Vertical Localization for Strongly Coupled Data Assimilation: Experiments in a Global Coupled Atmosphere-Ocean Model

Strongly coupled data assimilation allows observations of one Earth system component (e.g., the ocean) to directly update another component (e.g., the atmosphere). The majority of the information transfer in strongly coupled atmosphere-ocean systems is passed through vertical correlations between atmospheric boundary layer and ocean mixed layer fields. In this work we use correlations from a global, coupled model to study vertical observation-space localization techniques for strongly coupled data assimilation. We generate target correlations using a bootstrapping approach from a single 24 hr forecast from a realistic global, weakly coupled atmosphere-ocean cycling system with an 80-member ensemble, which is the ensemble size currently used by the NOAA operational global data assimilation system. We compare data assimilation methods with different localization schemes using single-update, offline experiments. We develop a new strategy for optimal observation space localization, called Empirical Optimal R-localization (EORL), to give an upper bound on the improvement we can expect with any localization scheme. We then evaluate Gaspari-Cohn localization, which is a commonly used parametric localization function and review its performance with respect to the optimal localization scheme. We investigate how the performance of these localization strategies changes with increasing ensemble sizes. Our results show that strongly coupled data assimilation has the potential to be an improvement over weakly coupled data assimilation when large ensembles are used. We also show that the Gaspari-Cohn localization function does not appear to be a particularly good choice for cross-fluid vertical localization. Accurate Earth system forecasts rely on accurate estimates of the current state of the system. This initial state is estimated through a process called data assimilation, which combines the previous forecast with current observations. This process relies on accurate estimates of uncertainty in both the model and the observations. Inaccurate estimates of model and observation uncertainty can result in a degradation of future forecasts. A technique called localization is widely used to minimize the impact of distant observations due to unreliable uncertainty estimates over long distances. In this work we study localization in the context of a global, coupled atmosphere-ocean model. In particular, we look at strongly coupled data assimilation, where observations of the ocean are used to update the atmospheric model state and vice versa. This is in contrast to weakly coupled data assimilation, where observations of the ocean are only used to update the ocean model state. We investigate different types of localization and find that a commonly used localization function does not perform particularly well in strongly coupled model initialization. We investigate vertical localization for strongly coupled atmosphere-ocean data assimilation in a realistic global model Strong coupling can improve data assimilation effectiveness over weak coupling when large ensembles are used We present a method for optimal observation space localization, called EORL, and demonstrate its performance in offline experiments