B43C-01 INVITED 13:40h
A Method for Measuring Subcanopy CO2 Advection
Underestimation of nocturnal CO2 respiration under calm conditions remains an unsolved problem at many forest flux stations, and several groups are currently investigating the direct measurement of horizontal advection of CO2. This presentation will describe a systematic, relatively low-cost methodology developed to determine whether horizontal mean transport of CO2 accounts for the missing CO2 at the Harvard Forest (Petersham, MA). This methodology includes the characterization of subcanopy motions, determining the appropriate size of the subcanopy network required to make the measurements, developing a method of integrating the measurements in the vertical, and determining the required averaging time. Measurements were conducted over 4 years and produced data for 310 nights covering all seasons. Subcanopy flows were decoupled from the flows aloft 75% of the time. Conditions conducive to the generation of negative buoyancy near the forest floor, necessary for drainage flows to develop, were given in 92% of all nights. The occurrence of nocturnal drainage flows correlated well with "missing flux" problems ("deficit nights"), prompting us to propose an improvement on the commonly used friction velocity criterion (which requires u* to be larger than some empirical cut-off for the eddy fluxes to be considered credible). The "negative buoyancy forcing fraction", i.e. negative buoyancy as a fraction of the sum of the dynamic driving forces, can be shown to predict deficit nights significantly better than the u* cut-off. The appropriate horizontal size of the network of wind and CO2 sensors at the Harvard Forest was shown to be on the order of 100 m, ensuring that sensors were generally observing coherent processes on this scale or larger and thus displaying some correlation. Horizontal transport of CO2 was found to be restricted to the bottom ~10 m of the forest, facilitating the development of a method of integrating the horizontal CO2 gradients in the vertical. Including the measured horizontal transport terms did not, on average, fully account for the observed difference in NEE of 1.2 +/- 0.3 umoles/m2/s between deficit and non-deficit nights, but decreased the difference to 0.7 +/- 0.5 umoles/m2/s.
B43C-02 INVITED 14:00h
Linking leaf photosynthesis to eddy-covariance measured fluxes on complex terrain
It is recognized that topography influences almost all aspects of forest-atmosphere carbon exchange, yet a limited number of studies considered the role of topography on the structure of turbulence within vegetation and its ultimate effect on photosynthesis and net ecosystem carbon fluxes. Here, we focus on the interplay between radiative transfer, flow dynamics, and eco-physiological controls on CO2 sources and sinks within a canopy volume on a gentle cosine hill. In particular, we address how does topography alter the forest-atmosphere CO2 exchange rate when compared to uniform flat terrain. Towards this end, we present results from a first order closure model that explicitly accounts for the flow dynamics, radiative transfer, and the nonlinear eco-physiological processes within the canopy volume. We compare the first order closure model calculations of advective fluxes with: 1) recent field experiments on CO2 transport conducted in Tharandt, Germany; and 2) recent flume experiments for momentum transport in the presence and absence of a dense canopy on a train of hills. Implications for CO2 flux monitoring on complex topography are also discussed in the context of the relative roles of advective terms and the local CO2 sources and sinks.
B43C-03 14:20h
A Novel Method for Measuring Carbon Dioxide Concentration From 0 to 2.5 Kilometers Above the Ground
This paper describes a combination of novel technologies that enable the collection of atmospheric data, including PTU, wind speed / direction, and carbon dioxide concentration, from ground level to a maximum of 2.5 kilometers height above the ground using temporarily placed equipment. The measurement platform is a portable, highly engineered tethered balloon system that is used to loft and recover one or more instrument packages at controlled ascent and descent rates into the atmospheric boundary layer. Carbon dioxide concentration is measured with a lightweight NDIR instrument that is diffusion aspirated, self contained, and requires only one watt of power for operation. A test flight of the system is described and atmospheric data from the flight is shown.
B43C-04 14:35h
CO2 Dynamics in Atmospheric Boundary Layer over Boreal Forest: Integrating Aircraft CO2 Concentration Profiles with Tower-Based Flux Measurements.
The atmospheric boundary layer (ABL) has an important role in the global carbon cycle since it acts as an interface between the global free troposphere and the surface sinks and sources of carbon dioxide. Multiple processes in a wide range of spatial and temporal scales are contributing to the dynamics of the CO2 concentration field in the ABL. While different research approaches provide insights on the processes in different scale domains, the relationship among the processes of different scales represents a significant challenge. On the diurnal time scale, eddy correlation CO2 flux measurements have footprints of about 1 km2, while CO2 concentration measurements have footprints of the order of 1000 km2. Therefore it is necessary to understand the relationship between the CO2 flux and concentration measurements to scale up the small scale flux measurements to the regional scale concentration measurements. In this study, we examined the flux-concentration relationship in the ABL at a boreal forest site in Saskatchewan, Canada; using a combination of tower flux measurements, tower CO2 concentration measurements, and aircraft profile concentration measurements. Some cases were illustrated using coupled biosphere-atmosphere model simulations. The results suggest that in the night time with typically low wind speed, the CO2 concentration in the shallow stably stratified ABL is dominated by the local CO2 flux. While comparison of boundary layer CO2 budget changes estimated from aircraft profiles with tower eddy covariance flux measurements indicates that the daytime CO2 concentration in the deep well-mixed ABL is strongly influenced by atmospheric transport, and the atmospheric transport is driven by mesoscale and synoptic scale processes. Therefore, understanding of the mesoscale and synoptic scale atmosphere-biosphere interaction is important in the flux-concentration relationship.
B43C-05 14:50h
Flux Footprint Simulation in Complex Terrain
Atmospheric measurements of forest exchange processes are often executed under non-ideal conditions. Observations show that energy fluxes over forest may even exceed the surface fluxes originating from the forest and the upwind terrain. The overshoot of atmospheric fluxes is caused by atmospheric flow obstruction at the forest edge. In order to calculate the forest surface flux from the atmospheric measurements, an estimate should be made of the corrections for upwind fluxes as well as for flux deformation by inhomogeneous turbulence. A new footprint model is presented with explicit calculation of the atmospheric turbulence field. The model is capable to simulate observed overshoot and undershoot of atmospheric fluxes downwind of a forest edge. Model results are sensitive to forest structural parameters. Values are given which part of the atmospheric flux originates from the forest and which part from the upwind terrain. It is shown that in conditions of inhomogeneous turbulence, the sum of these parts may deviate significantly from unity. It is concluded that perturbations of the atmospheric flow affect flux measurements in complex terrain. It is recommended to use a footprint model which accounts for the actual turbulence pattern in order to analyze atmospheric flux measurements in complex terrain.
B43C-06 INVITED 15:05h
European Forest Carbon Mass Balances Estimated with Remote Sensing and the Production Efficiency Model C-Fix: A hot Future Unfolds for Kyoto Protocol Implementation.
Carbon emission and -fixation fluxes are key variables to guide climate change stakeholders in the use of remediation techniques as well as in the follow-up of the Kyoto protocol. A common approach to estimate forest carbon fluxes is based on the forest harvest inventory approach. However, harvest and logging inventories have their limitations in time and space. Moreover, carbon inventories are limited to the estimation of net primary productivity (NPP). Additionally, no information is available when applying inventory based methods, on the magnitude of water limitation. Finally, natural forest ecosystems are rarely included in inventory based methods. To develop a Kyoto Protocol policy support tool, a good perspective towards a generalised and methodologically consistent application is offered by expert systems based on satellite remote sensing. They estimate vegetation carbon fixation using a minimum of meteorological inputs and overcome the limitations mentioned for inventory based methods. The core module of a typical expert system is a production efficiency model. In our case we used the C-Fix model. C-Fix estimates carbon mass fluxes e.g, gross primary productivity (GPP), NPP and net ecosystem productivity (NEP) for various spatial scales and regions of interest (ROI's). Besides meteorological inputs, the C-Fix model is fed with data obtained by vegetation RTF (Radiative Transfer Model) inversion. The inversion is based on the use of look-up tables (LUT's). The LUT allows the extraction of per pixel biome type (e.g. forests) frequencies and the value of a biophysical variable and its uncertainty at the pixel level. The extraction by RTF inversion also allows a land cover fuzzy classification based on six major biomes. At the same time fAPAR is extracted and its uncertainty quantified. Based on the biome classification, radiation use efficiencies are stratified according to biome type to be used in C-Fix. Water limitation is incorporated both at the GPP level using evaporative fraction (EF) and at the soil respiration level with soil moisture content (SMC). Water limitation is derived from optical and thermal information, extracted with NOAA/AVHRR and METEOSAT processing chains. A key feature of the approach outlined, is the direct inference of the bio-geophysical state of terrestrial forest cover from space borne observations. Since forest state is not estimated with deterministic models, the modelling of limiting factors like nutrient deficiency, pest and disease impacts, pollutant effects as well as any other long-term effect on forest carbon fluxes, is eliminated. The only processes still to be to be taken into account in C-Fix are by definition those that show temporal dynamics with response times shorter than the typical turnover time of chlorophyll metabolism. For example, short term water limitation effects on EF and SMC. Applications of the expert system include the mapping of the spatial and temporal patterns of GPP, NPP, NEP and soil respiration as well as the impact of water stress on water use efficiency and the carbon balance. Moreover an index indicating the ratio between carbon emission en its re-fixation by forests is mapped for Europe for non-water limiting conditions. It reveals large imbalances between carbon emissions (IPCC, 1997) and forest carbon re-fixation (1997) for almost all European countries. Keywords: Carbon relations, remote sensing, bio-geophysical variables, C-Fix, Kyoto protocol.
B43C-07 15:20h
A scaling analysis of vegetation carbon uptake for the North American Carbon Program: What are we missing and what can we do about it?
The ability to detect patterns of carbon assimilation by vegetation is a key component of the North American Carbon Program. To date, most efforts have focused on remote sensing of canopy leaf area index (LAI), which has been related to productivity across large resource gradients and can be estimated using spectral vegetation indices such as NDVI. However, a growing body of evidence suggests that approaches based solely on LAI may be problematic in dense plant canopies and in systems where variation in growth is driven to a greater extent by variation in physiological properties such as photosynthetic potential and light use efficiency. Because photosynthetic potential is strongly related to biochemical constituents such as nitrogen and chlorophyll concentrations in foliage, the ability to incorporate canopy chemistry into large-scale carbon cycling research would represent an important contribution to NACP research goals. Here, we present results from a study that examines the degree to which canopy nitrogen chemistry can serve as an integrator of C flux patterns over complex forested landscapes. The functional basis for using foliar N as a scalar of C uptake lies in the fact that the proteins responsible for CO2 capture by leaves (e.g. rubisco) account for the majority of nitrogen in plant canopies. This and other linkages between terrestrial C and N cycles has motivated several new advancements in the ability to detect canopy N using hyperspectral remote sensing. As part of an ongoing project, we have derived spatial coverages of canopy N at four eastern U.S. sites that are part of the AmeriFlux network and used these data to drive enhanced spatial estimates of gross carbon exchange. Results from these sites will be discussed along with plans for additional sites in other regions of North America.
B43C-08 15:35h
Constraining Night Time Ecosystem Respiration by Inverse Approaches
Estimating nighttime ecosystem respiration remains a key challenge in quantifying ecosystem carbon budgets. Currently, nighttime eddy-covariance (EC) flux measurements are plagued by uncertainties often attributed to poor mixing within the canopy volume, non-turbulent transport of CO$_{2}$ into and out of the canopy, and non-stationarity and intermittency. Here, we explore the use of second-order closure models to estimate nighttime ecosystem respiration by mathematically linking sources of CO$_{2}$ to mean concentration profiles via the continuity and the CO$_{2}$ flux budget equation modified to include thermal stratification. By forcing this model to match, in a root-mean squared sense, the nighttime measured mean CO$_{2}$ concentration profiles within the canopy the above ground CO$_{2}$ production and forest floor respiration can be estimated via multi-dimensional optimization techniques. We show that in a maturing pine and a mature hardwood forest, these optimized CO$_{2}$ sources are (1) consistently larger than the eddy covariance flux measurements above the canopy, and (2) agree well with chamber-based measurements. We also show that by linking the optimized nighttime ecosystem respiration to temperature measurements, the estimated annual ecosystem respiration from this approach agrees well with biometric estimates, at least when compared to eddy-covariance methods conditioned on a friction velocity threshold. The difference between the annual ecosystem respiration obtained by this optimization method and the friction-velocity thresholded night-time EC fluxes can be as large as 700 g C m$^{-2}$ (in 2003) for the maturing pine forest, which is about 40% of the ecosystem respiration. For 2001 and 2002, the annual ecosystem respiration differences between the EC-based and the proposed approach were on the order of 300 to 400 g C m$^{-2}$.