Climate Change Modeling and Prediction of Economic Impact

Global warming is widely recognized as one of the defining issues of our time, presenting one of the biggest challenges facing humanity in the 21st century, affecting lives, communities, and countries worldwide. To address this challenge, estimates of the impact of greenhouse gas emissions on Earth's climate and predictions of the responding climatic impact on society are an imperative prerequisite to successfully guide governmental and inter-governmental policies. Global Climate Models (GCMs) and Integrated Assessment Models (IAM) are designed to simulate the physical climate system and the economic impacts of climate change, respectively. While both modeling tools have evolved tremendously in recent years, GCMs still rely on IAMs for future emission scenarios, while IAMs rely on GCMs for key properties of the climate system. Each modeling tool is thus incomplete and suffers from its own shortcomings. In this research project, we seek to address the above issues, which currently preclude reliable estimates of future climate change and its economic impacts. Specifically, we are in the process of constraining Earth's climate sensitivity (that is, the global mean surface temperature increase per doubling of atmospheric carbon dioxide) through a cross-disciplinary approach that applies time series analysis that is common in the field of economics to climate data, and confronts GCMs with observational data in a new and innovative way. The goal is to reveal which models suffer from compensating errors that allow them to reproduce past observations for the wrong reasons. Secondly, we are working on the coupling of a state-of-the-art GCM with a recently developed IAM, in order to create a new and more complete tool for simulations of future climate and its economic impacts, and ultimately calculate economic damage using functions that can take climate variables beyond surface temperature into account. Result from the project have already emerged in the journals Geophysical Research Letters (2018) and Journal of Econometrics (2019). The latter received ample attention from Norwegian news media. The results indicate that Earth's climate sensitivity (measured in terms of the so-called transient climate response) is somewhat higher than previously thought. The results emerged from analysis that applied econometric methods (a type of statistics typically applied to economic data) on approximately 50 years of observations of temperature, radiation and greenhouse gas concentrations. The analysis is therefore completely independent of the global climate models that have traditionally served as the basis for knowledge on climate sensitivity. The new analysis supports the most sensitive of the state-of-the-art global climate models, and imply that in order to reach the goals set in the Paris agreement of 2015 (i.e. to limit global warming to 1.5-2 degrees Celsius), greenhouse gas emissions must be reduced even faster than previously thought. Two additional papers emerging from the project were recently published; a study using machine learning to fill gaps in the observational data sets described above was published in Journal of Climate, and a paper emerging in the journal Nature Geoscience that showed that climate sensitivity is not constant in time, but rather increases with global warming.

Prosjektets viktigste virkninger er at det har utviklet et nytt tverrfaglig modell-rammeverk for beregninger av de økonomiske konsekvensene av globale klima-endringer. Rammeverket er det første av sitt slag, og har potensiale til å transformere hvordan man utvikler klima-fremskrivinger i fremtiden. Basert på dette rammeverket er også et såkalt INTPART prosjekt finansiert, for å anvende det forsknings-baserte nye rammeverket i undervisning og formidling. Videre har prosjektet utviklet en ny (økonometrisk) metode for observasjons-basert beregning av global klima-følsomhet. Klimafølsomheten er et mål på hvor mye varmere kloden blir for en gitt økning av klimagasser i atmosfæren, og er derfor en helt avgjørende faktor for beregninger av fremtidens klima. Den nye metoden er komplementær til mer tradisjonelle beregninger basert på global klima-modeller, og er dermed velegnet for å teste hvordan disse samsvarer med observasjoner på en ny måte. Begge de to hoved-virkningene i prosjektet er direkte knyttet til effekter gjennom resultatenes viktige bidrag til kunnskapsgrunnlaget for klima-tiltak. Særlig er resultatene relevante for samfunnets forutsetning for å utvikle tiltak for å begrense global oppvarming, og for å tilpasse seg de fremtidige klima-endringene som er uunngåelig. To doktorgrader er gjennomført innenfor dette prosjektet, begge med en sterkt tverrfaglig profil. Denne typen tverrfaglighet er avgjørende for a kunne takle noen av de største samfunns-utfordringene, særlig relatert til bærekraft og klima, og derfor etterspurt i dagens arbeidsmarked. Prosjektet er gjennomført i tett samarbeid med to internasjonalt ledende universiteter (Yale University i USA og ETH-Zürich i Sveits), og har dermed styrket det internasjonale nettverket for de norske deltagerne og åpnet nye karriere-muligheter for de to PhD studentene .

Global warming is widely recognized as one of the defining issues of our time, presenting one of the biggest challenges facing humanity in the 21st century, affecting lives, communities, and countries worldwide. To address this challenge, estimates of the impact of greenhouse gas emissions on Earth's climate and predictions of the responding climatic impact on society are an imperative prerequisite to successfully guide governmental and inter-governmental policies. Global Climate Models (GCMs) and Integrated Assessment Models (IAM) are designed to simulate the physical climate system and the economic impacts of climate change, respectively. While both modeling tools have evolved tremendously in recent years, GCMs still rely on IAMs for future emission scenarios, while IAMs rely on GCMs for key properties of the climate system. Each modeling tool is thus incomplete and suffers from its own shortcomings. In the proposed research, we seek to address the above issues, which currently preclude reliable estimates of future climate change and its economic impacts. Specifically, we aim to firstly constrain Earth's climate sensitivity through a cross-disciplinary approach that applies econometric analysis to climate data, and confronts GCMs with observational data in a new and innovative way which is more likely to reveal which models suffer from compensating errors that allow them to reproduce past observations for the wrong reasons. Secondly, we intend to couple a state-of-the-art GCM with a recently developed gridded IAM, and ultimately calculate economic damage using functions that can take climate variables beyond surface temperature into account. The proposed research will be centered around two hypotheses, which we seek to test using the cross-disciplinary research approach outlined above. The project brings together climate scientists and economists with a record of successful cross-disciplinary collaboration, increasing the likelihood of a successful outcome.