Forecasting Global Surface Temperature

Introduction

One the most challenging issue of the 21st century is global warming (Jian-Bin, H., et al 2012). Global warming as a threat to human has been debated and beyond expectations, the debate engulfed practically every element of global warming, including its nature, causes, and implications, as well as whether global warming had actually occurred. Wang S. W. et al. (2010) evaluated three main themes in the present debate, including the present status of global warming, its implications, and the anthropogenic influence, pointing out that global warming is the consequence of a combination of human activity and natural variability. Climate change encompasses more than simply global warming. According to NASA (2021), the term "global warming" refers to the long term increase in global temperatures caused mostly by rising greenhouse gas concentrations in the atmosphere., while climatic change is defined as the change in the variability of climate variables through time, such as precipitation, temperature, and wind patterns.

How will the climate change in the next 50 or 100 years? No one can say for sure. However, there are legitimate concerns that fast climate change will have a significant impact on human life. The world has had a pretty steady climate for the last 10,000 years, and life has adapted to it. However, the earth's temperature has recently risen, and many scientists now believe there is a clear relationship between this warming and human-caused emissions of greenhouse gases such as carbon dioxide and nitrogen oxides. It is believed that the natural equilibrium will be disturbed if greenhouse gas levels continue to rise at their current rate. Fossil fuels take millions of years to develop, but just a few minutes to burn, releasing massive amounts of CO2 into the atmosphere. Natural reasons such as volcanic eruptions and the abundance of phytoplankton in the sea have previously been used to explain CO2 level changes. Other ideas explaining global warming include positive and negative feedback loops in ocean currents, as well as the Earth's location in space.

Climate change is extremely difficult to anticipate, which is the major issue for scientists. Goel, A. and Bhatt, R., (2012) argued that global temperatures are now higher than they have been in at least a millennium, and they are rising even faster than experts projected. Increase in sea level, extreme weather conditions, disease transmission, and mass extinctions are all consequences of global warming. In addition to changes in greenhouse gases, the chemistry of the atmosphere is altering owing to emissions of carbon monoxide, nitrogen oxides, and volatile organic compounds, among other species (Noam Mohr, 2005). Global warming can be induced by natural processes or by human intervention. A common example of natural cause is mount Etna, an active but currently dormant volcano in Sicily, is a prime example of volcanic eruptions. It is one of the most powerful natural carbon dioxide sources. It emits roughly 25 million tons of carbon dioxide into the atmosphere every year. As a result, the whole region surrounding the volcano is carbon dioxide-rich (Rajni Johar Chhatwal, 2009). Humans activity or anthropogenic cause is a major cause of global warming. Humans have been producing excess greenhouse gases as a result of the use of fossil fuels (like coal, oil and gas). CO2 created by humans will cause much greater warming in the next 100 years, ranging from three degrees Celsius to eight or ten degrees Celsius.

As the world is trying to reduce global temperature through various means, it is important that we are able to forecast future global temperature so as to guide our approach to combating climate change. This project is geared towards using machine learning to forecast global temperature for about 8 time steps into the future.

Problem Statement

Hansen, J. et al. (2016) argued that the year with highest global mean surface temperature is 2016 for years prior to 2017. In 2016, the global mean surface temperature was 1.26°C more than that between 1880 and 1920, the pre-industrial period considered. (Hansen, J. et al. 2016). According to Haldar, I. (2011), sea levels will rise as the global temperature rises, changing the amount and pattern of precipitation. The observed increase in global average air and ocean temperatures, widespread melting of snow and ice, and worldwide rise in sea level are all evidence of global warming (Haldar, I., 2011).

The Paris Agreement commits all countries to make aggressive measures to tackle climate change, with the goal of “keeping global mean temperature (GMT) well below 2 °C, relative to pre-industrial levels, and pursuing measures to limit warming to 1.5 °C” (Vicedo-Cabrera, A.M. 2018). There is therefore a need to tackle the most challenging problem of the 21st century (Jian-Bin, H., et al 2012).

Aims and Objectives

The aim of this project is to explore, predict and evaluate the global mean surface temperature from 2022 to 2030. The objectives of this research work are as follows:

1. To better understand and gain insight into the statistical patterns in global mean surface temperature.
2. To use machine learning to model the global mean surface temperature.
3. To forecast the global mean surface temperature from 2022 to 2030.
4. To estimate how close we will be to meeting the goal of the Paris agreement at the end of 2030.

Research Questions

The research questions for this project are as follows:

1. Is the global mean surface temperature increasing?
2. Can a machine learning model explain the variation of the global mean surface temperature?
3. What will be the global mean surface temperature in 2030?
4. How close we would be to meeting the goal of the Paris agreement by the end of 2030?

Deliverables

The deliverables of this project are exploratory data analysis, data mining, data munging, modeling, evaluation, forecasting and a project report. The report should thoroughly explain the source of the data used, exploratory data analysis, code implementation, feature selection, model selection as well as model performance.

Relevance

This project is mainly focused training a machine learning model to forecast the global mean surface temperature from 2022 to 2030 to the end of knowing how close we would be to meeting the goal of the Paris agreement.

Methodology

This project would consider a secondary research, modeling, and forecasting. They are further buttressed below:

Secondary research

The secondary research in this project will utilize a systematic approach (Johnson et al., 2016) to review the works of literature. The steps involved in the systematic review of the literature are provided below:

• Step 1: Identify the research questions that can be used for the project.
• Step 2: Identify the keywords that should be used to research the works of literature.
• Step 3: Extract the journals and books that are appropriate for this project.
• Step 4: Write the literature review chapter.

Modeling

This section involves sourcing for and making sense out of data. The steps involved are as follows:

• Data collection
• Exploratory Data Analysis
• Data Munging
• Model selection
• Model optimization
• Model evaluation

Forecasting

In this section, we will forecast the global mean surface temperature of the earth from 2022 to 2030. The steps in achieving these are:

• Forecasting from 2022 to 2030.
• Comparing the forecast to the goal of the Paris agreement.

Evaluation

The risk assessment conducted for this project is provided in the table below:

Table 1:  Risk assessment