Automatic modeling of socioeconomic drivers of energy consumption and pollution using Bayesian symbolic regression

Daniel Vázquez,Marta Sales-Pardo,Gonzalo Guillén-Gosálbez,Roger Guimerà

doi:10.1016/j.spc.2021.12.025

Abstract

Predicting countries’ energy consumption and pollution levels precisely from socioeconomic drivers will be essential to support sustainable policy-making in an effective manner. Current predictive models, like the widely used STIRPAT equation, are based on rigid mathematical expressions that assume constant elasticities. Using a Bayesian approach to symbolic regression, here we explore a vast amount of suitable mathematical expressions to model the link between energy-related impacts and socioeconomic drivers. We find closed-form analytical expressions that outperform the well-established STIRPAT equation and whose mathematical structure challenges the assumption of constant elasticities adopted in the literature. Our work unfolds new avenues to apply machine learning algorithms to derive analytical expressions from data in environmental studies, which could help find better models and solutions in energy-related problems.

Highlights

Today’s high living standards rely on the global trade, production, and use of resources, which results in a range of environmental impacts that could challenge the stability of the Earth system
Using a Bayesian approach to symbolic regression, here we explore a vast amount of suitable mathematical expressions to model the link between energy-related impacts and socioeconomic drivers
We study the effects of six environmental drivers, i.e., total population, GDP per capita, active population, population density, urbanization rate, and climate, on four environmental impact categories: CO2 emissions(CDE), energy consumption(EC), N2O emissions(NOE), and CH4 emissions(ME) at the country level, with data spanning 25 years, from 1990 to 2015

Summary

Introduction

Today’s high living standards rely on the global trade, production, and use of resources, which results in a range of environmental impacts that could challenge the stability of the Earth system. Energy consumption reached 583.9 EJ in 2020 (BP, 2020), leading to high levels of anthropogenic CO2 emissions (31.5 Gt (IEA, 2021)) due to the heavy reliance on fossil energy resources. There is, a clear need to curb emissions by shifting to more sustainable energy sources and reducing the current high energy demand. This will require deepening our knowledge on the specific forces driving these energy-related environmental stressors so more effective policies can be developed.

Objectives

Methods

Results

Conclusion