<p>This study aims to explain the temporal and spatial patterns of reedbed fires in conjunction with social dynamics, using Lake Eber as a case study. Within a mixed-methods design, semi-structured interviews with local residents (n = 15) and field observations were integrated with FIRMS (Fire Information for Resource Management System)–VIIRS (Visible Infrared Imaging Radiometer Suite)-based Fire Radiative Power (FRP; megawatts, MW) data and spatial statistics. FIRMS/VIIRS data for the period 2018–2025 (downloaded on 31 July 2025) were queried; however, as no active fires were detected for January–July 2025, all quantitative analyses were conducted using data from 2018 to 2024. Spatial clustering was evaluated using the Getis-Ord Gi* (hotspot analysis), global spatial autocorrelation was tested with Global Moran’s I, and IDW (inverse distance weighted interpolation) was applied for visualisation purposes. The FRP time series showed a pronounced increase in 2021 and, most notably, in 2024; in October, both the total and maximum FRP peaked at 607.81&#xa0;MW. While the Gi* results revealed significant clustering in March–April and November–December, Moran’s I did not identify significant global autocorrelation across the study area. These findings suggest that fire activity was characterised by seasonal and spatially distributed clusters rather than a single-centred pattern. Interviews with local residents indicate perceptions and experiences suggesting that fire spread is associated with easily accessible reed-dense areas and declining water levels. Overall, the findings suggest a potential need for an integrated wetland fire management approach that combines water management, land use, and local burning practices, adopts a cautious approach to controlled burning, and emphasises the integration of remote sensing with local knowledge.</p>