{{announcement.body}}
{{announcement.title}}

Visualizing Atmospheric Carbon Dioxide

DZone 's Guide to

Visualizing Atmospheric Carbon Dioxide

We take a look at how to visualize large data sets using Python, taking open data from the Earth System Research Laboratory as an example data set.

· Big Data Zone ·
Free Resource

Let's have a look at how to create a visualization that shows how CO2 concentrations have evolved in the atmosphere. First, we fetched from the Earth System Research Laboratory website as follows:

import pandas as pd

data_url = 'ftp://aftp.cmdl.noaa.gov/products/trends/co2/co2_weekly_mlo.txt'
co2_data = pd.read_csv(data_url, sep='\s+', comment='#', na_values=-999.99,
                       names=['year', 'month', 'day', 'decimal', 'ppm', 
                       'days', '1_yr_ago',  '10_yr_ago', 'since_1800'])

co2_data['timestamp'] = co2_data.apply(lambda x: pd.Timestamp(year=int(x.year),
                                                             month=int(x.month),
                                                             day=int(x.day)),
                                       axis=1)
co2_data = co2_data[['timestamp', 'ppm']].set_index('timestamp').ffill()

Then, we group the it by year and month while at the same time storing the result in a matrix where each element represents the concentration in a specific year and month:

import numpy as np
import matplotlib.pyplot as plt
from calendar import month_abbr

co2_data = co2_data['1975':'2018']
n_years = co2_data.index.year.max() - co2_data.index.year.min()
z = np.ones((n_years +1 , 12)) * np.min(co2_data.ppm)
for d, y in co2_data.groupby([co2_data.index.year, co2_data.index.month]):
  z[co2_data.index.year.max() - d[0], d[1] - 1] = y.mean()[0]

plt.figure(figsize=(10, 14))
plt.pcolor(np.flipud(z), cmap='hot_r')
plt.yticks(np.arange(0, n_years+1)+.5,
           range(co2_data.index.year.min(), co2_data.index.year.max()+1));
plt.xticks(np.arange(13)-.5, month_abbr)
plt.xlim((0, 12))
plt.colorbar().set_label('Atmospheric Carbon Dioxide in ppm')
plt.show()

This visualization makes us able to compare the CO2 levels month by month with a single glance. For example, we see that the period from April to June gets dark quicker than other periods, meaning that it contains the highest levels every year. Conversely, the period that goes from September to October gets darker more slowly, meaning that it's the period with the lowest CO2 levels. Also, looking at the color bar we note that in 43 years there was a 80 ppm increase.

Is this bad for the planet earth? Reading Hansen et al. (2008) we can classify CO2 levels less than 300 ppm as safe, levels between 300 and 350 ppm as dangerous, while levels beyond 350 ppm are considered catastrophic. According to this, the chart is a sad picture of how the levels transitioned from dangerous to catastrophic!

Concerned by this situation I created the CO2 Forecast twitter account where I'll publish short and long term forecasts of CO2 levels in the atmosphere.

Topics:
big data ,python tutorial ,data visualization python

Published at DZone with permission of

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}