')
+ htmlfile.write('
| '+season+'')
+ htmlfile.write(' |
|
')
diff --git a/arm_diags/src/diurnal_cycle.py b/arm_diags/src/diurnal_cycle.py
index 0e8c44b..d5dc8dd 100644
--- a/arm_diags/src/diurnal_cycle.py
+++ b/arm_diags/src/diurnal_cycle.py
@@ -113,6 +113,7 @@ def diurnal_cycle_data(parameter):
print(test_path,test_model,sites[0][:3]+test_model+'*hr' + sites[0][3:5].upper())
test_file = glob.glob(os.path.join(test_path,sites[0][:3]+test_model+'*hr' + sites[0][3:5].upper()+'*.nc' ))
+
if len(test_file) == 0:
print('No diurnal data for test model were found: '+sites[0])
@@ -123,6 +124,7 @@ def diurnal_cycle_data(parameter):
#initialize the indicator for temporal res of testmodel
test_tres = test_file[0].split(test_model)[-1][:3] #e.g., '3hr', '1hr'
+
fin = cdms2.open(test_file[0])
if test_tres == '1hr': test_tidlen=24
if test_tres == '3hr': test_tidlen=8
@@ -142,6 +144,7 @@ def diurnal_cycle_data(parameter):
print((variable+" "+season+" not processed for " + test_model))
print('!!please check the start and end year in basicparameter.py')
test_findex = 0
+
# Calculate for observational data
obs_var_season=np.empty([len(variables),24,len(seasons)])*np.nan
#obs_file = glob.glob(os.path.join(obs_path,'*ARMdiag_domain_diurnal*.nc')) #read in diurnal test data
@@ -225,6 +228,7 @@ def diurnal_cycle_data(parameter):
os.makedirs(os.path.join(output_path,'metrics',sites[0]))
for j, variable in enumerate(variables):
for k, season in enumerate(seasons):
+
if test_findex == 1: np.savetxt(output_path+'/metrics/'+sites[0]+'/'+variable+'_'+season+'_test_diurnal_cycle_'+test_tres+'_'+sites[0]+'.csv',test_var_season[j,:,k])
np.savetxt(output_path+'/metrics/'+sites[0]+'/'+variable+'_'+season+'_mmm_diurnal_cycle_'+sites[0]+'.csv',mmm_var_season[j,:,k])
np.savetxt(output_path+'/metrics/'+sites[0]+'/'+variable+'_'+season+'_cmip_diurnal_cycle_'+sites[0]+'.csv',cmip_var_season[:,j,:,k])
diff --git a/arm_diags/src/diurnal_cycle_LAcoupling.py b/arm_diags/src/diurnal_cycle_LAcoupling.py
new file mode 100644
index 0000000..774b2b4
--- /dev/null
+++ b/arm_diags/src/diurnal_cycle_LAcoupling.py
@@ -0,0 +1,155 @@
+#===========================================================================================================================
+# Program for generate the diurnal cycle on Land-Atmosphere Coupling
+#---------------------------------------------------------------------------------------------------------------------------
+# V4 Development
+ # ----------------------------------------------------------------------------------------------------
+ # Cheng Tao - Jul2024 @ LLNL
+ # ### Diurnal cycle of SH, LH, T_srf, RH_srf, LCL, PBL
+ # ----------------------------------------------------------------------------------------------------
+#===========================================================================================================================
+import os
+import pdb
+import glob
+import cdms2
+import cdutil
+import numpy as np
+import matplotlib.pyplot as plt
+from matplotlib.gridspec import GridSpec
+from .varid_dict import varid_longname
+from .taylor_diagram import TaylorDiagram
+from .utils import climo
+from .utils import get_diurnal_cycle_seasons
+import MV2
+import matplotlib.gridspec as gridspec
+import scipy.stats
+from scipy import interpolate
+import calendar
+import math
+
+def diurnal_cycle_LAcoupling_plot(parameter):
+ variables = parameter.variables
+ test_path = parameter.test_data_path
+ test_model = parameter.test_data_set
+ obs_path = parameter.obs_path
+ cmip_path = parameter.cmip_path
+ output_path = parameter.output_path
+ sites = parameter.sites
+ varnames = parameter.varnames
+ units = parameter.units
+ seasons = parameter.season
+ nseasons = len(seasons)
+ nvariables = len(variables)
+
+ print('test_path: ',test_path)
+
+ if not os.path.exists(os.path.join(output_path,'figures',sites[0])):
+ os.makedirs(os.path.join(output_path,'figures',sites[0]))
+
+ #==========================================================================
+ # Calculate for test model data
+ #==========================================================================
+ test_file = glob.glob(os.path.join(test_path,sites[0][:3]+'testmodel3hrLAcoupling'+sites[0][3:5].upper()+'*.nc' ))
+ print('Processing test_file: ',test_file)
+ fin1 = cdms2.open(test_file[0])
+
+ #==========================================================================
+ # Calculate for observational data
+ #==========================================================================
+ print('ARM data',sites[0])
+ obs_file = glob.glob(os.path.join(obs_path,sites[0][:3]+'armdiagsLAcoupling' + sites[0][3:5].upper()+'*c1.nc')) #read in data
+ print('Processing obs_file',obs_file)
+ fin = cdms2.open(obs_file[0])
+
+ for ivar in range(nvariables):
+ #============================
+ # Seasons: for observations
+ #============================
+ var = fin(variables[ivar])
+ years_obs = list(range(2004,2016))
+ nyears_obs = len(years_obs)
+
+ #var_seasons = get_seasonal(var,seasons,years_obs) #[year,season,90,24]
+ var_seasons = get_diurnal_cycle_seasons(var,seasons,years_obs)
+ narray = nyears_obs*365
+ var_seasons1 = np.empty([nseasons,narray,24])*np.nan
+ var_array_err = np.empty([nseasons,24])*np.nan
+ var_array = np.empty([nseasons,24])*np.nan
+ for iseason in range(nseasons):
+ for iyear in range(nyears_obs):
+ for iday in range(365):
+ var_seasons1[iseason,365*iyear+iday,:] = var_seasons[iyear,iseason,iday,:]
+
+ array_tmp = var_seasons1[iseason,:,:]
+ data0 = np.nanstd(array_tmp,axis=0) #stddev
+ var_array_err[iseason,:] = data0/(math.sqrt(narray))
+ var_array[iseason,:] = np.nanmean(array_tmp,axis=0)
+
+ #==========================
+ # Seasons: for models
+ #==========================
+ try:
+ var_mod = fin1(variables[ivar])
+ years_mod = list(range(2003,2015))
+ nyears_mod = len(years_mod)
+
+ #var_mod_seasons = get_seasonal_3hr(var_mod,seasons,years_mod) #[year,season,90,24]
+ var_mod_seasons = get_diurnal_cycle_seasons(var_mod,seasons,years_mod)
+ narray = nyears_mod*365
+ var_mod_seasons1 = np.empty([nseasons,narray,8])*np.nan
+ var_mod_array_err = np.empty([nseasons,8])*np.nan
+ var_mod_array = np.empty([nseasons,8])*np.nan
+ for iseason in range(nseasons):
+ for iyear in range(nyears_mod):
+ for iday in range(365):
+ var_mod_seasons1[iseason,365*iyear+iday,:] = var_mod_seasons[iyear,iseason,iday,:]
+
+ array_tmp = var_mod_seasons1[iseason,:,:]
+ data0 = np.nanstd(array_tmp,axis=0) #stddev
+ var_mod_array_err[iseason,:] = data0/(math.sqrt(narray))
+ var_mod_array[iseason,:] = np.nanmean(array_tmp,axis=0)
+ except:
+ var_mod_array_err = np.empty([nseasons,8])*np.nan
+ var_mod_array = np.empty([nseasons,8])*np.nan
+
+ #==========================================================================
+ # Plotting: Diurnal cycle (daytime)
+ #==========================================================================
+ for iseason in range(nseasons):
+ fig = plt.figure(figsize=[8,4], dpi=100)
+ xax = np.array([x for x in range(26)])-0.5
+ y_data = np.empty([26])*np.nan
+ y_data[0:19] = var_array[iseason,5:24]
+ y_data[19:26] = var_array[iseason,0:7]
+
+ e_data = np.empty([26])*np.nan
+ e_data[0:19] = var_array_err[iseason,5:24]
+ e_data[19:26] = var_array_err[iseason,0:7]
+ plt.errorbar(xax,y_data,e_data,color='black',label='OBS',linewidth=2)
+
+
+ if variables[ivar]=='SH' or variables[ivar]=='LH':
+ xax1 = np.array([x for x in range(10)])*3-1.5
+ else:
+ xax1 = np.array([x for x in range(10)])*3-3.0
+ y1_data = np.empty([10])*np.nan
+ y1_data[0:7] = var_mod_array[iseason,1:8]
+ y1_data[7:10] = var_mod_array[iseason,0:3]
+
+ e1_data = np.empty([10])*np.nan
+ e1_data[0:7] = var_mod_array_err[iseason,1:8]
+ e1_data[7:10] = var_mod_array_err[iseason,0:3]
+ plt.errorbar(xax1,y1_data,e1_data,color='red',label='MOD',linewidth=2)
+
+ plt.ylabel(units[ivar],fontsize=12)
+ plt.title(variables[ivar]+' ('+seasons[iseason]+')',fontsize=14)
+ plt.xticks([0,2,4,6,8,10,12,14,16,18,20,22,24],\
+ ["0","2","4","6","8","10","12","14","16","18","20","22","24"])
+ plt.xlabel('LST (hour)',fontsize=12)
+ plt.xlim([-0.1,24.1])
+ if variables[ivar]=='pbl':
+ plt.xlim([7.4,17.6])
+ plt.legend()
+ plt.savefig(output_path+'/figures/'+sites[0]+'/'+'Diurnal_cycle_'+seasons[iseason]+'_'+variables[ivar]+'_'+sites[0]+'.png')
+
+
+ fin.close()
diff --git a/arm_diags/src/pdf_daily.py b/arm_diags/src/pdf_daily.py
index f7fef55..3f2689a 100644
--- a/arm_diags/src/pdf_daily.py
+++ b/arm_diags/src/pdf_daily.py
@@ -431,3 +431,4 @@ def pdf_daily_plot(parameter):
#
#
#
+
diff --git a/arm_diags/src/twolegged_metric.py b/arm_diags/src/twolegged_metric.py
new file mode 100644
index 0000000..1081e39
--- /dev/null
+++ b/arm_diags/src/twolegged_metric.py
@@ -0,0 +1,425 @@
+#===========================================================================================================================
+# Program for generate the two-legged metrics: Terrestrial and Atmospheric Coupling Legs
+#---------------------------------------------------------------------------------------------------------------------------
+# V4 Development
+ # ----------------------------------------------------------------------------------------------------
+ # Cheng Tao - Jul2024 @ LLNL
+ # ### Scatter plot of soil moisture vs. SH/LH/EF
+ # ### Scatter plot of SH/LH/EF vs. LCL
+ # ----------------------------------------------------------------------------------------------------
+
+#===========================================================================================================================
+import os
+import pdb
+import glob
+import cdms2
+import cdutil
+import numpy as np
+import matplotlib.pyplot as plt
+from matplotlib.gridspec import GridSpec
+from .varid_dict import varid_longname
+from .taylor_diagram import TaylorDiagram
+from .utils import climo
+import MV2
+import matplotlib.gridspec as gridspec
+import scipy.stats
+from scipy import interpolate
+import calendar
+
+def get_seasonal_data_3hr(var,seasons,years):
+
+ '''Get seasonal data for each variable'''
+ nyears = len(years)
+ nseasons = len(seasons)
+ nsteps = int(nyears*90)
+ t0 = 0
+
+ var_seasons = MV2.zeros([nyears,nseasons,90])*np.nan #Daily mean for each season
+ var_seasons_f = MV2.zeros([nsteps,nseasons])*np.nan
+ for iyear in range(nyears):
+ if calendar.isleap(int(years[iyear]))==True:
+ nday = 366
+ else:
+ nday = 365
+ ntime = int(nday*8)
+ var1 = var[t0:t0+ntime]
+ var1_ext = np.concatenate((var1,var1),axis=0)
+ t0 = t0+ntime
+
+ for iseason in range(nseasons):
+ if seasons[iseason]=='MAM':
+ if nday==366: t1 = 60
+ else: t1 = 59
+ if seasons[iseason]=='JJA':
+ if nday==366: t1 = 152
+ else: t1 = 151
+ if seasons[iseason]=='SON':
+ if nday==366: t1 = 244
+ else: t1 = 243
+ if seasons[iseason]=='DJF':
+ if nday==366: t1 = 335
+ else: t1 = 334
+ var_seasons0 = var1_ext[int(t1*8):int(t1*8)+720]
+ var_seasons1 = np.reshape(var_seasons0,(90,8))
+ var_seasons2 = MV2.average(var_seasons1,axis=1) #daily mean for 90 days
+ #print(var_seasons2.shape)
+ var_seasons[iyear,iseason,:] = var_seasons2
+ var_seasons_f[iyear*90:iyear*90+90,iseason] = var_seasons2
+
+ return var_seasons_f
+
+def get_seasonal_data(var,seasons,years):
+
+ '''Get seasonal data for each variable'''
+ nyears = len(years)
+ nseasons = len(seasons)
+ nsteps = int(nyears*90)
+ t0 = 0
+
+ var_seasons = MV2.zeros([nyears,nseasons,90])*np.nan #Daily mean for each season
+ var_seasons_f = MV2.zeros([nsteps,nseasons])*np.nan
+ for iyear in range(nyears):
+ if calendar.isleap(int(years[iyear]))==True:
+ nday = 366
+ else:
+ nday = 365
+ ntime = int(nday*24)
+ var1 = var[t0:t0+ntime]
+ var1_ext = np.concatenate((var1,var1),axis=0)
+ t0 = t0+ntime
+
+ for iseason in range(nseasons):
+ if seasons[iseason]=='MAM':
+ if nday==366: t1 = 60
+ else: t1 = 59
+ if seasons[iseason]=='JJA':
+ if nday==366: t1 = 152
+ else: t1 = 151
+ if seasons[iseason]=='SON':
+ if nday==366: t1 = 244
+ else: t1 = 243
+ if seasons[iseason]=='DJF':
+ if nday==366: t1 = 335
+ else: t1 = 334
+ var_seasons0 = var1_ext[int(t1*24):int(t1*24)+2160]
+ var_seasons1 = np.reshape(var_seasons0,(90,24))
+ var_seasons2 = MV2.average(var_seasons1,axis=1) #daily mean for 90 days
+ #print(var_seasons2.shape)
+ var_seasons[iyear,iseason,:] = var_seasons2
+ var_seasons_f[iyear*90:iyear*90+90,iseason] = var_seasons2
+
+ return var_seasons_f
+
+def twolegged_metric_plot(parameter):
+ variables = parameter.variables
+ test_path = parameter.test_data_path
+ test_model = parameter.test_data_set
+ obs_path = parameter.obs_path
+ cmip_path = parameter.cmip_path
+ output_path = parameter.output_path
+ sites = parameter.sites
+ seasons = parameter.season
+ nseasons = len(seasons)
+
+ if not os.path.exists(os.path.join(output_path,'figures',sites[0])):
+ os.makedirs(os.path.join(output_path,'figures',sites[0]))
+ #==========================================================================
+ # Calculate for observational data
+ #==========================================================================
+ print('ARM data',sites[0])
+ obs_file = glob.glob(os.path.join(obs_path,sites[0][:3]+'armdiagsLAcoupling' + sites[0][3:5].upper()+'*c1.nc')) #read in data
+ print('Processing obs_file',obs_file)
+ fin = cdms2.open(obs_file[0])
+ sh0 = fin('SH')
+ lh0 = fin('LH')
+ LCL0 = fin('LCL')
+ sm_ebbr0 = fin('soil_moisture_ebbr')
+ sm_swats0 = fin('soil_moisture_swats')
+ fin.close()
+
+ #==========================================================================
+ # Calculate for model data
+ #==========================================================================
+ test_file = glob.glob(os.path.join(test_path,sites[0][:3]+'testmodel3hrLAcoupling' + sites[0][3:5].upper()+'*c1.nc')) #read in data
+ print('Processing test_file',test_file)
+ fin = cdms2.open(test_file[0])
+ sh1 = fin('SH')
+ lh1 = fin('LH')
+ LCL1 = fin('LCL')
+ print('sh1: ',np.min(sh1),np.max(sh1))
+ print('lh1: ',np.min(lh1),np.max(lh1))
+ print('LCL1: ',np.min(LCL1),np.max(LCL1))
+ fin.close()
+
+ #=================
+ # Seasons: obs
+ #=================
+ years_obs = list(range(2004,2016))
+ nday_obs = 90*len(years_obs)
+ sh = get_seasonal_data(sh0,seasons,years_obs)
+ lh = get_seasonal_data(lh0,seasons,years_obs)
+ LCL = get_seasonal_data(LCL0,seasons,years_obs)
+ sm_ebbr = get_seasonal_data(sm_ebbr0,seasons,years_obs)
+ sm_swats = get_seasonal_data(sm_swats0,seasons,years_obs)
+
+ EF = MV2.zeros([nday_obs,nseasons])*np.nan
+ for im in range(nday_obs):
+ for iseason in range(nseasons):
+ EF[im,iseason] = lh[im,iseason]/(lh[im,iseason]+sh[im,iseason])
+
+ #==================
+ # Seasons: models
+ #==================
+ years_mod = list(range(2003,2015))
+ nday_mod = 90*len(years_mod)
+ sh_mod = get_seasonal_data_3hr(sh1,seasons,years_mod)
+ lh_mod = get_seasonal_data_3hr(lh1,seasons,years_mod)
+ LCL_mod = get_seasonal_data_3hr(LCL1,seasons,years_mod)
+
+ EF_mod = MV2.zeros([nday_mod,nseasons])*np.nan
+ for im in range(nday_mod):
+ for iseason in range(nseasons):
+ EF_mod[im,iseason] = lh_mod[im,iseason]/(lh_mod[im,iseason]+sh_mod[im,iseason])
+
+ #==========================================================================
+ # Calculate daily_mean values (10-day means)
+ #==========================================================================
+ nt_10day = int(nday_obs/10)
+
+ sh_10day = MV2.zeros([nt_10day,nseasons])*np.nan
+ lh_10day = MV2.zeros([nt_10day,nseasons])*np.nan
+ LCL_10day = MV2.zeros([nt_10day,nseasons])*np.nan
+ EF_10day = MV2.zeros([nt_10day,nseasons])*np.nan
+ sm_ebbr_10day = MV2.zeros([nt_10day,nseasons])*np.nan
+ sm_swats_10day = MV2.zeros([nt_10day,nseasons])*np.nan
+ for im in range(nt_10day):
+ for iseason in range(nseasons):
+ tmp1 = sh[:,iseason]
+ tmp2 = np.mean(tmp1[10*im:10*im+10])
+ sh_10day[im,iseason] = tmp2
+
+ tmp1 = lh[:,iseason]
+ tmp2 = np.mean(tmp1[10*im:10*im+10])
+ lh_10day[im,iseason] = tmp2
+
+ tmp1 = LCL[:,iseason]
+ tmp2 = np.mean(tmp1[10*im:10*im+10])
+ LCL_10day[im,iseason] = tmp2
+
+ tmp1 = sm_ebbr[:,iseason]
+ tmp2 = np.mean(tmp1[10*im:10*im+10])
+ sm_ebbr_10day[im,iseason] = tmp2
+
+ tmp1 = sm_swats[:,iseason]
+ tmp2 = np.mean(tmp1[10*im:10*im+10])
+ sm_swats_10day[im,iseason] = tmp2
+
+ EF_10day[im,iseason] = lh_10day[im,iseason]/(lh_10day[im,iseason]+sh_10day[im,iseason])
+
+ #==========================================================================
+ # Calculate daily_mean values (10-day means) for models
+ #==========================================================================
+ nt_10day_mod = int(nday_mod/10)
+
+ sh_10day_mod = MV2.zeros([nt_10day_mod,nseasons])*np.nan
+ lh_10day_mod = MV2.zeros([nt_10day_mod,nseasons])*np.nan
+ LCL_10day_mod = MV2.zeros([nt_10day_mod,nseasons])*np.nan
+ EF_10day_mod = MV2.zeros([nt_10day_mod,nseasons])*np.nan
+ for im in range(nt_10day_mod):
+ for iseason in range(nseasons):
+ tmp1 = sh_mod[:,iseason]
+ tmp2 = np.mean(tmp1[10*im:10*im+10])
+ sh_10day_mod[im,iseason] = tmp2
+
+ tmp1 = lh_mod[:,iseason]
+ tmp2 = np.mean(tmp1[10*im:10*im+10])
+ lh_10day_mod[im,iseason] = tmp2
+
+ tmp1 = LCL_mod[:,iseason]
+ tmp2 = np.mean(tmp1[10*im:10*im+10])
+ LCL_10day_mod[im,iseason] = tmp2
+
+ EF_10day_mod[im,iseason] = lh_10day_mod[im,iseason]/(lh_10day_mod[im,iseason]+sh_10day_mod[im,iseason])
+
+ #==================================================
+ # Plots for observations
+ #==================================================
+ for iseason in range(nseasons):
+ #==========================================================================
+ # Plotting: Atmospheric coupling legs
+ # SH/LH/EF vs. LCL
+ #==========================================================================
+
+ fig = plt.figure(figsize=[14,4], dpi=100)
+ gs = gridspec.GridSpec(ncols=3,nrows=1,figure=fig)
+ gs.update(wspace=0.25, hspace=0.15)
+
+ ncols = 3
+ nrows = 1
+ for row in range(nrows):
+ for col in range(ncols):
+ tmp = ncols*row+col
+ ax = fig.add_subplot(gs[row,col])
+
+ data2 = LCL_10day[0:nt_10day,iseason]/1000.
+
+ if col==0:
+ data1 = sh_10day[0:nt_10day,iseason]
+ str_xlabel = 'Sensible heat flux (W/m\u00b2)'
+ if seasons[iseason]=='DJF': xmax = 100
+ if seasons[iseason]=='MAM': xmax = 150
+ if seasons[iseason]=='JJA': xmax = 200
+ if seasons[iseason]=='SON': xmax = 150
+ plt.xlim([0,xmax])
+ xx = xmax*0.6
+ xseq = np.linspace(5,0.8*xmax, num=500)
+ if col==1:
+ data1 = lh_10day[0:nt_10day,iseason]
+ str_xlabel = 'Latent heat flux (W/m\u00b2)'
+ if seasons[iseason]=='DJF': xmax = 100
+ if seasons[iseason]=='MAM': xmax = 150
+ if seasons[iseason]=='JJA': xmax = 200
+ if seasons[iseason]=='SON': xmax = 150
+ plt.xlim([0,xmax])
+ xx = xmax*0.6
+ xseq = np.linspace(5,0.8*xmax, num=500)
+ if col==2:
+ data1 = EF_10day[0:nt_10day,iseason]
+ str_xlabel = 'Evaporative fraction'
+ plt.xlim([0,1])
+ xx = 0.6
+ xseq = np.linspace(0.15, 0.85, num=500)
+
+ plt.scatter(data1, data2,s=30)
+ b, a = np.polyfit(data1, data2, deg=1)
+ r = scipy.stats.pearsonr(data1,data2) # Pearson's r
+ cc = '%.2f' % r[0]
+ ax.plot(xseq, a + b * xseq, color="k", lw=2.5)
+ plt.title('Obs. ('+seasons[iseason]+')',fontsize=14)
+ plt.xlabel(str_xlabel,fontsize=12)
+ plt.ylim([0,3.5])
+ plt.ylabel('LCL (km)',fontsize=12)
+ plt.text(xx,2.8,"R = "+cc,fontsize=15)
+
+ plt.savefig(output_path+'/figures/'+sites[0]+'/'+'Scatter_plot_'+seasons[iseason]+'_atmos_component_obs_'+sites[0]+'.png')
+
+ #===========================================================
+ # For models
+ #===========================================================
+
+ fig = plt.figure(figsize=[14,4], dpi=100)
+ gs = gridspec.GridSpec(ncols=3,nrows=1,figure=fig)
+ gs.update(wspace=0.25, hspace=0.15)
+
+ ncols = 3
+ nrows = 1
+ for row in range(nrows):
+ for col in range(ncols):
+ tmp = ncols*row+col
+ ax = fig.add_subplot(gs[row,col])
+
+ data2 = LCL_10day_mod[0:nt_10day_mod,iseason]/1000.
+
+ if col==0:
+ data1 = sh_10day_mod[0:nt_10day_mod,iseason]
+ str_xlabel = 'Sensible heat flux (W/m\u00b2)'
+ if seasons[iseason]=='DJF': xmax = 100
+ if seasons[iseason]=='MAM': xmax = 150
+ if seasons[iseason]=='JJA': xmax = 200
+ if seasons[iseason]=='SON': xmax = 150
+ plt.xlim([0,xmax])
+ xx = xmax*0.6
+ xseq = np.linspace(5,0.8*xmax, num=500)
+ if col==1:
+ data1 = lh_10day_mod[0:nt_10day_mod,iseason]
+ str_xlabel = 'Latent heat flux (W/m\u00b2)'
+ if seasons[iseason]=='DJF': xmax = 100
+ if seasons[iseason]=='MAM': xmax = 150
+ if seasons[iseason]=='JJA': xmax = 200
+ if seasons[iseason]=='SON': xmax = 150
+ plt.xlim([0,xmax])
+ xx = xmax*0.6
+ xseq = np.linspace(5,0.8*xmax, num=500)
+ if col==2:
+ data1 = EF_10day_mod[0:nt_10day_mod,iseason]
+ str_xlabel = 'Evaporative fraction'
+ plt.xlim([0,1])
+ xx = 0.6
+ xseq = np.linspace(0.15, 0.85, num=500)
+
+ plt.scatter(data1, data2,s=30)
+ b, a = np.polyfit(data1, data2, deg=1)
+ r = scipy.stats.pearsonr(data1,data2) # Pearson's r
+ cc = '%.2f' % r[0]
+ #print(seasons[iseason],'r = ',cc)
+ #print('data1: ',data1)
+ #print('data2: ',data2)
+ ax.plot(xseq, a + b * xseq, color="k", lw=2.5)
+ plt.title('Mod. ('+seasons[iseason]+')',fontsize=14)
+ plt.xlabel(str_xlabel,fontsize=12)
+ plt.ylim([0,3.5])
+ plt.ylabel('LCL (km)',fontsize=12)
+ plt.text(xx,2.8,"R = "+cc,fontsize=15)
+
+ plt.savefig(output_path+'/figures/'+sites[0]+'/'+'Scatter_plot_'+seasons[iseason]+'_atmos_component_testmod_'+sites[0]+'.png')
+
+ #==========================================================================
+ # Plotting: Terrestrial coupling legs
+ # soil moisture vs. SH/LH/EF
+ #==========================================================================
+
+ fig = plt.figure(figsize=[14,4], dpi=100)
+ gs = gridspec.GridSpec(ncols=3,nrows=1,figure=fig)
+ gs.update(wspace=0.25, hspace=0.15)
+
+ for row in range(nrows):
+ for col in range(ncols):
+ tmp = ncols*row+col
+ ax = fig.add_subplot(gs[row,col])
+
+ data1 = sm_swats_10day[0:nt_10day,iseason]
+
+ if col==0:
+ data2 = sh_10day[0:nt_10day,iseason]
+ str_ylabel = 'Sensible heat flux (W/m\u00b2)'
+ if seasons[iseason]=='DJF': ymax = 100
+ if seasons[iseason]=='MAM': ymax = 150
+ if seasons[iseason]=='JJA': ymax = 200
+ if seasons[iseason]=='SON': ymax = 150
+ plt.ylim([0,ymax])
+ yy = ymax*0.8
+ xseq = np.linspace(0.22,0.35,num=500)
+ if col==1:
+ data2 = lh_10day[0:nt_10day,iseason]
+ str_ylabel = 'Latent heat flux (W/m\u00b2)'
+ if seasons[iseason]=='DJF': ymax = 100
+ if seasons[iseason]=='MAM': ymax = 150
+ if seasons[iseason]=='JJA': ymax = 200
+ if seasons[iseason]=='SON': ymax = 150
+ plt.ylim([0,ymax])
+ yy = ymax*0.8
+ xseq = np.linspace(0.22,0.35,num=500)
+ if col==2:
+ data2 = EF_10day[0:nt_10day,iseason]
+ str_ylabel = 'Evaporative fraction'
+ plt.ylim([0,1])
+ yy = 0.098
+ xseq = np.linspace(0.22,0.35,num=500)
+
+ plt.scatter(data1, data2,s=30)
+ b, a = np.polyfit(data1, data2, deg=1)
+ r = scipy.stats.pearsonr(data1,data2) # Pearson's r
+ cc = '%.2f' % r[0]
+ ax.plot(xseq, a + b * xseq, color="k", lw=2.5)
+ plt.title('Obs. ('+seasons[iseason]+')',fontsize=14)
+ plt.ylabel(str_ylabel,fontsize=12)
+ plt.xlim([0.2,0.4])
+ plt.xlabel('Soil moisture (m\u00b3/m\u00b3)',fontsize=12)
+ plt.text(0.32,yy,"R = "+cc,fontsize=15)
+
+ plt.savefig(output_path+'/figures/'+sites[0]+'/'+'Scatter_plot_'+seasons[iseason]+'_land_component_obs_'+sites[0]+'.png')
+ #-------------------------------------------------------------------------
+#=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
+
+
+
diff --git a/arm_diags/src/utils.py b/arm_diags/src/utils.py
index 7b8761d..25da401 100644
--- a/arm_diags/src/utils.py
+++ b/arm_diags/src/utils.py
@@ -9,13 +9,14 @@
# ---------------------------------------------------------------------------------------
#===========================================================================================================================
-
import numpy as np
import numpy.ma as ma
import cdms2
import cdutil
import pdb
from copy import deepcopy
+import MV2
+import calendar
def climo(var, season):
"""
@@ -59,8 +60,6 @@ def climo(var, season):
# Compute time length
dt = tbounds[:,1] - tbounds[:,0]
-
-
# Convert to masked array
v = var.asma()
@@ -76,7 +75,7 @@ def climo(var, season):
climo = ma.zeros([ncycle])#+list(np.shape(v))[1:])
for n in range(ncycle):
idx = np.array( [ season_idx[cycle[n]][var_time_absolute[i].month-1]
- for i in range(len(var_time_absolute)) ], dtype=np.int).nonzero()
+ for i in range(len(var_time_absolute)) ], dtype=int).nonzero()
climo[n] = ma.average(v[idx], axis=0, weights=dt[idx])
# ---------------------------------------------------------------------------------------
@@ -84,7 +83,6 @@ def climo(var, season):
climo = climo.filled(np.nan)
# ---------------------------------------------------------------------------------------
-
if var.id == 'tas':
climo = climo - 273.15
@@ -93,3 +91,57 @@ def climo(var, season):
return climo
+#============================================================================================
+# Functions to get the diurnal cycle for particular seasons/years
+# Output: var_seasons [nyears,nseasons,365,nhour]
+#============================================================================================
+
+def get_diurnal_cycle_seasons(var,seasons,years):
+
+ '''Get seasonal data for each variable'''
+ nyears = len(years)
+ nseasons = len(seasons)
+ t0 = 0
+
+ time = var.getTime()[:]
+ if time[1]-time[0]==1: nhour=24
+ if time[1]-time[0]==3: nhour=8
+ #print('Time resolution: ', nhour)
+
+ var_seasons = MV2.zeros([nyears,nseasons,365,nhour])*np.nan #diurnal cycle for each day
+ for iyear in range(nyears):
+ if calendar.isleap(int(years[iyear]))==True:
+ nday = 366
+ else:
+ nday = 365
+ ntime = int(nday*nhour)
+ var1 = var[t0:t0+ntime]
+ var1_ext = np.concatenate((var1,var1),axis=0)
+ t0 = t0+ntime
+
+ for iseason in range(nseasons):
+ if seasons[iseason]=='ANN':
+ length = int(365*nhour)
+ var_seasons0 = var1[0:length]
+ var_seasons1 = np.reshape(var_seasons0,(365,nhour))
+ var_seasons[iyear,iseason,:,:] = var_seasons1
+ else:
+ if seasons[iseason]=='MAM':
+ if nday==366: t1 = 60
+ else: t1 = 59
+ if seasons[iseason]=='JJA':
+ if nday==366: t1 = 152
+ else: t1 = 151
+ if seasons[iseason]=='SON':
+ if nday==366: t1 = 244
+ else: t1 = 243
+ if seasons[iseason]=='DJF':
+ if nday==366: t1 = 335
+ else: t1 = 334
+ var_seasons0 = var1_ext[int(t1*nhour):int(t1*nhour)+90*nhour]
+ var_seasons1 = np.reshape(var_seasons0,(90,nhour))
+ var_seasons[iyear,iseason,0:90,:] = var_seasons1
+
+ return var_seasons
+
+
diff --git a/setup.py b/setup.py
index 68ee11a..f10ba16 100644
--- a/setup.py
+++ b/setup.py
@@ -1,11 +1,11 @@
from setuptools import find_packages,setup
setup(name='arm_diags',
- version='2.0',
+ version='4.0',
description='An ARM data-oriented package for climate model evaluation',
url='https://github.com/ARM-DOE/arm-gcm-diagnostics',
- author='Jill Chengzhu Zhang, Cheng Tao, Shaocheng Xie and Zeshawn Shaheen',
- author_email='zhang40@llnl.gov, tao4@llnl.gov and xie2@llnl.gov',
+ author='Cheng Tao, Jill Chengzhu Zhang, Shaocheng Xie and Zeshawn Shaheen',
+ author_email='tao4@llnl.gov, zhang40@llnl.gov, and xie2@llnl.gov',
license='LLNL and ARM',
packages=find_packages(exclude=["*.test", "*.test.*", "test.*", "test"]),
zip_safe=False)
+ 