Assessing global temperature anomaly using NASA’s space studies — Part I | by Himalaya Bir Shrestha | Oct, 2022
Exploring the underlying data, uncertainties, and science behind the historical global temperature anomaly
In recent years, extreme weather events have become a norm rather than rare occurrences. In 2022, torrential monsoon triggered the most severe flooding in Pakistan’s recent history displacing about 33 million people. On the other hand, China experienced the most severe heatwave in six decades, worsening the drought that affected food production, industry, and infrastructure in vast areas of the country. Fans have become commonplace in countries like Germany in Europe during summer, which was not the case a decade ago. While it is normal to attribute temperature rise to global warming, warming in the Arctic region has also been associated with extreme winter weather in Europe and North America, as the Arctic warming disturbs the circular pattern of wind known as the polar vortex.
In 2020, the global temperature reached 1.2 °C above pre-industrial levels (Climate Action Tracker, 2021). By the end of this century (2100), this temperature rise is projected to reach approx. 1.8 °C above pre-industrial levels under the optimistic scenario and 2.7 °C above pre-industrial levels under the current policies (Climate Action Tracker, 2021). The severity and frequency of extreme weather events increase with every decimal point of warming. Thus, scientists agree that the global temperature rise needs to be limited to 1.5 °C above pre-industrial levels by 2100 to prevent irrevocable changes to the climate system (Climate Action Tracker, 2021).
Temperature anomaly refers to the departure of the temperature from a reference value or average over a certain period of time and region (NASA GISS, 2022a). In this post, I am going to assess the historical global temperature anomaly of the past 142 years as provided by NASA’s Goddard Institute for Space Studies (GISS). I am going to explain the significance of temperature anomaly, how NASA derives this data, and the uncertainties involved. Next, I shall present the data using different visualization techniques based on matplotlib and seaborn packages in Python. Towards the end, I will explain how the global temperature anomaly is correlated with the concentration of CO₂ emissions in the atmosphere based on the data of the past six decades. Let’s get started.
Global temperature anomaly data
The temperature data is important to study the regional and global patterns and trends of weather and climate. While absolute temperatures vary enormously within short distances alone, the temperature anomalies are strongly correlated out to large distances of the order of 1000 km (NASA GISS, 2022b). This makes it significantly more accurate to estimate temperature anomalies between stations and in data-sparse areas, and hence temperature anomaly is a good indicator to study climate change.
NASA GISS Surface Temperature (GISTEMP) analysis analyses the temperature anomaly from 1880 to the present for regularly spaced virtual stations spread across the whole globe. A fixed base period is required to calculate temperature anomalies, that are consistent over time. NASA GISTEMP uses 1951–80 as the base period to get long-term temperature anomalies over years, decades and centuries.
Land-Ocean Temperature Index (L-OTI)
Weather stations that are positioned on land provide the Land Surface Air Temperatures (LSATs) that cover one-third of the planet. The Sea Surface Temperatures (SSTs) are available from ships, buoy reports, and satellite data. While SATs and SSTs may be very different, their anomalies are pretty similar (except in the presence of sea ice) (NASA GISS, 2022a). The Land-Ocean Temperature Index (L-OTI) maps the LSAT anomalies over land and sea ice, and SST anomalies over ice-free water.
The global annual mean surface air temperature change (L-OTI index) data is available 1880 onward when the recording began, from the NASA GISS website (NASA GISS, 2022c and Lenssen et al., 2019). I read this data as a pandas dataframe df:
In the given data, the Lowess Smoothing refers to the data obtained from LOWESS (Locally Weighted Scatterplot Smoothing), which is a tool used in regression analysis to create a smooth line through a time plot or scatter plot and help see the relationship between variables and foresee trends.
Preparation for data visualisation
To get consistency across visualisations, I customised the default rc (runtime configurations) settings of matplotlib as shown below:
#figure size and font size
rcParams["figure.figsize"] = (10, 6)
rcParams["font.size"] = 14#grid lines
rcParams["axes.grid"] = True
rcParams["axes.grid.axis"] = "y"
#Setting up axes
rcParams['axes.spines.bottom'] = True
rcParams['axes.spines.left'] = False
rcParams['axes.spines.right'] = False
rcParams['axes.spines.top'] = False
rcParams['axes.linewidth'] = 0.5
#Ticks
rcParams['ytick.major.width'] = 0
rcParams['ytick.major.size'] = 0
Line Plots
I start with plotting the line plots for df using the following code:
I get the following plot as a result:
In this plot, the solid black line refers to the global annual mean surface air temperature change relative to the 1951–80 mean. And the solid red line is the five-year lowess smooth. The temperature anomaly in 2020 relative to the 1951–1980 mean was 1.01 °C. The same value in 1880 was -0.17 °C. Hence, the global mean surface air temperature in 2020 was approx. 1.2 °C above 1880 (pre-industrial level).
There are uncertainties in the temperature anomaly data, which arise from measurement uncertainty, changes in spatial coverage of the station record, and systematic biases due to technology shifts and land cover changes (NASA GISS, 2022d). The gray shaded area in the plot represents the annual uncertainty at a 95% confidence interval for the total anomaly regarding both LSAT and SST.
The disaggregation of the temperature anomaly over land and ocean can be observed in the plot below. While there is an increase in temperature anomaly over both land and ocean, it can be observed that the earth’s land areas have warmed much higher than oceans in the past four decades. Scientists have been studying the reason behind this contrast.
Simple bar plot
In this step, I plot the global temperature anomaly as a simple bar plot using the code below:
df["Annual Mean"].plot(kind = "bar")
years = np.arange(1880, 2022, 20)
plt.xticks(ticks = np.arange(0, 142, 20), labels = years)
plt.title("Global Surface Temperature relative to 1950 to 1980 Mean")
plt.ylabel("Temperature Anomaly (°C)")
plt.show()
The height of the bar represents the annual temperature anomaly. As we can observe from the plot below, the temperature anomaly is zero on average during 1951–80 and is increasing steeply afterward.
Bar plot with the palette using seaborn
I wanted to represent the temperature anomaly in each stripe of the bar beside the height of the bar itself. Seaborn allows plotting this in a very simple manner using a palette. I used a diverging color map or palette called “coolwarm”, wherein, the blue end represents the cold/cool part, and the red end represents the hot/warm part.
The code used and the resulting plot are given below:
sns.barplot(x = df.index, y = df["Annual Mean"],
palette = "coolwarm")
years = np.arange(1880, 2022, 20)
plt.xticks(ticks = np.arange(0, 142, 20), labels = years)
plt.title("Global Surface Temperature relative to 1950 to 1980 Mean")
plt.ylabel("Temperature Anomaly (°C)")
plt.show()
Bar plots with colormap using matplotlib
The plot above can also be recreated using matplotlib using an additional step. First, the temperature anomaly values are scaled between 0 (min) and 1 (max) using a lambda function. The selected colormap (coolwarm) my_cmap contains 255 combinations where my_cmap([0]) refers to the blue color and my_cmap([1]) refers to the red color. Second, these color combinations are selected for each stripe/bar based on the rescaled value of y.
The code used and the resulting plots are given below:
x = df.index.tolist()
y = df["Annual Mean"].values.tolist()#rescale y between 0 (min) and 1 (max)
rescale = lambda y: (y - np.min(y)) / (np.max(y) - np.min(y))
plt.bar(x, y, color = my_cmap(rescale(y)), width = 0.8)years = np.arange(1880, 2022, 20)
plt.xticks(ticks = years, labels = years)
plt.title("Global surface temperature relative to 1951-1980 Mean")
plt.ylabel("Temperature Anomaly (°C)"); plt.xlabel("Year")
plt.show()
Warming stripes
The warming stripes were first introduced by Professor Ed Hawkins of the University of Reading, the UK in 2019 to represent global or regional warming in a very simple and concise manner. The warming stripes show how the average temperatures have risen over nearly two centuries in chronological order.
I am going to recreate the warming stripes in a way such that each stripe would represent the annual global temperature anomaly relative to 1951–80 mean. I read the annual average temperature anomaly as a pandas series anomaly and set up the figure and axes. Next, I created uniform stripes for each year between 1880 and 2021 using rectangular patches. The anomaly data is provided to the PatchCollection col, and is added to the Axes ax.
Similar to the colormap used in the plots above, shades of blue in the warming stripe indicate cooler-than-average years while red shows years that are hotter than average. The dotted white line represents a temperature anomaly of 0 °C, while the straight white lines represent the actual annual temperature anomaly.
Reasons behind the global temperature anomaly
In one of my previous stories, I depicted how global energy consumption has been dominated by fossil fuels such as coal, oil, and gas. The steep rise in the consumption of fossil fuels in the past few decades has led to a rise in greenhouse gas (GHG) emissions in the atmosphere. GHGs are the gases that absorb infrared radiation emitted from the earth’s surface and re-radiating it back to earth, contributing to temperature rise, which is called the greenhouse effect.
As of 2019, global GHG emissions hit a record high of approx. 50 GtCO₂ equivalent including Land Use, Land Use Change, and Forestry (LULUCF) (Climate Watch, 2022).
As of 2016, carbon dioxide (CO₂) accounted for three-quarters of global GHG emissions, followed by methane (CH₄, 17%), nitrous oxide (N₂O, 6%), and smaller trace gases such as hydrofluorocarbons (HFCs) and sulfur hexafluoride (SF₆) (Our World in Data, 2022). These gases have a different relative contribution to global warming over time, which is known as the Global Warming Potential (GWP). For example, one tonne of CH₄ would have about 25 times the warming effect of one tonne of CO₂ over 100 years. Similarly, the GWP of N₂O is 300, and that of Fluorinated gases (F-gases) tends to be more than 1000 over a 100-year time period (Our World in Data, 2022). This is depicted in the plot below:
Keeling Curve
The Keeling curve is a graph of the accumulation of the CO₂ concentration in the earth’s atmosphere based on daily measurements taken at Mouna Loa Laboratory in Hawaii from 1958 to date. It is named after Charles David Keeling, who developed this monitoring system and supervised it until his demise in 2005.
The atmospheric CO₂ concentration stayed fairly stable below 300 parts per million (ppm) for tens of thousands of years before the industrial revolution started (Scripps Institution of Oceanography, 2022a). It started rising exponentially from 1960 onward, which aligns very well with the trend of fossil energy consumption. The plot below illustrates that the CO₂ concentration started increasing from about 320 ppm in 1960 and reached approximately 370 ppm in 2000. It reached 400 ppm for the first time in the middle of the last decade and hit 420 ppm for the first time in human history in 2021 (Scripps Institution of Oceanography, 2022b). In the past three decades alone, atmospheric CO₂ concentration has increased by about 60 ppm.
The red line in the plot above is the monthly average CO₂ concentration, which shows that there is seasonal fluctuation every year. Levels of CO₂ in the atmosphere decline during spring and summer as plants take up more CO₂ for growth and reproduction. During fall and winter, the photosynthesis level declines, and the dominant process is the exhalation of CO₂ through the total ecosystem including plants, bacteria, and animals, increasing the CO₂ concentration. This effect is more dominant in the Northern hemisphere (Scripps Institution of Oceanography, 2013).
Correlation between global temperature anomaly and atmospheric CO₂ concentration
In this final step of the analysis, I plotted the global temperature anomaly alongside the atmospheric CO₂ concentration. The left y-axis is the temperature anomaly relative to the 1951–80 mean, which is a scale for the bars and the right y-axis is the atmospheric CO₂ concentration, which is a scale for the line plot. The data is plotted for 1959 onward based on the availability of CO₂ data.
The plot above shows that both the temperature anomaly and atmospheric CO₂ concentration have increased steeply since 1960. I found a correlation value between the two variables of 96%, which indicate a strong and positive correlation. Indeed, correlation does not necessarily imply causation. However, scientific evidence shows that current global warming cannot be explained by natural drivers such as total solar irradiance, orbital changes, or volcanic eruption (Bloomberg, 2015). Hence, the greenhouse effect discussed above is the testimony that the anthropogenic GHGs are the main culprit behind the global temperature rise (NASA Global Climate Change, 2022). It is also important to note that the atmospheric CO₂ concentration is the result of not only current emissions but the cumulative past emissions from all the sources.
Conclusion
In 2020, we reached 1.2 °C of warming above the 1880 level. The sixth assessment report (AR6) of the IPCC stated that a total of 2390 ± 240 GtCO₂ of anthropogenic CO₂ emissions have likely been emitted between 1850 and 2019. The world has a remaining carbon budget of 500 GtCO₂ from 2020 onward to stay below the 1.5 °C limits with a 50% probability (IPCC AR6, 2021). This implies that with the emissions levels of 2020, we only have about a decade before the temperature rise reaches 1.5 °C (Carbon Brief, 2021). Global warming of 1.5 °C and 2 °C will be exceeded during the 21st century unless deep reductions in CO₂ and other greenhouse gas emissions occur in the coming decades as every tonne of CO₂ emissions adds to global warming.
In this post, I discussed the underlying data, uncertainties, and science behind the global temperature anomaly. I presented different ways of visualising global temperature anomaly, and its relationship with the atmospheric concentration of CO₂. In the next part of this series, I intend to present different techniques for plotting temperature anomaly on world map. The notebooks for the analyses in this post are available in this GitHub repository. Thank you for reading!
References
Bloomberg, 2015. What’s really warming the world?
Carbon Brief, 2021. Analysis: What the new IPCC report says about when world may pass 1.5C and 2C.
Climate Action Tracker, 2021. The CAT Thermometer.
Climate Watch, 2022. Data explorer of historical emissions.
IPCC AR6, 2021. Climate Change 2021 — The Physical Science Basis. Summary for Policymakers.
Lenssen et al., 2019. Improvements in the GISTEMP uncertainty model.
NASA GISS, 2022a. GISS Surface Temperature Analysis (GISTEMP v4) Frequently Asked Questions.
NASA GISS, 2022b. The Elusive Surface Air Temperature (SAT).
NASA GISS, 2022c. GISS Surface Temperature Analysis (GISTEMP v4). Data accessed on 2022–07–01.
NASA GISS, 2022d. Surface Temperature Analysis: Uncertainty Quantification.
NASA Global Climate Change, 2022. The causes of climate change.
Our World in Data, 2022. Greenhouse gas emissions.
Scripps Institution of Oceanography, 2013. Why are Seasonal CO₂ Fluctuations Strongest at Northern Latitudes?
Scripps Institution of Oceanography, 2022a. The Keeling Curve.
Scripps Institution of Oceanography, 2022b. The Keeling Curve hits 420 ppm.
Exploring the underlying data, uncertainties, and science behind the historical global temperature anomaly
In recent years, extreme weather events have become a norm rather than rare occurrences. In 2022, torrential monsoon triggered the most severe flooding in Pakistan’s recent history displacing about 33 million people. On the other hand, China experienced the most severe heatwave in six decades, worsening the drought that affected food production, industry, and infrastructure in vast areas of the country. Fans have become commonplace in countries like Germany in Europe during summer, which was not the case a decade ago. While it is normal to attribute temperature rise to global warming, warming in the Arctic region has also been associated with extreme winter weather in Europe and North America, as the Arctic warming disturbs the circular pattern of wind known as the polar vortex.
In 2020, the global temperature reached 1.2 °C above pre-industrial levels (Climate Action Tracker, 2021). By the end of this century (2100), this temperature rise is projected to reach approx. 1.8 °C above pre-industrial levels under the optimistic scenario and 2.7 °C above pre-industrial levels under the current policies (Climate Action Tracker, 2021). The severity and frequency of extreme weather events increase with every decimal point of warming. Thus, scientists agree that the global temperature rise needs to be limited to 1.5 °C above pre-industrial levels by 2100 to prevent irrevocable changes to the climate system (Climate Action Tracker, 2021).
Temperature anomaly refers to the departure of the temperature from a reference value or average over a certain period of time and region (NASA GISS, 2022a). In this post, I am going to assess the historical global temperature anomaly of the past 142 years as provided by NASA’s Goddard Institute for Space Studies (GISS). I am going to explain the significance of temperature anomaly, how NASA derives this data, and the uncertainties involved. Next, I shall present the data using different visualization techniques based on matplotlib and seaborn packages in Python. Towards the end, I will explain how the global temperature anomaly is correlated with the concentration of CO₂ emissions in the atmosphere based on the data of the past six decades. Let’s get started.
Global temperature anomaly data
The temperature data is important to study the regional and global patterns and trends of weather and climate. While absolute temperatures vary enormously within short distances alone, the temperature anomalies are strongly correlated out to large distances of the order of 1000 km (NASA GISS, 2022b). This makes it significantly more accurate to estimate temperature anomalies between stations and in data-sparse areas, and hence temperature anomaly is a good indicator to study climate change.
NASA GISS Surface Temperature (GISTEMP) analysis analyses the temperature anomaly from 1880 to the present for regularly spaced virtual stations spread across the whole globe. A fixed base period is required to calculate temperature anomalies, that are consistent over time. NASA GISTEMP uses 1951–80 as the base period to get long-term temperature anomalies over years, decades and centuries.
Land-Ocean Temperature Index (L-OTI)
Weather stations that are positioned on land provide the Land Surface Air Temperatures (LSATs) that cover one-third of the planet. The Sea Surface Temperatures (SSTs) are available from ships, buoy reports, and satellite data. While SATs and SSTs may be very different, their anomalies are pretty similar (except in the presence of sea ice) (NASA GISS, 2022a). The Land-Ocean Temperature Index (L-OTI) maps the LSAT anomalies over land and sea ice, and SST anomalies over ice-free water.
The global annual mean surface air temperature change (L-OTI index) data is available 1880 onward when the recording began, from the NASA GISS website (NASA GISS, 2022c and Lenssen et al., 2019). I read this data as a pandas dataframe df:
In the given data, the Lowess Smoothing refers to the data obtained from LOWESS (Locally Weighted Scatterplot Smoothing), which is a tool used in regression analysis to create a smooth line through a time plot or scatter plot and help see the relationship between variables and foresee trends.
Preparation for data visualisation
To get consistency across visualisations, I customised the default rc (runtime configurations) settings of matplotlib as shown below:
#figure size and font size
rcParams["figure.figsize"] = (10, 6)
rcParams["font.size"] = 14#grid lines
rcParams["axes.grid"] = True
rcParams["axes.grid.axis"] = "y"
#Setting up axes
rcParams['axes.spines.bottom'] = True
rcParams['axes.spines.left'] = False
rcParams['axes.spines.right'] = False
rcParams['axes.spines.top'] = False
rcParams['axes.linewidth'] = 0.5
#Ticks
rcParams['ytick.major.width'] = 0
rcParams['ytick.major.size'] = 0
Line Plots
I start with plotting the line plots for df using the following code:
I get the following plot as a result:
In this plot, the solid black line refers to the global annual mean surface air temperature change relative to the 1951–80 mean. And the solid red line is the five-year lowess smooth. The temperature anomaly in 2020 relative to the 1951–1980 mean was 1.01 °C. The same value in 1880 was -0.17 °C. Hence, the global mean surface air temperature in 2020 was approx. 1.2 °C above 1880 (pre-industrial level).
There are uncertainties in the temperature anomaly data, which arise from measurement uncertainty, changes in spatial coverage of the station record, and systematic biases due to technology shifts and land cover changes (NASA GISS, 2022d). The gray shaded area in the plot represents the annual uncertainty at a 95% confidence interval for the total anomaly regarding both LSAT and SST.
The disaggregation of the temperature anomaly over land and ocean can be observed in the plot below. While there is an increase in temperature anomaly over both land and ocean, it can be observed that the earth’s land areas have warmed much higher than oceans in the past four decades. Scientists have been studying the reason behind this contrast.
Simple bar plot
In this step, I plot the global temperature anomaly as a simple bar plot using the code below:
df["Annual Mean"].plot(kind = "bar")
years = np.arange(1880, 2022, 20)
plt.xticks(ticks = np.arange(0, 142, 20), labels = years)
plt.title("Global Surface Temperature relative to 1950 to 1980 Mean")
plt.ylabel("Temperature Anomaly (°C)")
plt.show()
The height of the bar represents the annual temperature anomaly. As we can observe from the plot below, the temperature anomaly is zero on average during 1951–80 and is increasing steeply afterward.
Bar plot with the palette using seaborn
I wanted to represent the temperature anomaly in each stripe of the bar beside the height of the bar itself. Seaborn allows plotting this in a very simple manner using a palette. I used a diverging color map or palette called “coolwarm”, wherein, the blue end represents the cold/cool part, and the red end represents the hot/warm part.
The code used and the resulting plot are given below:
sns.barplot(x = df.index, y = df["Annual Mean"],
palette = "coolwarm")
years = np.arange(1880, 2022, 20)
plt.xticks(ticks = np.arange(0, 142, 20), labels = years)
plt.title("Global Surface Temperature relative to 1950 to 1980 Mean")
plt.ylabel("Temperature Anomaly (°C)")
plt.show()
Bar plots with colormap using matplotlib
The plot above can also be recreated using matplotlib using an additional step. First, the temperature anomaly values are scaled between 0 (min) and 1 (max) using a lambda function. The selected colormap (coolwarm) my_cmap contains 255 combinations where my_cmap([0]) refers to the blue color and my_cmap([1]) refers to the red color. Second, these color combinations are selected for each stripe/bar based on the rescaled value of y.
The code used and the resulting plots are given below:
x = df.index.tolist()
y = df["Annual Mean"].values.tolist()#rescale y between 0 (min) and 1 (max)
rescale = lambda y: (y - np.min(y)) / (np.max(y) - np.min(y))
plt.bar(x, y, color = my_cmap(rescale(y)), width = 0.8)years = np.arange(1880, 2022, 20)
plt.xticks(ticks = years, labels = years)
plt.title("Global surface temperature relative to 1951-1980 Mean")
plt.ylabel("Temperature Anomaly (°C)"); plt.xlabel("Year")
plt.show()
Warming stripes
The warming stripes were first introduced by Professor Ed Hawkins of the University of Reading, the UK in 2019 to represent global or regional warming in a very simple and concise manner. The warming stripes show how the average temperatures have risen over nearly two centuries in chronological order.
I am going to recreate the warming stripes in a way such that each stripe would represent the annual global temperature anomaly relative to 1951–80 mean. I read the annual average temperature anomaly as a pandas series anomaly and set up the figure and axes. Next, I created uniform stripes for each year between 1880 and 2021 using rectangular patches. The anomaly data is provided to the PatchCollection col, and is added to the Axes ax.
Similar to the colormap used in the plots above, shades of blue in the warming stripe indicate cooler-than-average years while red shows years that are hotter than average. The dotted white line represents a temperature anomaly of 0 °C, while the straight white lines represent the actual annual temperature anomaly.
Reasons behind the global temperature anomaly
In one of my previous stories, I depicted how global energy consumption has been dominated by fossil fuels such as coal, oil, and gas. The steep rise in the consumption of fossil fuels in the past few decades has led to a rise in greenhouse gas (GHG) emissions in the atmosphere. GHGs are the gases that absorb infrared radiation emitted from the earth’s surface and re-radiating it back to earth, contributing to temperature rise, which is called the greenhouse effect.
As of 2019, global GHG emissions hit a record high of approx. 50 GtCO₂ equivalent including Land Use, Land Use Change, and Forestry (LULUCF) (Climate Watch, 2022).
As of 2016, carbon dioxide (CO₂) accounted for three-quarters of global GHG emissions, followed by methane (CH₄, 17%), nitrous oxide (N₂O, 6%), and smaller trace gases such as hydrofluorocarbons (HFCs) and sulfur hexafluoride (SF₆) (Our World in Data, 2022). These gases have a different relative contribution to global warming over time, which is known as the Global Warming Potential (GWP). For example, one tonne of CH₄ would have about 25 times the warming effect of one tonne of CO₂ over 100 years. Similarly, the GWP of N₂O is 300, and that of Fluorinated gases (F-gases) tends to be more than 1000 over a 100-year time period (Our World in Data, 2022). This is depicted in the plot below:
Keeling Curve
The Keeling curve is a graph of the accumulation of the CO₂ concentration in the earth’s atmosphere based on daily measurements taken at Mouna Loa Laboratory in Hawaii from 1958 to date. It is named after Charles David Keeling, who developed this monitoring system and supervised it until his demise in 2005.
The atmospheric CO₂ concentration stayed fairly stable below 300 parts per million (ppm) for tens of thousands of years before the industrial revolution started (Scripps Institution of Oceanography, 2022a). It started rising exponentially from 1960 onward, which aligns very well with the trend of fossil energy consumption. The plot below illustrates that the CO₂ concentration started increasing from about 320 ppm in 1960 and reached approximately 370 ppm in 2000. It reached 400 ppm for the first time in the middle of the last decade and hit 420 ppm for the first time in human history in 2021 (Scripps Institution of Oceanography, 2022b). In the past three decades alone, atmospheric CO₂ concentration has increased by about 60 ppm.
The red line in the plot above is the monthly average CO₂ concentration, which shows that there is seasonal fluctuation every year. Levels of CO₂ in the atmosphere decline during spring and summer as plants take up more CO₂ for growth and reproduction. During fall and winter, the photosynthesis level declines, and the dominant process is the exhalation of CO₂ through the total ecosystem including plants, bacteria, and animals, increasing the CO₂ concentration. This effect is more dominant in the Northern hemisphere (Scripps Institution of Oceanography, 2013).
Correlation between global temperature anomaly and atmospheric CO₂ concentration
In this final step of the analysis, I plotted the global temperature anomaly alongside the atmospheric CO₂ concentration. The left y-axis is the temperature anomaly relative to the 1951–80 mean, which is a scale for the bars and the right y-axis is the atmospheric CO₂ concentration, which is a scale for the line plot. The data is plotted for 1959 onward based on the availability of CO₂ data.
The plot above shows that both the temperature anomaly and atmospheric CO₂ concentration have increased steeply since 1960. I found a correlation value between the two variables of 96%, which indicate a strong and positive correlation. Indeed, correlation does not necessarily imply causation. However, scientific evidence shows that current global warming cannot be explained by natural drivers such as total solar irradiance, orbital changes, or volcanic eruption (Bloomberg, 2015). Hence, the greenhouse effect discussed above is the testimony that the anthropogenic GHGs are the main culprit behind the global temperature rise (NASA Global Climate Change, 2022). It is also important to note that the atmospheric CO₂ concentration is the result of not only current emissions but the cumulative past emissions from all the sources.
Conclusion
In 2020, we reached 1.2 °C of warming above the 1880 level. The sixth assessment report (AR6) of the IPCC stated that a total of 2390 ± 240 GtCO₂ of anthropogenic CO₂ emissions have likely been emitted between 1850 and 2019. The world has a remaining carbon budget of 500 GtCO₂ from 2020 onward to stay below the 1.5 °C limits with a 50% probability (IPCC AR6, 2021). This implies that with the emissions levels of 2020, we only have about a decade before the temperature rise reaches 1.5 °C (Carbon Brief, 2021). Global warming of 1.5 °C and 2 °C will be exceeded during the 21st century unless deep reductions in CO₂ and other greenhouse gas emissions occur in the coming decades as every tonne of CO₂ emissions adds to global warming.
In this post, I discussed the underlying data, uncertainties, and science behind the global temperature anomaly. I presented different ways of visualising global temperature anomaly, and its relationship with the atmospheric concentration of CO₂. In the next part of this series, I intend to present different techniques for plotting temperature anomaly on world map. The notebooks for the analyses in this post are available in this GitHub repository. Thank you for reading!
References
Bloomberg, 2015. What’s really warming the world?
Carbon Brief, 2021. Analysis: What the new IPCC report says about when world may pass 1.5C and 2C.
Climate Action Tracker, 2021. The CAT Thermometer.
Climate Watch, 2022. Data explorer of historical emissions.
IPCC AR6, 2021. Climate Change 2021 — The Physical Science Basis. Summary for Policymakers.
Lenssen et al., 2019. Improvements in the GISTEMP uncertainty model.
NASA GISS, 2022a. GISS Surface Temperature Analysis (GISTEMP v4) Frequently Asked Questions.
NASA GISS, 2022b. The Elusive Surface Air Temperature (SAT).
NASA GISS, 2022c. GISS Surface Temperature Analysis (GISTEMP v4). Data accessed on 2022–07–01.
NASA GISS, 2022d. Surface Temperature Analysis: Uncertainty Quantification.
NASA Global Climate Change, 2022. The causes of climate change.
Our World in Data, 2022. Greenhouse gas emissions.
Scripps Institution of Oceanography, 2013. Why are Seasonal CO₂ Fluctuations Strongest at Northern Latitudes?
Scripps Institution of Oceanography, 2022a. The Keeling Curve.
Scripps Institution of Oceanography, 2022b. The Keeling Curve hits 420 ppm.
Denial of responsibility! Techno Blender is an automatic aggregator of the all world’s media. In each content, the hyperlink to the primary source is specified. All trademarks belong to their rightful owners, all materials to their authors. If you are the owner of the content and do not want us to publish your materials, please contact us by email – [email protected]. The content will be deleted within 24 hours.