Search

Northern peatlands store 300–600 Pg C, of which approximately half are underlain by permafrost. Climate warming and, in some regions, soil drying from enhanced evaporation are progressively threatening this large carbon stock. Here, we assess future CO2 and CH4 fluxes from northern peatlands using five land surface models that explicitly include representation of peatland processes. Under Representative Concentration Pathways (RCP) 2.6, northern peatlands are projected to remain a net sink of CO2 and climate neutral for the next three centuries. A shift to a net CO2 source and a substantial increase in CH4 emissions are projected under RCP8.5, which could exacerbate global warming by 0.21°C (range, 0.09–0.49°C) by the year 2300. The true warming impact of peatlands might be higher owing to processes not simulated by the models and direct anthropogenic disturbance. Our study highlights the importance of understanding how future warming might trigger high carbon losses from northern peatlands.

Africa is largely influenced by fires, which play an important ecological role influencing the distribution and structure of grassland, savanna and forest biomes. Here vegetation strongly interacts with climate and other environmental factors, such as herbivory and humans. Fire-enabled Dynamic Global Vegetation Models (DGVMs) display high uncertainty in predicting the distribution of current tropical biomes and the associated transitions, mainly due to the way they represent the main ecological processes and feedbacks related to water and fire. The aim of this study is to evaluate the outcomes of two state-of-the–art DGVMs, LPJ-GUESS and JSBACH, also currently used in two Earth System Models (ESMs), in order to assess which key ecological processes need to be included or improved to represent realistic interactions between vegetation cover, precipitation and fires in sub-Saharan Africa. To this end, we compare models and remote-sensing data, analyzing the relationships between tree and grass cover, mean annual rainfall, average rainfall seasonality and average fire intervals, using generalized linear models, and we compare the patterns of grasslands, savannas, and forests in sub-Saharan Africa. Our analysis suggests that LPJ-GUESS (with a simple fire-model and complex vegetation description) performs well in regions of low precipitation, while in humid and mesic areas the representation of the fire process should probably be improved to obtain more open savannas. JSBACH (with a complex fire-model and a simple vegetation description) can simulate a vegetation-fire feedback that can maintain open savannas at intermediate and high precipitation, although this feedback seems to have stronger effects than observed, while at low precipitation JSBACH needs improvements in the representation of tree-grass competition and drought effects. This comparative process-based analysis permits to highlight the main factors that determine the tropical vegetation distribution in model

Large uncertainty exists in model projections of the land carbon (C) sink response to increasing atmospheric CO2. Free-Air CO2 Enrichment (FACE) experiments lasting a decade or more have investigated ecosystem responses to a step change in atmospheric CO2 concentration. To interpret FACE results in the context of gradual increases in atmospheric CO2 over decades to centuries, we used a suite of seven models to simulate the Duke and Oak Ridge FACE experiments extended for 300 years of CO2 enrichment. We also determine key modeling assumptions that drive divergent projections of terrestrial C uptake and evaluate whether these assumptions can be constrained by experimental evidence. All models simulated increased terrestrial C pools resulting from CO2 enrichment, though there was substantial variability in quasi-equilibrium C sequestration and rates of change. In two of two models that assume that plant nitrogen (N) uptake is solely a function of soil N supply, the net primary production response to elevated CO2 became progressively N limited. In four of five models that assume that N uptake is a function of both soil N supply and plant N demand, elevated CO2 led to reduced ecosystem N losses and thus progressively relaxed nitrogen limitation. Many allocation assumptions resulted in increased wood allocation relative to leaves and roots which reduced the vegetation turnover rate and increased C sequestration. In addition, self-thinning assumptions had a substantial impact on C sequestration in two models. Accurate representation of N process dynamics (in particular N uptake), allocation, and forest self-thinning is key to minimizing uncertainty in projections of future C sequestration in response to elevated atmospheric CO2.

Global scale dynamic vegetation models simulate the global C cycle and atmosphere-vegetation interactions, an essential component in the global climate system. The important role of the N-cycle in determining fluxes of carbon and climate dynamics is unequivocally evident. The current generation of ecosystem models include progressively carbon-nitrogen interactions but vary in their representation of important processes. We contribute to this development by evaluating predictions of the newly implemented N-cycle in LPJ-GUESS with direct observations. Modelled C-fluxes and vegetation characteristics in LPJ-GUESS will be compared to EC-data for 75 FLUXNET forest sites. We assess the inclusion of the N-cycle in LPJ-GUESS by comparing the C-only with the CN-version of the model. Further we compare simulated C and N pool sizes and key biological characteristics (biomass, foliar N and LAI) between the model versions, and compare to site data. Site-specific parameterization of LPJ-GUESS include local meteorology, plant functional type and time of last major disturbance of the sites. The inclusion of local conditions allows predicting C-fluxes and pool sizes with greater accuracy. We hypothesize that the inclusion of the N-cycle improves model predictions. The benefit of including the N-cycle is expected to differ between forest types, ecosystem types and/or climate regions. This effort will allow identifying the conditions in which dynamic global vegetation models potentially under- or overestimate C fluxes when ignoring the N-cycle dynamics and interactions. The results will contribute to the development of ecosystem models with fully coupled carbon-nitrogen cycles, and a better understanding of interactions between the C and N cycle in forest ecosystems. These are essential steps towards better and more reliable predictions of the present and future global C and N cycle in times of global change.