A Study on One-way Nesting Technique for Oceanic Regional Circulation Modeling through Small-scale Information Restoration

An Oceanic Regional Circulation Model (ORCM) is generally constructed by downscaling the large-scale information of an Oceanic Global Circulation Model (OGCM) through a nesting technique. Since the OGCM has relatively low resolution and so only has large-scale information, ORCM can only be built with large-scale information. Even if the ORCM is run with only large-scale information, small-scale fluctuations can be developed by “stimulating sources” such as the self-interaction of large-scale motions. However, it is not sufficient to generate small-scale fluctuations by itself in a short time, and if other stimulating sources, such as islands, coastlines, and strong advection, are absent, errors can occur (Van Tuyl and Errico, 1989; Pham and Hwang, 2020). Pham and Hwang (2020) suggested a method to introduce artificial small-scale fluctuations for boundary conditions that can reduce errors and generate small-scale features in a short period. This method was applied to the ocean of 330 km  220 km with a resolution of 1 km  1 km, and the accuracy of the applied model was higher than that of the original model. In this study, this small-scale addition method is developed to be applied to the coast smaller than the ocean, and the generated small-scale fluctuations are added in the vertical direction as well.

To construct the artificial small-scale fluctuations, we first need to analyze the regional energy spectrum. A high-resolution ORCM model is built using the OGCM data and used as reference data. The regional energy spectrum is defined in the high-resolution ORCM results, and small signals below a certain wavelength are removed with a low-pass filter using the Discrete Cosine Transform to mimic the low-resolution OGCM data. We then assume that energies cascade monotonically from the large scale. After the energy spectrum is extended from the larger scales to the smaller scales, this spectrum is transformed to realistic values using the Inverse Discrete Cosine Transform. The realistic values with small-scale features are added to the original values with large-scale features, which apply to all vertical layers. When this small-scale addition method is applied to the coast where the strong advection is present, the method does not significantly improve the accuracy, unlike when applied to the ocean. Therefore, in coastal areas where strong advection exists, another method is needed to increase accuracy in a short time.

 

References

Pham, V. S., & Hwang, J. H. (2020). Effects and recovery of small-scale fluctuations in one-way nesting for regional ocean modeling. Ocean Modelling, 145, 101524.

Van Tuyl, A. H., & Errico, R. M. (1989). Scale interaction and predictability in a mesoscale model. Monthly weather review, 117(3), 495-517.

 

Acknowledgement

This work was supported by the project entitled “Development of living shoreline technology based on blue carbon science toward climate change adaptation [grant number 20220526]” funded by the Ministry of Oceans and Fisheries (MOF), South Korea.