Claudia Dislich

Breaking functional connectivity into components: a novel approach using an individual-based model, and first outcomes.

Landscape connectivity is a key factor determining the viability of populations in fragmented landscapes. Predicting ‘functional connectivity’, namely whether a patch or a landscape functions as connected from the perspective of a focal species, poses various challenges. First, empirical data on the movement behaviour of species is often scarce. Second, animal-landscape interactions are bound to yield complex patterns. Lastly, functional connectivity involves various components that are rarely assessed separately. We introduce the spatially explicit, individual-based model FunCon as means to distinguish between components of functional connectivity and to assess how each of them affects the sensitivity of species and communities to landscape structures. We then present the results of exploratory simulations over six landscapes of different fragmentation levels and across a range of hypothetical bird species that differ in their response to habitat edges. i) Our results demonstrate that estimations of functional connectivity depend not only on the response of species to edges (avoidance versus penetration into the matrix), the movement mode investigated (home range movements versus dispersal), and the way in which the matrix is being crossed (random walk versus gap crossing), but also on the choice of connectivity measure (in this case, the model output examined). ii) We further show a strong effect of the mortality scenario applied, indicating that movement decisions that do not fully match the mortality risks are likely to reduce connectivity and enhance sensitivity to fragmentation. iii) Despite these complexities, some consistent patterns emerged. For instance, the ranking order of landscapes in terms of functional connectivity was mostly consistent across the entire range of hypothetical species, indicating that simple landscape indices can potentially serve as valuable surrogates for functional connectivity. Yet such simplifications must be carefully evaluated in terms of the components of functional connectivity they actually predict.

A review of the ecosystem functions in oil palm plantations, using forests as a reference system.

Oil palm plantations have expanded rapidly in recent decades. This large‐scale land‐use change has had great ecological, economic, and social impacts on both the areas converted to oil palm and their surroundings. However, research on the impacts of oil palm cultivation is scattered and patchy, and no clear overview exists. We address this gap through a systematic and comprehensive literature review of all ecosystem functions in oil palm plantations, including several (genetic, medicinal and ornamental resources, information functions) not included in previous systematic reviews. We compare ecosystem functions in oil palm plantations to those in forests, as the conversion of forest to oil palm is prevalent in the tropics. We find that oil palm plantations generally have reduced ecosystem functioning compared to forests: 11 out of 14 ecosystem functions show a net decrease in level of function. Some functions show decreases with potentially irreversible global impacts (e.g. reductions in gas and climate regulation, habitat and nursery functions, genetic resources, medicinal resources, and information functions). The most serious impacts occur when forest is cleared to establish new plantations, and immediately afterwards, especially on peat soils. To variable degrees, specific plantation management measures can prevent or reduce losses of some ecosystem functions (e.g. avoid illegal land clearing via fire, avoid draining of peat, use of integrated pest management, use of cover crops, mulch, and compost) and we highlight synergistic mitigation measures that can improve multiple ecosystem functions simultaneously. The only ecosystem function which increases in oil palm plantations is, unsurprisingly, the production of marketable goods. Our review highlights numerous research gaps. In particular, there are significant gaps with respect to socio‐cultural information functions. Further, there is a need for more empirical data on the importance of spatial and temporal scales, such as differences among plantations in different environments, of different sizes, and of different ages, as our review has identified examples where ecosystem functions vary spatially and temporally. Finally, more research is needed on developing management practices that can offset the losses of ecosystem functions. Our findings should stimulate research to address the identified gaps, and provide a foundation for more systematic research and discussion on ways to minimize the negative impacts and maximize the positive impacts of oil palm cultivation.

Technical Note: Approximate Bayesian parameterization of a process-based tropical forest model

Inverse parameter estimation of process-based models is a long-standing problem in many scientific disciplines. A key question for inverse parameter estimation is how to define the metric that quantifies how well model predictions fit to the data. This metric can be expressed by general cost or objective functions, but statistical inversion methods require a particular metric, the probability of observing the data given the model parameters, known as the likelihood. For technical and computational reasons, likelihoods for process-based stochastic models are usually based on general assumptions about variability in the observed data, and not on the stochasticity generated by the model. Only in recent years have new methods become available that allow the generation of likelihoods directly from stochastic simulations. Previous applications of these approximate Bayesian methods have concentrated on relatively simple models. Here, we report on the application of a simulation-based likelihood approximation for FORMIND, a parameter-rich individual-based model of tropical forest dynamics. We show that approximate Bayesian inference, based on a parametric likelihood approximation placed in a conventional Markov chain Monte Carlo (MCMC) sampler, performs well in retrieving known parameter values from virtual inventory data generated by the forest model. We analyze the results of the parameter estimation, examine its sensitivity to the choice and aggregation of model outputs and observed data (summary statistics), and demonstrate the application of this method by fitting the FORMIND model to field data from an Ecuadorian tropical forest. Finally, we discuss how this approach differs from approximate Bayesian computation (ABC), another method commonly used to generate simulation-based likelihood approximations. Our results demonstrate that simulation-based inference, which offers considerable conceptual advantages over more traditional methods for inverse parameter estimation, can be successfully applied to process-based models of high complexity. The methodology is particularly suitable for heterogeneous and complex data structures and can easily be adjusted to other model types, including most stochastic population and individual-based models. Our study therefore provides a blueprint for a fairly general approach to parameter estimation of stochastic process-based models.

Water scarcity and oil palm expansion: social views and environmental processes

Conversions of natural ecosystems, e.g., from rain forests to managed plantations, result in significant changes in the hydrological cycle including periodic water scarcity. In Indonesia, large areas of forest were lost and extensive oil palm plantations were established over the last decades. We conducted a combined social and environmental study in a region of recent land-use change, the Jambi Province on Sumatra. The objective was to derive complementary lines of arguments to provide balanced insights into environmental perceptions and eco-hydrological processes accompanying land-use change. Interviews with villagers highlighted concerns regarding decreasing water levels in wells during dry periods and increasing fluctuations in stream flow between rainy and dry periods. Periodic water scarcity was found to severely impact livelihoods, which increased social polarization. Sap flux measurements on forest trees and oil palms indicate that oil palm plantations use as much water as forests for transpiration. Eddy covariance analyses of evapotranspiration over oil palm point to substantial additional sources of evaporation in oil palm plantations such as the soil and epiphytes. Stream base flow from a catchment dominated by oil palms was lower than from a catchment dominated by rubber plantations; both showed high peaks after rainfall. An estimate of erosion indicated approximately 30 cm of topsoil loss after forest conversion to both oil palm and rubber plantations. Analyses of climatic variables over the last 20 years and of a standardized precipitation evapotranspiration index for the last century suggested that droughts are recurrent in the area, but have not increased in frequency or intensity. Consequently, we assume that conversions of rain forest ecosystems to oil palm plantations lead to a redistribution of precipitated water by runoff, which leads to the reported periodic water scarcity. Our combined social and environmental approach points to significant and thus far neglected eco-hydrological consequences of oil palm expansion.

Natural Landslides Which Impact Current Regulating Services: Environmental Preconditions and Modeling

Recurrent landslide activity in the natural mountain forest is assumed to be a major factor for maintaining its high biodiversity. It is hypothesized that abiotic–biotic interactions are a prerequisite for natural landslides. A statistical model solely driven by topographic predictors can explain areas prone to landslides but also shows that other factors (e.g., geology, soil, climate, vegetation) than topography might play an important role to improve model performance. Thus, the chapter also shows approaches to derive spatial information on soil properties and wind stress as potential driving predictors for the model. Furthermore, it can be shown that even changes in the biogeochemical cycle and the regulation between nutrient input and biomass production might influence the risk of landslides.

Climate Change and Its Impact on Current and Future Vegetation Dynamics and Carbon Cycling

In this chapter, pasture and forest productivity and dynamics are investigated under global climate change impacts. Due to unsustainable management, pasture areas, mostly covered by Setaria sphacelata, are invaded by the aggressive Southern Bracken (Pteridium arachnoideum). The Southern Bracken Competition Model (SoBraCoMo) was applied to predict the development of the pasture-weed competition under IPCC-SRES A1B conditions. The model and respective physiological observations reveal an equal gain in performance of Setaria and bracken under global warming. In the forest, a potential increase in precipitation could boost landslide activity and thus affect growth dynamics. The forest-gap FORMIND model predicts a higher fraction of early successional species in tree species composition, which would reduce the aboveground carbon stocks. In summary, climate warming might not only improve regulating and supporting services on the pasture side (increased carbon sequestration, higher pasture yield), but also lead to a reduction of aboveground carbon stocks in the natural forest.

Land-use change in oil palm dominated tropical landscapes-An agent-based model to explore ecological and socio-economic trade-offs.

Land-use changes have dramatically transformed tropical landscapes. We describe an ecological-economic land-use change model as an integrated, exploratory tool used to analyze how tropical land-use change affects ecological and socio-economic functions. The model analysis seeks to determine what kind of landscape mosaic can improve the ensemble of ecosystem functioning, biodiversity, and economic benefit based on the synergies and trade-offs that we have to account for. More specifically, (1) how do specific ecosystem functions, such as carbon storage, and economic functions, such as household consumption, relate to each other? (2) How do external factors, such as the output prices of crops, affect these relationships? (3) How do these relationships change when production inefficiency differs between smallholder farmers and learning is incorporated? We initialize the ecological-economic model with artificially generated land-use maps parameterized to our study region. The economic sub-model simulates smallholder land-use management decisions based on a profit maximization assumption. Each household determines factor inputs for all household fields and decides on land-use change based on available wealth. The ecological sub-model includes a simple account of carbon sequestration in above-ground and below-ground vegetation. We demonstrate model capabilities with results on household consumption and carbon sequestration from different output price and farming efficiency scenarios. The overall results reveal complex interactions between the economic and ecological spheres. For instance, model scenarios with heterogeneous crop-specific household productivity reveal a comparatively high inertia of land-use change. Our model analysis even shows such an increased temporal stability in landscape composition and carbon stocks of the agricultural area under dynamic price trends. These findings underline the utility of ecological-economic models, such as ours, to act as exploratory tools which can advance our understanding of the mechanisms underlying the trade-offs and synergies of ecological and economic functions in tropical landscapes.