• Please note this new thread in the Announcements forum regarding updated management of the site. Jeff Duda has agreed to assume ownership and lead management responsibilities during what is anticipated to be a 3-12-month interim period while we perform some updates to the site, including changing the hosting. The staff will do their best to provide timely and transparent updates throughout this transition period.

  • A website update occured last week that rendered some of the page display styles obsolete. If you have logged onto the site and are experiencing a strange or messy appearance, it is likely due to this factor. An inadvertent mistake in the meantime likely made this issue worse. It has hopefully now been fixed. See this thread for details on the fix. --Jeff

Strange forecast sounding?

Just trying to get an idea of environmental conditions around the storms in central Oklahoma overnight, I happened to click upon a strange sounding on PivotalWeather:

Apparent supersaturation around 900 mb???

WeatherNerds and TropicalTidbits don't show the same anomaly:

1645030019009.png 1645029965071.png

But they all show notable differences in the sounding thermodynamics and wind field for what appears to be the same lat/lon/run/fhour, unless I made a mistake somewhere? And being the same to the hundredth of degree, that's only ~ 1 km, so generally would think the same grid point. I tried to search up on the topic, but only found another anomalous supersaturation sounding from a Twitter post in 2019: So the interest becomes... is there significant difference in how these soundings are generated? Is there major extrapolation taking place that can cause such supersaturation? Is it actually realistic that strong downdrafts/sharp boundaries can produce such soundings in real life (doubting it??), or at least affect the models locally so that they have nonrealities, at least in extrapolation?

The point I chose was ironically intended to be an environmental sounding, well out ahead of the convection at the fhour:


Any input on these oddities?

Jeff Duda

site owner, PhD
Staff member
This has something to do with post-processing and spatial interpolation of output HRRR fields. We have seen this behavior before, but it does not happen in the WRF output. So, somewhere between WRF output and dissemination, some kind of interpolation or correction is causing the anomaly.