Sustaining biodiverse and productive landscapes in tropical swidden forests of the Peruvian Amazon

Background/Question/Methods

Tropical lowland forests are highly productive ecosystems with a capacity to regenerate quickly, but the rate and direction of forest regeneration is often contingent on the type and intensity of previous land use. Swidden-agroforestry is the predominant agricultural system practiced by small farming communities throughout Amazonia and its long-term productivity relies on fallow periods to restore soil fertility. Fallow lengths can vary substantially between large and small landholders. Small landholders re-cultivate sites more frequently than large landholders, creating a mosaic of land-use intensity across the landscape. Moreover, large landholders often substitute fruit tree orchards in place of naturally regenerating forest during their longer fallow periods. Our study examines the historical effects of land use (type and intensity) on soil fertility and regenerating tree species biodiversity in large and small landholders’ fallow forests. We use historic (50 yrs) cultivation patterns from a long-term study community in the Peruvian Amazon to evaluate i) soil fertility restoration, ii) tree community similarity and species diversity between sites experiencing short versus long fallow periods, and iii) to those fallows currently or previously planted as orchards. Using field collected data and a GIS of farmer landowning we look for patterns of biodiversity and soil degradation across the community landscape. 

Results/Conclusions

SOM (p=0.01), P0₄⁺ (p=0.05), K⁺ levels (p=0.02) all increased with fallow age at similar in rates across large and small landholding fallow sites. No difference was seen in soil parameters below forest and orchard fallows, suggesting little immediate consequence of substituting fruit orchards as fallows for future agricultural production. Yet, a biological legacy associated with planting orchards was detected. Fallows re-growing on former orchard sites showed a depauperate tree community assemblage with tree species richness (p = 0.02) and diversity (p = 0.006) increasing more slowly than in fallows with an orchard history. Despite this, stem growth was not hindered. Similar rates of basal area accrual were seen in fallows regardless of an orchard history, providing farmers a reliable source of timber for extraction and charcoal production at the end of the cycle.  This implies that orchard sites, which make up ~23% of all landholdings (predominantly of large landowners), will re-grow timber quickly but take longer to recover biodiversity once re-fallowed. These results suggest that differences in fallow management may little affect farmers reliant on fallows for timber production and soil fertility restoration, but pose a conservation concern as they may contribute to overall landscape biodiversity loss.