Oil Movements and Energy Consumption Insights - Using Python, Pandas, Matplotlib and Bokeh¶
By Younes Soliman
The fuel tankers war happening mostly for those moving along strait of Hormuz raises several questions about what could be the effect of those activities and the increasing risk on the fuel prices around the world. Some of the questions are:
- What is the quantity of Oil movements between areas, and what specifically is the amount passing through the risky area.
- Does Oil form a big percentage of the needed energy for consumption by the world.
- How much is the total world energy consumption and how the amount/type of energy changed over years
- Is the global energy consumption increasing? and what are the top energy consuming countries?
- Calculate the energy consumption per country per person to see what countries should put plans for their energy use.
I will try to answer the above questions using python data science and presentation libraries.
Let's not talk more, and let's dig deeper and catch python by our hands. I will not show the full code in order not to make you bored, but few explanations about the output
So first of all I will show a data sample of the crude oil movements from one area to another. The amount is in million tonnes. I will not take into account here the oil products movements.¶
Notes: Does not include biofuels trade. Bunker fuel use is not included as exports. Intra-area movements (for example, between countries within Europe) are excluded.
In [2]:
#Import needed libraries
import numpy as np
import pandas as pd
#Import plotting libraries and initialize params
import matplotlib.pyplot as plt
The first column contains the areas from which the crude oil is going, while the other columns show the area to where the oil is going. The total amount is about 2263 million tonnes in year 2018.¶
In [4]:
Out[4]:
We can have a better view using a heatmap with blue color from/to area¶
In [9]:
Out[9]:
And here we calculate the approximate percentage passing through Hormuz by summing the amount moving from Middle east gulf countries. The amount comes to about 35% of world crude oil is passing through this strait.¶
In [10]:
PercentHormuz=((Crude1['Total']['Saudi Arabia']+Crude1['Total']['Iraq']+Crude1['Total']['UAE']+Crude1['Total']['Kuwait'])*100)/(Crude1['Total']['Total imports'])
PercentHormuz
Out[10]:
But since oil is not the only form of energy that people depend on, what is its percentage from overall forms of energy nowadays? For this i will show a sample data of energy consumption by form of energy in TeraWatt-Hours¶
In [15]:
csvf['totalconsumption (terawatt-hours)']=csvf['Coal (terawatt-hours)']+csvf['Solar (terawatt-hours)']+csvf['Crude oil (terawatt-hours)']+csvf['Natural gas (terawatt-hours)']+csvf['Traditional biofuels (terawatt-hours)']+csvf['Other renewables (terawatt-hours)']+csvf['Hydropower (terawatt-hours)']+csvf['Nuclear (terawatt-hours)']+csvf['Wind (terawatt-hours)']
In [16]:
csvf.tail()
Out[16]:
So in this table, Crude oil consumption comes around 53K TWH out of total 153K TWH consumed in 2017. which is around 35% of total energy. So what passes by Hormuz is 35% of world needed crude oil which inturn forms 35% of total world energy needed. This means that about 12% of total needed energy is in danger.¶
Now let's see in a matplotlib graph the change of energy consumption by form of energy since the year of 1800 until 2017. The graph shows that previously people depended on traditional biofuels for energy while with years, another forms were introduced while the main are coal, crude oil, natural gas and few traditional biofuels. While other forms of energy like solar, nuclear, hydro, wind are still forming low percenatge of total world percentage. The graph also shows how much the amount consumed had changed over the last 70 years, let's say from 1950 until now. World used to consume 25K twh in 1950 while they consume 150K in 2017 which is 6 times higher. The increase could be little justified when we know the increase in world population from 1950 until 2017 from around 2.5 billion up to 7.5 billion which is 3 times higher.¶
In [17]:
Let's plot the same graph using Bokeh. We can see that Bokeh has more interactive graphs where user can zoom in or out or move to a specific location, and also can add more functionalities.¶
In [22]:
from bokeh.core.properties import value
from bokeh.plotting import figure, show, output_notebook
from bokeh.palettes import brewer
from bokeh.models import HoverTool
from bokeh.models import NumeralTickFormatter
output_notebook()
show(p)
We can draw it also as a bokeh line graph with tooltips on mouse hovering over the lines¶
In [24]:
p = figure(x_axis_type="linear")
p.add_tools(hover)
p.title.text = 'Energy Consumption by type'
p.xaxis.axis_label = 'Year'
p.yaxis.axis_label = 'Energy Consumption'
p.legend.location = "top_left"
show(p)
Now let's see the energy consumption by country and some group of countries split by form of energy¶
Here is a data sample¶
In [30]:
df_esr['Total (TWH)']=df_esr['Oil (TWH)']+df_esr['Gas (TWH)']+df_esr['Coal (TWH)']+df_esr['Solar PV (TWH)']+df_esr['Other renewables (TWH)']+df_esr['Wind (TWH)']+df_esr['Nuclear (TWH)']+df_esr['Hydropower (TWH)']
df_esr_ds=ColumnDataSource(df_esr)
df_esr.head()
Out[30]:
In [31]:
df_esr1=df_esr[df_esr['Year']==2016].groupby(['Entity']).agg({'Oil (TWH)':'sum','Gas (TWH)':'sum','Coal (TWH)':'sum','Solar PV (TWH)':'sum','Other renewables (TWH)':'sum','Wind (TWH)':'sum',
'Nuclear (TWH)':'sum','Hydropower (TWH)':'sum','Total (TWH)':'sum'})
See only year =2016¶
In [32]:
Out[32]:
We see that China and US are the highest countries in energy consumption and are much higher than any other country. Remember that even the energy consumption should be relative because a country could be higher in area than other, higher population than other..and other factors. So in the next paragraphs, I will calculate the energy consumption relative to the population to see what countries people are consuming more than other countries people.¶
So here is a sample of the data after transformation to show the consumption of each year by countries and sorted by the top down in the year of 2016¶
In [40]:
Out[40]:
Let's draw the top 10 in a matplotlib bar graph.¶
In [41]:
df_esr_pop_filter_pvt.head(10).plot.bar(stacked=False, figsize=(15,15))
Out[41]:
and here is the same graph for top 30 but only for year=2016¶
In [42]:
df_esr_pop_filter1.sort_values('Energy by citizen (KWH)', ascending=False).head(30).plot.bar(rot=45)
Out[42]:
So what we see here is that people in Qatar are the top consumers of energy and they are much higher than most of the world countries. It shows that also that almost all Arab gulf countries are high consumers of energy, and that would be probably because of the high temperatures and use of air conditioners almost the whole year because of the weather. Other reason is the cheap price of energy in those countries as they are oil producers. The same theory probably applies to Canada where the weather needs conditioning.¶
I hope, at the end of this demonstartion, i was able to find the answers for the questions in your mind too. If you are interested in more insights, studies or work, please don't hesitate to contact.¶
If you like this analysis, please like and follow "Bee Eyes" page.¶
https://www.linkedin.com/company/bee-eyes¶
In [ ]: