Research during Phase 1 of the CRC concentrated on analysing a large set of ecological and socioeconomic functions of the main agricultural land-use systems in Jambi – namely oil palm, rubber monoculture, and rubber agroforestry (jungle rubber) – in comparison to forest as the reference. The widely accepted concept of sustainable development encompasses ecological, economic and social dimensions, which we all included in our research program. For brevity, we often talk about ecological and socioeconomic functions of land-use systems, whereby in this terminology socioeconomic always includes both economic and social (including cultural) aspects. This research will be continued in Phase 2, putting greater emphasis on the analysis of heterogeneity.
As expected and detailed in the hypotheses of the application of Phase 1 of the CRC, consequences of rainforest conversion are associated with major changes in abiotic factors, as well as plant and animal diversity, but in parallel with increased income and access to education, especially for migrants from Central Indonesia who heavily influence the agricultural sector in Jambi. Results from our core plots and socioeconomic surveys showed in all aspects significant differences between the four land-use systems studied. In terms of most of the environmental and diversity parameters, oil palm and rubber plantations were similar, but differed significantly from forest and jungle rubber (Fig. 1; Clough et al., 2016, Drescher et al., 2016; for details see applications of individual projects). The results confirmed our initial hypothesis that the economic benefits of rainforest transformation systems were in stark contrast with most biodiversity and ecosystem function indicators. Oil palm and rubber monocultures have low naturalness, low biodiversity (B01 Brose: litter communities; B06 Kreft: plants; B07 Polle: mycorrhiza; B08 Scheu: decomposer system, B09 Tscharntke/Clough: aboveground biodiversity), and low carbon stocks (A04 Kuzyakov) compared to forests and agroforests, while oil palm stood out due to high nutrient leaching fluxes threatening freshwater ecosystems (A05 Corre/Veldkamp) (see Fig. 1, joint results). Interestingly, the biodiversity changes were associated with reductions in energy fluxes between the trophic groups indicating lowered ecosystem functions in the non-forest transformation systems.
Highlights of scientific results Fig. 1
Decomposition and herbivory were reduced (from forest to jungle rubber, rubber and oil palm plantations), associated with a strong shift in top-down control from predators to omnivores. Land-use transformation from forests to oil-palm plantations causes 51% reductions in energy fluxes that exceed the 45% decreases in species richness (B01 Brose, Barnes et al. 2015). However, the diversity of the prokaryotic communities are more diverse in plantations than in rainforest soils, which could be partially related to different soil attributes such as P content (Bacteria) and N content (Archaea) (B02 Daniel). Similarly, oil palm plantations had the highest epiphyte abundance, whereas mean epiphyte species diversity did not differ significantly from forest to jungle rubber and oil palm (B06 Kreft). Ant species richness did also not change with land-use system (B09 Tscharntke/Clough). In B05 Kleinn, current and historical land-use maps for Jambi Province were generated for the years 1990, 2000, 2011 and 2013 from Landsat imagery with 30 m spatial resolution, quantifying a marked decrease of secondary forests and huge increase of oil palm and rubber plantations. The systematic literature review by Dislich et al. (2015) on all ecosystem functions in oil palm plantations identifies major research gaps in our understanding of a land-use system that is currently receiving major interest of the scientific community (B10 Wiegand/Lay).
Using structural equations modelling, direct land-use effects and indirect plant-mediated effects on above- and below-ground taxa including microbes, invertebrates and birds were separated (Fig. 2; Barnes, Allen et al., 2017). It was shown that cascading bottom-up effects played an important role in structuring communities of consumer organisms, but most of these effects did not propagate from plant communities. Furthermore, it was shown that land-use change to plantation agriculture dominated in terms of the number and strength of ecological impacts by directly altering most trophic groups.
Highlights of scientific results Fig. 2
The study revealed that land-use directly and indirectly drives large-scale ecological shifts which can entirely bypass plant communities, and this may have important consequences for conservation efforts for whole ecosystems.
Reconciling biodiversity conservation, ecosystem functioning and increasing human demands in tropical landscapes is a major goal of the CRC. During Phase 1 we studied changes and trade-offs between and among numerous ecological and socioeconomic functions across the major land-use systems in Jambi province. Our synthesis provides strong evidence for losses in biodiversity and ecological functions with rainforest transformation, whereby expansion of monocultures of rubber and oil palm, associated with higher yields and economic benefits, occurs at the expense of forests and agroforest jungle rubber (Fig. 3; Clough et al., 2016).
Highlights of scientific results Fig. 3
The central achievements of the socioeconomic studies of Phase 1 are based on different levels of analysis ranging from the micro-farm and household to the macro-national and international level. Quantitative and qualitative analytical tools have been developed, applied and adapted to the specific Indonesian context (Faust et al. 2013). The successful implementation is documented in a cross-cutting paper on the growing role of smallholders for land-use change in Jambi disentangling factors influencing smallholders’ crop choice between rubber and oil palm plantations (Fig. 4). The socioeconomic datasets were used to analyse smallholders’ decision-making by combining qualitative, quantitative, and experimental methods (Schwarze et al. 2015). Farmer’s land-use choices are strongly influenced by asset endowments (land, labour, capital) and access to processing and market infrastructure. Important reasons for decisions to cultivate oil palm are higher returns to labour and shorter immature crop phases (see Fig. 1). However, for credit-constrained farmers the high investment costs associated with oil palm adoption pose a barrier. In comparison, continued cultivation of rubber, which has been an established crop in Jambi for many decades, seems to be the more secure choice for many smallholders. Indeed, the village and household surveys reveal that – in spite of the rapid expansion of oil palm – rubber remains the dominant crop in Jambi (Gatto et al. 2015a, Krishna et al. 2015). Analysis of longitudinal data shows that oil palm adoption in the small farm sector started in the early-1990s through outgrower schemes in which farmers were contracted by a public or private company and received financial and technical support (“Supported smallholders”). Since the late-1990s, independent adoption (without company contracts) has dominated (Fig. 4). Differences in the speed of oil palm adoption between migrant and non-migrant households and unequal access to land bears certain risks of social tensions (Gatto et al. 2015, Euler et al. 2015).
Further, results on trade-offs between ecological and socioeconomic functions in Phase 1 indicate that the role of markets and infrastructure affect the trade-offs at different spatial scales (Klasen et al. 2015). Improvements in infrastructure and markets enable greater specialization of production at increasingly higher spatial scales thus creating a dilemma for policy as it seeks to promote income growth through the development of infrastructure and markets. Those policies and associated specialization can then lead to substantial losses of ecosystem functions which depend on some diversity at lower spatial scales.
Highlights of scientific results Fig. 4
The detailed assessment of the social, economic and ecological significance of land-use change in Jambi allowed for more general evaluations of the social, economic and ecological goals of policies as documented in two review papers (Brümmer et al., unpubl. data, Tscharntke et al., unpubl. data). The first of these papers focuses on (direct) land-use policy, trade policy, standards and guidelines as well as environmental policy, comparing inventory of policies at local, regional and national scales; it identifies policies impacting land-use decisions in Indonesia with respect to changes of context, interdependencies as well as temporal and spatial scales.
The second review paper focuses on scenarios of socioeconomic-ecological trade-offs in land use. Here, we discuss six land-use scenarios (Fig. 5 A-F) from positive to negative socioeconomic-ecological relationships, which are illustrated with real-world case studies: (A) Socioeconomic goods increase despite destruction of ecological goods in the environment (“the environmentalist’s paradox”), (B) agroecological best-practice management results in approaches combining low losses of socioeconomic goods with high gains in ecological goods, i.e. low cost-high benefit approaches (“agroecological intensification”), (C) land-use intensification does not harm ecological goods, (D) agroecological intensification continuously promotes both socioeconomic and ecological goods, e.g. both yield and biodiversity (“general win-win solutions), (E) socioeconomic and ecological goods increase, but only up to a turning point (“win-win solutions up to a tipping point”), and (F) an U-shaped relationship between ecological and socioeconomic goods, which can be observed over long temporal scales from societies’ early industrialization to higher living standards (“the environmental Kuznets curve”). Which of these socioeconomic-ecological relationships applies in the real world depends on the multifunctionality of land use, the spatial and temporal scales considered, the cultural context and the multifaceted meanings of socioeconomic and ecological goods that are in place. In conclusion, the relationships between socioeconomic and ecological goods are manifold reflecting individual situations and are typically non-linear. Depending on the observed relationship, there are opportunities to combine socioeconomic goods with benefits from ecological goods, which is one of the greatest challenges for future land-use intensification and the major topic of Phase 2 of the CRC.
Highlights of scientific results Fig. 5