Operational Applications of Historical Lightning Data Patterns to Predict Convection in Phoenix, Arizona

While most of Arizona is classified as having a desert climate, during the summer months frequent and intense rainfall occurs in connection with the North American Monsoon. It is well established that convective patterns in the Phoenix, AZ metropolitan area follow atypical diurnal patterns compared to most other locations on Earth. During the day, skies are often cloud free as weak subsidence/a capping inversion suppresses convection. Meanwhile, the areas within eyesight to the north, east, and south, all of significantly higher elevation, typically see showers and thunderstorms develop between 17 UTC and 22 UTC. As the afternoon progresses to evening, density and gravity driven outflow airmasses rush toward the Phoenix area. If the capping inversion is weak enough, the mechanical lift produced by these gust fronts can prompt continued or new convective development. In some cases, gust fronts from different directions and areas converge over Phoenix itself, exciting rapid convective development. This unusual setup results in a peak convective period for the Phoenix area of 23 UTC to 05 UTC.

This talk will utilize lightning data from Vaisala’s National Lightning Detection Network and Earth Network’s Lightning Detection Network to demonstrate the previously described climatological patterns in higher details than previously completed. Additional analysis will apply statistical methods to assess what predictive patterns may exist in the lightning data to determine convective activity over the Phoenix area itself and discuss operational implications. Presentation will showcase research methods and data visualization techniques in a NOAA/NWS Weather Forecast Office utilizing mainstream python libraries (matplotlib, numpy, pandas) and the process of operationalizing findings.