In [3]:
%matplotlib inline
%config InlineBackend.figure_format = 'png'
%config InlineBackend.print_figure_kwargs = {'dpi':300, 'bbox_inches': 'tight'}
import numpy as np
import pandas as pd
import xarray as xr
import xarray.ufuncs as xu
import cmocean as cm
from scipy import stats
import matplotlib.pyplot as plt
import matplotlib as mpl
from matplotlib.ticker import AutoMinorLocator
import cartopy.crs as ccrs
import cartopy.feature as cfeature
import os
In [4]:
dir_raw = '/local/data/artemis/workspace/gloege/SOCAT-LE/data/raw/'
dir_clean = '/local/data/artemis/workspace/gloege/SOCAT-LE/data/clean/'
dir_figs = '/local/data/artemis/workspace/gloege/SOCAT-LE/results/figures_MS'
### SOCAT directory
#data_dir = f'{dir_raw}SOCAT'
In [5]:
reconstruction='SOMFFN'
ds = xr.open_dataset(f'{dir_raw}SOCAT/SOCATv5_tracks_gridded_monthly.nc')
ds_mask = xr.open_dataset(f'{dir_raw}SOCAT/SOCATv5_mask_198201-201701.nc')
#ds_stats = xr.open_dataset(f'{dir_clean}statistics_SOMFFN_CESM.nc')
ds_stats = xr.open_mfdataset(f'{dir_clean}/CO2_flux_stats/stats_FCO2_*_SOMFFN.nc', concat_dim='ensemble')
Normalized Standard Deviation¶
In [49]:
plot_style = 'seaborn-paper'
with plt.style.context(plot_style):
fig = plt.figure(figsize=(4,4))
ax = fig.add_subplot(111)
cmap = plt.cm.nipy_spectral
cmap = cm.cm.oxy
## ------------------------------------
## Define data variables
## -----------------------------------
corr_dec = ds_stats['corr_dec'].mean(['ensemble','lon'])
aae = ds_stats['aae'].mean(['ensemble','lon'])
bias = ds_stats['bias'].mean(['ensemble','lon'])
std_star_seas = ds_stats['std-star_seasonal'].mean(['ensemble','lon'])
std_star_res = ds_stats['std-star_residual'].mean(['ensemble','lon'])
std_star_dec = ds_stats['std-star_dec'].mean(['ensemble','lon'])
lat = ds_stats['lat']
## ------------------------------------
## 0 reference line
## -----------------------------------
#ax.plot([0,168],[0,0],'--',color=[0.2, 0.2, 0.2], linewidth=1)
ax.plot([-100,100],[-35,-35],'--',color=[0.5, 0.5, 0.5], linewidth=1)
ax.plot([-100,100],[35,35],'--',color=[0.5, 0.5, 0.5], linewidth=1)
ax.plot([0,0],[-90,90],'-',color=[0.5, 0.5, 0.5], linewidth=1)
#sub = ax.scatter(nmonths, bias, s=20,
# c=corr_dec, cmap=cmap, vmin=0, vmax=1,
# linewidth=0.75, marker='o')
#sub.set_facecolor("none")
ax.plot(std_star_dec*100, lat,color=[0,0,0], linestyle='-', label='Decadal')
ax.plot(std_star_seas*100, lat, color=[0.7,0.7,0.7], linestyle='-', label='Seasonal')
#ax.plot(std_star_res*100, lat, color=[0.9,0.9,0.9], linestyle='-', label='Sub-decadal')
## ------------------------------------
## axis limits and labels
## -----------------------------------
ax.set_xlim(-100,100)
ax.set_xticks(np.arange(-100, 100, 25))
ax.set_ylim(-90,90)
# fongsize was 20
ax.set_xlabel('Normalized STD (%)')
ax.set_ylabel('Latitude')
#ax.xaxis.set_tick_params(labelsize=18)
#ax.yaxis.set_tick_params(labelsize=18)
## ------------------------------------
## Modify axes
## -----------------------------------
# Turn on the minor ticks
# Turn off the display of all ticks.
ax.tick_params(which='both', # Options for both major and minor ticks
top='off', # turn off top ticks
left='off', # turn off left ticks
right='off', # turn off right ticks
bottom='off') # turn off bottom ticks
### Hide the right and top spines
ax.spines['right'].set_visible(False)
#ax.spines['left'].set_visible(False)
ax.spines['top'].set_visible(False)
#ax.spines['bottom'].set_visible(False)
### major / minor tick lines
#ax.minorticks_on()
ax.yaxis.set_minor_locator(AutoMinorLocator(4))
ax.xaxis.set_minor_locator(AutoMinorLocator(5))
#ax.grid(axis='y', which='major', color=[0.8,0.8,0.8], linestyle='-')
### Only show ticks on the left and bottom spines
ax.yaxis.set_ticks_position('left')
ax.xaxis.set_ticks_position('bottom')
## ------------------------------------
## Make plot square
## and add grid lines
## -----------------------------------
x0,x1 = ax.get_xlim()
y0,y1 = ax.get_ylim()
ax.set_aspect(abs(x1-x0)/abs(y1-y0))
ax.get_position().bounds
ax.grid(axis='x', which='major', color=[0.8,0.8,0.8], linestyle='-')
# Don't allow the axis to be on top of your data
ax.set_axisbelow(True)
ax.legend(frameon=False)
#fig.savefig(f'{dir_figs}/bias_vs_nobs.pdf',bbox_inches='tight', dpi=300, pad_inches=0)
/home/gloege/miniconda3/envs/tensorflow/lib/python3.6/site-packages/dask/array/numpy_compat.py:41: RuntimeWarning: invalid value encountered in true_divide x = np.divide(x1, x2, out) /home/gloege/miniconda3/envs/tensorflow/lib/python3.6/site-packages/dask/array/numpy_compat.py:41: RuntimeWarning: invalid value encountered in true_divide x = np.divide(x1, x2, out)
Correlation¶
In [58]:
plot_style = 'seaborn-paper'
with plt.style.context(plot_style):
fig = plt.figure(figsize=(4,4))
ax = fig.add_subplot(111)
cmap = plt.cm.nipy_spectral
cmap = cm.cm.oxy
## ------------------------------------
## Define data variables
## -----------------------------------
corr_dec = ds_stats['corr_dec'].mean(['ensemble','lon'])
corr_res = ds_stats['corr_residual'].mean(['ensemble','lon'])
corr_seas = ds_stats['corr_seasonal'].mean(['ensemble','lon'])
aae = ds_stats['aae'].mean(['ensemble','lon'])
bias = ds_stats['bias'].mean(['ensemble','lon'])
std_star_seas = ds_stats['std-star_seasonal'].mean(['ensemble','lon'])
std_star_res = ds_stats['std-star_residual'].mean(['ensemble','lon'])
std_star_dec = ds_stats['std-star_dec'].mean(['ensemble','lon'])
lat = ds_stats['lat']
## ------------------------------------
## 0 reference line
## -----------------------------------
#ax.plot([0,168],[0,0],'--',color=[0.2, 0.2, 0.2], linewidth=1)
ax.plot([-100,100],[-35,-35],'--',color=[0.5, 0.5, 0.5], linewidth=1)
ax.plot([-100,100],[35,35],'--',color=[0.5, 0.5, 0.5], linewidth=1)
ax.plot([0,0],[-90,90],'-',color=[0.5, 0.5, 0.5], linewidth=1)
#sub = ax.scatter(nmonths, bias, s=20,
# c=corr_dec, cmap=cmap, vmin=0, vmax=1,
# linewidth=0.75, marker='o')
#sub.set_facecolor("none")
ax.plot(corr_dec, lat,color=[0,0,0], linestyle='-', label='Decadal')
ax.plot(corr_seas, lat, color=[0.7,0.7,0.7], linestyle='-', label='Seasonal')
#ax.plot(corr_res, lat, color=[0.9,0.9,0.9], linestyle='-', label='Sub-decadal')
## ------------------------------------
## axis limits and labels
## ----------------------------------
ax.set_xticks(np.arange(-1, 1.2, 0.25))
ax.set_xlim(0,1)
ax.set_ylim(-90,90)
# fongsize was 20
ax.set_xlabel('Correlation')
ax.set_ylabel('Latitude')
#ax.xaxis.set_tick_params(labelsize=18)
#ax.yaxis.set_tick_params(labelsize=18)
## ------------------------------------
## Modify axes
## -----------------------------------
# Turn on the minor ticks
# Turn off the display of all ticks.
ax.tick_params(which='both', # Options for both major and minor ticks
top='off', # turn off top ticks
left='off', # turn off left ticks
right='off', # turn off right ticks
bottom='off') # turn off bottom ticks
### Hide the right and top spines
ax.spines['right'].set_visible(False)
#ax.spines['left'].set_visible(False)
ax.spines['top'].set_visible(False)
#ax.spines['bottom'].set_visible(False)
### major / minor tick lines
#ax.minorticks_on()
ax.yaxis.set_minor_locator(AutoMinorLocator(4))
ax.xaxis.set_minor_locator(AutoMinorLocator(5))
#ax.grid(axis='y', which='major', color=[0.8,0.8,0.8], linestyle='-')
### Only show ticks on the left and bottom spines
ax.yaxis.set_ticks_position('left')
ax.xaxis.set_ticks_position('bottom')
## ------------------------------------
## Make plot square
## and add grid lines
## -----------------------------------
x0,x1 = ax.get_xlim()
y0,y1 = ax.get_ylim()
ax.set_aspect(abs(x1-x0)/abs(y1-y0))
ax.get_position().bounds
ax.grid(axis='x', which='major', color=[0.8,0.8,0.8], linestyle='-')
# Don't allow the axis to be on top of your data
ax.set_axisbelow(True)
ax.legend(frameon=False)
#fig.savefig(f'{dir_figs}/bias_vs_nobs.pdf',bbox_inches='tight', dpi=300, pad_inches=0)
/home/gloege/miniconda3/envs/tensorflow/lib/python3.6/site-packages/dask/array/numpy_compat.py:41: RuntimeWarning: invalid value encountered in true_divide x = np.divide(x1, x2, out) /home/gloege/miniconda3/envs/tensorflow/lib/python3.6/site-packages/dask/array/numpy_compat.py:41: RuntimeWarning: invalid value encountered in true_divide x = np.divide(x1, x2, out)
Bias¶
In [63]:
plot_style = 'seaborn-paper'
with plt.style.context(plot_style):
fig = plt.figure(figsize=(4,4))
ax = fig.add_subplot(111)
cmap = plt.cm.nipy_spectral
cmap = cm.cm.oxy
## ------------------------------------
## Define data variables
## -----------------------------------
corr_dec = ds_stats['corr_dec'].mean(['ensemble','lon'])
corr_res = ds_stats['corr_residual'].mean(['ensemble','lon'])
corr_seas = ds_stats['corr_seasonal'].mean(['ensemble','lon'])
aae = ds_stats['aae'].mean(['ensemble','lon'])
bias = ds_stats['bias'].mean(['ensemble','lon'])
std_star_seas = ds_stats['std-star_seasonal'].mean(['ensemble','lon'])
std_star_res = ds_stats['std-star_residual'].mean(['ensemble','lon'])
std_star_dec = ds_stats['std-star_dec'].mean(['ensemble','lon'])
lat = ds_stats['lat']
## ------------------------------------
## 0 reference line
## -----------------------------------
#ax.plot([0,168],[0,0],'--',color=[0.2, 0.2, 0.2], linewidth=1)
ax.plot([-100,100],[-35,-35],'--',color=[0.5, 0.5, 0.5], linewidth=1)
ax.plot([-100,100],[35,35],'--',color=[0.5, 0.5, 0.5], linewidth=1)
ax.plot([0,0],[-90,90],'-',color=[0.5, 0.5, 0.5], linewidth=1)
#sub = ax.scatter(nmonths, bias, s=20,
# c=corr_dec, cmap=cmap, vmin=0, vmax=1,
# linewidth=0.75, marker='o')
#sub.set_facecolor("none")
ax.plot(bias, lat,color=[0,0,0], linestyle='-', label='Bias')
#ax.plot(aae, lat, color=[0.7,0.7,0.7], linestyle='-', label='Mean-Absoluste-Error')
#ax.plot(corr_res, lat, color=[0.9,0.9,0.9], linestyle='-', label='Sub-decadal')
## ------------------------------------
## axis limits and labels
## ----------------------------------
ax.set_xticks(np.arange(-0.2, 0.22, 0.05))
ax.set_xlim(-0.2,0.2)
ax.set_ylim(-90,90)
# fongsize was 20
ax.set_xlabel('Bias')
ax.set_ylabel('Latitude')
#ax.xaxis.set_tick_params(labelsize=18)
#ax.yaxis.set_tick_params(labelsize=18)
## ------------------------------------
## Modify axes
## -----------------------------------
# Turn on the minor ticks
# Turn off the display of all ticks.
ax.tick_params(which='both', # Options for both major and minor ticks
top='off', # turn off top ticks
left='off', # turn off left ticks
right='off', # turn off right ticks
bottom='off') # turn off bottom ticks
### Hide the right and top spines
ax.spines['right'].set_visible(False)
#ax.spines['left'].set_visible(False)
ax.spines['top'].set_visible(False)
#ax.spines['bottom'].set_visible(False)
### major / minor tick lines
#ax.minorticks_on()
ax.yaxis.set_minor_locator(AutoMinorLocator(4))
ax.xaxis.set_minor_locator(AutoMinorLocator(5))
#ax.grid(axis='y', which='major', color=[0.8,0.8,0.8], linestyle='-')
### Only show ticks on the left and bottom spines
ax.yaxis.set_ticks_position('left')
ax.xaxis.set_ticks_position('bottom')
## ------------------------------------
## Make plot square
## and add grid lines
## -----------------------------------
x0,x1 = ax.get_xlim()
y0,y1 = ax.get_ylim()
ax.set_aspect(abs(x1-x0)/abs(y1-y0))
ax.get_position().bounds
ax.grid(axis='x', which='major', color=[0.8,0.8,0.8], linestyle='-')
# Don't allow the axis to be on top of your data
ax.set_axisbelow(True)
#ax.legend(frameon=False)
#fig.savefig(f'{dir_figs}/bias_vs_nobs.pdf',bbox_inches='tight', dpi=300, pad_inches=0)
/home/gloege/miniconda3/envs/tensorflow/lib/python3.6/site-packages/dask/array/numpy_compat.py:41: RuntimeWarning: invalid value encountered in true_divide x = np.divide(x1, x2, out)
In [ ]: