Scaling-up of ecological and socioeconomic functions from local to landscape and broader scales

The development of sustainable land-use concepts requires (i) a spatially explicit analysis of the land-use mosaic, (ii) understanding how multiple functions and their trade-offs or synergies vary with and across scales, and (iii) scaling up of insights on ecological and socioeconomic functions from local to landscape and broader scales. Especially the latter is not a trivial task, since ecological patterns as well as ecological and socioeconomic functions can emerge from lower-level processes or may be imposed by higher-level constraints (Levin 1992). Hence, understanding the relationship between land-use trans-formation and ecological and socioeconomic functions at the local scale does not necessarily result in the same linkages at the landscape scale (Lindborg et al. 2017). This illustrates the strong need for inves-tigating how these relationships change across scales, how they are connected across scales (Meyer et al. 2010) and what spatial scale is most important in driving landscape effects on socioeconomic and ecological functions. Throughout the project, the research efforts combined in Focus 3 centre around the following four core Hypotheses:

  1. Ecological and socioeconomic functions and their trade-offs change from local to landscape scales.

  2. The spatial configuration of the land-use mosaic affects ecological and socioeconomic functions as well as their variability differently at local versus landscape scales.

  3. Land-use transformation and land-use intensification affect the scale-dependency of ecological func-tions, socioeconomic functions and their trade-offs, for example in species-area relationships.

  4. Decisions taken at different levels (e.g., by households, farmers, companies, policy makers) affect land-use patterns (and associated implications for biodiversity and carbon emissions) at local, regional and national scales.

In Phases 1 and 2, we demonstrated that agricultural specialization decreased from the household to the province level (Klasen et al. 2016; Hypothesis 1) and that the labor savings achieved through palm oil expansion resulted in increasing land use (at the cost of deforestation) at all scales of analysis (Kubitza et al. 2019, Hypotheses 1 and 2) and worked on quantifying the effects of spatial, temporal and social heterogeneity on ecological and socioeconomic functions (Hypothesis 3).
The newly developed EFForTS-ABM land-use model linked local decisions to landscape patterns (B10, Dislich et al. 2018, hypothesis 2). Together with various integrative modelling tools (see Fig. 21), this will enable us to further analyse the above hypotheses in Phase 3 with an emphasis on hypotheses 2-4. With the Landscape Assessment (see above) and new acquisitions of scale-explicit data (e.g., LiDAR data), we will explore further the spatial scales that are most important for landscape or regional effects on socioeconomic and ecological functions (hypotheses 2 to 4). Moreover, we will analyse the trade-offs and scale-dependent interactions between multiple socioeconomic and ecological functions (hypotheses 3 and 4). Different modeling approaches that make use of the data collected during Phases 1 and 2 and within the framework of the Landscape Assessment will be used for scaling up various socio-ecological outcomes (Fig. 25; A05 Veldkamp/Corre, A07 Knohl/Veldkamp, B09 Westphal/Grass, C11 Lay et al.). This will allow for analyzing trade-offs at different spatial scales (B10 Wiegand/Lay, C10 Kis-Katos, C11 Lay et al., C12 Paul). For instance, we will explore biodiversity patterns (e.g., species-area relationships; Kunin et al. 2018) or regional habitat connectivity/fragmentation using graph models (Urban et al. 2009) (B09 Westphal/Grass, B11 Hölscher/Kraft/Wollni) and will link changes in land-use cover to changing ecologi-cal functions and socioeconomic outcomes (C10 Kis-Katos). Scaling-up will thus provide us with an inte-grated and dynamic view on ecological and socioeconomic functions across scales.