Topography may play two different roles in affecting the vorticity in a mesocyclone or even a tornado. First of all, wind channeling between mountains and in valleys produces a local jet of increased wind speed and change of direction to align with the valley. In a well-mixed BL, this jet may extend upward, above 1 km AGL. The change in wind direction and speed creates local maxima (minima) of ambient vertical vorticity to the left (right) of the channeled wind vector. This background vorticity may be tapped by a rotating thunderstorm, enhancing or weakening the mesocyclone.
Secondly, it has been observed in numerous cases, primarily at a tornado damage survey, that tornadoes sometimes form, or intensify, as the elevation of the ground surface below the parent storm suddenly decreases. Also, tornadoes have weakened or dissipated in several cases as the elevation of the terrain below suddenly increases. It is possible that a simple stretching process is responsible, as potential vorticity in the subcloud layer is conserved. Local storm-relative helicity may also change.
Finally, horizontal gradients in roughness length (friction) also produce local, sometimes large, gradients in wind speed in the BL. This is especially true along land/water interfaces. Similar to wind channeling, a maximum (minimum) in vertical vorticity is produced to the left (right) of the wind vector. This vorticity may also be ingested by a mesocyclone, causing it to intensify (weaken).
Several case studies are examined, using radar data, geographical data (topography and land use), storm data, and storm damage surveys. 1) Wind channeling appears to have played a role in two cases of tornadogenesis on the west side of the Tennessee River Valley in northeastern Alabama. One tornado, an EF-1, developed there on 6 February 2008, and another, an EF-3, developed in the same area on 10 April 2009. 2) Numerous cases of tornado damage increase/decrease have been observed as the tornadoes went over valleys and mountains. However, four cases will be examined herein. In the first, a long-lived supercell producing no tornado suddenly produced a tornado as the storm crossed a steep downslope of 100 m vertical terrain change on 6 February 2008. The tornado began producing EF-4 damage within 2 minutes of its touchdown. In the second, a tornado in eastern Tennessee developed in a deep valley, then dissipated on the uphill side, producing no damage on either hilltop but extensive damage in the valley, as shown by aerial photography. In the third, a tornado moving through Huntsville, Alabama on 21 January 2010 quickly dissipated as it moved up Monte Sano, a 300 m increase in terrain elevation. In the fourth, a supercell that had not produced a tornado over its > 100 km path produced an EF-2 tornado as it descended the Appalachian Mountains into Atlanta, GA on 14 March 2008. 3) Horizontal gradients in friction appear to have played a role in weakening supercell storms as they moved onshore in Mississippi in the strong easterly flow north of Hurricane Katrina in 2005, even though storm-relative helicity onshore was higher. As Hurricane Ivan approached the Alabama coast in 2004, a similar situation was occurring along the Florida Panhandle coast, with strong alongshore flow. However, in this case, a tornado touched down as a supercell moved onshore near Panama City Beach, FL, perhaps due to the local frictional effect of a large bay and lagoon to the north and condominiums to the south.