EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Using Synchronization to Model Twins for Eddy‐Rich State Estimation–The Impact of Eddy‐Noise

Armin Köhl

Earth and Space Science · 2026-11 · 13 (11)

DOI: 10.1029/2026ea005504 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Earth and Space Science, online erschienen am 15.11.2026. Verfasst von Armin Köhl. Die Arbeit nennt 81 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Deutsche Forschungsgemeinschaft.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Based on linearized models, the adjoint method is limited by the chaotic behavior of dynamical systems. A further challenge is the technical effort required to develop an adjoint of a specific model. To address both challenges, we employ synchronization in a twin‐model framework that exchanges information between a chaotic and a nonchaotic model that has an existing adjoint. For a perturbed Lorenz system, it has previously been shown that an adjoint model developed for one system can be efficiently applied to optimize other systems, provided that they represent the same physical processes. Here, the method is employed to test the feasibility of long‐window state estimation with an eddy‐rich 1/12 global ocean model. The adjoint was derived from the GECCO3 synthesis configuration (0.4 resolution). For a 1‐year assimilation period, we demonstrate moderate efficiency of the method in assimilating hydrographic observations. However, assimilating sea level anomalies (SLA) shows lower performance due to a low signal‐to‐noise ratio. Because the exact positions of the simulated eddies cannot be reconstructed with long assimilation windows, two‐thirds of the SLA cost contribution is associated with different realizations of eddy fields arising from parameter perturbations. Different filtering methods were tested to reduce the impact of eddy noise. Although convergence could be recovered, the achieved performance remains limited, constraining prospects for long‐window SLA assimilation in eddy‐rich models. However, the 1‐year window is challenging because non‐eddy‐related anomalies are relatively small. Over longer timescales, the signal‐to‐noise ratio is expected to increase.

Förderung

  • Deutsche Forschungsgemeinschaft

Zitieren

APA

Köhl, A. (2026). Using Synchronization to Model Twins for Eddy‐Rich State Estimation–The Impact of Eddy‐Noise. Earth and Space Science, 13(11). https://doi.org/10.1029/2026ea005504

Vancouver

Köhl A. Using Synchronization to Model Twins for Eddy‐Rich State Estimation–The Impact of Eddy‐Noise. Earth and Space Science. 2026;13(11). doi:10.1029/2026ea005504

BibTeX
@article{Khl2026,
  author = {Köhl, Armin},
  title = {Using Synchronization to Model Twins for Eddy‐Rich State Estimation–The Impact of Eddy‐Noise},
  journal = {Earth and Space Science},
  year = {2026},
  volume = {13},
  number = {11},
  doi = {10.1029/2026ea005504}
}
Kostenlos starten →