Abstracts – Session C3
Understanding the response of terrestrial ecosystems to climate change and rising atmospheric CO2 concentrations
C301 - ORAL-0247: Constrained projections of high northern latitudinal photosynthesis increase by satellite observations of vegetation greenness
Alexander Winkler1, 2, Victor Brovkin1, Ranga Myneni3
1Max Planck Institute for Meteorology, Hamburg, Germany 2International Max Planck Research School on Earth System Modelling, Hamburg, Germany 3Boston University, Boston, The United States of America
Satellite observations of the last three decades provide strong evidence that the Earth is greening. Especially in northern high latitudes, a substantial increase of the leaf area index (LAI), an indicator of greening, is observed. For these regions, it is assumed that plant growth benefits from higher temperature (radiative effect) as well as rising atmospheric CO2 concentration (CO2 fertilization effect). This greening trend, in terms of increasing LAI, is also simulated by various global ecosystem models. We also found a persistent greening trend analyzing historical simulations of Earth system models (ESM) participating in Phase 5 of the Coupled Model Intercomparison Project (CMIP5). However, a wide spread in magnitude of an associated increase of terrestrial gross primary production (GPP) among the ESMs is found, and thus contributes to pronounced uncertainties in projections of future climate change.
Here we demonstrate that the tight correlation between projections of GPP of high northern latitudinal ecosystems and their LAI sensitivity to both key environmental factors, temperature and CO2 concentration, opens up the possibility of an Emergent Constraint on projected plant photosynthesis. Combining this model-based linear relationship across the CMIP5 ensemble with the LAI trends in the long-term satellite records, we are able to substantially reduce the uncertainty of projections of vegetation growth increase for respective ecosystems.
C302 - ORAL-0049: Driving mechanisms and feedbacks of the land greening
1Environmental Sciences Division and Climate Change Science Institute, Oak Ridge National Laboratory, Oak Ridge, The United States of America
The land greening, characterized by the change of vegetation indices, has been documented to significantly increase over the past 3 decades, especially in the northern extratropical land surface. Drivers of this enhanced vegetation growth have been extensively investigated using multiple estimates from observed and modeled datasets and various statistical methods. Spatialtemporal changes of the main climate drivers (e.g., temperature, precipitation and radiation) have been widely accessed to modulate the variation in vegetation growth. The combined anthropogenic effects, particularly the rising atmospheric concentrations of well-mixed GHGs, was recently clearly identified as the dominant factor controlling this observed land greening. On the other hand, the enhancement of vegetation activity can potentially accelerate the photosynthetic removal of atmospheric CO2 and decrease the surface air temperature causing a negative forcing on climate system. Also, through the complicated biophysical feedbacks (e.g., the increased evapotranspiration and absorption of solar radiation), the land greening can intensify or diminish negative climate forcing induced via its biogeochemical feedbacks. Future vegetation growth and feedbacks, however, remain to be determined because of nonlinear human-ecosystem-climate interactions under global warming.
C303 - ORAL-0111: Endurance of larch forest ecosystems in eastern Siberia under warming trends
Hisashi Sato1, Hideki Kobayashi1, Go Iwahana2, Takeshi Ohta3
1Japan Agency for Marine-Earth Science and Technology (JAMSTEC), Yokohama, Japan 2International Arctic Research Center, University of Alaska Fairbanks, Alaska Fairbanks, The United States of America 3Nagoya University, Nagoya, Japan
We examined endurance of larch forests ecosystems on permafrost areas in eastern Siberia under forecasted warming trends. The team developed an integrated simulation model by combining a dynamic vegetation model (SEIB-DGVM) and a land physics model (NOAH-LSM). Simulations with the model demonstrated that plant productivity of the larch forests in eastern Siberia is likely to increase due to global warming effects despite rapid disappearance of near-surface permafrost for most part of the larch forests.
The area of larch forests in eastern Siberia is about ten times larger than the Japan's terrestrial area. The endurance of Siberian larch forests is considered to be dependent on near-surface permafrost as it helps retain soil water in near-surface soil layers by inhibiting its percolation to deeper soil layers. Existing land surface models have indicated a trend of surface permafrost decay due to rapid global warming around high latitudinal areas; however, its influences on the persistence of Siberian larch forest had been rarely examined.
Simulations with the model presented, however, that the flow of soil water from land surface to deeper layer would be limited to only several ten millimeters per year. Furthermore, the entire area of eastern Siberia will be moist with increase in annual precipitation by 100-300mm. As a result, it is more likely that plant productivity, biomass and leaf area index will increase in the entire larch forest.
The research team is further working on improvement of simulation accuracy by incorporating various factors such as plant stress caused by heat and excessive moisture, growth limitation due to nitrogen availability in soil, and impacts of topographic conditions on soil moisture and radiation availability. It is expected to lead to more sophisticated climate prediction at a local level and also on a global scale.
C304 - ORAL-0160: Fast and persistent soil carbon reductions in naturally-warmed sub-arctic grasslands
Ivan Janssens1, Niki Leblans1, Christopher Poeplau2, Dajana Radujkovic1, Tom Walker3, Sara Marañón-Jiménez4, Jennifer Soong1, Sara Vicca1, James Weedon5, Erik Verbruggen1, Josep Penuelas6, Andreas Richter3, Bjarni Sigurdsson7
1University of Antwerp; Research Group Plants and Ecosystems (PLECO), Antwerpen, Belgium 2Thuenen Institute of Climate-Smart Agriculture, Braunschweig, Germany 3University of Vienna, Vienna, Austria 4University of Granada, Granada, Spain 5Vrije Universiteit Amsterdam, Department of Ecological Science, Amsterdam, The Netherlands 6CREAF, Barcelona, Spain 7Agricultural University of Iceland, Bogarnes, Iceland
Global warming is expected to lead to transfers of carbon from soils to the atmosphere, thereby exacerbating the warming. This paper presents soil carbon stock changes observed along natural geothermal soil temperature gradients in Icelandic grasslands, five of which were at least 50 years old, while five others were established in 2008. These temperature gradients (between 0°C and +15°C) encompass the full range of IPCC projections for the northern region. There were large (4.1 % °C-1) and linear soil organic carbon reductions upon both short-term (5 years) and long-term (≥ 50 years) geothermal warming of northern grassland soils, but only in the topsoil (subsoils did not lose carbon). Central to this observed soil carbon reduction was a decline of physical stabilization of SOC in soil aggregates. The similar soil carbon reductions after 5 and ≥ 50 years of soil warming indicate that soil carbon reductions do not continue long after a temperature increase, despite the sustained warmer conditions. These results imply a fast and strong positive feedback to climate warming, which can be halted if climate warming itself is stopped.
C305 - ORAL-0043: Putting land ecosystem models on firmer foundations
Iain Colin Prentice1
1Imperial College London, ASCOT, United Kingdom
Current vegetation and land-surface models suffer from a severe and worsening lack of transparency and reproducibility. They are not actually improving either in consistency among models, or agreement with observations. Complexification, with the attendant multiplication of poorly known parameters, appears to be the only accessible route for model ‘improvement’. But this approach is liable to provide right answers for the wrong reasons, thus further degrading the models’ real predictive power.
One principal (but seldom discussed) reason for this dire situation is that key plant and ecosystem processes, including the most fundamental – photosynthesis, respiration and transpiration – have been almost universally misrepresented, above all through the widespread disregard of the adaptive nature of acclimation to spatial and temporal variations in the environment. I will argue that acclimation is not an unwelcome complication for models, but rather presents an opportunity for radical simplification through the systematic application and evaluation of eco-evolutionary optimality principles. These have a long history and yet have never become mainstream in model development. Given the unprecedented availability of data now on all relevant space and time scales, the time has come to revisit the foundations of land ecosystem modelling. I will show how optimality principles allow testable, and demonstrably accurate, predictions of plant and ecosystem properties ranging from photosynthetic capacity and stable carbon isotope composition to gross primary production as inferred from flux measurements and the responses of photosynthetic capacity, assimilation rates and stomatal conductance to CO2 enhancement as observed in Free Air Carbon dioxide Enrichment experiments. These advances have been achieved despite a dramatic reduction in the numbers of parameters to be estimated and, in particular, the abolition of most distinctions among plant functional types.
C306 - ORAL-0192: Convergence of canopy properties across climate gradients informs economic modelling of vegetation responses to
Mat Williams1, Quinn Thomas2, Molly Cavaleri3, J-F Exbrayat1, Luke Smallman1, Lorna Street1
1University of Edinburgh, Edinburgh, United Kingdom 2Virginia Tech, Blacksburg, VA, The United States of America 3Michigan Technological University, Houghton, The United States of America
Rates of photosynthesis and respiration are related to canopy properties, including leaf area index (LAI), and total canopy nitrogen (TCN). These canopy properties emerge from populations of leaves and their traits, which are linked to biodiversity, climate and disturbance. However, observations reveal a highly conservative LAI-TCN relationship across plant functional types and climate gradients from arctic to tropical. We argue that understanding the nature of the LAI-TCN connection can improve our ability to predict vegetation responses to global change. We use these observations to calibrate an economic model for canopy carbon balance, to determine marginal economic returns on C and N investment into canopies. The model defines the LAI-TCN relationships that are economically viable – i.e. generating net canopy C export - dependent on climate and leaf traits. Similar sets of LAI-TCN relationships, matching observations, can arise from different but constrained combinations of leaf traits, consistent also with observed trade-offs in traits in the leaf economics spectrum. The modelling identifies knowledge gaps for traits controlling leaf respiration, and informs the development of earth system models.
C307 - ORAL-0063: Development of an Environment-related Establishment Scheme for a DGVM
Xiang Song1, Xiaodong Zeng1, Jia-wen Zhu1, Pu Shao1
1International Center for Climate and Environment Sciences, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China
Environmental changes not only shift vegetation distribution, but also alter their abundance by determining seedling establishment and success. However, most ecosystem models mainly focus on the impacts of climate or environment on biogeophysics, ignoring their roles in the population dynamics of ecological communities. In this work, a new establishment scheme for introducing soil water as a function rather than a threshold was developed and validated, using version 1.0 of the IAP-DGVM as a test bed. Compared with the original scheme, the new scheme significantly improved simulations of tree population density and fractional coverage, especially in the peripheral areas of forests and transition zones. Furthermore, the relative biases in the global simulated areas of tree, shrub, grass and bare soil were reduced from 34.3% to 4.8%, from 27.6% to 13.1%, from 55.2% to 9.2%, and from 37.6% to 3.6%, respectively.
C308 - ORAL-0153: Tropical forest response to elevated CO2: Model-based hypotheses for the AmazonFACE experiment
Katrin Fleischer1, Anja Rammig1, David M. Lapola2
1Technical University of Munich, Munich, Germany 2UNICAMP, Campinas, Brazil
Direct evidence for the existence and strength of the tropical CO2 fertilization effect is scarce, being a major impediment for prediction reliability of global Earth System Models. The implications of the tropical CO2 effect are far-reaching, as it strongly influences the global carbon and water cycle, and hence future global climate. In the case of the Amazon tropical rainforest, it is believed to potentially stabilize the ecosystem as a whole and prevent major forest dieback due to climate change. The ecosystem scale experiment AmazonFACE proposes to address these uncertainties, being the first tropical FACE, as well as the first FACE in an old-growth, highly diverse tropical rainforest. A priori model-based hypotheses for the experiment are derived from a set of ecosystem models (n=12) and presented here. Model simulations identify key uncertainties in our understanding of limiting processes and derive model-based hypotheses of expected ecosystem changes that can directly be tested during the experiment, aiming to maximize the scientific output of the experiment. Ambient model simulations compare satisfactorily with in situ measurements of ecosystem carbon fluxes, as well as carbon, nitrogen and phosphorus stocks. Models consistently predict an increase in photosynthesis with elevated CO2, which lowers over time due to developing limitations. The conversion of enhanced photosynthesis into biomass, and hence ecosystem carbon sequestration, varied strongly among the models due to different assumptions on nutrient limitation. Models with flexible allocation schemes consistently predicted a heightened investment in belowground structures to alleviate nutrient limitation, in turn accelerating turnover rates of soil organic matter. The models diverge on the prediction for carbon accumulation after 10 years of elevated CO2, mainly due to contrasting assumptions in their phosphorus cycle representation. These differences define the expected response ratio to elevated CO2 at the AmazonFACE site and identify priorities for experimental work and model development.
C309 - ORAL-0195: Combining CO2 enrichment experiments with model-data fusion approaches to guide ecosystem model process development
Luke Smallman1, David Milodowski1, J-F Exbrayat1, Mat Williams1
1School of GeoSciences, University of Edinburgh, Edinburgh, United Kingdom
The carbon (C) and water cycles of terrestrial ecosystems are integrally linked and are a requirement for understanding many ecosystem responses (e.g. photosynthate allocation patterns) to perturbation of their environment (e.g. climate change). Data rich field scale manipulation experiments, such as Duke FACE, offer the opportunity to quantify the ability of models to capture changes in ecosystem dynamics in response to large shifts in environmental conditions. Recent model inter-comparisons have demonstrated that many current model structures and parameterisation are unable to accurately represent shifts in ecosystem carbon partitioning in response to CO2 enrichment. Necessitating the identification of both missing and poorly described processes in ecosystem models. Here we combine a simple terrestrial ecosystem model representing C and water with a model-data fusion (MDF) approach and time varying carbon stock information from the Duke FACE experiment. The MDF retrieves an ensemble of ecosystem traits (e.g. C allocation and root structural information) which we compare with field derived estimates. Using a simple framework allows for rapid testing of multiple hypothesised model formulations to allow identification of both the most likely process representation but also uncertainties surrounding poorly known and difficult to observe ecosystem traits.
C310 - ORAL-0360: Modelling the sensitivity of the Zambezi teak forests to climate change, Zambia
Justine Ngoma1, 2, Bart Kruijt1, Eddy Moors1, 3, James H. Speer4, Royd Vinya2, Rik Leemans1
1Wageningen University and Research, Wageningen, The Netherlands 2The copperbelt University, Kitwe, Zambia 3VU University Amsterdam , Armsterdam, The Netherlands 4Indiana State University, Terre Haute, Indiana, The United States of America
Understanding the sensitivity of the forests to changes in temperature and rainfall is important following the tree’s mitigating role of climate change. Forest’s current response to different climatic conditions provides insight on the possible response of these forests to the projected future climate change. The aim of this study was to determine the sensitivity of the Zambezi teak forests to climate change along a rainfall gradient. We adapted a dynamic vegetation model (Lund Potsdam Jena General Ecosystem Simulator, LPJ GUESS model) and ran it at Kabompo, Namwala and Sesheke sites in Zambia using climate data from the local weather stations. We also used local parameter values for allometric relations, Specific Leaf Area, Maximum crown area, wood density, and C/N ratios. Modelled biomass was compared to measured biomass (15 hectares) at each sites. We found significant positive correlations between modelled and observed biomass with the highest correlation observed at the wetter Kabompo (r2 =0.71, p <0.001) compared to Namwala (r2 =0.34, p <0.02) and the drier Sesheke (R2 =30, p <0.02) sites . A 10% rainfall reduction reduced biomass by 7% at the wetter Kabompo site and by 3% at the drier Sesheke site. These results indicate that the effects of the projected rainfall decrease for southern Africa, as projected by IPCC, would reduce biomass of the Zambezi teak forests and the reduction would be more at the wetter Kabompo site than at the drier Sesheke site.
C311 - ORAL-0404: What happens to CO2 after photosynthesis and how this process is linked to ecosystem's productivity?
Nicolas Raab1, Alexander Shenkin1, Christopher Doughty2, Hugh Dickinson3, Yadvinder Malhi1
1Environmental Change Institute, University of Oxford, Oxford, United Kingdom 2Northern Arizona University, Flagstaff, Arizona, The United States of America 3Plant Sciences, University of Oxford, Oxford, United Kingdom
It has been long hypothesised that the increase in atmospheric CO2 concentration will enhance primary plant productivity around the globe by facilitating carbon diffusion from the surrounding air to the carboxylation site in leaves. However, experimental data and theoretical background have suggested that ecosystem productivity is not only limited by environmental factors driving photosynthesis, but by plant’s carbon growing needs. In other words, carbon is not “pushed” from the atmosphere to the leaf, but “pulled” by growth and metabolic processes, thus representing plants’ carbon demand. When carbon supply from leaves exceeds carbon demand, an imbalance is reached, producing a cascade of signals that end up affecting photosynthesis, hence ecosystems productivity. There has been a lack of empirical studies on the effect of sugar imbalance on photosynthesis; here we present the first such study for tropical and temperate species.
As big and strong runners are better at sprinting compared to small and lean athletes that perform better on long distance courses, different traits in plants yield different growing strategies, therefore different capacities to scrub CO2 from the atmosphere. Our results show that a combination of these traits, such as leaves' investement in a vein network and specialised cells, which pump sugars from the mesophyll into these veins to be later translocated across the plant, can explain these different growing strategies and tell us what kind of plants will be better adapted at coping with increasing CO2 atmospheric concentration.
C312 - ORAL-0373: Modelling future nutrient constraints on Amazon forest productivity
Michelle Johnson1, Emanuel Gloor, David Galbraith, Tim Baker, Sarah Batterman
1University of Leeds, Leeds, United Kingdom
The response of net primary productivity (NPP) to increasing atmospheric CO2 concentrations contributes the largest uncertainty to predicting future carbon-climate feedbacks. In temperate forests, FACE experiments predict increases in productivity under elevated atmospheric CO2 concentrations. However, the response of tropical forests to rising CO2 concentrations is uncertain. Nutrient availability, specifically phosphorus is particularly important in the tropics and may constrain the predicted CO2 fertilization effect on productivity. Although tropical soils are undoubtedly depleted in mineral-derived nutrients such as phosphorus, it is still unclear to what extent tropical forest growth is limited by phosphorus supply. Field observations indicate that above-ground woody productivity across the Amazon is not related to available soil phosphorus. However, a positive relationship between productivity and total soil phosphorus is observed. This suggests that processes operating on less-readily available pools (such as sorbed P) and over longer timescales are important for supplying P to vegetation.
Here we introduce a phosphorus and nitrogen cycle model for the Amazon that can simulate these processes. We model the major pools of phosphorus and nitrogen in the soil and vegetation and combine the model with soil and vegetation observations from permanent sample plots across the basin. We predict whether the future demand for extra nutrients can be met and the associated carbon cost of acquiring nutrients. We indicate in which forests growth is likely to be constrained by either phosphorus or nitrogen and which forests are ably to supply and cycle the extra nutrients needed. We also investigate the importance of plant acquisition strategies such as biological nitrogen fixation and phosphatase production for overcoming nutrient limitation.
C313 - ORAL-0263: Mycorrhizal N acquisition strategies determine ecosystem responses to elevated CO2 in the GFDL global land model
Benjamin Sulman1, Edward Brzostek2, Sergey Malyshev1, Elena Shevliakova3
1Princeton University, Princeton, NJ, The United States of America 2West Virginia University, Morgantown, WV, The United States of America 3Geophysical Fluid Dynamics Laboratory, Princeton, NJ, The United States of America
Nutrients constrain the ability of terrestrial ecosystems to sequester CO2. While Earth System Models (ESMs) include intricate representations of the photosynthetic machinery of carbon fixation in plants, they lack sophisticated representations of the nutrient acquisition machinery necessary to sustain photosynthesis and growth, calling into question their ability to predict the future land C sink. We conducted simulations using a new model of terrestrial carbon (C) and nitrogen (N) cycling that uses a return on investment framework to simulate N acquisition via fixation of N2 from the atmosphere, scavenging of inorganic N from soil solution, or mining of organic N from soil organic matter (SOM). We show that these strategies drive divergent C cycle responses to elevated CO2. At the ecosystem scale, mining of organic N sustained enhanced vegetation growth under elevated CO2 for decades. By contrast, reliance on inorganic N drove increases in soil N immobilization, ultimately trapping N in inaccessible physically protected soil pools and leading to N limitation of plant growth. These results agreed with contrasting observed responses to elevated CO2 at the Duke and Oak Ridge National Lab FACE experiments. In global simulations, the return on investment framework successfully reproduced global patterns of N fixation and soil N acquisition via inorganic and organic strategies. Because inorganic N availability did not increase in response to higher N demands, elevated CO2 drove a widespread shift from inorganic N toward organic N acquisition at global scales. Collectively, our results indicate that the ability of the land C sink to mitigate atmospheric CO2 levels is tightly coupled to the distribution of N acquisition strategies and their capacity to change over time.
C314 - ORAL-0053: Assessing the condition of forests exposed to climate change and nitrogen deposition by use of monitoring and modelling data
Winfried Schröder1, Stefan Nickel1
1Chair of Landscape Ecology, University of Vechta, Vechta, Germany
By example of Germany, a comprehensive and spatial explicit methodology for evaluating forest condition was developed. The approach integrates data on vegetation, chemical and physical soil condition as well as on climate change and atmospheric deposition of nitrogen. Key component for evaluating forst integrity is a classification of forest containing information on six ecological functions. Respective data covering 1961-1990 was regarded as reference. The assessment of ecological integrity relies on comparing current and future forest condition with the respective reference. Whilst current forest condition was quantified by measurements, potential future developments were projected by Classification and Regression Trees (CART) at the national level and pedo-chemical soil modelling and data from a regional climate change model for selected forest stands.
The current forest types were related to geo-data (elevation a.s.l., soil texture, air temperature, humidity, evapotranspiration, precipitation 1961-1990) by CART. The relations determined by this were applied to the above mentioned geo-data and then used to map the spatial pattern of forest type clusters for 1961-1990. Then, the climate data 1961-1990 were replaced by results from a regional climate model for 1991-2010, 2011-2040, and 2041-2070. The relations identified for the reference were then applied to the modified datasets so that for each period one map of ecosystem type clusters was produced and evaluated with regard to the development of areal coverage of ecosystem clusters across time. This evaluation of structural aspects of ecological integrity in terms of bio-geographical coverage on the national level was added by projecting potential future values of indicators for ecological functions at site-level. This was achieved by using the Very Simple Dynamics soil modelling technique using the above mentioned climate data and two scenarios of atmospheric nitrogen deposition as input. The results were compared to the reference and enabled evaluating site-specifically ecosystem integrity across time.
C315 - ORAL-0177: The impact of heat waves and drought on emissions of monoterpenes from Douglas Fir
Ines Bamberger1, Nadine K Ruehr1, Andrea Ghirardo2, Andreas Gast1, Georg Wohlfahrt3, Almut Arneth1
1KIT IMK-IFU, Garmisch-Partenkirchen, Germany 2Institute of Biochemical Plant Pathology, EUS, Helmholtz Centre Munich, Neuherberg, Germany 3Institute of Ecology, University of Innsbruck, Innsbruck, Austria
Forests are the major emitters of biogenic volatile organic compounds of which monoterpenes are quantitatively the second most important compound group emitted from trees. Under future climatic conditions trees will be exposed to more extreme weather conditions, including heat waves and drought, which is expected to alter emission and composition of monoterpenes with potential consequences for air quality and atmospheric chemistry. Yet, the knowledge on the magnitude of monoterpene emissions and composition changes in response to heat waves, especially in combination with drought is limited.
Monoterpene emissions and composition of Douglas fir during heat stress and in combination with drought were observed by proton-transfer-reaction mass spectrometry (PTR-MS) and by gas chromatography mass-spectrometry (GC-MS). The heat waves lasted for 14 days (ca. 10°C above ambient) followed by 7 days of recovery. During severe heat stress or heat stress in combination with drought, Douglas fir trees showed a strong emission burst of monoterpenes that resulted in two orders of magnitude larger daytime emissions compared to values measured before stress. Towards the end of the first heat wave and during the second heat wave the emissions declined and were just moderately higher than the control, suggesting that a major fraction of the emissions originated from storage pools which depleted quickly after being affected by high temperatures (> 45°C) at the onset of the stress. Alpha- and beta-pinene dominated the monoterpene emissions. The relative contribution of alpha-pinene to total monoterpenes decreased during stress compared to before stress conditions, while the contribution of beta-pinene did not change significantly. This observation suggests that alpha-pinene is partially synthesized de-novo during heat and heat-drought stress, while beta-pinene originated mainly from storage-pools. Generally, heat and heat-drought stress caused very similar responses on monoterpene emission patterns of Douglas Fir trees.
C316 - ORAL-0329: Examining the terrestrial mechanisms behind the O3/temperature relationship
William Porter1, 2, Colette Heald2
1UC Riverside, Riverside, The United States of America 2MIT, Cambridge, The United States of America
Tropospheric ozone (O3) pollution levels have widely observed correlations to daytime surface temperatures, especially in highly polluted regions. This correlation is nonlinear and results from a variety of temperature dependent mechanisms related to O3 precursor emissions, lifetimes, and reaction rates, making the reproduction of temperature sensitivities – and the projection of associated human health risks – a complex problem.
Here we use the chemical transport model GEOS-Chem alongside observations to explore two terrestrial mechanisms (biogenic VOC emissions and soil NOx emissions) that contribute to the summertime O3 climate penalty in the United States and Europe. By removing the modeled temperature dependency of these two emission sources, we quantify the contribution of each mechanism to the overall correlation. We then use commonality analysis to examine the contribution of other meteorological influences, categorizing each location by relative contribution of temperature-dependent mechanisms and covarying meteorology.
C317 - POSTER-0099: High temperatures and the terrestrial carbon cycle: assessing the drivers of Net Ecosystem Exchange under
hot-wet conditions in a semi-arid savanna
Caitlin Ransom1, Sally Archibald1
1University of the Witwatersrand, Johannesburg, South Africa
Terrestrial ecosystems play an important role in the global carbon cycle as well as regulating climate change, but there is uncertainty about the future of terrestrial carbon cycle dynamics with the increased temperature. The influence of temperature on carbon fluxes, Net Ecosystem Exchange (NEE), in semi-arid savanna’s, when soil moisture was not limiting was explored using eddy covariance data from Skukuza, Kruger National Park, South Africa. Previous studies conducted in Skukuza have found soil moisture to be a strong influence of carbon fluxes, but not temperature, this may be because the response to temperature was confounded by soil moisture. In this study, diurnal summer time NEE was grouped into temperature classes to compare the response of NEE to temperature. Soil from the Skukuza site was incubated at different temperatures and the accumulated carbon dioxide concentrations were measured; to elucidate the soil respiration response of NEE. Summer diurnal NEE was found to decrease with increasing temperatures up until 26⁰C where after NEE increased. Soil respiration increases with temperature until 35⁰C, then levels off and shows no sign of decreasing at high temperatures (50⁰C). Increases in soil respiration are driving the increase in NEE at high temperatures; as at Skukuza NEE only increases after 25⁰C well before 35⁰C where photosynthesis is compromised. Respiration is greatest above 35⁰C therefore photosynthesis must still be occurring at high temperatures, or NEE would be positive. Understanding how these carbon fluxes respond to temperature is important for our understanding of the climate-carbon cycle feedbacks.
C318 - POSTER-0126: Contrasting the response of gross primary productivity to climate change in northern and southern China during 1982-2015
Shaoqiang Wang1, Miaomiao Wang1, Junbang Wang1, Hao Yan2
1Key Lab of Ecosystem Network Observation and Modelling, Institute of Geographic Sciences and Natural Resources Research, Beijing, China 2National Climate Center, Beijing, China
Climate change has affected the terrestrial ecosystem’s structure and function, and China is one of the most sensitive regions to climate change in world. Accuracy estimation terrestrial ecosystem productivity in China is critical for the response of ecosystem carbon budget to climate change. In this study, we utilize Boreal Ecosystem Productivity simulator model (BEPS), Global Production Efficiency Model (GLOPEM) and Terrestrial Ecosystem Carbon Flux model (TEC) to simulate the terrestrial ecosystem’s gross primary productivity (GPP) in China during 1982-2015 by combining remote sensing data. And we applied Mann-Kendall method to analyze the spatial and temporal variations of GPP in China’s terrestrial ecosystem and explore its control mechanism. Furthermore, we analyzed the trend of GPP in northern and southern China and their response to climate change. The results indicated that the mean of annual GPP in whole China was about 5.99±0.21 PgC, and southern China was accounted for 60%. The terrestrial GPP was significantly increasing (0.02Pg C yr-1) during 1982-2015 in China, and the increasing trend in southern China (9.17TgC yr-1) was higher than northern China (5.69TgC yr-1). And the significantly increased areas were generally detected over the forests and cropland regional. For every degree increase in temperature, GPP increased by 0.22Tg yr-1 in southern China, while northern China increased by 0.14TgC yr-1, respectively.
C319 - POSTER-0223: Ground-based high resolution near-infrared heterodyne measurements of atmospheric carbon dioxide column
Jingjing Wang1,Tu Tan1, Yanan Cao1, Yang Dong1, Weidong Chen2, Xiaoming Gao1, *
1 Laboratory of Atmospheric Physico-Chemistry, Anhui Institute of Optics & Fine Mechanics, Chinese Academy of Sciences, Hefei 230031, China; 2 Laboratoire de Physicochimie de l’Atmosphère, Université du Littoral Côte d’Opale, 189A, Av, Maurice Schumann, 59140, Dunkerque, France
We present our recent development of a novel laser heterodyne radiometer (LHR) operating in the near-infrared (NIR) near 1.573 μm for ground-based in situ measurements of carbon dioxide (CO2) in the atmospheric column. The LHR instrument is mainly consisted of an optical fiber solar tracker, a core optical system for mixing laser beam with sunlight, and a signal processing circuit with a bandwidth of 45 MHz for heterodyne measurement of CO2 absorption spectrum.
C320 - POSTER-0398: Predicting dominant plant traits and community assembly in a demographic dynamic global vegetation model designed
for Earth System Models
Ensheng Weng1, Nancy Kiang2
1Center for Climate Systems Research, Columbia University, New York, NY, The United States of America 2NASA Goddard Institute for Space Studies, New York, NY, The United States of America
Current Earth System Models (ESMs) lack the representation of plant community dynamics in their land models, which hampers their ability to capture long-term ecosystem carbon dynamics. Consequently, they are unable realistically to predict transient changes in vegetation and the feedbacks between the terrestrial carbon cycle and climate at decadal to century scales. We have developed a theoretical model of an "evolutionarily stable strategy" (ESS) based on plant competition and the feedbacks between plant traits and biogeochemical cycles, which then predicts dominant vegetation types by identifying the most competitive combinations of plant traits. We identify wood density, leaf mass per area, and height allometry as major independent traits that express the ESS for plants to compete for light, water, and nutrients. This scheme is implemented in the Ent Terrestrial Biosphere Model (Ent TBM), the dynamic global vegetation model (DGVM) coupled to the NASA Goddard Institute for Space Studies (GISS) ESM. We perform simulations of ensembles of trait values to see which sets of values ultimately dominate in the course of community assembly and the competition of individuals. The optimal trait values thus predict the amount of ecosystem carbon storage, vegetation structure, and dynamics. We show preliminary tests of this model on competition of broadleaf deciduous and evergreen needleleaf forests to reproduce their transition zone in North America. Because it incorporates the mechanisms of competition and trait shifts, the model has the potential to help to understand the impacts of climate change on terrestrial ecosystems and address the scientific questions about ecosystem vulnerability and resilience. Future development will investigate responses to climate change and disturbances (e.g, fires and human activities) at multi-decadal time scales.