

In [2]:

    
%matplotlib inline
from __future__ import division
from ghgforcing import CO2, CH4
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

Example uses of ghgforcing

Define the time axis along which all emissions and calculations take place, from year 0 to 100. The time-step (tstep) is set to 0.01 years to improve the accuracy of calcuations.



In [3]:

    
tstep = 0.01
time = np.linspace(0, 100, 100/tstep+1)

Deterministic results

Start with a pulse emission of CO₂



In [4]:

    
co2_emission = np.zeros_like(time)
co2_emission[0] = 1



In [5]:

    
co2_forcing = CO2(co2_emission, time, kind='RF')

The CO2 and CH4 functions return values at 1-year timesteps - 101 values from year 0 to 100. The calculations are done at a default time-step of 0.01 years, but that level of detail seems unnecessary and burdensome for an output.



In [6]:

    
co2_forcing.size









    Out[6]:





101

Having the outputs on an annual basis makes it really easy to plot.



In [7]:

    
plt.plot(co2_forcing)
plt.xlabel('Time', size=14)
plt.ylabel('$W \ m^{-2}$', size=14)









    Out[7]:





<matplotlib.text.Text at 0x114873bd0>

Pulse emissions of fossil CH₄ with and non-fossil CH₄ without climate-carbon feedbacks.



In [9]:

    
ch4_emission = co2_emission.copy()



In [15]:

    
ch4_forcing_with = CH4(ch4_emission, time, kind='RF', cc_fb=True)
ch4_forcing_without = CH4(ch4_emission, time, kind='RF', 
                          cc_fb=False, decay=False)



In [19]:

    
plt.plot(ch4_forcing_with, label = 'Fossil CH$_4$ with cc-fb')
plt.plot(ch4_forcing_without, label = 'Non-fossil CH$_4$ without cc-fb')
plt.xlabel('Time', size=14)
plt.ylabel('$W \ m^{-2}$', size=14)
plt.legend(fontsize=12)









    Out[19]:





<matplotlib.legend.Legend at 0x115ae4fd0>

Continuous CO₂ emissions - 1kg/year

Emissions at each time-step should be equal to the annual emission rate.



In [20]:

    
co2_cont = np.ones_like(time)
co2_cont_rf = CO2(co2_cont, time)



In [21]:

    
plt.plot(co2_cont_rf)
plt.xlabel('Time', size=14)
plt.ylabel('$W \ m^{-2}$', size=14)









    Out[21]:





<matplotlib.text.Text at 0x115aa4510>

Stochastic (Monte Carlo) results

Just using the continuous CO₂ emissions scenario.

Mean and +/- 1 sigma

When full_output is False, the functions return a Pandas DataFrame



In [22]:

    
runs = 500



In [23]:

    
co2_cont_mc = CO2(co2_cont, time, kind='RF',
                 runs=runs, full_output=False)



In [24]:

    
co2_cont_mc.head()









    Out[24]:






  
    
      
      mean
      -sigma
      +sigma
    
  
  
    
      0
      1.732239e-17
      1.612672e-17
      1.851805e-17
    
    
      1
      1.663821e-15
      1.545853e-15
      1.781789e-15
    
    
      2
      3.189961e-15
      2.956823e-15
      3.423099e-15
    
    
      3
      4.638858e-15
      4.291089e-15
      4.986627e-15
    
    
      4
      6.032293e-15
      5.569073e-15
      6.495513e-15



In [25]:

    
plt.plot(co2_cont_mc['mean'])
plt.fill_between(x=np.arange(0,101), 
                 y1=co2_cont_mc['+sigma'],
                 y2=co2_cont_mc['-sigma'],
                 alpha=0.4)
plt.xlabel('Time', size=14)
plt.ylabel('$W \ m^{-2}$', size=14)









    Out[25]:





<matplotlib.text.Text at 0x115cbab10>

Full Monte Carlo in addition to mean and +/- 1 sigma

When full_output is True, the functions return a Pandas DataFrame with the mean and +/- 1 sigma, and a numpy array with the full Monte Carlo results.



In [26]:

    
co2_cont_mc, co2_cont_full = CO2(co2_cont, time, kind='RF',
                                 runs=runs, full_output=True)

The co2_cont_mc results are the same as above



In [27]:

    
co2_cont_mc.head()









    Out[27]:






  
    
      
      mean
      -sigma
      +sigma
    
  
  
    
      0
      1.732239e-17
      1.612672e-17
      1.851805e-17
    
    
      1
      1.663821e-15
      1.545853e-15
      1.781789e-15
    
    
      2
      3.189961e-15
      2.956823e-15
      3.423099e-15
    
    
      3
      4.638858e-15
      4.291089e-15
      4.986627e-15
    
    
      4
      6.032293e-15
      5.569073e-15
      6.495513e-15

I'm putting the co2_cont_full into a DataFrame just to make showing some of the results easier.



In [49]:

    
df = pd.DataFrame(co2_cont_full)



In [51]:

    
df.loc[:,:200].plot(legend=False, c='b', alpha=0.2)
plt.xlabel('Time', size=14)
plt.ylabel('$W \ m^{-2}$', size=14)









    Out[51]:





<matplotlib.text.Text at 0x12178d510>



In [ ]:

	mean	-sigma	+sigma
0	1.732239e-17	1.612672e-17	1.851805e-17
1	1.663821e-15	1.545853e-15	1.781789e-15
2	3.189961e-15	2.956823e-15	3.423099e-15
3	4.638858e-15	4.291089e-15	4.986627e-15
4	6.032293e-15	5.569073e-15	6.495513e-15