DADA: Detection and Attribution of climate change based on Data Assimilation

DADA is a theoretical project aimed at developing new strategies and methods for the detection and attribution of climate change based on data assimilation.


New methods, based on data assimilation, designed to timely attribute climate change and climate related extreme events.

Project Summary

How can observations be used to best evidence the influence on climate of human activities, among other forcings ? Statistical methods of Detection and Attribution (D&A) were designed to answer this question which is of high societal relevance when it comes to adaptation and mitigation policy. Conventional D&A methods are based on linear regression of spatial or temporal patterns extracted from one or several climate models (‘optimal fingerprinting’). They are quite successful in doing so for a variety of situations. Yet, the D&A community is increasingly tackling a number of variables and scales for which these methods are less effective, and these methods also have some limitations. So the quest to improve upon available D&A methods is heating up.

How can observations be used to best constrain a numerical model’s state variables and parameters ? Methods of Data Assimilation (DA) meet this general purpose. This now firmly established methodological field has grown out of its original application in numerical weather forecasting to reach a wide variety of applications in geophysics. While DA methods were at first mostly oriented towards the problem of initializing and updating a model’s state variables, the recent extensions of DA towards the problem of calibrating a model’s parameters have shown exciting results. So the quest to elaborate and extend DA methods is going on.

How could D&A methods take advantage of recent progresses in DA? Could D&A be a fruitful field of application for DA? One may hope so. Indeed, under an inverse problem formulation of D&A, observations can be seen as a complex function of the forcings consisting of the full climate model itself. Under this perspective, D&A consists ‘merely’ in reconstructing forcings from available observations by inverting the full climate model itself – a challenge for which recent DA- based parameter estimation schemes might be an answer.

This project will explore the potential for cross-fertilization between these two methodological research fields in climate science that have been so far somewhat isolated from one another. In this purpose, the project team will gather together experts from both communities. Our purpose will be to establish a « proof of concept » for this idea, which will address the most fundamental questions associated to relevance, utility and feasibility of such an alternative methodological paradigm in D&A. The project is ambitious because the approach is novel and requires to bridge between two research fields of climate dynamics that have their own vocabulary and concept. But it is reasonable because we will restrict this exploration to simplified and idealized conditions, with low to intermediate complexity models, and different well-targeted situations.

Project Details
Funding Agency: 
Agence Nationale Francaise de la Recherche
NERSC Principal Investigator: 
Alberto Carrassi
Coordinating Institute: 
Nansen Environmental and Remote Sensing Center
Project Status: