Lightning activity could influence regional air quality and terrestrial ecosystems through physicochemical processes such as lightning-generated nitrogen oxides and wildfire risks. However, the characteristics and causes of lightning activity across high latitudes such as the pan-Arctic region are still unclear. Here we explore spatiotemporal variations and climatic drivers of lightning stroke density (LSD) across the pan-Arctic (45° N –90° N) during 2010–2024, based on the latest global lightning datasets. Results show that LSD was up to 9.29 strokes 100 km−2 yr−1 over pan-Arctic regions during 2010–2024, with positive trends of 0.556 strokes 100 km−2 year−1 for land grids. Regionally, LSD shows significant positive linear trends in North America (NAM), central Europe (CEU) and western Russia (WRU), with highs up to 2.159 strokes 100 km−2 yr−1, 3.612 strokes 100 km−2 yr−1 and 2.071 strokes 100 km−2 yr−1, respectively. However, LSD decreases significantly by −2.042 strokes 100 km−2 yr−1 in northern China (NCH). The possible climatic causes of LSD trends are examined across the four regions through multiple linear regression and random forest models. Increasing sea surface temperature dominates positive LSD trends up to 1.125 strokes 100 km−2 yr−1 and 2.119 strokes 100 km−2 yr−1 over NAM and CEU, accounting for 52% and 59% of the observed trends. Increased surface air temperature and reduced aerosols jointly control the increasing LSD trend over WRU, up to 1.280 strokes 100 km−2 yr−1 and 1.112 strokes 100 km−2 yr−1, respectively. The negative LSD trend in NCH are mainly related to increases in local wind shear, resulting in a decline of −0.934 strokes 100 km−2yr−1 in LSD trends. Moreover, the use of random forest models as cross-validation also revealed similar climatic causes over these regions. Our findings highlight increasing trends in lightning activity over high-latitude regions in recent decades, suggesting the need to fully consider the impacts of climate change on future lightning activities under various emission scenarios.