We read with enthusiasm the whitepaper by Berry et al. proposing an NACP Intensive at the ARM "Cloud and Radiation Testbed" (CART) site in Oklahoma/Kansas. Their whitepaper highlights the breadth and depth of atmospheric parameters which are already being monitored at the ARM CART, and the long data record that exists for this location in the form of the ARM Archive. This whitepaper suggests that NACP should leverage the more-than-10-year investment that DOE has made in instrumenting this facility with state-of-the-art sensors, including exotic millimeter radars and Raman lidars. They also point out the conceptual similarity of the regularly-held ARM "Intensive Observation Periods" (IOPs) with the proposed NACP Intensive.
Berry et al. emphasize that atmospheric transport processes, including entrainment of CO2 and methane in convective systems and the hypothesized "atmospheric rectifier effect" represent major uncertainties in the global carbon cycle and serve as a logical science theme for an NACP Intensive in the ARM CART region. We heartily agree with this proposal and would like to further develop some of the carbon science investigations which would be made possible by locating an NACP Intensive at ARM CART.
Understanding the Planetary Boundary Layer (PBL) is important to couple terrestrial carbon budgets with the greater climate system. The height reached by the PBL ranges from 100 to 3000m, and varies with time, location, and weather conditions. Because this layer of air is associated closely (at least temporarily) with a particular part of the land surface, the boundary layer is directly influenced by the presence of terrestrial ecosystems, responding to factors like vegetation roughness, solar heating, and evapotranspiration.
As the interface between the land, ocean, atmosphere and biosphere, the PBL is receiving increased emphasis and scrutiny. Boundary layer clouds play an important role in climate. Contemporaneous intensive PBL studies are needed within a single area to accurately connect all of these components, and to represent the role of clouds in mediating the transfer of carbon-containing gases between the PBL and the greater troposphere. There is increasing emphasis on the development of PBL-transport and PBL-cloud parameterizations for GCMs, especially for coupled-climate-system studies, global climate change and weather forecasting.
Under many (but not all) conditions, the PBL has very little mixing with the overlying free atmosphere. Because of its close association with land surface properties, characteristics of the PBL under these circumstances are affected by terrestrial and oceanic ecosystems. Carbon fixation (assuming that it exceeds autotrophic and heterotrophic respiration) acts to deplete carbon dioxide from the PBL, lowering the concentration and altering the isotopic ratios of CO2 by measurable amounts. The height and volume of the PBL, along with horizontal and vertical mixing, serves to integrate these ecosystem-mediated signatures over time and across space. Thus, PBL measurements could potentially provide an integrated means by which large-scale carbon cycle processes could be monitored.
The PBL is spatially and temporally dynamic. Vertical mixing can serve to replenish depleted carbon dioxide concentrations in the PBL from equilibrium levels in the lower troposphere. Horizontal mixing and diffusion also occur within the PBL itself. Changes in the PBL occur on diurnal and seasonal cycles, as well as with the frequency of passing weather systems. These dynamics transform and confound potential ecosystem carbon signals, and represent a research challenge to the use of PBL for carbon flux monitoring.
As detailed by Berry et al., the ARM CART region has existing infrastructure and measurement programs that can be built upon for an NACP Intensive. However, the full strength of integrating an NACP Intensive with ARM-CART will be realized through the application and comparison of multiple techniques for measuring the carbon dynamics of the region. ARM-CART offers an unparalleled opportunity to investigate the importance of deep convective processes in controlling PBL depth and transporting CO2 into an out of the PBL (the "rectifier effect"). Furthermore, the ARM-CART region has existing and planned instrumentation that will contribute to measurement and understanding of the carbon cycle using techniques ranging from space-borne imaging, to atmospheric inversion and forward models, with the potential for ground and atmospheric datasets to drive and validate those approaches.
The existing eddy covariance measurements at the Central Facility, at nine extended facility locations, and two roving systems will contribute greatly in this regard. However, while eddy covariance is a favored technique for carbon cycle investigations and contributes greatly to our understanding of the processes involved, it is not without limitations in terms of accuracy and reliability, particularly in nighttime and complex terrain measurements. Further, the cost and technical requirements of the method mean that spatial coverage and representativeness of eddy flux sites is rarely complete. The use of independent and complementary alternative techniques is therefore of particular importance. Some of these were discussed by Berry et al. and further below:
The diurnal pattern in CO2 concentrations measured at flux tower sites can be interpreted within the framework of a simple atmospheric boundary layer model to predict surface CO2 flux. Concentration measurements have a much larger footprint than flux measurements, so this method is suitable for regional flux estimates. In combination with remote sensing data to assess spatial heterogeneity, a network of sites monitoring concentrations could provide continental-scale flux estimates.
Ecosystem fluxes estimated from draw-down in PBL carbon dioxide concentration seem to be qualitatively consistent with local flux tower measurements. An extended network of mixing ratio observations could be used to further constrain and improve the accuracy of inversion-based estimates of net ecosystem exchange (NEE) of carbon dioxide.
Terrestrial ecosystems affect the composition of stable carbon isotopes within atmospheric CO2, because of photosynthetic discrimination against the heavier stable carbon isotope (13C). Isotopic fractionation is affected by the type of photosynthetic pathway (C3/C4), the process of stomatal gas diffusion, and the recycling of respired CO2 within canopy.
Measurements of carbon dioxide isotopes in the PBL and the free troposphere offer the potential for regionally integrated estimates of isotope discrimination. Tall-tower and aircraft measurements of carbon dioxide and carbon and oxygen isotopes have been used successfully to develop estimates of PBL-scale isotope discrimination. One can then utilize ecosystem-level measurements of the isotope ratio of respiration and land surface model estimates of photosynthetic discrimination to deconvolve net carbon dioxide fluxes into the gross components of photosynthesis and respiration at the regional scale.
The ratio of the dry heat flux to the evaporative energy flux, referred to as the Bowen ratio, is a key parameter for the determination of evapotranspiration. At a larger scale, the Bowen ratio can also be used to quantify the degree of convective coupling between land and atmosphere, and hence the role of terrestrial ecosystems in climate variability and change.
Recent studies have directly compared CO2 fluxes measured by eddy covariance and Bowen ratio. USDA-ARS currently has a network of sites estimating CO2 and water vapor fluxes on rangeland ecosystems in the western USA using the Bowen Ratio approach, while the larger AmeriFlux network uses the eddy covariance technique. Both systems give very similar estimates of water vapor fluxes. The agreement for CO2 fluxes is weaker, but still quite good, with agreement varying at different times during the season. Agreement is improved when energy balance closure is forced for the eddy covariance fluxes. Like eddy covariance, Bowen ratio sensors work best in simple terrain.
The PBL can persist as a coherent structure over several days. The PBL during a sequence of fair weather days can be viewed as a steady-state structure forced on the one hand by surface exchanges and on the other by atmospheric subsidence coupled to the large-scale energy balance of the troposphere. Surface-modified air in the PBL is replaced by air descending from areas of subsidence or high pressure, and this flow diverges to areas with storm activity. Given sufficient time, an equilibrium may establish such that the transport of flux by divergent flow balances the surface exchange fluxes. Under these conditions the flux of any trace gas is proportional to the change in mixing ratio of that gas in the PBL relative to the free troposphere multiplied by the mean vertical velocity through the top of the PBL.
This equilibrium relationship can be used to calculate net carbon dioxide flux over the area of integration of the PBL, using as inputs measurements of the carbon dioxide mixing ratio made from tall towers that sample the PBL and aircraft measurements that sample the free troposphere. There are at least two tracer methods for resolving the mean vertical velocity through the top of the PBL at large temporal and spatial scales. Candidate trace gases which have been experimentally used include methane, NOx, radon and water vapor itself. Such tracer estimates, along with measurements of carbon dioxide gradients, can be used to estimate PBL-scale net carbon dioxide flux. Initial tracer-based estimates agree well with eddy-covariance based estimates of regional flux over monthly to annual periods.
The new techniques for flux estimation outlined above require ancillary information about the PBL and its entrainment and detrainment, venting, and vertical and horizontal convective mixing. As a result the carbon community is showing renewed interest in the scientific characterization of the PBL. Serendipitously, study of the formation of clouds and their impact on upwelling and downwelling radiation, which has been underway for several decades, also requires the same types of information. The ARM Cloud And Radiation Testbed was designed, constructed and instrumented specifically for such studies of clouds and radiation. Existing instrumentation, along with the long period of historical measurement records available from the ARM archive, makes ARM CART a perfect place for NACP to conduct intensive research on the PBL.
To subport and evaluate these new techniques, measurements and observations would be intensively collected across ARM CART on all aspects of the PBL. The objective is not to decide on any single technique for estimation of carbon flux, but to explore the contribution that each of these experimental flux techniques can make to further constrain regional and continental estimates of the carbon flux from terrestrial ecosystems. Such a science theme essentially restates the ultimate goal of NACP. Additional measurements would, in turn, fuel data assimilation and inverse modeling approaches, with consequential feedbacks to the NACP mission.
The proposed Intensive should concentrate on the complete elucidation of vertical and horizontal transport, PBL height characterization, and free tropospheric mixing, along with simultaneous concentration and isotopic ratio measurements. Such mixing and transport are also important with respect to cloud formation and evolution. The homogeneous flat terrain of the SGP should minimize terrain and surface-induced effects on the PBL height.
ARM CART has eddy covariance and Bowen ratio systems currently operating at sites across ARM CART, and an additional nine eddy covariance systems are to add CO2 measurements this year. These new sensors, in addition to the existing tall tower at the Central Facility and several mobile towers, will make the ARM CART the most intensively monitored location of its size on the planet. This new network of eddy covariance towers across the ARM CART will provide contemporaneous measurements, establishing estimates of carbon exchange that can be compared to other measurement techniques to quantify uncertainty in source-sink relationships and identify potential sources of error in a multiple-constraints approach.
Other possibilities for coordination exist. In 2007, NASA will launch the Orbiting Carbon Observatory (OCO). Water vapor interferes with infrared signatures for carbon-containing gases measured by the OCO. Because ARM CART has specialized instruments designed specifically to detect water vapor, NASA representatives have expressed interest in using the ARM CART as a testbed for the calibration of initial OCO measurements.
Investigation of alternative methods for measuring carbon flux will be important as NACP wrestles with decisions on the number, placement, and instrumentation appropriate for the Tier 3 (soon to be renamed Tier 2) sites within its NACP carbon observation network. Although NACP is a multi-agency effort, resources to subport NACP goals will come largely from within each agency, and may involve reprogramming rather than new dollars. As such, it will be of particular advantage if NACP can position Intensives at sites like ARM CART, where considerable measurement infrastructure already exists. The recent COBRA 2003 airborne carbon sampling project recognized this leveraging potential, including special stops at ARM CART in their North American continental "racetrack."
DOE and the ARM program would welcome the opportunity to host an NACP Intensive at the ARM CART (Wanda Ferrell, personal communication), and have been sponsoring work to specifically encourage carbon modelers to make more use of data within the ARM Archive. The recent designation of ARM as a User Facility also positions the ARM CART well for a role as host to this NACP activity.
We believe that the proposal by Berry et al. for an NACP Intensive at the ARM CART is an idea which makes sense scientifically, politically, and logistically. We have suggested here that such an Intensive should be centered on characterization of the PBL and inter-comparison of alternative techniques for measuring terrestrial carbon flux across this well-characterized area. ARM CART is well-suited for such a science theme, and the results obtained would be both exciting and relevant for NACP objectives.
Contributors to this whitepaper review:
William Hargrove, Raymond McCord, Niall Hanan, Yetta Jager, Craig Brandt
Contact Information:
William W. Hargrove, 865/241-2748, hnw@fire.esd.ornl.gov
Raymond McCord, 865/574-7827, mccordra@ornl.gov
Niall Hanan, 970/491-0240 , niall@nrel.colostate.edu
Henriette Jager, 865/574-8143, zij@ornl.gov
Craig Brandt, 865/574-1921, fcb@ornl.gov