Identifying dominant environmental predictors of freshwater wetland methane fluxes across diurnal to seasonal time scales

While wetlands are the largest natural source of methane (CH4) to atmosphere, they represent a large uncertainty in global CH4 budget due complex biogeochemical controls on dynamics. Here we present, our knowledge, first multi-site synthesis how predictors fluxes (FCH4) freshwater vary across wetland types at diel, multiday (synoptic), and seasonal time scales. We used several statistical approaches (correlation analysis, generalized additive modeling, mutual information, random forests) wavelet-based multi-resolution framework assess importance environmental predictors, nonlinearities lags FCH4 23 eddy covariance sites. Seasonally, soil air temperature were dominant sites with smaller variation water table depth (WTD). In contrast, WTD was predictor for variations (e.g., tropical/subtropical wetlands). Changes lagged fluctuations by ~17 ± 11 days, median values 8 16 5 15 respectively. Temperature also scale. Atmospheric pressure (PA) another important scale peat-dominated sites, drops PA coinciding synchronous releases CH4. At diel scale, relationships latent heat flux vapor deficit suggest that physical processes controlling evaporation boundary layer mixing exert similar volatilization, influence pressurized ventilation aerenchymatous vegetation. addition, 1- 4-h ecosystem photosynthesis indicate recent carbon substrates, such as root exudates, may control FCH4. By addressing issues asynchrony, nonlinearity, this work improves understanding timing can inform future studies models, help constrain emissions.