Inquiry Regarding Eddy Detection Parameters in META3.2 for Higher Resolution Data

[441837297]

Hello, my name is Zhisheng, and I am currently conducting research on mesoscale eddy detection using the META3.2 dataset.

I have been using the Global Ocean Gridded L4 Sea Surface Heights And Derived Variables Reprocessed dataset (cmems_obs-sl_glo_phy-ssh_my_allsat-l4-duacs-0.125deg_P1D), which has a resolution of 0.125°. However, I noticed that META3.2 was developed using an older version of the dataset with a resolution of 0.25°.

When I applied the detection and tracking parameters from the META3.2 paper (e.g., shape_error=0.7 and min_overlap=0.05) to the higher-resolution 0.125° data, I observed significant differences in the detected eddy characteristics. Specifically, the results seem to include more elongated or non-eddy mesoscale structures.

Would you recommend adjusting the detection and tracking parameters for higher-resolution datasets? For instance, could stricter thresholds (e.g., reducing shape_error or increasing min_overlap) help mitigate the inclusion of non-eddy features?

Thank you for your time and insights. I greatly appreciate your guidance on this matter.