Maarten Boudewijn Eppinga

Humans have inhabited Bornean tropical forests for at least 45,000 years, yet the impact of that long‐term presence is poorly understood. Borneo's extensive archaeological record and high biodiversity offer a unique laboratory to examine patterns of site selection and the effects of human activity on tropical forest ecosystems. This paper establishes a synthetic literature‐based data set of 73 archaeological sites, 47 with radiocarbon or radiometric records dating during the last 6 ka BP. Palaeoclimate, current climate, soil, and terrain variables of these 47 sites are compared against a set of simulated random locations across Borneo. We also consider spatial proximity as a factor influencing human settlement, clustering sites based on location and examining the timing and duration of habitation of sites within the clusters. Finally, we compare the 47 site records against the HYDE‐3.2 land‐use model to assess correspondence. Our results suggest that during the last 6000 years BP, Bornean human settlements tended to be located in the forests at lower elevations, near coasts and river networks, in higher temperatures and lower precipitation. Comparative sedimentary analysis also suggests preferential use of locations exhibiting lower clay and higher sand content. This combination of conditions likely improved food production and resource supply. Our study highlights the value of integrating archaeological data with global historical land‐use and climate models to uncover long‐term human–environment interactions. By establishing a cross‐site environmental baseline, these findings provide insights into past human settlement patterns and likely human legacies in Borneo's tropical forests.

The rapid environmental changes of the Anthropocene create legacy effects that may shape future Earth system responses. One significant legacy effect is the species extinction debt caused by past habitat destruction. As biodiversity underpins ecosystem services vital to human societies, social-ecological systems may, in turn, be subjected to biodiversity-dependent ecosystem service debts. While biodiversity-dependent ecosystem service debts have been quantified with analytical approaches, less attention has been paid to their potential impact on social-ecological system trajectories. We performed a theoretical study of a dynamical systems model that includes the possibility of biodiversity-dependent ecosystem service debts emerging from past habitat destruction. Our results suggest that these debts reduce systems’ safe operating spaces and create environmental tipping points associated with critical transitions in system states. These transitions, however, may include long transients of apparent stability, making it difficult to identify cause and effect. Notably, biodiversity-dependent ecosystem service debts may drive initial phases of apparent recovery after disturbance, still followed by system collapse. Our theoretical findings highlight the need to consider biodiversity-dependent ecosystem service debts for sustainable management of social-ecological systems. Furthermore, these results suggest that social-ecological systems’ safe operating spaces cannot be reliably inferred from recent observations of apparent system stability.

Fires are generally considered to promote biodiversity, although the exact relationship is unclear, because it can be affected by several factors, including fire regime and ecosystem type. Given the ongoing global change, a better understanding of this connection is needed to assess the extent to which projected increases in fire frequency may affect current biodiversity trends. A major challenge lies in vegetation-fire feedback, which often mediates changes in fire regimes. To shed light on the role of fires in promoting or limiting biodiversity, we studied the compositional and functional diversity of simulated plant communities along a gradient of fire frequencies. We extended an existing model to include a large number of species. The model reproduces plant successional dynamics and is parameterized to represent Boreal and Mediterranean communi-ties. Fire events are stochastic, with frequencies that depend on community flammability, and plants have different fire responses, thus creating a vegetation-fire feedback. For both ecosys-tems, we found that fires generally had a positive effect on both compositional and functional diversity. Furthermore, in most cases, both peaked at intermediate fire frequencies. Interestingly, compositional and functional diversity were correlated but did not reach their maximum values in the same communities. This seemingly underlines that a certain degree of functional similarity may be necessary to achieve maximum species richness. These results stem from the vegetation-fire feedback, highlighting its importance for predicting ecosystem responses to global change, including biodiversity losses and wildfire regime shifts.

Navigating social‐ecological systems toward sustainable trajectories is an important challenge of the Anthropocene. Models of social‐ecological systems can increase our understanding of how social and ecological subsystems interact, their response to environmental changes, and how their dynamics may be altered by management interventions. However, the level of representational detail required for models to describe a particular social‐ecological system with high fidelity (i.e., accurately quantifying system dynamics) may hamper both the interpretability of model results and our ability to identify key processes and feedbacks within the system. In contrast, stylized models describe simplified interactions between a small subset of social‐ecological system elements. Stylized models are a useful tool to identify potential consequences of specific key processes and feedbacks on system functioning. However, the relatively low level of representational detail in these models limits their ability to deliver concrete management options for a particular social‐ecological system. Here, we describe how an exploratory modeling approach can utilize the strengths of stylized models before the construction of social‐ecological system models with high fidelity and representational detail. This exploratory modeling approach is an iterative strategy, with the initial steps comprising the development of stylized models informed by empirical observations. We illustrate this with two examples of stylized modeling of isolated and connected social‐ecological systems. Through repeated confrontation of alternative models with empirical data, exploratory modeling provides useful stepping stones toward the development of models that describe social‐ecological systems in increasingly specific settings with increasing levels of representational detail. When these latter types of models reach a high level of fidelity, they could be used for scenario‐based analyses and participatory decision‐making processes. At this stage, the conceptual insights previously obtained during the exploratory modeling phase may aid in the interpretation and communication of the outcomes of scenario‐based analyses. Hence, exploratory modeling aims to create a synergy between the insights obtained from stylized models and system‐specific, high‐fidelity models in order to generate a deep understanding of the drivers of social‐ecological system dynamics, and how to leverage these drivers to initiate desired changes.

By 2050, 70% of the global population will reside in urban areas, raising concerns about food security and biodiversity loss. Tropical homegardens, a form of agroforestry, may have the potential to align concurrent demands for biodiversity conservation and food security in urban environments. Nevertheless, this depends on land-use decision-making strategies that are enabled by household livelihood assets. We examined which livelihood assets are associated with homegarden agrobiodiversity and food availability and accessibility in Kerala, India, the nature of these associations, and whether they vary along urbanisation gradients using data from 304 homegardens. We observed higher plant species richness closer to the urban centre and in smaller homegardens. Meanwhile, yields and share of household food consumption from homegardens tend to be higher farther away from urban centres. Assets like fertilizer use, irrigation and kinship affiliation are positively associated with both homegarden agrobiodiversity and food outcomes. Conversely, homegarden size is negatively associated with biodiversity outcomes across urbanisation gradients and positively associated with yield outcomes furthest away from urban centre. Some assets like family labour and farming association membership were only positively associated with food outcomes, while others like external labour availability were positively associated only with plant diversity in the urban centre. Overall, synergies between homegarden agrobiodiversity and household food security requires inputs to natural processes, such as irrigation and soil fertility, and social processes like knowledge sharing and social cohesion, especially in urban peripheries. We emphasize the relevance of homegardens and enhancing livelihood assets for food-secure households in biodiverse urban environments.

Meeting restoration targets may require active strategies to accelerate natural regeneration rates or overcome the resilience associated with degraded ecosystem states. Introducing desired ecosystem patches in degraded landscapes constitutes a promising active restoration strategy, with various mechanisms potentially causing these patches to become foci from which desired species can re‐establish throughout the landscape. This study considers three mechanisms previously identified as potential drivers of introduced patch dynamics: autocatalytic nucleation, directed dispersal, and resource concentration. These mechanisms reflect qualitatively different positive feedbacks. We developed an ecological model framework that compared how the occurrence of each mechanism was reflected in spatio‐temporal patch dynamics. We then analyzed the implications of these relationships for optimal restoration design. We found that patch expansion accelerated over time when driven by the autocatalytic nucleation mechanism, while patch expansion driven by the directed dispersal or resource concentration mechanisms decelerated over time. Additionally, when driven by autocatalytic nucleation, patch expansion was independent of patch position in the landscape. However, the proximity of other patches affected patch expansion either positively or negatively when driven by directed dispersal or resource concentration. For autocatalytic nucleation, introducing many small patches was a favorable strategy, provided that each individual patch exceeded a critical patch size. Introducing a single patch or a few large patches was the most effective restoration strategy to initiate the directed dispersal mechanism. Introducing many small patches was the most effective strategy for reaching restored ecosystem states driven by a resource concentration mechanism. Our model results suggest that introducing desirable patches can substantially accelerate ecosystem restoration, or even induce a critical transition from an otherwise stable degraded state toward a desired ecosystem state. However, the potential of this type of restoration strategy for a particular ecosystem may strongly depend on the mechanism driving patch dynamics. In turn, which mechanism drives patch dynamics may affect the optimal spatial design of an active restoration strategy. Each of the three mechanisms considered reflects distinct spatio‐temporal patch dynamics, providing novel opportunities for empirically identifying key mechanisms, and restoration designs that introduce desired patches in degraded landscapes according to these patch dynamics.

Biodiversity loss presents a growing threat to the global environment, and systematic and spatially contiguous monitoring is needed to inform mitigation strategies. Monitoring of genetic diversity within species, a key factor when assessing biodiversity loss, is laborious and could be supported by scalable phenotypic observations allowing inferences about genetic variation. We studied genetic and phenotypic variation in one of Europe's most prevalent forest-forming trees, the common beech Fagus sylvatica L., using whole-genome sequence data and spectral phenotypes from 219 individuals at 23 sites across the species' natural range. Spectral phenotypes were collected under standardized illumination and observation conditions from the same top-of-canopy leaves used for nuclear DNA extraction. We found that spectral and environmental information accounted for 77% of the variance along the first two principal coordinates representing genetic structure among sampled individuals, where spectral phenotypes contributed 12% to the prediction of genetic structure. Further, we identified 14 SNPs (single nucleotide polymorphisms), of which two were located within annotated genes, that showed significant associations with variation in leaf reflectance. Our study demonstrates how linking spectral and genomic variation in tree species may be upscaled to the remote observations to support monitoring, understanding and mitigating loss of genetic diversity within species.