Interannual Variability of Tropical and Subtropical Lightning Production from TRMM LIS: Locations, Magnitudes, and Mechanisms

The amount of lightning activity at a given location can vary significantly with the hour, day, month, and season given the nature of lightning and the convective mechanisms at that location. What is harder to explain is the variability of lightning production at the inter-annual time scale, where normal cycles of convective mechanisms can no longer describe the differences in the amount of observed lightning. This study uses a climatological lightning record from the space-based Tropical Rainfall Measuring Mission Lightning Imaging Sensor (TRMM, LIS) to compare individual years of lightning activity for all locations in the tropics and subtropics.

As a first guess from previous literature, the El-Niño/Southern Oscillation (ENSO) was chosen as the most likely cause of inter-annual variability within the TRMM LIS domain. Individual years from the lightning record beginning in 1998 and ending in 2014 were thus classified based on their corresponding phase of ENSO, i.e. warm, cold, or neutral, using the Oceanic Niño Index (ONI) for ENSO phase. Doing so identified several locations where inter-annual variability existed during that period and was seemingly explained by ENSO phase. Other locations of interest were identified that had significant variability with no apparent correlation to ENSO phase, where no variation from year to year was observed at all, and where one or two anomaly years split from an otherwise repeated annual pattern. These locations were treated as ENSO-null cases. The general patterns are described in the study, including areas previously identified with ENSO sensitivities such as Brazil and Argentina, and potentially new locations such as Central and Southeastern Africa.

To further investigate what mechanisms were causing the lightning variance, some of the above regions were chosen for a ‘deep dive’ analysis. Ultimately 6 ENSO sensitive and 6 ENSO-null locations were selected based on geographic position, ENSO phase sensitivity, total annual lightning activity, the magnitude of the signal, and the geographic size of the pattern. The annual lightning patterns were compared to NCEP/NCAR reanalysis data and rainfall observations from TRMM to determine what may have led to the changes in lightning activity. In other words, are there any changes to the environment during a particular phase or year that may explain the observed variation in the lightning production at that location. If there is no variability in the lightning record, then the reanalysis data are used to determine which mechanisms control the activity at that location and what keeps it so consistent. In conjunction with conditional flash rates, these analyses are used to determine if the changes in lightning production are attributable to changes in the frequency of thunderstorms, the intensity of the thunderstorms, or a change in the frequency of transient synoptic and tropical weather systems.