Agricultural expansion and intensification are major threats to tropical biodiversity and ecosystem functioning. While large-scale commercial agriculture receives considerable attention, smallholders manage a significant portion of agricultural land in many tropical regions, even for crops like oil palm and rubber. Despite their influence on tropical landscapes, the consequences of their land-use choices for socioeconomic and ecological functions are poorly understood. Tropical smallholder landscapes often consist of a mosaic of land uses, potentially combining high yields and biodiversity. However, many landscapes are undergoing transitions to more profitable monocultures, potentially creating economic-ecological trade-offs where economic profit increases at the expense of biodiversity and ecosystem functioning. The UN Sustainable Development Goals aim to manage landscapes for improved livelihoods while conserving ecosystems, highlighting the need to understand these trade-offs. This study explores trade-offs and synergies between multidiversity, multifunctionality, and profitability across multiple land uses in Jambi Province, Sumatra, Indonesia, a region experiencing significant agricultural expansion. Unlike previous work, this study explicitly models relationships between profits and ecological functions to determine the shape of these relationships and examines an unprecedented number of taxonomic groups and ecosystem functions, using indices of multidiversity and multifunctionality to characterize the whole-ecosystem state. Finally, the study scales up from plot to landscape level by identifying optimized landscape compositions that mitigate trade-offs.

The existing literature highlights the significant threats posed by agricultural expansion and intensification to tropical biodiversity and ecosystem services. Studies have focused on large-scale commercial agriculture, particularly the impacts of oil palm, but the role of smallholders, who manage a large proportion of agricultural land in many tropical regions, is less well-understood. Previous research in the study system has shown that land-use choices prioritize profitability over ecological functions. However, a comprehensive analysis of whole-ecosystem biodiversity (multidiversity) and ecosystem functioning (multifunctionality) across multiple land uses and transitional stages has been lacking. This study builds upon this previous work by incorporating a more detailed and comprehensive assessment of biodiversity and ecosystem functions and by explicitly modeling the relationships between economic profitability and ecological outcomes.

This study was conducted in the tropical lowlands of Jambi Province, Sumatra, Indonesia, a region characterized by significant land-use change and smallholder agriculture. Two landscapes were selected, representing different soil types, with four land-use systems: primary degraded lowland rainforest, jungle rubber (agroforestry), rubber monoculture, and oil palm monoculture. Extensive ecological and socioeconomic field studies were conducted, including continuous records of rubber and oil palm yields over two years and data from 700 smallholder farm households. Profitability was calculated as annual profits per hectare after deducting production costs from revenues. The study also included 150 stakeholder interviews with government representatives, NGOs, and corporate actors. Biodiversity assessments included 26,894 species across 14 taxonomic groups. Ecosystem functioning was assessed using 36 indicators of 10 ecosystem functions. Multidiversity and multifunctionality indices were calculated to characterize whole-ecosystem state. A genetic algorithm was used to simulate and optimize landscape compositions for maximizing multidiversity/multifunctionality under different profit demands. Statistical analyses involved simulation-extrapolation (SIMEX) to address measurement error in profit data and generalized additive models to account for non-linear relationships between variables. Specific data collection methods included canopy fogging for ants and wasps, point counts and sound recordings for birds, mist nets and harp traps for bats, sweep netting for butterflies, litter sieving for invertebrates, soil sampling for protists, microarthropods, fungi, bacteria, and archaea, litterbag experiments for decomposition, sap flux measurements for transpiration, and meteorological stations for microclimatic data.

The study revealed strong evidence for non-linear losses in species richness with increasing profits. Losses were most pronounced at the transition from forest and jungle rubber to monocultures. Monocultures, while highly profitable (up to 1000 USD ha⁻¹ year⁻¹), harbored the lowest biodiversity levels. Negative richness-profitability relationships were pervasive even when considering species present in rainforest areas, indicating that even though some species may replace forest species in monocultures, the conversion negatively affects forest biodiversity. Analyses of ecosystem functions showed trade-offs for many key functions. Carbon stocks, soil respiration, and decomposition declined, while nutrient leaching and greenhouse gas fluxes increased with higher profits. Multidiversity and multifunctionality exhibited consistent trade-offs with increasing profits across a wide range of thresholds, indicating that increased profitability always comes at the expense of overall ecosystem diversity and functioning. Landscape-scale optimization using a genetic algorithm demonstrated that maintaining high multidiversity required substantial rainforest cover. With higher profit demands, unavoidable trade-offs resulted in rainforest replacement by plantations and corresponding reductions in multidiversity. Highly profitable landscapes were dominated by oil palm plantations, suggesting that even with less intensive smallholder management, significant biodiversity losses occur.

The findings demonstrate clear economic-ecological trade-offs associated with land-use transitions in tropical smallholder landscapes. Increased profits from monoculture systems are achieved at the cost of significant biodiversity loss and ecosystem degradation. This challenges the long-term sustainability of current economic development models. The loss of multifunctionality has far-reaching consequences, including carbon loss and increased greenhouse gas emissions, which contribute to climate change. The results underscore the need for more targeted landscape planning that considers both economic and ecological dimensions. While the specific results apply to Jambi Province, the general findings on economic-ecological trade-offs are likely applicable to other tropical lowland regions.

This study highlights the significant trade-offs between economic profit and ecosystem multifunctionality in tropical smallholder landscapes. Intensive monocultures consistently yield higher profits but lead to substantial biodiversity loss and ecosystem degradation. Mitigating these trade-offs requires changes in economic incentive structures through policies such as payment for environmental services or premium prices for sustainably produced products. Future research should focus on developing and evaluating specific policy interventions to support multifunctional landscapes while ensuring equitable economic benefits for smallholder farmers.

The study focuses on a specific region in Sumatra, Indonesia, and the findings may not be directly generalizable to all tropical smallholder landscapes. The profitability analysis relies on data from smallholder farms, and the results may not fully capture the impacts of large-scale commercial plantations. The study’s time frame was limited, and long-term monitoring would be valuable to fully assess the ecological and socioeconomic consequences of land-use change. The genetic algorithm used for landscape optimization is a simplified representation of complex ecological and social systems, and the results should be interpreted as conceptual examples rather than precise predictions.