Micromet Methods for Determining Fluxes of Nitrogen Species
Tilden P. Meyers NOAA/ARL Atmospheric Turbulence and Diffusion Division Oak Ridge, TN
Presentation Objectives
Discuss various methodologies to measure NH3 exchange
Highlight specific sampling considerations and guidelines for each method
Present results from various field experiments
All micrometeorological methods are constrained by:
1. Fluxes don't vary in X,Y plane in the flux “footprint” 2. The exchange at the surface is same as that at sensor
For NH3, we often are interested in characterizing deposition as well as emission
Emission sources of NH3 generally are characterized by small areas that have spatial scales less than a typical flux footprint (swine lagoons, poultry farms, etc.)
Deposition targets are vegetation (grasslands, forests, crops, soil) that can have considerable spatial extent, but are often located in areas that have small NH3 concentrations
For characterizing NH3 emissions Integrated Mass Balance
Average emission rate = what is advected past tower
Flux =
u z c z dz
0
z
r
Similarly, inverse Lagrangian methods have been used to estimate the mean source area emission strength from measurements of NH3 concentrations downwind of the source area.
(Flesch et al, 1995, J. Applied Meteorology) (Kljun, et al., 2002, Boundary-Layer Meteorology)
The application of micromet methods is not limited by the micromet state of the art but by the current methods used to measure the trace gas of interest (NH3).
Fast time response (> 1 Hz) eddy covariance Slow but accurate gradient methods (AM,MBR)
Accumulation methods (conditional sampling (REA)
Eddy covariance
Emission flux+ w+,c+, w-,c-, Deposition fluxw-,c+, w+,c-,
w ' c'
+
w ' c' -
Eddy covariance is inherently a noise rejection method as the high number of samples in an averaging period (30 min – 60 min) will average out if errors are random with a mean of 0. Useful results can be obtained even with low
Advantages of Eddy Covariance
Good time resolution
Inherently a noise rejection method > many samples
Consideration s Usually requires major power for pumps, etc.
Usually not “all” weather instrumentation If using sampling tubes, tubes, and inlet losses
Measure
Gradient (Modified Bowen Ratio Method)
the flux (eddy covariance) and
vertical gradient of constituent (heat, water vapour, CO2) over specified height.
Compute
effective transfer coefficient (flux/gradient) vertical gradient of NH3 over same height interval and apply computed transfer coefficient to obtain= H ΔC FNH3 a measure of the flux
Δθ
Measure
Water vapor and CO2 gradients To remove bias error, use same analyzer for both heights,
switching at 30 sec to 5 min intervals and allowing for representative sampling. Tc = cycletime
Relative error
ε = 6 Tc / τ
0.8
τ = turbulencetimescale
For Tc = 60 s, = 30 s, = 10%
Advantages of MBR
Good for slow response trace gas sensors Adequate time resolution (30 min -> hour)
Considerations
Bias tests on surrogate scalars Bias test on trace gas gradient systems
Sampling tube and inlet losses
Conditional Sampling Relaxed Eddy Accumulation
For sampling gases and aerosols in accumulation devices like annular denuders, filterpacks, etc. Flux = w(Cup - Cdn) = empirical coefficient, 0.6 w = standard deviation vertical velocity Cup, Cdn = average concentration of updrafts, downdrafts
What constitutes an updraft, etc.?
Separate “w” into three bins
deadband +/- 0.10 m/s dead accumulator updraft > 0.1 m/s downdraft < -0.1 m/s accumulator up accumulator down
Updraft Deadband (mid)
Downdraft
Ammonia Fluxes (REA)
● ●
USDA/ARS-BARC J. Meisinger NOAA/ARL W. Luke
20 l/min flow cyclone/impactor 2.5 m cut-point citric acid (phosphoric) coated denuders 3-4 hour sample intervals
These plants could use a drink.......mmh
Average Loss 1.5 kg N/ha/day
Advantages of REA
When used with denuders and filterpacks, can sample several species at once (NH3, NH4, SO2, | SO4, HNO3,Consideration NO3)
s Very manual intensive with denuders (cleaning, coating, exposing, extracting, IC analysis
Sample flows and extraction volumes need to be