TanDEM-X data for aboveground biomass retrieval in a tropical peat swamp forest

Abstract

Forests play an important role in the global carbon cycle as a carbon sink. Deforestation and degradation of forests lead to carbon emissions, which should be prevented or minimized by protecting forests. Radar remote sensing has proven to be particularly useful to monitor forests especially in the tropics due to weather and daytime independence. Radar data from the TanDEM-X mission provide a potential opportunity to monitor large areas of tropical forests due to the multi-temporal global coverage and the high resolution.Tropical peat swamp forests are difficult to access and thus high effort to conduct field measurements is necessary. Therefore, aboveground biomass was estimated from a limited amount of in-situ measurements of relatively undisturbed peat swamp forest and a LiDAR based canopy height model to achieve a representative amount of biomass estimates for radar analysis. The LiDAR and field measurements resulted in an identical estimate of mean biomass and thus provided a reliable source to correlate with SAR (synthetic aperture radar) features from the TanDEM-X mission and ultimately up-scale the found relation to the entire study site.The relationship of interferometric coherence of the bistatic TanDEM-X data showed a moderate to high correlation with the biomass (R2=0.5) and RMSE of 53t/ha corresponding to a biomass range from 183 to 495t/ha. Thus, it could be used to indicate forest degradation areas, which are characterized by larger opening of the canopy cover and thus lower biomass. The results indicate that interferometric coherence is useful for quantification of aboveground biomass in tropical peat swamp forest. TanDEM-X coherence can at minimum serve as a stratification to assess spatial distribution of qualitative biomass classes in the context of REDD+ monitoring, reporting, and verification schemes and for the identification of forest degradation areas.