A study of climate projections of lightning frequency over Italy

Continuous high-resolution lightning observations from geostationary satellites, together with ground-base networks, provide today a detailed description  of lightning activity across the globe. At the same time, such information cannot be directly transferred to global meteorological or climate models (GCMs), due to their limited spatial resolution and simplified physical processes, leading to the need for parameterizations and scaling methods. This study presents a novel trend-based scaling approach that uses the high resolution of modern detectors to improve the performances of GCMs in simulating  lightning activity trends through  the end of the century. The scaling method identifies  atmospheric parameters that best reproduce current lightning activity, which are then combined with coarser GCM trends to project future lightning changes.

We focus on  Italy as a case study,  the past 14 years of lightning observations from the LINET network to calibrate  lightning predictors from  the ERA5 reanalysis.  We found  that the best predictors consist of combinations of convective available potential energy, temperature gradients, relative humidity, geopotential height, wind velocity and shear, and freezing level. The model demonstrates a high degree of accuracy in capturing both the spatial distribution and temporal variability of contemporary lightning activity. Using CMIP6 projections, we then applied trend scaling based on nine future climate scenarios to estimate the  future evolution of lightning activity over different time spans.

Our results show that trend-based scaling significantly improves the projection of lightning flash rates in terms of spatial distribution and intensity compared to traditional parameterizations. This work provides a practical framework for integrating lightning projections into climate impact studies, enhancing the reliability of lightning future changes under various climate scenarios. Moreover, the possibility of applying  our model to reanalysis datasets of any resolution makes this approach  a versatile  tool  for assessing lightning-related risks in a warming world.