These data were collected from different treatments. Once we added the larvae and Bti, we gas-tight sealed all microcosms, and the experiment started. We vigorously shook the three additional control microcosms to ensure equilibration of gas between pore water, surface water, and headspace, and gas samples were collected for determining the initial amount of CO2 and CH4. At 24 h, 72 h, and 120 h after the start of the experiment, dissolved O2 concentration, and the CO2 and CH4 mixing ratios in the headspace were measured to determine O2 consumption and CO2 and CH4 emission rates of the sediment. We ended the experiment after 120 h when the first adult was observed. The dissolved O2 concentration in the overlying water was 81 ± 8 % saturation at the end of the experiment. After sampling the headspace at 120 h, we vigorously shook the microcosms to ensure full equilibration between porewater, surface water, and headspace, and collected gas samples from the headspace to determine the total net production of CO2 and CH4 during the experimental period, which includes gas that has accumulated in the pore water. We collected 100 µL of headspace gas from each microcosm at 24 h, 72 h, and 120 h after the start of the experiment using a gastight syringe (Hamilton, USA). The mixing ratios (ppmv) of CO2 and CH4 were measured by injecting the samples into a gas analyzer (Ultra-portable Greenhouse Gas Analyzer; UGGA, Los Gatos Research Inc., Mountain View, CA, USA) in closed-loop operation (Wilkinson et al. 2018). By assuming full equilibration between the headspace and the overlying water, we determined the amount of CH4 and CO2 (µmol) in the headspace and overlying water at each sampling time (gas-specific Henry coefficients at incubation temperature were estimated following (International Hydropower Association 2010)). We then calculated the emissions rates of CH4 and CO2 as the difference in mass between two subsequent samplings divided by the elapsed time. The emission rates include fluxes across the sediment-water interface and potential oxidation of CH4 and CO2 production by respiration in the surface water.After each headspace sampling, we measured dissolved O2 concentration in the overlying water. We calculated the amount of O2, assuming equilibrium between the water and the headspace, as the sum of O2 gas in the headspace and dissolved O2 in the overlying water (µmol). We calculated O2 consumption rates (i.e. respiration) from the difference between two subsequent samplings divided by the elapsed time.We estimated the total net production rate of CH4 and CO2, including gas that has accumulated in the sediment porewater and bubbles during the five-day experiment, from the difference in the amounts estimated from measurements after shaking of the additional control microcosms at the beginning of the experiment, and those estimated after the final shaking of each experimental microcosm at the end of the experiment. The net CH4 and CO2 production rates include gas that was produced in the sediment but not emitted to the water and headspace during the incubation period.