Picking YSU from the literature
We ship the YSU planetary boundary layer scheme. We picked it the way most operational regional setups do - from the published evidence on what works at this resolution. Here is the reasoning.
The PBL scheme decides how the boundary layer grows through the day. It is the piece of the model most directly responsible for everything a pilot cares about: boundary layer height, surface heat flux, low-level winds, thermal triggering, moisture mixing up into the cumulus level. If you change nothing else, changing the PBL scheme will move your forecast more than changing any other single piece of the physics.
There are roughly a dozen PBL schemes in WRF. For daytime convective conditions in the mid-latitudes, two are usually shortlisted: YSU (Yonsei University) and MYNN-2.5. YSU is a non-local first-order scheme - it represents large convective eddies as a bulk transport term that reaches through most of the PBL in one step. MYNN-2.5 is a local TKE-closure scheme - it evolves a prognostic turbulent kinetic energy field and mixes properties based on the local TKE.
We ship YSU. The reasoning is straightforward: it is the most-documented operational PBL scheme, the one most regional weather centres use as their default, and the one whose failure modes are best understood in the published literature. The literature consensus is that YSU produces deeper, more well-mixed boundary layers than MYNN-2.5 in convective regimes, and that it tends to fire convective onset slightly earlier (a known feature of non-local schemes). For soaring, deeper boundary layers and earlier onset are both useful failure modes if you are going to be off in some direction.
When something looks wrong in a forecast, you want to be able to look up other people's experience with the scheme. YSU has by far the broadest published validation, and that practical pull toward the well-trodden path is part of why it is a defensible v1 choice rather than a coin flip.
The bake-off we have not yet run. A head-to-head of YSU against MYNN-2.5 on UK convective days is on the roadmap. Doing that properly requires a verification setup we are still building - radiosonde comparisons, pilot-reported convective onset times, and a long enough sample of days that the differences are not lost in the noise of one synoptic regime. The bake-off lives downstream of the validation pipeline, and the companion post on MYNN-2.5 lays out what that test will look like.
When we would revisit. Three triggers. First, when the validation pipeline is in place and we can run a proper comparison without making numbers up. Second, when we extend coverage to terrain regimes that stress YSU's known weak points - particularly stable nocturnal boundary layers and complex orography, where MYNN-2.5 tends to do better. Third, if any pilot or club integrating Convek reports systematic timing errors on the YSU early-onset bias that they cannot live with.
For now, YSU is the shipping choice. It is a defensible default by every standard the published literature offers. The model page lists the full physics stack alongside it.