The full methodology to build this dataset is detailed in Herry et al. (2025). In short:

* The ERA5 wind field is first filtered within a 200km-radius from the TC track, using an 11-day low-pass filter designed to remove the underestimated ERA5 TC, but keeping the synoptic flow.

* The parametric wind field is computed over a 1/4° grid from the Wood et al. (2013) parametric formulation, and best-track TC parameters, Vmax, Rmax, and the latitude. The coefficient of the parametric formulation are expressed as a function of best-track parameters also, following a fit to SAR data detailed in Herry et al. (2025).

* Then the blending is performed between two radii R1 and R2. Within R1 of the TC track position, the zonal and meridional wind components of the symmetric parametric vortex are added to the asymmetric ERA5 filtered wind field. Between R1 and R2, the wind transition from that to ERA5 wind with a linear ponderation. Out of R2, the ERA5 wind field remains unchanged. R1 and R2 are adjusted for each TC position:

- if an 34-kt wind radius (R34) can be found in ERA5 and the ratio Rmax/R34 is less than 1/3, then the blending is performed between R1 = R34/2 and R2 = R34,

- if that ratio is greater than 1/3, then R1 = 2.Rmax and R2 = R34

- if no R34 can be found, the transition zone is defined between 100 km and 200 km.

The pressure field is similarly blended. The parameterization of the pressure profile is derived from Knaff and Zehr (2007) and Holland (1980).