Proximate Causes of Forest Degradation in the Democratic Republic of the Congo Vary in Space and Time

Abstract

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.

Are deforestation and degradation in the Congo Basin on the rise? An analysis of recent trends and associated direct drivers.

“…Small-scale agricultural activities are generally spatially limited around inhabited areas due to practical reasons -minimal travel time to fields is more efficient, and secondary forests and fallow areas are preferred to primary forests, meaning that the agricultural expansion is not endless. Political instability, conflicts and insecurity have long driven migration patterns in the region, which could force people away from forests 50,82,83 . In the Central African Republic, civil war, combined with a lack of infrastructure, low population density has driven many people from rural areas to cities, and limited the expansion of industrial agriculture which could potentially deforest large areas once stability returns 84,85 .…”

“…The central premise of our approach is the identification of multiple overlapping drivers, which is representative of DD occurring at national and sub-national scales 50,56,82,113 and the result of the actions of multiple actors, multiple processes and motivations 33,104,114 . Global assessments, or postdisturbance land use cannot adequately discern multiple drivers 56,104 and are not relevant for decision making which requires national context 41 .…”

“…Most studies identify expanding small-scale agriculture as the primary cause of forest loss 41,47,48 while others cite forestry activities as having the greatest impact on forest areas 2 . It is important to comprehensively assess drivers because not all will have the same impact on carbon stocks, biodiversity 49,50 or communities 51,52 .…”

Economic benefits of reduced deforestation are offset by excess deforestation – a counterfactual simulation

“…Knowing which deforestation rates would be observed under expected, often profit oriented human land-use allocation 7 , is a precondition for sound evaluation of policies with the potential to accelerate or decelerate rates of deforestation 8 . Commonly-cited influences on this rate include command-and-control policies 4 , pre-election promises 9 , armed conflicts 10 or mining 11 , as well as climate anomalies 12 .…”

“…A similar 2016 high of observed forest losses was found in Indonesia, which may be attributed to the same El Niño event, which also caused high forest losses in Brazil. (c) Armed conflicts in the DR Congo were associated with an AC phase of deforestation from 2012 onwards10 . Armed conflicts likely accelerate deforestation by land-use activities associated with movements of refugees and displaced people, and a loss in confidence of land users about their ability to exploit future rents from avoided deforestation today.…”