04-781   Data Science for Energy, Air, and Inequality

Course description

Energy systems and the atmosphere are deeply interconnected. How we generate, distribute, and use energy shapes the air we breathe, the climate we inhabit, and the health and economic outcomes of communities around the world. However, the benefits and harms of these energy systems are not equally distributed across global society. Closing this gap requires both a clear understanding of how these systems work and the computational tools to analyze them rigorously.

This course trains students to use data science to address problems of inequality in energy and atmosphere. The course does so by building the scientific, computational, and social science frameworks needed to carry out rigorous analysis. Students with strong programming backgrounds will learn how energy and atmospheric systems function, then apply statistical modeling and machine learning to real-world datasets from communities facing challenges today. Case studies from West Africa, the Americas, South Asia, and beyond will serve as examples for engaging contemporary topics in energy and environmental justice, such as air pollution exposure, energy poverty, and extreme weather risk.

Coursework includes problem sets, bi-weekly quizzes, and a final project on a topic of one’s choosing. By the end of the course, students will have developed the conceptual and quantitative toolkit to become strong contributors to research, policy, and industry work that can create positive societal change in energy and atmosphere.

Learning objectives

• Introduce energy and atmospheric science foundations, including topics such as emissions, aerosols, solar irradiance, and air pollution health pathways, that are needed to support rigorous data analysis of inequality problems at the energy-air nexus
• Develop students' ability to work with real-world, heterogeneous data, including satellite products, population datasets, and energy system measurements
• Engage with case-studies to examine how the benefits and burdens of energy and atmospheric systems are distributed unequally across global society, using examples spanning West Africa, the Americas, South Asia, and beyond
• Prepare students to apply statistical modeling and machine learning to energy and atmospheric challenges in ways that can inform research, policy, and community-based efforts to improve societal outcomes 

Outcomes

By the end of this course, students will be able to:

• Describe the core mechanisms and concepts that cause energy system benefits and burdens, including air pollution, energy access, and climate hazards, to be distributed unequally across global society - drawing on frameworks such as energy justice, environmental justice, and energy sovereignty
• Design and execute data analysis strategies by acquiring, merging, and analyzing heterogeneous datasets, such as gridded satellite products, population data, and energy system measurements
• Build, evaluate, and interpret quantitative models, including trend estimation, forecasting, and machine learning, to characterize patterns in energy and atmospheric data
• Quantify the effects of environmental hazards related to energy and atmosphere on populations, such as air pollution exposure and extreme heat, and compare how these inequalities manifest across different geographic, political, and historical contexts
• Critically assess the limitations, biases, and uncertainties in environmental and energy datasets, particularly in data-scarce regions
• Communicate quantitative findings in terms of their implications for policy, research, or community decision-making

Content details

Atmospheric science

• Greenhouse gas emissions and warming
• Aerosol formation, transport, and deposition
• Solar irradiance transmission through the atmosphere
• Health impact pathways of energy-related air pollution

Data science skills

• Programming fundamentals
• Working with geospatial data (xarrays, interpolation, regridding)
• Accessing and using APIs for environmental and population data
• Population-weighted exposure analysis
• Statistical modeling and trend estimation
• Machine learning model development and evaluation

Social concepts

• Energy justice
• Energy sovereignty
• Energy access
• Energy poverty/burden
• Air pollution exposure disparities
• Extreme weather risks

Prerequisite

Demonstrated programming proficiency. Students should have taken Programming for Data Analytics, Data Analytics, and Data Inference for Machine Learning courses or their equivalents.