Comparison of Methods to Infer Gas Fluxes off a Lagoon

Field techniques in agro-meteorology are always predicated on assumptions (eg. spatial symmetry, and/or the "similarity" of transport efficiencies of different species). When these methods spread to related fields of study, there is a danger that users may pick up the technical procedures (instruments and algorithms), but lose sight of the crucial assumptions.

We performed a "synthetic experiment" to determine the accuracy with which gas fluxes (Q, kg m^-2 s^-1) off a small lagoon may be estimated from measured gas concentrations C (or concentration differences DC) over water, using a local advection model to generate the fields of windspeed, temperature and gas concentration. From these "data" we deduced several micro-meteorological estimators of the (known) source strength Q. Estimates by integration of the horizontal flux (Q^IHF) are the most satisfactory, followed by estimates using a source-receptor relationship based on a backward Lagrangian stochastic method (Q^bLS). Flux-gradient estimates Q^FG, which posit horizontally-uniform wind and stability, can be very seriously in error, and should not be used in advective flows.