Impact of Data Resolution on Tracking Southern Ocean Cyclones

Abstract

Abstract The ERA5 new generation of high-resolution reanalysis provides a possibility to obtain more accurate cyclone tracks in the Southern Ocean. With a commonly used cyclone tracking algorithm, this study evaluates the impact of data resolution on Southern Ocean cyclone tracks for the period from 1980 to 2020 by preprocessing the ERA5 dataset at different spatial and temporal resolutions. A new track-matching method is proposed to assure an accurate comparison of different track datasets, considering the multiple match pairs and best match pair for each track. It is found that the number, distribution, and characteristics of cyclones are considerably different for various resolution scenarios. The higher spatial resolution captures more tracks, while the increased temporal resolution decreases the number, as well as the lifetime and the moving distance of tracks. The shared cyclones of different track datasets show different characteristics, influenced by both spatial and temporal resolutions. Higher spatial resolution schemes tend to identify more additional track points after the overlapping time of shared tracks rather than before. The spatial distribution pattern of additional track points is consistent when increasing temporal or spatial resolution separately. These results are a reference for the application of objective tracking algorithms in the Southern Ocean using input data with higher resolution. Significance Statement Automatic tracking algorithms are important tools for the research of cyclones and their associated weather phenomena. High-resolution input data are available now for cyclone tracking but it also brings more noise that may affect the results. The problem can be solved by a preprocess (smoothing) process using suitable spatial and temporal resolutions. This study compared the features of Southern Ocean cyclones obtained by different spatial and temporal resolution schemes. We further study the shared cyclones in different track datasets with a new track-matching method and discuss the different impacts of spatial and temporal resolution, as well as provide a reference for the application and improvement of cyclone tracking in the future.