Originally posted by **christophfroehlich** December 13, 2021
As described in the [Wiki ](https://github.com/SpaceTeam/ortools/wiki/Multilevel-Wind-Model), the current wind speed + direction is calculated at the current altitude by using a linear interpolation separately in North/East direction.
Considering changing directions of different wind layers, is the cartesian component-wise interpolation the best choice? If the wind changes directions by 180°, is there zero wind in between?
- Should we use polar coordinates for interpolation, i.e., wind speed and direction? How does one solve the correct 2-pi interpolation?
- Is there a more elaborate theory on how the wind can change with altitude?
Discussed in https://github.com/SpaceTeam/ortools/discussions/7