<p>Forests are important regulators of carbon dioxide fluxes, whereas overall greenhouse gas (GHG) budgets, in particular, nitrous oxide (N<sub>2</sub>O), are still largely unknown. No studies on ecosystem-level N<sub>2</sub>O budgets (soil and tree stem fluxes with eddy covariance (EC) measurements above the canopy) are found. Only a few examples are available on N<sub>2</sub>O emissions from tree stems. Nevertheless, estimation of the N<sub>2</sub>O and the full GHG balance in different forest ecosystems under various environmental conditions is essential to understand their impact on climate.</p><p>During the period of August 2017 to December 2019, we measured the N<sub>2</sub>O budget of a 40-yr old hemiboreal grey alder (Alnus incana) forest stand on former agricultural land in Estonia considering fluxes from the soil, tree stems and whole ecosystem. Grey alder (Alnus incana) is a fast-growing tree species typically found in riparian zones, with great potential for short-rotation forestry. Their symbiotic dinitrogen (N<sub>2</sub>) fixation ability makes alders important for the regulation of nitrogen (N) cycle in forested areas.</p><p>We measured the N<sub>2</sub>O budget considering fluxes from the soil surface (12 automated chambers; Picarro 2508), tree stems (60 manual sampling campaigns from 12 model trees with chambers at 0.1, 0.8 and 1.7 m; gas chromatographic analysis in lab) and whole ecosystem (EC technique: Aerodyne TILDAS). Simultaneously, soil water level, temperature and moisture were measured automatically, and composite soil samples were taken for physico-chemical analysis. Potential N<sub>2</sub> flux in intact soil cores was measured in the lab using the He-O incubation method.</p><p>Average N<sub>2</sub>O fluxes from the soil and tree stems varied from 1.2 to 3.0 and 0.01 to 0.03 kg N<sub>2</sub>O-N ha<sup>–1</sup> yr<sup>–1</sup>, respectively, being the highest during the wet periods, peaking during the freezing-thawing, and being the lowest in dry periods. The average annual potential N<sub>2</sub> flux in the soil was 140 kg N<sub>2</sub> ha<sup>–1</sup> yr<sup>–1</sup> which made the average N<sub>2</sub>:N<sub>2</sub>O-N ratio in the soil about 60. According to the EC measurements, the forest was a net annual source of N<sub>2</sub>O (3.4 kg N<sub>2</sub>O ha<sup>–1</sup>). Thus, the main gaseous nitrogen flux in this forest was N<sub>2</sub> emission. Our carbon (C) budget showed that the forest was a significant net annual C sink.</p><p>Results of our long-term study underline the high N and C buffering capacity of riparian alder forests. For better understanding of C and nutrient budgets of riparian forests, we need long-term, high-frequency measurements of N<sub>2</sub>O fluxes from the soil and tree stems in combination with ecosystem-level EC measurements. The identification of microorganisms and biogeochemical pathways associated with N<sub>2</sub>O production and consumption is another future challenge.</p>

<p>Peatlands are an enormous sink of carbon and nitrogen. Natural and human disturbances may release them as greenhouse gases (GHGs) or water pollutants. Tropical peatlands have especially intensive matter cycling. Amazonia holds almost a half of tropical peatlands. Most of it is inaccessible to current forestry and drainage machinery and thus untouched by man. Tropical rainforest has been labelled ’lungs of the Earth’. While photosynthesis in mature forests does sequester carbon in biomass, they respire an equal amount of carbon dioxide (CO<sub>2</sub>). Only swamp forests may sequester carbon in wet anoxic peat for centuries. However, anoxic decomposition of peat yields methane (CH<sub>4</sub>) and suboxic processes release nitrous oxide (N<sub>2</sub>O). Both have high global warming potential. In undisturbed peatlands, carbon sequestration outweighs GHG emissions. GHG budgets are more complicated in disturbed peatlands.</p><p>With an objective to clarify the greenhouse gas budget of tropical peatlands, the Department of Geography, University of Tartu held a measurement campaign in Iquitos, Peruvian Amazon in September 2019. We observed fluxes of the three GHGs using opaque chambers and measured potential environmental factors in three sites under various disturbance histories: 1) a Mauritia flexuosa palm-dominated swamp forest, 2) toe-slope swamp forest grown in 12 years on fallow pasture and banana plantation, and 3) slash-and-burn cassava field.</p><p>The toe-slope swamp respired the largest amounts of CO<sub>2</sub> while site differences were small and may have been offset by photosynthesis (which we did not measure). The wet swamp forest sites, especially palm trunks, emitted large amounts of CH<sub>4</sub>. The dry slash-and-burn cassava field emitted little methane. The CH<sub>4 </sub>emissions were strongly correlated with nitrogen content of the peat. Previous literature links high soil nitrogen content with lability of soil organic carbon and high microbial activity. The swamp forest floor emitted an average of 390 µg N<sub>2</sub>O-N m<sup>–2</sup> h<sup>–1</sup> after torrential rainfall. The downpour may have carried just enough oxygen into the peat to trigger N<sub>2</sub>O production by nitrification or hamper the full pathway of denitrification to N<sub>2</sub>. High peat Ca<sup>++</sup> and Mg<sup>++</sup> content and pH>4 favoured nitrification. High NH<sub>4</sub><sup>+</sup>-N concentration in the swamp peat (190 mg kg<sup>–1</sup>), which can be related to N<sub>2</sub> fixation and litter from three species of leguminous trees, formed a solid base for nitrification. The slash-and-burn cassava field emitted a sizable 37 µg N<sub>2</sub>O-N m<sup>–2</sup> h<sup>–1</sup>. In conclusion, the variety of disturbances produced an interesting pattern of GHG emissions in relationship with environmental conditions. Thus, Amazonian peatlands demand elevated attention.</p>