A cheap data assimilation approach for expensive numerical simulations

TL;DR

This paper introduces a cost-effective data assimilation method as an alternative to traditional approaches for numerical climate models, aiming to improve long-term simulations and reduce biases using observational data.

Contribution

The paper proposes a novel, inexpensive data assimilation approach tailored for large-scale climate models, addressing their sensitivity and bias issues.

Findings

The method effectively reduces model bias using observational data.

It is computationally cheaper than classical data assimilation techniques.

The approach improves long-term climate simulation accuracy.

Abstract

Using a very cheap Data Assimilation (DA) method, I show an alternative approach to classical DA for numerical climate models which produce a large amount of "big data". The problematic features of state-of-the-art high resolution Regional Climate Models are highlighted. One of the shortcomings is the sensitivity of such models to the slightly different initial and boundary conditions which could be corrected by assimilating scattered observational data. This method might help to reduce the bias of numerical models based on available observations within the model domain, especially for the time-averaged observations and the long-term simulations.