Velocity Radar: November 5, 2017 KBUF

November 6, 2017—Buffalo, NY

A storm system moved through the area of Buffalo, New York on November 6, 2017 between 0029Z and 0200Z. A cold front was prominent near Saint Catharines, Ontario, Canada to Erie, Pennsylvania and crossed the region from west to east between 0100Z and sunrise. In advance of this cold front, there was moderately strong south-southwest low-level flow with surface gusts up to 30 MPH (fig.1). In radar imagery from GIF1 and GIF2, there seems to be a mesoscale convective vortex (MCV) present over northwest Pennsylvania. Towards the beginning of GIF1, there was actually an EF-1 tornado reported at 23:22 UTC (6:05 PM EST) by the National Weather Service (NWS) storm survey team, just four miles southwest of the town of Erie in the Millcreek Township. The tornado track was roughly 2.4 miles in length by 100 yards in width, with maximum winds estimated to be 90 MPH. Fortunately, no one was killed or injured during this event that brought nearly three inches of rain to the area. However, there was not really any indication on radar that there was a tornado. Next to the green inbound winds, there was no red outbound winds near the area, instead there was purple haze, indicating that the radar was unable to determine the wind’s velocity, which is referred to as range folding (RF) or velocity folding. Depending on the radar’s operation mode and PRF (pulse repetition frequency), the range folded data may occasionally obscure large portions of the radar’s image. As you may notice in GIF3, there seems to be a decent amount of range folding near the southern end of this storm system. At any rate, another reason I would not have thought there was a tornado just from radar alone was due to the lack of intensity of the inbound winds. Generally, intense circulations with diameter of one mile or less are tornadic, while large couplets are associated with mesocyclones. There is no specific value of diameter and magnitude that differentiates between the two due to the variety of circulations that may occur and due to gates having a variety of volumes and height, which depend on range and beam width. However, in this case, I suspect that the radar probably overshot the tornado circulation completely, given that the tornado occurred roughly 8,351 feet from the radar.

Figure 1

GIF 1

GIF 2

Figure 2

GIF 3