This fabric is examined in three ways. The most obvious one is with reference to radar reflectivity over the depth of the CBL, since cloud streets commonly are identified by the albedo pattern on visible satellite imagery or by the reflectivity pattern on radar maps. However echo plumes' have remarkably little buoyancy or rising air motion, irrespective of the linear vs. cellular organization of CBL echoes.
Secondly, we isolate rising columns or radar updraft plumes' in the CBL. These are more buoyant, especially at low levels. They have the dynamic characteristics of entraining thermals. On the day without linear cloud/echo organization, convective updrafts tend to be weaker and narrower, and their buoyancy is slightly less. They more clearly coincide with echo plumes, compared to the two days with cloud streets. The latter suggests that they have a longer lifecycle, or at least that snow crystals have a longer residence time in the weaker updrafts, enhancing growth by deposition.
Thirdly, we isolate local domes in the CBL top, as inferred from WCR zenith reflectivity. These domes correspond well with both rising air in the column below and with positive buoyancy within the CBL, both on the two days with cloud streets, and the one day without. This finding confirms that cloud streets are a manifestation of horizontal convective rolls within the CBL, rather than due to gravity waves above the CBL. Even though horizontal Doppler velocity patterns are suggestive of a secondary roll circulation, the dominant dynamics are convective. The roll circulation certainly affects the location of the convective cells, but our study does not provide evidence that it affects their intensity.