SFR picked up band of heavier snow across some areas…but not as well in others.

Another arctic front and associated precipitation moved across the region on Tuesday (Jan 12, 2016). The airmass was cold enough that the precipitation generally began as snow and remained as snow during the event.

A band of heavier snow developed along the arctic front as it pushed across Ohio into Pennsylvania and West Virginia. When comparing the SFR image and the radar, the SFR product picked up the band of heavier snow quite well across southwest Pennsylvania. Some places in the heavier band of snow in Pennsylvania reported visibilities of 1/4 mile.

The band of heavier snow extended southwest along the Ohio River across southeast Ohio into Kentucky with some visibilities of a mile or less. However, the SFR product did not seem to pick up the band in southeast Ohio or Kentucky as well. In fact, it seemed to miss the snow in Kentucky all together.

Based on the radar mosaic, it appears the band of snow across southeast Ohio and portions of Kentucky was generally as intense as that across southwest Pennsylvania. However, the band in southeast Ohio and Kentucky was not as wide. Could this be part of the reason for the SFR not indicating as heavy snow in southeast Ohio and Kentucky?

1840Z 12_Jan_16 SFR product

1842Z 12_Jan_16 Radar Mosaic

19Z 12_Jan_16 Surface Observations

7 thoughts on “SFR picked up band of heavier snow across some areas…but not as well in others.

  1. Another nice post, Jeff! I’ve forwarded your question along to the NESDIS developers to address your hypothesis that the SFR product missed the snow in KY was due to the thin banded nature of the event. I’m not 100% sure myself, so I wanted to see their thoughts!

  2. Our first-step snowfall detection model did catch the snow band in Kentucky, but the second-step filters removed it. One of the filters is 2-m temperature < 33.8 F (1 C). It appears that the surface temperature was too warm to pass this filter, at least by GFS forecast. I spot checked a point in the snow belt. The GFS 18Z 2m temperature was 35.3 F. It may lead to too high a FAR if we relax the temperature threshold.

    • I understand the need to have such a filter such as 2-m temperature < 33.8 DegF as a cutoff for snow. But when we have really cold 850 mb and/or 925 mb temperatures, we often have all snow even with surface temperatures as high as 35 or 36 DegF. I could ask our forecasters what the appropriate temperatures should be.

      Could another filter be added to include looking at 850 mb and/or 925 mb temperatures? If these temperatures are colder than a certain value, a different surface temperature could be used as a cutoff?

      • Yes. We can add another check for temperature at 850 mb/925 mb and relax the surface temperature threshold. It would be a great help if you could provide a temperature that we should use for the check. Thanks!

      • I have asked our forecasters for appropriate threshold 850/925 mb temperatures and will get back with you as soon as I receive some responses. However, I have seen all snow at the surface with temperatures of 35 or 36 DegF when 850 temperatures were around -10 DegC or colder.

  3. One more comment…Could the RAP or HRRR 2m temperatures be used instead of the GFS? These models are run much more frequently than the GFS.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s