Experimental Ecosystem Eutrophication Causes Offsetting Effects on Emissions of CO2, CH4, and N2O from Agricultural Reservoirs.

Despite decades of research and management efforts, eutrophication remains a persistent threat to inland waters. As nutrient pollution intensifies in the coming decades, the implications for aquatic greenhouse gas (GHG) emissions are poorly defined, particularly the responses of individual GHGs: carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O). The biogeochemical controls of each gas can differ, making it difficult to predict the overall effect of nutrient pollution on the net radiative forcing of aquatic ecosystems. Here, we induced eutrophication of small nitrogen (N)-limited agricultural reservoirs and measured changes in diffusive GHG emissions within a before-after-control-impact (BACI) study design during June to September 2021. Each gas exhibited a unique response to 300% increases in primary production, with a shift from an overall CO2 source to a sink, a modest increase in N2O flux, and, unexpectedly, no significant change in CH4 emissions. The lack of net directional change in CO2-equivalent GHG emissions in fertilized reservoirs during the summer contrasts findings from empirical studies of eutrophic lakes. Our findings illustrate the difficulty in extrapolating among different sized ecosystems and suggest that forecast 2-fold increases in agricultural N fertilization by 2050 may not result in consistently elevated GHG emissions during summer, at least from small reservoirs in continental grassland regions.

Salinity causes widespread restriction of methane emissions from small inland waters.

Inland waters are one of the largest natural sources of methane (CH4), a potent greenhouse gas, but emissions models and estimates were developed for solute-poor ecosystems and may not apply to salt-rich inland waters. Here we combine field surveys and eddy covariance measurements to show that salinity constrains microbial CH4 cycling through complex mechanisms, restricting aquatic emissions from one of the largest global hardwater regions (the Canadian Prairies). Existing models overestimated CH4 emissions from ponds and wetlands by up to several orders of magnitude, with discrepancies linked to salinity. While not significant for rivers and larger lakes, salinity interacted with organic matter availability to shape CH4 patterns in small lentic habitats. We estimate that excluding salinity leads to overestimation of emissions from small Canadian Prairie waterbodies by at least 81% ( ~ 1 Tg yr-1 CO2 equivalent), a quantity comparable to other major national emissions sources. Our findings are consistent with patterns in other hardwater landscapes, likely leading to an overestimation of global lentic CH4 emissions. Widespread salinization of inland waters may impact CH4 cycling and should be considered in future projections of aquatic emissions.

CO2 dynamics in a small and old subtropical reservoir in East Asia: Environmental controls driving seasonal and spatial variability.

Inland waters have been increasingly viewed as hotspots for greenhouse gas (GHG) emissions owing to their strong capability to intercept and mineralize carbon from the terrestrial environment. Although small waterbodies in humid subtropical climates have the potential to emit considerable amounts of GHG, their emission patterns have remained understudied. This study involved intensive measurements of carbon dioxide (CO2) emissions from a small reservoir and its upstream and downstream reaches located in subtropical Hong Kong. Our results revealed that a variety of metabolic, hydrological, and hydrochemical processes play a critical role in regulating its CO2 dynamics. The reservoir was an overall source of CO2 to the atmosphere with an average areal flux of 24.6 mmol m-2 d-1, and it occasionally functioned as a sink for atmospheric CO2 under intense solar radiation when primary productivity was high. This flux is on the low side relative to that of global (sub)tropical reservoirs, which was likely attributable to the prolonged history of the reservoir (>150 years) and the occasional undersaturation of CO2 in the water column. We also noticed pronounced differences in the underlying controls of CO2 dynamics between the reservoir and its upstream and downstream reaches, emphasizing the importance of taking into account the distinct characteristics of both lentic and lotic waters when evaluating catchment-scale CO2 fluxes.