This dataset comprises a series of CSV files, each of which catalogues a suite of information for an individual cyclone track for the Northern Hemisphere. Detection and tracking are conducted using version 13.3 of the Lagrangian cyclone detection and tracking algorithm described by Crawford et al. (2021). This algorithm is applied to sea-level pressure data from climate model simulations that were run as part of the Coupled Model Intercomparison Project Phase 6 (CMIP6) at a 6-hour temporal resolution and 100-km spatial resolution. Of the 22 models included, 13 provided the sufficient inputs for five experiments: historical, SSP1-2.6 (low emissions), SSP2-4.5 (medium emissions), SSP3-7.0 (high emissions), and SSP5-8.5 (very high emissions). The other 9 models are only present from historical and SSP5-8.5. The historical simulation is 1950-2014. The CMIP6 simulations are from 2015-2100. Also included is a comparable set of historical cyclone tracks (1950-2014) derived from the 5th version of the European Centre for Medium-Range Weather Forecasts (ECMWF) Retrospective Analysis (ERA5). The cyclone detection and tracking algorithm was applied to sea-level pressure inputs with the same spatial and temporal resolution as the CMIP6 data.

Each CSV file contains information regarding the location, propagation, intensity, and size of the storm for each observation time. Whether a storm is a multi-center cyclone or interacting (splitting or merging) with other storm systems is also noted. Only cyclones that last at least 24 hours and travel at least 1000 km are included. Storms must also be observed at least once over an elevation under 500 m and at least once at a distance at least 500 km from their point of origin. These standard criteria are necessary to remove spurious systems (i.e., reduce the false positive rate).