Site-specific scaling of remote sensing-based estimates of woody cover and aboveground biomass for mapping long-term tropical dry forest degradation status

Fremout, Tobias; Cobián-De Vinatea, Jorge; Thomas, Evert; Huaman-Zambrano, Wilson; Salazar-Villegas, Mike; Limache-de la Fuente, Daniela; Bernardino, Paulo N.; Atkinson, Rachel; Csaplovics, Elmar; Muys, Bart


Remote sensing-based approaches are important for evaluating ecosystem degradation and the efficient planning of ecosystem restoration efforts. However, the large majority of remote sensing-based degradation assessments are trend-based, implying that they can only detect degradation that occurred after medium or high-resolution satellite imagery became available. This makes them less suitable to map long-term degradation in ecosystems that have been under high human pressure since before. The main goal of this study was to develop a robust operational approach to map forest degradation status in heterogeneous landscapes with a long-standing degradation history to inform the planning of restoration interventions. We hereby use the tropical dry forests of Lambayeque, Peru, as a case study. Instead of using a trend-based assessment, we evaluated forest degradation status by comparing current woody cover (WC) and aboveground biomass (AGB) estimates obtained from remote sensing imagery with benchmark values consisting of the 95th percentile WC and AGB values inside environmentally homogenous land capability classes. Using boosted regression tree models and a combination of optical (Sentinel-2) and synthetic aperture radar (Sentinel-1) data of different seasons, we mapped WC and AGB, using training data obtained through very high-resolution imagery and field measurements. Further, we aimed at assessing (i) whether the inclusion of Sentinel-1 data improves mapping accuracy in comparison to using only Sentinel-2 data, and (ii) whether the use of multi-seasonal data improves accuracy in comparison to single-season data. Models combining multi-seasonal Sentinel-1 and Sentinel-2 data resulted in the most accurate WC predictions (mean absolute error (MAE): 16%; MAE normalized by dividing by the inter-quartile range of training data: 26%) and AGB predictions (MAE: 28.6 t/ha; normalized MAE: 65%), but differences in predictive accuracy with single season models or models using only Sentinel-2 data were small. The most accurate models estimated an average WC of 41% and an average AGB of 23.4 t/ha. Average WC and AGB reduction due to degradation was 35% and 36%, respectively, indicating that these forests are highly degraded. The site-specific scaling of WC and AGB allows to efficiently estimate forest degradation status irrespective of the time when this degradation occurred, and to express degradation status against site-specific benchmarks. On the condition that there are still some areas that are sufficiently undegraded to be used as a benchmark, the approach can be used to prioritize forest restoration actions and inform targets for restoration in heterogeneous landscapes suffering the impacts of undocumented long-term degradation.