Fencing farm ponds to limit livestock access reduces nitrous oxide and methane emissions.

Farm ponds are globally abundant features in agricultural landscapes and are increasingly recognised as important sources of greenhouse gases. However, greenhouse gas assessments of these systems have largely focused on methane (CH4), while nitrous oxide (N2O) emissions remain poorly understood. Furthermore, it remains unclear whether management interventions that reduce CH4 emissions also reduce N2O emissions. Here, we quantified carbon dioxide (CO2), CH4, and N2O fluxes from 54 fenced and unfenced farm ponds across 495 km in southeastern Australia, amounting to 16,773 hourly greenhouse gas measurements, to assess the effects of livestock exclusion on multi-gas emissions. Fenced ponds exhibited 44% lower N2O and 52% lower CH4 emissions than unfenced ponds, resulting in 44% lower total CO2-eq emissions. In contrast, CO2 fluxes did not differ between treatments. Reductions in CH4 and N2O emissions were not consistently coupled across sites, with only 52% of sites showing concurrent reductions in both gases. Measured water-quality variables did not strongly explain greenhouse gas variability, suggesting that standard water-quality metrics alone may be insufficient to predict mitigation outcomes. Despite contributing little to total emissions in mass units, N2O accounted for more than three-quarters (76.7%) of total CO2-eq emissions owing to its high global warming potential. Together, our findings show that livestock exclusion can reduce CH4 and N2O emissions from farm ponds, but mitigation outcomes are strongly context-dependent and cannot be inferred from single-gas responses alone.