Seasonal Predictability of Lightning Over the Global Hotspot Regions

Skillful seasonal prediction of lightning is crucial over several global hotspot regions, as it causes severe damages to infrastructures and losses of human life. While major emphasis has been given for predicting rainfall, prediction of lightning in one season advance remained uncommon, owing to the nature of problem, which is short‐lived local phenomenon. Here we show that on seasonal time scale, lightning over the major global hotspot regions is strongly tied with slowly varying global predictors (e.g., El Niño and Southern Oscillation). Moreover, the sub‐seasonal variance of lightning is highly correlated with global predictors, suggesting a seminal role played by the global climate modes in shaping the local land‐atmosphere interactions, which eventually affects seasonal lightning variability. It is shown that seasonal predictability of lightning over the hotspot is comparable to that of seasonal rainfall, opens up an avenue for reliable seasonal forecasting of lightning for special awareness and preventive measures. Plain Language Summary Lightning, atmospheric hazards have an impact on the loss of human life, forest fire, health, agriculture, and economy across the globe. However, due to its chaotic nature, the tendency of seasonal forecasting of lightning is considered not viable as the understanding of the predictability of lightning is still incomplete. Here, we have explored the possibility of seasonal forecasting of lightning activity and provided a scientific basis as the lightning flashes are found to be tied with slowly varying remote forcings (e.g., El Niño and Southern Oscillation, or other global predictors). Correlation of flash count with different indices (Nino, Pacific decadal oscillation, North Atlantic oscillation, and Extra tropics, etc.) demonstrate the potential of seasonal forecasting of lightning. The multiple regression analysis enhances the skill. The climatology of lightning flash density from the Goddard Earth observing system model is compared with observation. The pattern correlation between observation and model is very high (∼0.7) over global tropics hints at the predictability of lightning flashes on a seasonal time scale. Therefore, better climate models that capture crucial couplings between ocean, atmosphere, and land processes could make skillful predictions of lightning and opens up a possibility for lightning forecast in one season advance. Key Points Possibility of reliable seasonal forecasting of lightning over