Individual-based Dynamic Vegetation Model Research Articles

Nitrogen restricts future sub-arctic treeline advance in an individual-based dynamic vegetation model

Abstract. Arctic environmental change induces shifts in high-latitude plant community composition and stature with implications for Arctic carbon cycling and energy exchange. Two major components of change in high-latitude ecosystems are the advancement of trees into tundra and the increased abundance and size of shrubs. How future changes in key climatic and environmental drivers will affect distributions of major ecosystem types is an active area of research. Dynamic vegetation models (DVMs) offer a way to investigate multiple and interacting drivers of vegetation distribution and ecosystem function. We employed the LPJ-GUESS tree-individual-based DVM over the Torneträsk area, a sub-arctic landscape in northern Sweden. Using a highly resolved climate dataset to downscale CMIP5 climate data from three global climate models and two 21st-century future scenarios (RCP2.6 and RCP8.5), we investigated future impacts of climate change on these ecosystems. We also performed model experiments where we factorially varied drivers (climate, nitrogen deposition and [CO2]) to disentangle the effects of each on ecosystem properties and functions. Our model predicted that treelines could advance by between 45 and 195 elevational metres by 2100, depending on the scenario. Temperature was a strong driver of vegetation change, with nitrogen availability identified as an important modulator of treeline advance. While increased CO2 fertilisation drove productivity increases, it did not result in range shifts of trees. Treeline advance was realistically simulated without any temperature dependence on growth, but biomass was overestimated. Our finding that nitrogen cycling could modulate treeline advance underlines the importance of representing plant–soil interactions in models to project future Arctic vegetation change.

Modelling Holocene peatland dynamics with an individual-based dynamic vegetation model

Abstract. Dynamic global vegetation models (DGVMs) are designed for the study of past, present and future vegetation patterns together with associated biogeochemical cycles and climate feedbacks. However, most DGVMs do not yet have detailed representations of permafrost and non-permafrost peatlands, which are an important store of carbon, particularly at high latitudes. We demonstrate a new implementation of peatland dynamics in a customized Arctic version of the LPJ-GUESS DGVM, simulating the long-term evolution of selected northern peatland ecosystems and assessing the effect of changing climate on peatland carbon balance. Our approach employs a dynamic multi-layer soil with representation of freeze–thaw processes and litter inputs from a dynamically varying mixture of the main peatland plant functional types: mosses, shrubs and graminoids. The model was calibrated and tested for a sub-Arctic mire in Stordalen, Sweden, and validated at a temperate bog site in Mer Bleue, Canada. A regional evaluation of simulated carbon fluxes, hydrology and vegetation dynamics encompassed additional locations spread across Scandinavia. Simulated peat accumulation was found to be generally consistent with published data and the model was able to capture reported long-term vegetation dynamics, water table position and carbon fluxes. A series of sensitivity experiments were carried out to investigate the vulnerability of high-latitude peatlands to climate change. We found that the Stordalen mire may be expected to sequester more carbon in the first half of the 21st century due to milder and wetter climate conditions, a longer growing season, and the CO2 fertilization effect, turning into a carbon source after mid-century because of higher decomposition rates in response to warming soils.

Predicting vegetation type through physiological and environmental interactions with leaf traits: evergreen and deciduous forests in an earth system modeling framework.

Earth system models are incorporating plant trait diversity into their land components to better predict vegetation dynamics in a changing climate. However, extant plant trait distributions will not allow extrapolations to novel community assemblages in future climates, which will require a mechanistic understanding of the trade-offs that determine trait diversity. In this study, we show how physiological trade-offs involving leaf mass per unit area (LMA), leaf lifespan, leaf nitrogen, and leaf respiration may explain the distribution patterns of evergreen and deciduous trees in the temperate and boreal zones based on (1) an evolutionary analysis of a simple mathematical model and (2) simulation experiments of an individual-based dynamic vegetation model (i.e., LM3-PPA). The evolutionary analysis shows that these leaf traits set up a trade-off between carbon- and nitrogen-use efficiency at the scale of individual trees and therefore determine competitively dominant leaf strategies. As soil nitrogen availability increases, the dominant leaf strategy switches from one that is high in nitrogen-use efficiency to one that is high in carbon-use efficiency or, equivalently, from high-LMA/long-lived leaves (i.e., evergreen) to low-LMA/short-lived leaves (i.e., deciduous). In a region of intermediate soil nitrogen availability, the dominant leaf strategy may be either deciduous or evergreen depending on the initial conditions of plant trait abundance (i.e., founder controlled) due to feedbacks of leaf traits on soil nitrogen mineralization through litter quality. Simulated successional patterns by LM3-PPA from the leaf physiological trade-offs are consistent with observed successional dynamics of evergreen and deciduous forests at three sites spanning the temperate to boreal zones.

Nitrogen feedbacks increase future terrestrial ecosystem carbon uptake in an individual-based dynamic vegetation model

Abstract. Recently a considerable amount of effort has been put into quantifying how interactions of the carbon and nitrogen cycle affect future terrestrial carbon sinks. Dynamic vegetation models, representing the nitrogen cycle with varying degree of complexity, have shown diverging constraints of nitrogen dynamics on future carbon sequestration. In this study, we use LPJ-GUESS, a dynamic vegetation model employing a detailed individual- and patch-based representation of vegetation dynamics, to evaluate how population dynamics and resource competition between plant functional types, combined with nitrogen dynamics, have influenced the terrestrial carbon storage in the past and to investigate how terrestrial carbon and nitrogen dynamics might change in the future (1850 to 2100; one representative "business-as-usual" climate scenario). Single-factor model experiments of CO2 fertilisation and climate change show generally similar directions of the responses of C–N interactions, compared to the C-only version of the model as documented in previous studies using other global models. Under an RCP 8.5 scenario, nitrogen limitation suppresses potential CO2 fertilisation, reducing the cumulative net ecosystem carbon uptake between 1850 and 2100 by 61%, and soil warming-induced increase in nitrogen mineralisation reduces terrestrial carbon loss by 31%. When environmental changes are considered conjointly, carbon sequestration is limited by nitrogen dynamics up to the present. However, during the 21st century, nitrogen dynamics induce a net increase in carbon sequestration, resulting in an overall larger carbon uptake of 17% over the full period. This contrasts with previous results with other global models that have shown an 8 to 37% decrease in carbon uptake relative to modern baseline conditions. Implications for the plausibility of earlier projections of future terrestrial C dynamics based on C-only models are discussed.

Implications of incorporating N cycling and N limitations on primary production in an individual-based dynamic vegetation model

Abstract. The LPJ-GUESS dynamic vegetation model uniquely combines an individual- and patch-based representation of vegetation dynamics with ecosystem biogeochemical cycling from regional to global scales. We present an updated version that includes plant and soil N dynamics, analysing the implications of accounting for C–N interactions on predictions and performance of the model. Stand structural dynamics and allometric scaling of tree growth suggested by global databases of forest stand structure and development were well reproduced by the model in comparison to an earlier multi-model study. Accounting for N cycle dynamics improved the goodness of fit for broadleaved forests. N limitation associated with low N-mineralisation rates reduces productivity of cold-climate and dry-climate ecosystems relative to mesic temperate and tropical ecosystems. In a model experiment emulating free-air CO2 enrichment (FACE) treatment for forests globally, N limitation associated with low N-mineralisation rates of colder soils reduces CO2 enhancement of net primary production (NPP) for boreal forests, while some temperate and tropical forests exhibit increased NPP enhancement. Under a business-as-usual future climate and emissions scenario, ecosystem C storage globally was projected to increase by ca. 10%; additional N requirements to match this increasing ecosystem C were within the high N supply limit estimated on stoichiometric grounds in an earlier study. Our results highlight the importance of accounting for C–N interactions in studies of global terrestrial N cycling, and as a basis for understanding mechanisms on local scales and in different regional contexts.

Tundra shrubification and tree-line advance amplify arctic climate warming: results from an individual-based dynamic vegetation model

One major challenge to the improvement of regional climate scenarios for the northern high latitudes is to understand land surface feedbacks associated with vegetation shifts and ecosystem biogeochemical cycling. We employed a customized, Arctic version of the individual-based dynamic vegetation model LPJ-GUESS to simulate the dynamics of upland and wetland ecosystems under a regional climate model–downscaled future climate projection for the Arctic and Subarctic. The simulated vegetation distribution (1961–1990) agreed well with a composite map of actual arctic vegetation. In the future (2051–2080), a poleward advance of the forest–tundra boundary, an expansion of tall shrub tundra, and a dominance shift from deciduous to evergreen boreal conifer forest over northern Eurasia were simulated. Ecosystems continued to sink carbon for the next few decades, although the size of these sinks diminished by the late 21st century. Hot spots of increased CH4 emission were identified in the peatlands near Hudson Bay and western Siberia. In terms of their net impact on regional climate forcing, positive feedbacks associated with the negative effects of tree-line, shrub cover and forest phenology changes on snow-season albedo, as well as the larger sources of CH4, may potentially dominate over negative feedbacks due to increased carbon sequestration and increased latent heat flux.

Modelling Tundra Vegetation Response to Recent Arctic Warming

The Arctic land area has warmed by > 1 °C in the last 30 years and there is evidence that this has led to increased productivity and stature of tundra vegetation and reduced albedo, effecting a positive (amplifying) feedback to climate warming. We applied an individual-based dynamic vegetation model over the Arctic forced by observed climate and atmospheric CO(2) for 1980-2006. Averaged over the study area, the model simulated increases in primary production and leaf area index, and an increasing representation of shrubs and trees in vegetation. The main underlying mechanism was a warming-driven increase in growing season length, enhancing the production of shrubs and trees to the detriment of shaded ground-level vegetation. The simulated vegetation changes were estimated to correspond to a 1.75 % decline in snow-season albedo. Implications for modelling future climate impacts on Arctic ecosystems and for the incorporation of biogeophysical feedback mechanisms in Arctic system models are discussed.

Simulation of the vegetation structure and function in a Malaysian tropical rain forest using the individual-based dynamic vegetation model SEIB-DGVM

An individual-based Dynamic Global Vegetation Model, the SEIB-DGVM, was adapted to a Malaysian tropical rain forest by incorporating formulas and parameters from a gap dynamics model, FORMIX3. After calibration, the model reconstructed forest structure (i.e., size structure, leaf area index, and woody biomass) and carbon fluxes (i.e., gross and net primary productivity) of a dipterocarp forest in Pasoh, Peninsular Malaysia. Sensitivity analysis demonstrated that the model was robust; forest structure and ecosystem functions moderately fluctuated due to changes in parameters and climatic environments. Sensitivity analysis also indicated that the success and decay of a dominant species group that monopolized the canopy layer greatly affected those of a less abundant, shade-intolerant group. This result indicates that even if environmental changes do not exhibit clear effects on dominant canopy species and/or whole forest structure, such changes may still substantially impact the biodiversity of subdominant species. In simulations without gap formation, woody biomass was overestimated and a shade-intolerant species group was eliminated. This finding indicates that incorporating gap formation into the individual-based model is essential for the appropriate simulation of forest biomass and biodiversity in this Malaysian tropical rain forest.