Data-model fusion approach using tree-ring width series, a mechanistic growth model and an Earth system model to reconstruct the large-scale climate variability

Tree-ring widths represent the most commonly used proxy to reconstruct the climate of the last millennium at high resolution, thanks to their large-scale availability. The approach often relies on a relationship between tree-ring width series and climate estimated on the basis of a linear regression. However, the underlying linearity and stationarity assumptions may be inadequate and dendroclimatic process-based models may be able to overcome some of those limitations. In this thesis, we have included for the first time a dendroclimatic process-based model, MAIDEN, into a data assimilation procedure in order to improve large-scale climate reconstructions. We have first developed a protocol to calibrate and apply MAIDEN to potentially any sites with tree-ring width data in the extratropical region. We have shown that MAIDEN can be applied at global scale with a good performance and that the clear benefits obtained make it an appealing candidate for large-scale paleoclimate reconstructions. Subsequently, the MAIDEN model has been successfully incorporated into a data assimilation procedure as a proxy system model allowing a robust comparison of the outputs of an Earth system model with tree-ring width observations. The procedure has been tested to reconstruct the climate variability in the Southern Hemisphere over the last four centuries with encouraging results. Even though additional analysis are needed to fully take advantage of the new development, important steps have been made to demonstrate that using MAIDEN as a proxy system model is a promising way to improve the large-scale climate reconstructions with data assimilation.(SC - Sciences) -- UCL, 202