Global sensitivity analysis of a dynamic vegetation model: Model sensitivity depends on successional time, climate and competitive interactions

The importance of regeneration processes on forest biodiversity in old-growth forests in the Pacific Northwest

Forest diversity is the outcome of multiple species-specific processes and tolerances, from regeneration, growth, competition and mortality of trees. Predicting diversity thus requires a comprehensive understanding of those processes. Regeneration processes have traditionally been overlooked, due to high stochasticity and assumptions that recruitment is not limiting for forests. Thus, we investigated the importance of seed production and seedling survival on forest diversity in the Pacific Northwest (PNW) using a forest gap model (ForClim). Equations for regeneration processes were fit to empirical data and added into the model, followed by simulations where regeneration processes and parameter values varied. Adding regeneration processes into ForClim improved the simulation of species composition, compared to Forest Inventory Analysis data. We also found that seed production was not as important as seedling survival, and the time it took for seedlings to grow into saplings was a critical recruitment parameter for accurately capturing tree species diversity in PNW forest stands. However, our simulations considered historical climate only. Due to the sensitivity of seed production and seedling survival to weather, future climate change may alter seed production or seedling survival and future climate change simulations should include these regeneration processes to predict future forest dynamics in the PNW. This article is part of the theme issue 'Ecological novelty and planetary stewardship: biodiversity dynamics in a transforming biosphere'.

An evaluation of multi-species empirical tree mortality algorithms for dynamic vegetation modelling

Tree mortality is key for projecting forest dynamics, but difficult to portray in dynamic vegetation models (DVMs). Empirical mortality algorithms (MAs) are often considered promising, but little is known about DVM robustness when employing MAs of various structures and origins for multiple species. We analysed empirical MAs for a suite of European tree species within a consistent DVM framework under present and future climates in two climatically different study areas in Switzerland and evaluated their performance using empirical data from old-growth forests across Europe. DVM projections under present climate showed substantial variations when using alternative empirical MAs for the same species. Under climate change, DVM projections showed partly contrasting mortality responses for the same species. These opposing patterns were associated with MA structures (i.e. explanatory variables) and occurred independent of species ecological characteristics. When comparing simulated forest structure with data from old-growth forests, we found frequent overestimations of basal area, which can lead to flawed projections of carbon sequestration and other ecosystem services. While using empirical MAs in DVMs may appear promising, our results emphasize the importance of selecting them cautiously. We therefore synthesize our insights into a guideline for the appropriate use of empirical MAs in DVM applications.

Tree recruitment is determined by stand structure and shade tolerance with uncertain role of climate and water relations

Tree regeneration is a key process for long-term forest dynamics, determining changes in species composition and shaping successional trajectories. While tree regeneration is a highly stochastic process, tree regeneration studies often cover narrow environmental gradients only, focusing on specific forest types or species in distinct regions. Thus, the larger-scale effects of temperature, water availability, and stand structure on tree regeneration are poorly understood.We investigated these effects in respect of tree recruitment (in-growth) along wide environmental gradients using forest inventory data from Flanders (Belgium), northwestern Germany, and Switzerland covering more than 40 tree species. We employed generalized linear mixed models to capture the abundance of tree recruitment in response to basal area, stem density, shade casting ability of a forest stand as well as site-specific degree-day sum (temperature), water balance, and plant-available water holding capacity. We grouped tree species to facilitate comparisons between species with different levels of tolerance to shade and drought.Basal area and shade casting ability of the overstory had generally a negative impact on tree recruitment, but the effects differed between levels of shade tolerance of tree recruitment in all study regions. Recruitment rates of very shade-tolerant species were positively affected by shade casting ability. Stem density and summer warmth (degree-day sum) had similar effects on all tree species and successional strategies. Water-related variables revealed a high degree of uncertainty and did not allow for general conclusions. All variables had similar effects independent of the varying diameter thresholds for tree recruitment in the different data sets.Synthesis: Shade tolerance and stand structure are the main drivers of tree recruitment along wide environmental gradients in temperate forests. Higher temperature generally increases tree recruitment rates, but the role of water relations and drought tolerance remains uncertain for tree recruitment on cross-regional scales.

Stand-scale climate change impacts on forests over large areas: transient responses and projection uncertainties

The increasing impacts of climate change on forest ecosystems have triggered multiple model-based impact assessments for the future, which typically focused either on a small number of stand-scale case studies or on large scale analyses (i.e., continental to global). Therefore, substantial uncertainty remains regarding the local impacts over large areas (i.e., regions to countries), which is particularly problematic for forest management. We provide a comprehensive, high-resolution assessment of the climate change sensitivity of managed Swiss forests (~10,000 km2 ), which cover a wide range of environmental conditions. We used a dynamic vegetation model to project the development of typical forest stands derived from a stratification of the Third National Forest Inventory until the end of the 22nd century. Two types of simulations were conducted: one limited to using the extant local species, the other enabling immigration of potentially more climate-adapted species. Moreover, to assess the robustness of our projections, we quantified and decomposed the uncertainty in model projections resulting from the following sources: (1) climate change scenarios, (2) local site conditions, and (3) the dynamic vegetation model itself (i.e., represented by a set of model versions), an aspect hitherto rarely taken into account. The simulations showed substantial changes in basal area and species composition, with dissimilar sensitivity to climate change across and within elevation zones. Higher-elevation stands generally profited from increased temperature, but soil conditions strongly modulated this response. Low-elevation stands were increasingly subject to drought, with strong negative impacts on forest growth. Furthermore, current stand structure had a strong effect on the simulated response. The admixture of drought-tolerant species was found advisable across all elevations to mitigate future adverse climate-induced effects. The largest uncertainty in model projections was associated with climate change scenarios. Uncertainty induced by the model version was generally largest where overall simulated climate change impacts were small, thus corroborating the utility of the model for making projections into the future. Yet, the large influence of both site conditions and the model version on some of the projections indicates that uncertainty sources other than climate change scenarios need to be considered in climate change impact assessments.

A Step-by-Step Guide to Initialize and Calibrate Landscape Models: A Case Study in the Mediterranean Mountains

The use of spatially interactive forest landscape models has increased in recent years. These models are valuable tools to assess our knowledge about the functioning and provisioning of ecosystems as well as essential allies when predicting future changes. However, developing the necessary inputs and preparing them for research studies require substantial initial investments in terms of time. Although model initialization and calibration often take the largest amount of modelers’ efforts, such processes are rarely reported thoroughly in application studies. Our study documents the process of calibrating and setting up an ecophysiologically based forest landscape model (LANDIS-II with PnET-Succession) in a biogeographical region where such a model has never been applied to date (southwestern Mediterranean mountains in Europe). We describe the methodological process necessary to produce the required spatial inputs expressing initial vegetation and site conditions. We test model behaviour on single-cell simulations and calibrate species parameters using local biomass estimations and literature information. Finally, we test how different initialization data—with and without shrub communities—influence the simulation of forest dynamics by applying the calibrated model at landscape level. Combination of plot-level data with vegetation maps allowed us to generate a detailed map of initial tree and shrub communities. Single-cell simulations revealed that the model was able to reproduce realistic biomass estimates and competitive effects for different forest types included in the landscape, as well as plausible monthly growth patterns of species growing in Mediterranean mountains. Our results highlight the importance of considering shrub communities in forest landscape models, as they influence the temporal dynamics of tree species. Besides, our results show that, in the absence of natural disturbances, harvesting or climate change, landscape-level simulations projected a general increase of biomass of several species over the next decades but with distinct spatio-temporal patterns due to competitive effects and landscape heterogeneity. Providing a step-by-step workflow to initialize and calibrate a forest landscape model, our study encourages new users to use such tools in forestry and climate change applications. Thus, we advocate for documenting initialization processes in a transparent and reproducible manner in forest landscape modelling.

Bayesian calibration of a growth-dependent tree mortality model to simulate the dynamics of European temperate forests

Dynamic vegetation models (DVMs) are important tools to understand and predict the functioning and dynamics of terrestrial ecosystems under changing environmental conditions. In these models, uncertainty in the description of demographic processes, in particular tree mortality, is a persistent problem. Current mortality formulations lack realism and are insufficiently constrained by empirical evidence. It has been suggested that empirically estimated mortality submodels would enhance DVM performance, but due to the many processes and interactions within a DVM, the claim has rarely been tested. Here, we compare the performance of three alternative growth-dependent tree mortality submodels in the DVM ForClim: (1) a mortality function with theoretical foundation (ForClim v3.3); (2) a mortality function with parameters directly estimated based on forest inventory data; and (3) the same function, but with parameters estimated using an inverse approach through Bayesian calibration (BC). Time series of inventory data from 30 ecologically distinct Swiss natural forest reserves collected over 35+ yr, including the main tree species of Central Europe, were used for the calibration and subsequent validation of the mortality functions and the DVM. The recalibration resulted in mortality parameters that differed from the direct empirical estimates, particularly for the relationship between tree size and mortality. The calibrated parameters outperformed the direct estimates, and to a lesser extent the original mortality function, for predicting decadal-scale forest dynamics at both calibration and validation sites. The same pattern was observed regarding the plausibility of their long-term projections under contrasting environmental conditions. Our results demonstrate that inverse calibration may be useful even when direct empirical estimates of DVM parameters are available, as structural model deficiencies or data problems can result in discrepancies between direct and inverse estimates. Thus, we interpret the good performance of the inversely calibrated model for long-term projections (which were not a calibration target) as evidence that the calibration did not compensate for model errors. Rather, we surmise that the discrepancy was mainly caused by a lack of representativeness of the mortality data. Our results underline the potential for learning more about elusive processes, such as tree mortality or recruitment, through data integration in DVMs.

Indicators sensitivity analysis for environmental engineering geological patterns caused by underground coal mining with integrating variable weight theory and improved matter-element extension model

This research utilized the variable weight theory (VWT) and improved the matter-element extension model (MEEM) to conduct an indicators sensitivity analysis of environmental engineering geological patterns (EEGPs) related to underground coal mining. First, four types of EEGPs, i.e., basically unaffected model, gradually restored model after destruction, gradually deteriorated model and disaster model, were defined. Subsequently, 13 indicators were selected from different spheres. Then, a sensitivity analysis for EEGPs was conducted by VWT and improved MEEM. On the basis of changes in indicator values by ±10-50%, indicators sensitivity was separately determined in four types of EEGPs. In conclusion, rainfall capacity, evaporation capacity and normalized difference vegetation index (NDVI) are sensitive indicators in a basically unaffected model, while the thicknesses of the coal seam and laterite are sensitive indicators in the other three EEGPs. For comparison, types of EEGPs were tested by both VWT and traditional MEEM, verifying the accuracy of indicators sensitivity results and reasonability by VWT and improved MEEM method. Finally, mining measures such as layered, stripe, filling mining and grouting reinforce method corresponding to thicknesses of coal seam and laterite were proposed. Therefore, specific mining methods under types of EEGPs can be provided to decision-makers in the mining industry and to environmental protection departments.

High Resolution Maps of Climatological Parameters for Analyzing the Impacts of Climatic Changes on Swiss Forests

Assessing the impacts of climatic changes on forests requires the analysis of actual climatology within the forested area. In mountainous areas, climatological indices vary markedly with the micro-relief, i.e., with altitude, slope, and aspect. Consequently, when modelling potential shifts of altitudinal belts in mountainous areas due to climatic changes, maps with a high spatial resolution of the underlying climatological indices are fundamental. Here we present a set of maps of climatological indices with a spatial resolution of 25 by 25 m. The presented dataset consists of maps of the following parameters: average daily temperature high and low in January, April, July, and October as well as of the year; seasonal and annual thermal continentality; first and last freezing day; frost-free vegetation period; relative air humidity; solar radiation; and foehn conditions. The parameters represented in the maps have been selected in a knowledge engineering approach. The maps show the climatology of the periods 1961–1990 and 1981–2010. The data can be used for statistical analyses of forest climatology, for developing tree distribution models, and for assessing the impacts of climatic changes on Swiss forests.

The sensitivity of the forest carbon budget shifts across processes along with stand development and climate change

The future trajectory of atmospheric CO₂ concentration depends on the development of the terrestrial carbon sink, which in turn is influenced by forest dynamics under changing environmental conditions. An in-depth understanding of model sensitivities and uncertainties in non-steady-state conditions is necessary for reliable and robust projections of forest development and under scenarios of global warming and CO₂ enrichment. Here, we systematically assessed if a biogeochemical process-based model (3D-CMCC-CNR), which embeds similarities with many other vegetation models, applied in simulating net primary productivity (NPP) and standing woody biomass (SWB), maintained a consistent sensitivity to its 55 input parameters through time, during forest ageing and structuring as well as under climate change scenarios. Overall, the model applied at three contrasting European forests showed low sensitivity to the majority of its parameters. Interestingly, model sensitivity to parameters varied through the course of >100 yr of simulations. In particular, the model showed a large responsiveness to the allometric parameters used for initialize forest carbon and nitrogen pools early in forest simulation (i.e., for NPP up to ~37%, 256 g C·m^-2 ·yr^-1 and for SWB up to ~90%, 65 Mg C/ha, when compared to standard simulation), with this sensitivity decreasing sharply during forest development. At medium to longer time scales, and under climate change scenarios, the model became increasingly more sensitive to additional and/or different parameters controlling biomass accumulation and autotrophic respiration (i.e., for NPP up to ~30%, 167 g C·m^-2 ·yr^-1 and for SWB up to ~24%, 64 Mg C/ha, when compared to standard simulation). Interestingly, model outputs were shown to be more sensitive to parameters and processes controlling stand development rather than to climate change (i.e., warming and changes in atmospheric CO₂ concentration) itself although model sensitivities were generally higher under climate change scenarios. Our results suggest the need for sensitivity and uncertainty analyses that cover multiple temporal scales along forest developmental stages to better assess the potential of future forests to act as a global terrestrial carbon sink.