Regional Ecosystem-Atmosphere CO2 Exchange Via Atmospheric Budgets

Davis, K J ; Richardson, S J ; Miles, N L

关键词: ACCURACY;    AIR;    CARBON CYCLE;    CLIMATES;    ECOSYSTEMS;    LAGRANGIAN FUNCTION;    MIXING RATIO;    PLANTS;    RESOLUTION;    SIMULATION;    TRANSPORT;    CARBON DIOXIDE carbon dioxide;    mixing ratio;    flux;    regional network;   

DOI  :  10.2172/900475
RP-ID  :  DOE/ER63475
PID  :  OSTI ID: 900475
Others  :  Other: 47AM
Others  :  TRN: US200722%%262

学科分类：地球科学(综合)

美国|英语

来源: SciTech Connect

【 摘 要 】

Inversions of atmospheric CO2 mixing ratio measurements to determine CO2 sources and sinks are typically limited to coarse spatial and temporal resolution. This limits our ability to evaluate efforts to upscale chamber- and stand-level CO2 flux measurements to regional scales, where coherent climate and ecosystem mechanisms govern the carbon cycle. As a step towards the goal of implementing atmospheric budget or inversion methodology on a regional scale, a network of five relatively inexpensive CO2 mixing ratio measurement systems was deployed on towers in northern Wisconsin. Four systems were distributed on a circle of roughly 150-km radius, surrounding one centrally located system at the WLEF tower near Park Falls, WI. All measurements were taken at a height of 76 m AGL. The systems used single-cell infrared CO2 analyzers (Licor, model LI-820) rather than the siginificantly more costly two-cell models, and were calibrated every two hours using four samples known to within Âą 0.2 ppm CO2. Tests prior to deployment in which the systems sampled the same air indicate the precision of the systems to be better than Âą 0.3 ppm and the accuracy, based on the difference between the daily mean of one system and a co-located NOAA-ESRL system, is consistently better than Âą 0.3 ppm. We demonstrate the utility of the network in two ways. We interpret regional CO2 differences using a Lagrangian parcel approach. The difference in the CO2 mixing ratios across the network is at least 2-3 ppm, which is large compared to the accuracy and precision of the systems. Fluxes estimated assuming Lagrangian parcel transport are of the same sign and magnitude as eddy-covariance flux measurements at the centrally-located WLEF tower. These results indicate that the network will be useful in a full inversion model. Second, we present a case study involving a frontal passage through the region. The progression of a front across the network is evident; changes as large as four ppm in one minute are captured. Influence functions, derived using a Lagrangian Particle Dispersion model driven by the CSU Regional Atmospheric Modeling System and nudged to NCEP reanalysis meteorological fields, are used to determine source regions for the towers. The influence functions are combined with satellite vegetation observations to interpret the observed trends in CO2 concentration. Full inversions will combine these elements in a more formal analytic framework.