A shortwave trough with moderate low-level flow spawned storms across the northern U.S. Below is the convective outlook discussion from the SPC at 16:30 UTC.

...Northern High Plains into the northern Plains...
   Recent satellite imagery shows a well-defined shortwave trough
   moving eastward across southern Saskatchewan and western MT/eastern
   ID. A belt of stronger westerly flow aloft likely exists within the
   base of this trough from eastern ID into western WY. Recent
   mesoanalysis estimates 50 to 60 kt at 500 mb, but its actual
   magnitude is uncertain. Showers and isolated thunderstorms are
   already ongoing downstream of this wave, and the general expectation
   is for these showers and thunderstorms to persist as the wave moves
   eastward, eventually encountering a notably more unstable airmass
   downstream.

   Current surface observations show mid 60s dewpoints into the western
   Dakotas, with upper 50s/low 60s dewpoints farther west in eastern
   MT. Afternoon high temperatures are expected range from the low to
   mid 80s along the MT/ND border into the low 90s farther south into
   southwest SD. Despite relatively poor pre-frontal mid-level lapse
   rates, the resulting temperature/dewpoint combination will support
   MLCAPE ranging from around 500-1000 J/kg across eastern MT and the
   western Dakotas to closer to 2000 J/kg across the central Dakotas.
   Some low-level stability may linger, but should be overcome by the
   strong forcing, and the development of a line of strong to severe
   storms is anticipated. Strong to severe gusts will be the primary
   risk, with some instance of gusts from 65 to 80 mph possible.
   Maximum storm intensity and highest severe coverage is currently
   forecast to occur from the ND/SD/MT border intersection vicinity
   eastward into south-central ND and north-central SD. Decreasing
   buoyancy is expected from the central Dakotas eastward, with
   gradually decreasing storm intensity expected as the line of storms
   moves into this region this evening.

   ...Central High Plains...
   Increased coverage and depth of the thunderstorms across UT indicate
   the presence of secondary area of ascent south of the primary
   shortwave trough, perhaps tied to a convectively augmented vorticity
   maximum. Expectation is for this vorticity maximum to progress
   eastward, persisting across the central Rockies before entering
   deeply mixed boundary-layer in place over the central High Plains.
   High-based storm character coupled with steep low-level lapse rates
   will support the development of strong outflow, and gusts from 60 to
   75 mph are possible.

The spatially based ProbSevere v4 had moderate probabilities ahead of the supercell in far northwest South Dakota. However, smaller, short-lived cells in the county to its south had < 10% probabilities, yet produced numerous wind reports. The NWS severe thunderstorm warning also was a bit late, indicating the first reports spurred the warning.

Further south, towards Rapid City, SD, a congealed outflow boundary also produced many damaging or measured wind gusts. While the NEXRAD level-II reflectivity looked fairly benign, the doppler velocity showed the strength of the winds. Surprisingly, a warning was not issued this far south.

We can easily see that a situation like this is difficult for PSv4. The MRMS quality-controlled out the gust front in reflectivity, though the low-level MRMS azimuthal shear should have been able to help a bit. This is an example where the level-II data may have helped produce more reasonable probabilities of severe wind.