Karin Rebel

Sensitivity analysis of a wetland methane emission model based on temperate and arctic wetland sites

Modelling of wetland CH4 fluxes using wetland soil emission models is used to determine the size of this natural source of CH4 emission on local to global scale. Most process models of CH 4 formation and soil-atmosphere CH 4 transport processes operate on a plot scale. For large scale emission modelling (regional to global scale) upscaling of this type of model requires thorough analysis of the sensitivity of these models to parameter uncertainty. We applied the GLUE (Generalized Likelihood Uncertainty Analysis) methodology to a well-known CH 4 emission model, the Walter-Heimann model, as implemented in the PEATLANDVU model. The model is tested using data from two temperate wetland sites and one arctic site. The tests include experiments with different objective functions, which quantify the fit of the model results to the data. The results indicate that the model 1) in most cases is capable of estimating CH4 fluxes better than an estimate based on the data avarage, but does not clearly outcompete a regression model based on local data; 2) is capable of reproducing larger scale (seasonal) temporal variability in the data, but not the small-scale (daily) temporal variability; 3) is not strongly sensitive to soil parameters, 4) is sensitive to parameters determining CH 4 transport and oxidation in vegetation, and the temperature sensitivity of the microbial population. The GLUE method also allowed testing of several smaller modifications of the original model. We conclude that upscaling of this plot-based wetland CH4 emission model is feasible, but considerable improveCorrespondence to: J. van Huissteden (ko.van.huissteden@geo.falw.vu.nl) ments of wetland CH4 modelling will result from improvement of wetland vegetation data.

Vegetation-mediated feedback in water, carbon, nitrogen and phosphorus cycles

Since the industrial revolution, industry, traffic and the manufacture and application of nitrogenous fertilizers have increased carbon dioxide emissions and accelerated the nitrogen (N) cycle. The combined effects of a warming climate, CO2 fertilization, land-use change and increased N availability may be responsible for primary productivity increases in many parts of the world. Enhanced productivity may lead to shifts in albedo and transpiration, which feed back to the water cycle through heat fluxes and precipitation. Plants may also respond to elevated CO2 by closing their stomata or by structurally adapting their stomatal density and size, which potentially diminishes transpiration. Intensification of agriculture has also led to an increase in both nitrogenous (N) and phosphorus (P) fertilization. The combined effect of atmospheric N deposition and P fertilization has distorted the balance between N and P availability in many ecosystems. The active role of plants in accessing nutrients from the soil may trigger switches in nutrient availability, triggering shifts in plant productivity and species composition in these ecosystems and therefore also in the carbon (C) cycle. In response to global change, the above plant responses may influence each other positively or negatively and may impact on the elemental cycles of C, N and P and the water cycle. We are only beginning to understand how these four cycles interact, the role of plant processes and vegetation in these interactions, and the net outcome for plant competition, vegetation distribution, landscape development and directions of global change. In this paper we have integrated a number of recent research findings into known relationships that together elucidate interactions between these cycles through vegetation, and could potentially have unexpected effects on landscapes and larger-scale systems (continental, global). These interactions include processes operating at very distinct temporal and spatial scales, in which terrestrial ecosystems and their spatial organization in the landscape are key. We argue that to better understand the effects of changes in land cover and land use on biogeochemical and biogeophysical fluxes, it is necessary to account for feedbacks via vegetation and how these interfere with elemental cycles. Finally, we suggest directions for further research to fill the current knowledge gaps.

Low historical nitrogen deposition effect on carbon sequestration in the boreal zone

Nitrogen (N) cycle dynamics and N deposition play an important role in determining the terrestrial biospheres carbon (C) balance. We assess global and biome-specific N deposition effects on C sequestration rates with the dynamic global vegetation model LPJ-GUESS. Modeled CN interactions are evaluated by comparing predictions of the C and CN version of the model with direct observations of C fluxes from 68 forest FLUXNET sites. N limitation on C uptake reduced overestimation of gross primary productivity for boreal evergreen needleleaf forests from 56% to 18%, presenting the greatest improvement among forest types. Relative N deposition effects on C sequestration (dC/dN) in boreal, temperate, and tropical sites ranged from 17 to 26kgCkgN(-1) when modeled at site scale and were reduced to 12-22kgCkgN(-1) at global scale. We find that 19% of the recent (1990-2007) and 24% of the historical global C sink (1900-2006) was driven by N deposition effects. While boreal forests exhibit highest dC/dN, their N deposition-induced C sink was relatively low and is suspected to stay low in the future as no major changes in N deposition rates are expected in the boreal zone. N deposition induced a greater C sink in temperate and tropical forests, while predicted C fluxes and N-induced C sink response in tropical forests were associated with greatest uncertainties. Future work should be directed at improving the ability of LPJ-GUESS and other process-based ecosystem models to reproduce C cycle dynamics in the tropics, facilitated by more benchmarking data sets. Furthermore, efforts should aim to improve understanding and model representations of N availability (e.g., N fixation and organic N uptake), N limitation, P cycle dynamics, and effects of anthropogenic land use and land cover changes. (Less)

Terrestrial nitrogen cycling in Earth system models revisited

Understanding the degree to which nitrogen (N) availability limits land carbon (C) uptake under global environmental change represents an unresolved challenge. First-generation ‘C-only’ vegetation models, lacking explicit representations of N cycling, projected a substantial and increasing land C sink under rising atmospheric CO2 concentrations. This prediction was questioned for not taking into account the potentially limiting effect of N availability, which is necessary for plant growth (Hungate et al., 2003). More recent global models include coupled C and N cycles in land ecosystems (C–N models) and are widely assumed to be more realistic. However, inclusion of more processes has not consistently improved their performance in capturing observed responses of the global C cycle (e.g. Wenzel et al., 2014). With the advent of a new generation of global models, including coupled C, N, and phosphorus (P) cycling, model complexity is sure to increase; but model reliability may not, unless greater attention is paid to the correspondence of model process representations and empirical evidence. It was in this context that the ‘Nitrogen Cycle Workshop’ at Dartington Hall, Devon, UK was held on 1–5 February 2016. Organized by I. Colin Prentice and Benjamin D. Stocker (Imperial College London,UK), the workshopwas funded by theEuropeanResearchCouncil, project ‘Earth systemModelBias Reduction and assessing AbruptClimate change’ (EMBRACE).We gathered empirical ecologists and ecosystem modellers to identify key uncertainties in terrestrial C–N cycling, and to discuss processes that are missing or poorly represented in current models.

Exploring the use of vegetation indices to sense canopy nitrogen to phosphorous ratio in grasses

Abstract Reduced availability of plant nutrients such as nitrogen (N) and phosphorous (P) has detrimental effects on plant growth. Plant N:P ratio, calculated as the quotient of N and P concentrations, is an ecological indicator of relative N and P limitation. Remote sensing has already been widely used to detect plant traits in foliage, particularly canopy N and P concentrations and could be used to detect canopy N:P faster and at lower cost than traditional destructive methods. Despite the potential opportunity of applying remote sensing techniques to detect canopy N:P, studies investigating canopy N:P remote detection are scarce. In this study, we examined if vegetation indices developed for canopy N or P detection can also be used for canopy N:P detection. Using in situ spectrometry, we measured the reflectance of a common grass species, Yorkshire fog ( Holcus lanatus L.), grown under different nutrient ratios and levels. We calculated 60 VIs found in literature and compared them to optimized VIs developed specifically for this study. The VIs were calculated using both the original narrow band spectra and the spectra resampled to the band properties of six satellite sensors (MSI – Sentinel 2, OLCI – Sentinel 3, MODIS – Terra/Aqua, OLI – Landsat 8, WorldView 4 and RapidEye) to investigate the influence of bandwidths and band positions. The results showed that canopy N:P was significantly related to both existing VIs (r 2  = 0.16 - 0.48) and optimized VIs (r 2  = 0.59 – 0.72) with correlations similar to what was observed for canopy N or canopy P. Existing VIs calculated with MSI and OLI sensors bands showed higher correlation with canopy N:P compared to the other sensors while the correlation with optimized VIs was not affected by the differences in sensors’ bands. This study might lead to future practical applications using in situ reflectance measurements to sense canopy N:P in grasslands.

Integrating Mycorrhizas Into Global Scale Models : A Journey Toward Relevance in the Earth's Climate System

Abstract Associations between plants and mycorrhizal fungi are ubiquitous in nature and are among the most important trophic interactions affecting ecosystem services and global change. Despite their evolutionary history and current ecological importance, mycorrhizal dynamics have rarely been included in the process-based ecosystem models commonly used to predict vegetation responses to, and mediation of, climate change. Here we provide a framework for incorporating mycorrhizal dynamics into global models, building on well-developed theories of optimal allocation and stoichiometrically-explicit plant–microbe interactions. First we discuss the strengths of existing model frameworks used at the ecosystem scale that can inform global model development. Then we identify mycorrhizal functions that are critical to model at the global scale and highlight how using mycorrhizal fungi as trait integrators may be an important first step in integrating mycorrhizae into global models. We conclude by describing the unique challenges that modeling mycorrhizae at global scales presents.