdo.selindexbox(1,128,1,64,input =
filename),
returnMaArray = field))
#print '@@ getting memory errors here...try using CDO to select appropriate lons for the zm calc'
#fld = ncfile.variables[field][...] # have to get field before removing lon
os.system('rm -rf /tmp/cdoPy*')
## if remlon:
## # remove extra lon
## if ndims==4:
## fld = np.squeeze(fld[:,:,:,0:-1])
## elif ndims==3:
## fld = np.squeeze(fld[:,:,0:-1])
## else: # shouldn't really get here, not expecting 2D (time x lon?)
## fld = np.squeeze(fld[:,0:-1])
## lastdimidx = ndims-1
## fld = np.mean(fld,lastdimidx)
else:
print "huh? timesel and calc combo doesn't make sense"
####### TIME AVERAGE the VARIABLE ##########
# fld has to be 3d by the time it is passed to func
# (time,lev,lat) or (time,lat,lon)
if seas != None:
#print 'getNCvar(): seas!=None: fld.shape: ' + str(fld.shape) # @@@
## if fld.ndim != 3:
## ## if 1 in fld.shape:
## ## fld=fld.squeeze() # attempting to deal with spurious dims of 1 @@@
## ## if fld.ndim != 3:
## ## print 'data must be 3 dimensional to seasonalize()'
## ## return
## ## else:
## print 'data must be 3 dimensional to seasonalize()'
## return
if monsel != None:
print "Can't do seasonal average when monsel != None"
return
elif seas == 'climo':
fld,stddev = cutl.climatologize(fld)
elif type(seas) == int: # @@ does this work?
#elif seas not in ('ANN','DJF','JJA','MAM','SON','NDJ'):
# means seas is an int value for a month
#fld = cutl.seasonalize_monthlyts(fld,mo=seas)
fld = cutl.seasonalize(fld,mo=seas)
else:
#print 'seasonalizing'
#fld = cutl.seasonalize_monthlyts(fld,season=seas)
fld = cutl.seasonalize(fld,season=seas)
#print fld.shape
# Apply any scaling and offsetting needed:
try:
var_offset = ncvar.add_offset
except:
var_offset = 0
try:
var_scale = ncvar.scale_factor
print 'var_scale ' + str(var_scale)
except:
var_scale = 1
fld = fld*var_scale + var_offset
ncfile.close()
return fld
plat = platform.system()
if plat == 'Darwin': # means I'm on my mac
basepath = '/Users/kelly/CCCma/CanSISE/RUNS/' # @@ needs updating
subdir = '/'
else: # on linux workstation in Vic
basepath = '/home/rkm/work/DATA/' + model + '/'
subdir = '/ts/'
# get sea ice concentration for averaging purposes
field = 'sicn'
fnamec = basepath + casenamec + subdir + casenamec + '_' + field + '_' + timstr + '_ts.nc'
sicnc = cnc.getNCvar(fnamec,field.upper(),timesel=timesel)
sicnc,siccstd = cutl.climatologize(sicnc) # use to mask the flux region
fnamep = basepath + casenamep + subdir + casenamep + '_' + field + '_' + timstr + '_ts.nc'
sicnp = cnc.getNCvar(fnamep,field.upper(),timesel=timesel)
sicnp,sicpstd = cutl.climatologize(sicnp) # use to mask the flux region
# get second set of sims:
fnamecb = basepath + casenamecb + subdir + casenamecb + '_' + field + '_' + timstrb + '_ts.nc'
sicncb = cnc.getNCvar(fnamecb,field.upper(),timesel=timeselb)
sicncb,siccstdb = cutl.climatologize(sicncb) # use to mask the flux region
fnamepb = basepath + casenamepb + subdir + casenamepb + '_' + field + '_' + timstrb + '_ts.nc'
sicnpb = cnc.getNCvar(fnamepb,field.upper(),timesel=timeselb)
sicnpb,sicpbstd = cutl.climatologize(sicnpb) # use to mask the flux region
plt.figure()
plt.plot(mons,pvalp1,'b')
plt.plot(mons,pvalp2,'r')
plt.plot(mons,pvalp3,'g')
plt.plot((1,12),(siglevel,siglevel),'k')
plt.xlim((1,12))
plt.title('pval')
# first try just the polar mean
fldcclimpm = cutl.polar_mean_areawgted3d(fldcclim,lat,lon,latlim=latlim)
fldp1climpm = cutl.polar_mean_areawgted3d(fldp1clim,lat,lon,latlim=latlim)
fldp2climpm = cutl.polar_mean_areawgted3d(fldp2clim,lat,lon,latlim=latlim)
fldp3climpm = cutl.polar_mean_areawgted3d(fldp3clim,lat,lon,latlim=latlim)
# get stddev of the already polar averaged data!
tmpc,fldcstdpm = cutl.climatologize(fldcpm)
tmpp1,fldp1stdpm = cutl.climatologize(fldp1pm)
tmpp2,fldp2stdpm = cutl.climatologize(fldp2pm)
tmpp3,fldp3stdpm = cutl.climatologize(fldp3pm)
## fldcstdpm = cutl.polar_mean_areawgted3d(fldcstd,lat,lon,latlim=latlim) # this is wrong metric
## fldp1stdpm = cutl.polar_mean_areawgted3d(fldp1std,lat,lon,latlim=latlim)
## fldp2stdpm = cutl.polar_mean_areawgted3d(fldp2std,lat,lon,latlim=latlim)
## fldp3stdpm = cutl.polar_mean_areawgted3d(fldp3std,lat,lon,latlim=latlim)
# mean values
fig = plt.figure()
plt.plot(mons,fldcclimpm,'k',linewidth=2)
plt.plot(mons,fldp1climpm,'b',linewidth=2)
plt.plot(mons,fldp2climpm,'r',linewidth=2)
def pattcorr_ensemble(ename, field, latlim=60):
# @@@@@@@@@@@@ is this fully implemented? Don't think so. 12/2/14
if ename=='ANT':
ename='HistIC'
elif ename=='TOT':
ename='HistBC'
enssims = con.build_ensemblesims(ename)
ensnum=len(enssims)
# ======= copied from Canam4_BCpatterncorr-Copy0.py ===========
#ensnum=5
diffdict = {}
pcmeandict = {} # fldp-fldc pattern corr compared to mean BC
pchaddict = {} # fldp-fldc pattern corr compared to hadisst
seadiffdict = {} # seasonal mean
pcseameandict = {}
pcsea2meandict = {} # to test the other pattern corr calc
pcsea2pvalmeandict = {} # to test the other pattern corr calc
# generate weights for the pattern corr
lat = con.get_t63lat()
lon = con.get_t63lon()
areas = cutl.calc_cellareas(lat,lon)
areas = areas[lat>latlim,:]
weights = areas / np.sum(np.sum(areas,axis=1),axis=0)
#for eii in range(1,ensnum+1):
for skey in enssims:
#skey = etype + str(eii)
#casenamec = bcasenamec + skey
#casenamep = bcasenamep + skey
#fnamec = basepath + casenamec+ subdir + casenamec + '_' + field + '_001-121_ts.nc'
#fnamep = basepath + casenamep+ subdir + casenamep + '_' + field + '_001-121_ts.nc'
fnamec,fnamep = con.build_filepathpair(skey,field)
# monthly calc
fldc = cnc.getNCvar(fnamec,ncfield,timesel=timesel)*conv
fldp = cnc.getNCvar(fnamep,ncfield,timesel=timesel)*conv
fldd = fldp-fldc
# take the pattern correlation
flddclimo,flddstd = cutl.climatologize(fldd) # climo first (don't need to do for BCs technically)
flddcclimo,flddcstd = cutl.climatologize(flddc) # climo first. baseline diff data
diffdict[skey] = flddclimo
# for each month, compute pattern corr
pc = np.zeros((12))
for mii,mon in enumerate(con.get_mon()):
tmp = np.squeeze(flddclimo[mii,lat>latlim,...])
tmpcmp = np.squeeze(flddcclimo[mii,lat>latlim,...])
pc[mii] = cutl.pattcorr(tmp.flatten()*weights.flatten(),tmpcmp.flatten()*weights.flatten())
pcmeandict[skey] = pc # monthly
# seasonal calc
fldcsea = np.zeros((4,len(lat),len(lon)))
fldpsea = np.zeros((4,len(lat),len(lon)))
flddsea = np.zeros((4,len(lat),len(lon)))
pcsea = np.zeros((4))
pcsea2 = np.zeros((4)) # test pattcorr_pearson() @@
pcsea2pval = np.zeros((4)) # test pattcorr_pearson()
for seaii,sea in enumerate(seasons):
fldcsea[seaii,...] = np.mean(cnc.getNCvar(fnamec,ncfield,timesel=timesel,seas=sea)*conv,axis=0)
fldpsea[seaii,...] = np.mean(cnc.getNCvar(fnamep,ncfield,timesel=timesel,seas=sea)*conv,axis=0)
flddsea[seaii,...] = fldpsea[seaii,...]-fldcsea[seaii,...]
tmp = np.squeeze(flddsea[seaii,lat>latlim,...])
tmpcmp = np.squeeze(flddcsea[seaii,lat>latlim,...])
pcsea[seaii] = cutl.pattcorr(tmp.flatten()*weights.flatten(),
tmpcmp.flatten()*weights.flatten())
pcsea2[seaii],pcsea2pval[seaii] = cutl.pattcorr_pearson(tmp.flatten()*weights.flatten(),
tmpcmp.flatten()*weights.flatten())
seadiffdict[skey] = flddsea
pcseameandict[skey] = pcsea
pcsea2meandict[skey] = pcsea2
pcsea2pvalmeandict[skey] = pcsea2pval
def pattcorr_withinensemble(ename,fdict,latlim=60,timesel='0002-01-01,0121-12-31'):
""" pattcorr_withinensemble(ename,field,latlim=60)
pattern corr each member of ensemble with each other one
return pctable, pctablesea (DataFrames)
"""
# @@ need diffdict
field=fdict['field']
ncfield=fdict['ncfield']
conv=fdict['conv']
seasons=('SON','DJF','MAM','JJA')
if ename=='ANT':
ename='histIC'
elif ename=='TOT':
ename='histBC'
enssims = con.build_ensemblesims(ename)
ensnum=len(enssims)
print 'ENSSIMS: ' # @@@
print enssims # @@
# generate weights for the pattern corr
lat = con.get_t63lat()
lon = con.get_t63lon()
areas = cutl.calc_cellareas(lat,lon)
areas = areas[lat>latlim,:]
weights = areas / np.sum(np.sum(areas,axis=1),axis=0)
# ========= create diffdict first =====
diffdict={}
seadiffdict={}
for skey in enssims:
fnamec,fnamep = con.build_filepathpair(skey,field)
# monthly calc
fldc = cnc.getNCvar(fnamec,ncfield,timesel=timesel)*conv
fldp = cnc.getNCvar(fnamep,ncfield,timesel=timesel)*conv
fldd = fldp-fldc
# Monthly
flddclimo,flddstd = cutl.climatologize(fldd) # climo first (don't need to do for BCs technically)
#flddcclimo,flddcstd = cutl.climatologize(flddc) # climo first. baseline diff data
diffdict[skey] = flddclimo
print skey + ' ' + str(flddclimo.shape) # @@@
# Seasonal
flddsea = np.zeros((4,len(lat),len(lon)))
for seaii,sea in enumerate(seasons):
fldcsea = np.mean(cnc.getNCvar(fnamec,ncfield,timesel=timesel,seas=sea)*conv,axis=0)
fldpsea = np.mean(cnc.getNCvar(fnamep,ncfield,timesel=timesel,seas=sea)*conv,axis=0)
flddsea[seaii,...] = fldpsea-fldcsea
seadiffdict[skey] = flddsea
# ======= Now do pattern corrs within ensemble ====
# ======= copied from Canam4_BCpatterncorr-Copy0.py ===========
outterdict= dict.fromkeys(enssims)
for skey1 in enssims:
outfld = diffdict[skey1]
innerdict = dict.fromkeys(enssims)
for skey2 in enssims:
#print skey1 + ' compared to ' + skey2
infld = diffdict[skey2]
# for each month, compute pattern corr
pc = np.zeros((12))
for mii,mon in enumerate(con.get_mon()):
tmp = np.squeeze(infld[mii,lat>latlim,...])
tmpcmp = np.squeeze(outfld[mii,lat>latlim,...])
pc[mii] = cutl.pattcorr(tmp.flatten()*weights.flatten(),
tmpcmp.flatten()*weights.flatten())
innerdict[skey2] = pc
outterdict[skey1] = innerdict
pctable = pd.DataFrame(outterdict) # 5x5
# seasonal
outterdictsea= dict.fromkeys(enssims)
for skey1 in enssims:
outfld = seadiffdict[skey1]
innerdictsea = dict.fromkeys(enssims)
for skey2 in enssims:
#print skey1 + ' compared to ' + skey2
infld = seadiffdict[skey2]
# for each season, compute pattern corr
pcsea = np.zeros((4))
for seaii,sea in enumerate(seasons):
tmp = np.squeeze(infld[seaii,lat>latlim,...])
tmpcmp = np.squeeze(outfld[seaii,lat>latlim,...])
pcsea[seaii] = cutl.pattcorr(tmp.flatten()*weights.flatten(),
tmpcmp.flatten()*weights.flatten())
innerdictsea[skey2] = pcsea
outterdictsea[skey1] = innerdictsea
pctablesea = pd.DataFrame(outterdictsea) # 5x5
return pctable, pctablesea
# # Compare 2xco2 with the 2xco2 nudge: RMSE
# calculate RMSE
rmsedt = {}
rmseclimdt = {}
annrmseclimdt = {}
climrmsedt = {}
region = "polcap60"
fldclimdt = {}
flddiffdt = {}
flddiffclimdt = {}
# get controls
fldc2x = flddt["gregory_2xco2"]
fldc2xclim, std = cutl.climatologize(fldc2x) # climo mean (2xco2)
fldclimdt["gregory_2xco2"] = fldc2xclim
fldc = flddt["iga"]
fldcclim, std = cutl.climatologize(fldc) # climo mean (iga)
fldclimdt["iga"] = fldcclim
for skey in sims[2:10]: # Group I
fld = flddt[skey]
# timeseries
flddiffdt[skey] = fldc2x - fld
# climos
fldclimdt[skey], std = cutl.climatologize(fld)
flddiffclimdt[skey] = fldc2xclim - fldclimdt[skey]
rmse = np.sqrt(np.square(flddiffdt[skey]))
ctldt = dict.fromkeys(models)
p1dt = dict.fromkeys(models) # 2002-2005
for mod in cm5.models:
print mod
meta=cm5.allmodeldt[mod]
if meta==None:
print mod + ' not yet implemented!' # @@@
elif type(meta['fyears'])!=str:
# skip
print ' skipping models w/ annoying years for now'
else:
fyears=meta['fyears']
enum=meta['numens'] # number of ens members
ctlmoddt={}; p1moddt={}
for eii in range(1,enum+1):
if 'skipens' in meta.keys() and eii in meta['skipens']:
print 'skipping ensemble: ' + str(eii)
else:
fname=bp+mod+'/r'+str(eii)+'i1p1/' + field + '_' + comp + '_' +\
mod + '_historical_r' + str(eii)+'i1p1_' + fyears +'.nc'
print fname
ctlmoddt[eii],junk = cutl.climatologize(cnc.getNCvar(fname,field,timesel=timeselc))
p1moddt[eii],junk = cutl.climatologize(cnc.getNCvar(fname,field,timesel=timeselp1))
ctldt[mod]=ctlmoddt
p1dt[mod]=p1moddt
