Evaluation of the Relationship between NSSL MRMS Rotational Tracks and Tornadoes in Iowa

Evaluation of azimuthal shear is a common technique used by National Weather Service meteorologists when interrogating the strength of mesocyclones in supercells where tornadoes or other types of severe weather are considered a threat. While shear-evaluating algorithms such as the tornado detection algorithm (TDA) and mesocyclone detection algorithm (MDA) are also available, typically meteorologists who are aggressively interrogating storms perform similar assessments manually rather than waiting for algorithm output. This can be a labor intensive process and is especially challenging when several storms need interrogation. Moreover, all three methods suffer from the inability to easily visualize spatial and temporal trends in the data, and they focus on single radar data unless forecasters can rapidly compare data from neighboring radar sites.

Since 2011, the NWS Davenport, Iowa office has been informally evaluating several National Severe Storms Laboratory (NSSL) Multi-Radar Multi-Sensor (MRMS) products during severe weather operations. The low-level rotational tracks product has generated interest because it provides a frequently updated (every two minutes) assessment of azimuthal shear. The product displays a temporal and spatial history of the shear magnitude while utilizing data from all radars observing the storm.

This study formally evaluates the utility of the rotation tracks output via comparison to observed tornadoes in Iowa between late 2007 and 2013. Rotational tracks output and tornado observations were combined in a GIS framework to assess their spatial and temporal relationships. In addition, quantitative shear values generated by the rotational tracks algorithm were compared to the occurrence and strength of tornadoes. Iowa was selected as the area of study since the climatology includes tornadoes generated by supercells, QLCSs, and non-supercell storms which permits evaluation of the algorithm in a variety of storm modes.