Skip to main content
eScholarship
Open Access Publications from the University of California

UC Irvine

UC Irvine Electronic Theses and Dissertations bannerUC Irvine

Identifying Gaps in Climate Impacts Research and Addressing them through a Software-based Workflow

Creative Commons 'BY' version 4.0 license
Abstract

Over the last several years, research has increasingly used econometrics-style statistical models to investigate relationships between climate change and macroeconomic trends. Such work, in aggregate, has shaped both decision-making at the policy level and public perception of the risks related to climate change. The primary goals of the field of climate econometrics are to: 1) examine how climate change, specifically global warming, has affected economic growth and agricultural productivity, and 2) examine how economic activity contributes to greenhouse gas emissions. Using a systematic literature review, this dissertation identifies several methodological gaps that hinder the ability of a large portion of existing climate econometrics research to convincingly present the threat to global macroeconomies posed by climate change. To demonstrate the importance of addressing these gaps for our future understanding of climate change impacts, two case studies are presented, involving: 1) an investigation of the historical impact of drought on agriculture, and 2) reproductions of several other commonly explored climate/macroeconomic relationships. To enable support for researchers to focus on oft-omitted methodological components such as out-of-sample model evaluation, quantification of uncertainty in results, computation of impacts from model coefficients, and sharing of source code and data, the dissertation then presents a standardized climate econometrics workflow, which is operationalized in the form of a software toolkit. Evaluation of the toolkit was carried out via qualitative, semi-structured interviews with several researchers in the field, and feedback based on these interviews was incorporated into the toolkit. Overall, this dissertation presents a novel, software-based approach towards the carrying out of robust and reproducible climate impacts studies, with a focus on improving the transparency, forecasting, and representation of uncertainty in climate econometrics research.