Greenhouse gas. Remote Sensing. Floodplain. Seasonality

Studies indicate that the floodplain is a significant natural source of methane to the atmosphere, however, there are still gaps regarding seasonality, spatiality and contribution of different types of methane transport mechanisms. The objective of this work is to characterize the biophysical parameters of a fragment of floodplain forest and estimate the flux of methane on the surface of this forest, as well as evaluate the seasonality and different mechanisms of methane transport from the surface to the atmosphere. The characterization of the forest's biophysical parameters was carried out through processing of Sentinel-2 multispectral bands and LiDAR point cloud. It was found that the forest was sensitive to the phases of the hydrological cycle, height heterogeneity in the vertical profile and that 95% of the variability in forest biomass could be explained through structural variables derived from LiDAR. Methane concentration measurements on the surface were carried out using the chamber method, coupled to an ultra-portable greenhouse gas analyzer. It was found that the floodplain forest fragment showed a significant difference in flux intensity between the flooded phase and the non-flooded phase. In the non- flooded phase, the forest floor was a sink and source of methane. During the flooded phase, the surface of the forest was a source of methane. The months with the greatest intensity of boiling flux were February and August, corresponding to the flood and low water phases, respectively. The Nemenyi test (p < 0.05) did not indicate a difference in the diffusive flux of methane between the flood x ebb hydrological periods and indicated a difference in the boiling flux between the dry hydrological period and the other periods. It was found that the boiling flux corresponded to around 70% of the total methane emission and the diffusive flux corresponded to around 30%. This work showed that the floodplain forest responded physiologically to the variation in the waters of the Amazon River, and that hydrological periods significantly influenced the emissive behavior of methane from the forest surface. These results point to the importance of further studies of methane dynamics in different types of floodplain forests and to the importance of integrating gas flow results with ecosystem information, signaling that ecosystem methods, such as tower methods, can be more suitable for estimating methane flux in these flooded forests.