Jakob Beran, Talha Bin Masood, Wito Engelke, Gunilla Svensson, Rodrigo Caballero, Ingrid Hotz
This dataset contains historical paths of the cyclones produced by using a novel cyclone tracking algorithm. The data consist of cyclone tracks based on global ERA5 data at the spatial/temporal resolution 0.25°/1-hourly, for the years 1979 – 2020. Data are provided in the form of text files.
The data was created in order to provide better representation of multi-centre cyclones, which were previously proven to be difficult to accurately track. It can be used for variety of studies that involve cyclone analysis — from individual case studies to more broader climatological studies.
The data has been generated with TopoTrack, an objective feature tracking code using novel topological methods. The cyclone tracks has a global coverage over the time period and are based on MSLP (mean sea-level pressure) data from the ERA5 reanalysis datasets (the fifth generation ECMWF atmospheric reanalysis of the global climate).
The analysis was performed on resources provided by the National Academic Infrastructure for Supercomputing in Sweden (NAISS).
Download data
Citation
Jakob Beran, Talha Bin Masood, Wito Engelke, Gunilla Svensson, Rodrigo Caballero, Ingrid Hotz (2024) Hourly ERA5 cyclone tracking data from 1979 – 2020 using topological methods. Dataset version 1. Bolin Centre Database. https://doi.org/10.17043/beran-2024-era5-cyclone-tracking-1
References
Engelke W, Masood TB, Beran J, Caballero R, Hotz I (2021) Topology-based feature design and tracking for multi-center cyclones. In: Hotz I, Bin Masood T, Sadlo F, Tierny J (eds) Topological methods in data analysis and visualization VI: Theory, applications, and software. Springer, Cham, pp 71 – 85. https://doi.org/10.1007/978-3-030-83500-2_5
Nilsson E, Lukasczyk J, Engelke W, Masood TB, Svensson G, Caballero R, Garth C, Hotz I (2022) Exploring Cyclone Evolution with Hierarchical Features. 2022 Topological Data Analysis and Visualization (TopoInVis). Oklahoma City, OK, USA, pp 92 – 102, https://doi.org/10.1109/TopoInVis57755.2022.00016
Hanley J, Caballero R (2011) Objective identification and tracking of multicentre cyclones in the ERA‐Interim reanalysis dataset. Quart J Royal Meteoro Soc 138:612–625. https://doi.org/10.1002/qj.948
Data description
The dataset is provided in one zip file containing 60 text files with file extension .dat. It also contains two csv files for mapping global trackIDs to local trackIDs used in the individual files.
The dataset consists of 1-hourly cyclone tracks for the time period 1979 – 2020 with global coverage. The entire time period is partitioned into 60 chunks (files), with the default names of tracks_*.dat, where * goes from 0 to 59.
Each row in the files provides information about a cyclone, while columns at the beginning of each file provide information about the data that is represented in each column.
Individual file size varies, but is in the order of 240 MB each. Total file size: 14.6 GB (3.3 GB compressed).
Each file contains information about all global tracks for a period of about 8.5 months. There is an overlap of the tracks with the tracks in the preceding file to ensure correct matching of tracks at the interfaces. Tracks that show up in two different files have local trackIDs in the two track files and the file localGlobalMap.csv holds information about the associated global trackID. The file globalTrackMap.csv holds information about the local trackIDs for a given global trackID. Thus, the dataset provides a seamless continuum of global tracks for a period of 42 years.
The tracks have been produced with the tracking algorithm TopoTrack, a Python package currently not yet published. The TopoTrack code provides Python classes and functions for reading and writing the data. The code also provides tools for analysing and plotting tracks. Contact the author to access the tools needed for reading the tracks.
Tracks could also be read through standard text programming languages that process text, like Python.
Comments
The dataset was produced based on ERA5 mean sea-level pressure (MSLP) data fields with 1-hourly temporal resolution and 0.25 degrees spatial resolution.
The cyclone tracking algorithm, primarily developed by Jakob Beran, uses topological methods (Engelke et al. 2021, Nilsson et al. 2022) to track cyclones and it has the added benefit of better representing multi-centre cyclones. It extends the work previously done by Hanley & Caballero (2012).