-
Notifications
You must be signed in to change notification settings - Fork 11
/
analysis.py
327 lines (262 loc) · 8.94 KB
/
analysis.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
import os
import numpy as np
import xarray as xr
import cftime
import pandas as pd
import matplotlib
import matplotlib.pyplot as plt
import glob
import dask
def get_files(exp,tape='h0',yy=()):
top='/glade/campaign/asp/djk2120/PPEn11/'
d=top+exp+'/hist/'
oaats=['CTL2010','C285','C867','AF1855','AF2095','NDEP']
key={oaat:'/glade/campaign/asp/djk2120/PPEn11/csvs/surv.csv' for oaat in oaats}
yys={oaat:(2005,2014) for oaat in oaats}
key['transient']='/glade/campaign/asp/djk2120/PPEn11/csvs/lhc220926.txt'
yys['transient']=(1850,2014)
key['EmBE']='/glade/work/linnia/CLM-PPE-LAI_tests/exp1_EmBE/psets_exp1_EmBE_230419.txt' #LRH
yys['EmBE']=(1850,2014) #LRH
df=pd.read_csv(key[exp])
if not yy:
yr0,yr1=yys[exp]
else:
yr0,yr1=yy
if exp=='transient' or exp=='EmBE': #LRH
keys = df.member.values
appends={}
params=[]
for p in df.keys():
if p!='member':
appends[p]=xr.DataArray(np.concatenate(([np.nan],df[p].values)),dims='ens')
params.append(p)
appends['params']=xr.DataArray(params,dims='param')
if exp=='transient': #LRH
keys=np.concatenate((['LHC0000'],keys))
else: #LRH
keys=np.concatenate((['exp1_EmBE0001'],keys)) #LRH
appends['key']=xr.DataArray(keys,dims='ens')
else:
keys=df.key.values
appends={v:xr.DataArray(df[v].values,dims='ens') for v in ['key','param','minmax']}
fs = np.array(sorted(glob.glob(d+'*'+tape+'*')))
yrs = np.array([int(f.split(tape)[1][1:5]) for f in fs])
#bump back yr0, if needed
uyrs=np.unique(yrs)
yr0=uyrs[(uyrs/yr0)<=1][-1]
#find index to subset files
ix = (yrs>=yr0)&(yrs<=yr1)
fs = fs[ix]
#organize files to match sequence of keys
ny=len(np.unique(yrs[ix]))
if exp=='EmBE': #LRH
fkeys=np.array([f.split('transient_')[1].split('.')[0] for f in fs]) #LRH
else: #LRH
fkeys=np.array([f.split(exp+'_')[1].split('.')[0] for f in fs])
if ny==1:
files=[fs[fkeys==k][0] for k in keys]
dims = 'ens'
else:
files=[list(fs[fkeys==k]) for k in keys]
dims = ['ens','time']
#add landarea information
if exp=='transient' or 'EmBE': #LRH
fla='landarea_transient.nc'
else:
fla='landarea_oaat.nc'
la=xr.open_dataset(fla)
appends['la']=la.landarea
if tape=='h1':
appends['lapft']=la.landarea_pft
return files,appends,dims
def get_ds(files,dims,dvs=[],appends={},singles=[]):
if dvs:
def preprocess(ds):
return ds[dvs]
else:
def preprocess(ds):
return ds
ds = xr.open_mfdataset(files,combine='nested',concat_dim=dims,
parallel=True,
preprocess=preprocess)
f=np.array(files).ravel()[0]
htape=f.split('clm2')[1][1:3]
#add extra variables
tmp = xr.open_dataset(f)
for v in tmp.data_vars:
if 'time' not in tmp[v].dims:
if v not in ds:
ds[v]=tmp[v]
#fix up time dimension, swap pft
if (htape=='h0')|(htape=='h1'):
yr0=str(ds['time.year'][0].values)
nt=len(ds.time)
ds['time'] = xr.cftime_range(yr0,periods=nt,freq='MS',calendar='noleap') #fix time bug
if (htape=='h1'):
ds['pft']=ds['pfts1d_itype_veg']
for append in appends:
ds[append]=appends[append]
return ds
def get_exp(exp,dvs=[],tape='h0',yy=(),defonly=False):
'''
exp: 'transient','CTL2010','C285','C867','AF1855','2095','NDEP'
dvs: e.g. ['TLAI'] or [] returns all available variables
tape: 'h0','h1',etc.
yy: e.g. (2005,2014) or () returns all available years
'''
files,appends,dims=get_files(exp,tape=tape,yy=yy)
if defonly:
files=files[0]
dims='time'
ds=get_ds(files,dims,dvs=dvs,appends=appends)
f,a,d=get_files(exp,tape='h0',yy=yy)
singles=['RAIN','SNOW','TSA','RH2M','FSDS','WIND']
tmp=get_ds(f[0],'time',dvs=singles)
for s in singles:
ds[s]=tmp[s]
if len(yy)>0:
ds=ds.sel(time=slice(str(yy[0]),str(yy[1])))
ds['PREC']=ds.RAIN+ds.SNOW
t=ds.TSA-273.15
rh=ds.RH2M/100
es=0.61094*np.exp(17.625*t/(t+234.04))
ds['VPD']=((1-rh)*es).compute()
ds['VPD'].attrs={'long_name':'vapor pressure deficit','units':'kPa'}
ds['VP']=(rh*es).compute()
ds['VP'].attrs={'long_name':'vapor pressure','units':'kPa'}
whit = xr.open_dataset('./whit/whitkey.nc')
ds['biome']=whit.biome
ds['biome_name']=whit.biome_name
#get the pft names
pfts=xr.open_dataset('/glade/campaign/asp/djk2120/PPEn11/paramfiles/OAAT0000.nc').pftname
pfts=[str(p)[2:-1].strip() for p in pfts.values][:17]
ds['pft_name']=xr.DataArray(pfts,dims='pft_id')
return ds
def amean(da,cf=1/365):
#annual mean
m = da['time.daysinmonth']
xa = cf*(m*da).groupby('time.year').sum().compute()
xa.name=da.name
return xa
def gmean(da,la,g=[],cf=None,u=None):
'''
g defines the averaging group,
g=[] is global, otherwise use ds.biome or ds.pft
'''
if len(g)==0:
g=xr.DataArray(np.tile('global',len(da.gridcell)),dims='gridcell')
if not cf:
cf=1/la.groupby(g).sum()
with dask.config.set(**{'array.slicing.split_large_chunks': True}):
x=cf*(da*la).groupby(g).sum()
x.name=da.name
x.attrs=da.attrs
if u:
x.attrs['units']=u
if 'group' in x.dims:
x=x.isel(group=0)
if len(x.dims)>0:
if x.dims[0]!='ens':
x=x.T
return x.compute()
def get_ix(ds,pft):
ix=ds.pfts1d_itype_veg==pft
a=ds.pfts1d_lat[ix]
o=ds.pfts1d_lon[ix]
nlon=len(ds.lon)
nlat=len(ds.lat)
nx=len(a)
lats=xr.DataArray(np.tile(ds.lat.values.reshape([-1,1,1]),[1,nlon,nx]),dims=['lat','lon','pft'])
lons=xr.DataArray(np.tile(ds.lon.values.reshape([1,-1,1]),[nlat,1,nx]),dims=['lat','lon','pft'])
ix=((abs(lats-a)<0.25)&(abs(lons-o)<0.25)).sum(dim='pft')
return ix==1
def pftgrid(da,ds):
#set up dims for outgoing data array
dims=[]
s=[]
ix=get_ix(ds,1)
for dim in da.dims:
if dim !='pft':
dims.append(dim)
s.append(len(da[dim]))
dims=[*dims,*ix.dims]
s=[*s,*ix.shape]
ndims=len(dims)
das=[]
ix0=[slice(None) for i in range(ndims-2)]
pfts=np.unique(ds.pfts1d_itype_veg)
for pft in pfts:
out=np.zeros(s)+np.nan
if pft>0:
ix=get_ix(ds,pft)
ix2=tuple([*ix0,ix])
ixp=ds.pfts1d_itype_veg==pft
out[ix2]=da.isel(pft=ixp)
das.append(xr.DataArray(out.copy(),dims=dims))
da_out=xr.concat(das,dim='pft')
da_out['pft']=pfts
da_out['lat']=ds.lat
da_out['lon']=ds.lon
for dim in da.dims:
if dim !='pft':
da_out[dim]=da[dim]
return da_out
def get_map(da,sgmap=None):
if not sgmap:
sgmap=xr.open_dataset('sgmap.nc')
return da.sel(gridcell=sgmap.cclass).where(sgmap.notnan).compute()
def find_pair(da,params,minmax,p):
'''
returns a subset of da, corresponding to parameter-p
the returned pair corresponds to [p_min,p_max]
'''
ixmin = np.logical_and(params==p,minmax=='min')
ixmax = np.logical_and(params==p,minmax=='max')
#sub in default if either is missing
if ixmin.sum().values==0:
ixmin = params=='default'
if ixmax.sum().values==0:
ixmax = params=='default'
emin = da.ens.isel(ens=ixmin).values[0]
emax = da.ens.isel(ens=ixmax).values[0]
return da.sel(ens=[emin,emax])
def top_n(da,nx):
''' return top_n by param effect '''
dx=abs(da.sel(minmax='max')-da.sel(minmax='min'))
ix=dx.argsort()[-nx:].values
x=da.isel(param=ix)
return x
def rank_plot(da,nx,ax=None):
x = top_n(da,nx)
xdef = da.sel(param='default',minmax='min')
if not ax:
fig=plt.figure()
ax=fig.add_subplot()
ax.plot([xdef,xdef],[0,nx-1],'k:',label='default')
ax.scatter(x.sel(minmax='min'),range(nx),marker='o',facecolors='none', edgecolors='r',label='low-val')
ax.plot(x.sel(minmax='max'),range(nx),'ro',label='high-val')
i=-1
for xmin,xmax in x:
i+=1
ax.plot([xmin,xmax],[i,i],'r')
ax.set_yticks(range(nx))
ax.set_yticklabels([p[:15] for p in x.param.values])
def brown_green():
'''
returns a colormap based on colorbrewer diverging brown->green
'''
# colorbrewer colormap, diverging, brown->green
cmap = np.zeros([11,3]);
cmap[0,:] = 84,48,5
cmap[1,:] = 140,81,10
cmap[2,:] = 191,129,45
cmap[3,:] = 223,194,125
cmap[4,:] = 246,232,195
cmap[5,:] = 245,245,245
cmap[6,:] = 199,234,229
cmap[7,:] = 128,205,193
cmap[8,:] = 53,151,143
cmap[9,:] = 1,102,94
cmap[10,:] = 0,60,48
cmap = matplotlib.colors.ListedColormap(cmap/256)
return cmap