Operational demands and the natural inflow of water actively drive biweekly fluctuations in water levels in hydropower reservoirs. These daily to weekly fluctuations could have major effects on methane (CH4) and carbon dioxide (CO2) emissions via release of bubbles from reservoir bottom sediments (ebullition) or organic matter inputs, respectively. The impact of transient fluctuations in water levels on GHG emissions is poorly understood and particularly so in tropical run-of-the-river reservoirs. These reservoirs, characterized by high temperatures and availability of labile organic matter, are usually associated with extensive CH4 generation within bottom sediments. The aim of this study is to determine how water level fluctuations resulting from the operation of the Belo Monte hydropower plant on the Xingu River, eastern Amazon River Basin, affect local CO2 and CH4 emissions. Between February and December 2022, we monitored weekly fluxes and water concentrations of CO2 and CH4 in a site on the margin of the Xingu reservoir. Throughout the study period, fluxes of CO2 and CH4 were 118 +/- 137 and 3.62 +/- 8.47 mmol m- 2 d-1 (average +/- 1SD) while concentrations were 59 +/- 29.77 and 0.30 +/- 0.12 mu M, respectively. The fluxes and water concentrations of CO2 were clearly correlated with the upstream discharge, and the variation observed was more closely associated with a seasonal pattern than with biweekly fluctuations in water level. However, CH4 fluxes were significantly correlated with biweekly water level fluctuations. The variations observed in CH4 fluxes occurred especially during the high-water season (February-April), when biweekly water level fluctuations were frequent and had higher amplitude, which increased CH4 ebullition. Reducing water level fluctuations during the high-water season could decrease ebullitive pulses and, consequently, total flux of CH4 (TFCH4) in the reservoir margins. This study underscores the critical role of water level fluctuations in near-shore CH4 emissions within tropical reservoirs and highlights significant temporal variability. However, additional research is necessary to understand how these findings can be applied across different spatial scales.