Suchergebnisse

Background: Recent studies have shown that fragmentation is an increasing threat to global forests, which has major impacts on biodiversity and the important ecosystem services provided by forested landscapes. Several tools have been developed to evaluate global patterns of fragmentation, which have potential applications for REDD+. We study how canopy height and above ground biomass (AGB) change across several categories of forest edges determined by fragmentation analysis. We use Democratic Republic of Congo (DRC) as an example. Results: An analysis of variance of different edge widths and airborne estimated canopy height found that canopy heights were significantly different in forest edges at a distance of 100 m from the nonforest edge. Biomass was significantly different between fragmentation classes at an edge distance of 300 m. Core forest types were found to have significantly higher canopy height and greater AGB than forest edges and patches, where height and biomass decrease significantly as the level of fragmentation increases. A change analysis shows that deforestation and degradation are increasing over time and biomass loss associated with degradation account for at least one quarter of total loss. We estimate that about 80 % of primary forests are intact, which decreases 3.5 % over the 15 year study period, as primary forest is either deforested or transitioned to forest edge. While the carbon loss per hectare is lower than that of deforestation, degradation potentially affects up to three times more area than deforestation alone. Conclusions: When defining forest degradation by decreased biomass without any loss in forest area, assessing transitions of core forest to edges over time can contribute an important element to REDD+MRV systems. The estimation of changes between different forest fragmentation types and their associated biomass loss can provide an estimate of degradation carbon emission factors. Forest degradation and emissions due to fragmentation are often underestimated and should ...

Forest degradation, generally defined as a reduction in the delivery of forest ecosystem services, can have long-term impacts on biodiversity, climate, and local livelihoods. The quantification of forest degradation, its dynamics and proximate causes can help prompt early action to mitigate carbon emissions and inform relevant land use policies. The Democratic Republic of the Congo is largely forested with a relatively low deforestation rate, but anthropogenic degradation has been increasing in recent years. We assess the impact of eight independent variables related to land cover, land use, infrastructure, armed conflicts, and accessibility on forest degradation, measured by the Forest Condition (FC) index, a measure of forest degradation based on biomass history and fragmentation that ranges from 0 (completely deforested) to 100 (intact). We employ spatial panel models with fixed effects using regular 25 × 25 km units over five 3-year intervals from 2002 to 2016. The regression results suggest that the presence of swamp ecosystems, low access (defined by high travel time), and forest concessions are associated with lower forest degradation, while built up area, fire frequency, armed conflicts result in greater forest degradation. The impact of neighboring units on FC shows that all variables within the 50 km spatial neighborhood have a greater effect on FC than the on-site spatial determinants, indicating the greater influence of drivers beyond the 25 km2 unit. In the case of protected areas, we unexpectedly find that protection in neighboring locations leads to higher forest degradation, suggesting a potential leakage effect, while protected areas in the local vicinity have a positive influence on FC. The Mann-Kendall trend statistic of occurrences of fires and conflicts over the time period and until 2020 show that significant increases in conflicts and fires are spatially divergent. Overall, our results highlight how assessing the proximate causes of forest degradation with spatiotemporal analysis can support targeted interventions and policies to reduce forest degradation but spillover effects of proximal drivers in neighboring areas need to be considered. ; Peer Reviewed

Forest degradation, generally defined as a reduction in the delivery of forest ecosystem services, can have long-term impacts on biodiversity, climate, and local livelihoods. The quantification of forest degradation, its dynamics and proximate causes can help prompt early action to mitigate carbon emissions and inform relevant land use policies. The Democratic Republic of the Congo is largely forested with a relatively low deforestation rate, but anthropogenic degradation has been increasing in recent years. We assess the impact of eight independent variables related to land cover, land use, infrastructure, armed conflicts, and accessibility on forest degradation, measured by the Forest Condition (FC) index, a measure of forest degradation based on biomass history and fragmentation that ranges from 0 (completely deforested) to 100 (intact). We employ spatial panel models with fixed effects using regular 25 × 25 km units over five 3-year intervals from 2002 to 2016. The regression results suggest that the presence of swamp ecosystems, low access (defined by high travel time), and forest concessions are associated with lower forest degradation, while built up area, fire frequency, armed conflicts result in greater forest degradation. The impact of neighboring units on FC shows that all variables within the 50 km spatial neighborhood have a greater effect on FC than the on-site spatial determinants, indicating the greater influence of drivers beyond the 25 km² unit. In the case of protected areas, we unexpectedly find that protection in neighboring locations leads to higher forest degradation, suggesting a potential leakage effect, while protected areas in the local vicinity have a positive influence on FC. The Mann-Kendall trend statistic of occurrences of fires and conflicts over the time period and until 2020 show that significant increases in conflicts and fires are spatially divergent. Overall, our results highlight how assessing the proximate causes of forest degradation with spatiotemporal analysis can support targeted interventions and policies to reduce forest degradation but spillover effects of proximal drivers in neighboring areas need to be considered.

The forests of Central Africa contain some of Earth's few remaining intact forests. These forests are increasingly threatened by infrastructure development, agriculture, and unsustainable extraction of natural resources (e.g. minerals, bushmeat, and timber), all of which is leading to deforestation and forest degradation, particularly defaunation, and hence causing declines in biodiversity and a significant increase in carbon emissions. Given the pervasive nature of these threats, the global importance of Central African forests for biodiversity conservation, and the limited resources for conservation and sustainable management, there is a need to identify where the most important areas are to orientate conservation efforts. We developed a novel approach for identifying spatial priorities where conservation efforts can maximize biodiversity benefits within Central Africa's most intact forest areas. We found that the Democratic Republic of Congo has the largest amount of priority areas in the region, containing more than half, followed by Gabon, the Republic of Congo and Cameroon. We compared our approach to one that solely prioritizes forest intactness and one that aims to achieve only biodiversity representation objectives. We found that when priorities are only based on forest intactness (without considering biodiversity representation), there are significantly fewer biodiversity benefits and vice versa. We therefore recommend multi-objective planning that includes biodiversity representation and forest intactness to ensure that both objectives are maximized. These results can inform various types of conservation strategies needed within the region, including land-use planning, jurisdictional REDD + initiatives, and performance related carbon payments, protected area expansion, community forest management, and forest concession plans.

National forest inventories in tropical regions are sparse and have large uncertainty in capturing the physiographical variations of forest carbon across landscapes. Here, we produce for the first time the spatial patterns of carbon stored in forests of Democratic Republic of Congo (DRC) by using airborne LiDAR inventory of more than 432,000 ha of forests based on a designed probability sampling methodology. The LiDAR mean top canopy height measurements were trained to develop an unbiased carbon estimator by using 92 1-ha ground plots distributed across key forest types in DRC. LiDAR samples provided estimates of mean and uncertainty of aboveground carbon density at provincial scales and were combined with optical and radar satellite imagery in a machine learning algorithm to map forest height and carbon density over the entire country. By using the forest definition of DRC, we found a total of 23.3 ± 1.6 GtC carbon with a mean carbon density of 140 ± 9 MgC ha-1 in the aboveground and belowground live trees. The probability based LiDAR samples capture variations of structure and carbon across edaphic and climate conditions, and provide an alternative approach to national ground inventory for efficient and precise assessment of forest carbon resources for emission reduction (ER) programs. ; SCOPUS: ar.j ; info:eu-repo/semantics/published