Echo Size and Asymmetry: Impact on NEXRAD Storm Identification

Dennis E. Buechler Universities Space Research Association, Marshall Flight Center Huntsville, Alabama

Search for other papers by Dennis E. Buechler in
Current site
Google Scholar
PubMed
Close
and
Steven J. Goodman NASA, Marshall Space Flight Center, Huntsville, Alabama

Search for other papers by Steven J. Goodman in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

The effects of echo shape and radar viewing angle on detecting small thunderstorms with the NEXRAD storm identification algorithms are examined. The amorphous low level echo shapes are modeled as ellipses with major axes ranging from 5–15 km and minor axes varying between 2–5 km. The model echoes are then used to create a “probability of detection” chart that demonstrates the impact of storm asymmetry on cell identification. Moreover, we examine the algorithm performance on small thunderstorms observed near Huntsville, Alabama and Kennedy Space Center, Florida. The two thunderstorms observed near Huntsville also produced microbursts. The probability of storm detection using the NEXRAD default values for both Huntsville cases is less than 0,5 at the time of the first lightning discharge and less than 0.4 at microbursts onset. The Kennedy Space Center storms were already electrically active when the probability of detection was 0.5 or less. A new algorithm based on the analysis of 15 storms observed in Florida, Alabama and New Mexico is proposed that would identify storms as having Lightning if 40 dBZ reflectivity is present at the −10°C level and the echo top exceeds 9 km. This algorithm would have a 100% probability of detecting lightning producing storms 4–33 min before the first flash, a 7% false alarm rate and a critical success index of 93%.

Abstract

The effects of echo shape and radar viewing angle on detecting small thunderstorms with the NEXRAD storm identification algorithms are examined. The amorphous low level echo shapes are modeled as ellipses with major axes ranging from 5–15 km and minor axes varying between 2–5 km. The model echoes are then used to create a “probability of detection” chart that demonstrates the impact of storm asymmetry on cell identification. Moreover, we examine the algorithm performance on small thunderstorms observed near Huntsville, Alabama and Kennedy Space Center, Florida. The two thunderstorms observed near Huntsville also produced microbursts. The probability of storm detection using the NEXRAD default values for both Huntsville cases is less than 0,5 at the time of the first lightning discharge and less than 0.4 at microbursts onset. The Kennedy Space Center storms were already electrically active when the probability of detection was 0.5 or less. A new algorithm based on the analysis of 15 storms observed in Florida, Alabama and New Mexico is proposed that would identify storms as having Lightning if 40 dBZ reflectivity is present at the −10°C level and the echo top exceeds 9 km. This algorithm would have a 100% probability of detecting lightning producing storms 4–33 min before the first flash, a 7% false alarm rate and a critical success index of 93%.

Save