First validation of the Lightning Imager aboard Meteosat Third Generation satellite with Earth Networks Total Lightning Network

• LI shows good spatial–temporal agreement with ENTLN in dense network regions. • LI achieves > 0.8 detection efficiency and < 0.3 FAR in dense network regions. • LI overestimates ENTLN flash detection at night and underestimates it during the day. • Logarithmic A–R regression enables peak current estimates from LI data. The Meteosat Third Generation Lightning Imager (MTG-LI), launched aboard the MTG-Imager-1 satellite, is the first geostationary optical lightning sensor to cover Europe and Africa while also covering parts of South America. It enables continuous monitoring of total lightning (both intra-cloud (IC) and cloud-to-ground (CG)) with high temporal and spatial resolution. This study presents the first validation of LI observations over South America and adjacent regions during October 2024, using data from the ground-based Earth Networks Total Lightning Network (ENTLN). Three objectives are addressed: (i) evaluation of LI’s detection capabilities across varying spatiotemporal thresholds; (ii) assessment of the temporal accuracy of LI; and (iii) investigation of the empirical relationship between ENTLN peak current (A) and LI optical radiance (R). Our results show that in areas with dense ground network coverage, detection efficiency exceeds 0.8 and FAR is below 0.3 using moderate thresholds (±2 s, 1° x 1°). Relatively high FAR values, particularly at night, are in-part due to LI detecting lightning that ENTLN misses in areas where its performance is reduced. In addition, a regression model of the A–R relationship is proposed enabling the peak current to be estimated based solely on satellite data, which is advantageous in areas with insufficient ground-based observations. The results show that this relationship is more suitable in logarithmic transformation, with positive discharges exhibiting twice the regression slope than negative discharges.