VERTICAL-DISTRIBUTION AND FLUXES OF AMMONIA AT GREAT DUN FELL

As part of the study of the ammonia budget over Great Dun Fell, measur ements of fluxes of gaseous ammonia (NH3) with the hill surface (grass moorland and blanket bog) were made using micrometeorological techniq ues, to provide information on NH3 removal by the hill surface and on vertical concentration gradients. Measurements of vertical concentrati on, chi, profiles of NH3 concentration were coupled with turbulent dif fusivities to determine fluxes, F-g, deposition velocities, and canopy resistances, R-e, to uptake by the ground. Consistent with published measurements for this site, NH3 was generally found to deposit efficie ntly to the vegetation canopy, with mean R-c of 5 and 27 s m(-1) for e xample days shown. However, short periods of NH3 emission from the moo rland were also observed at small chi(0.3 mu gm(-3)). Under these cond itions of bidirectional exchange, the R-c model does not adequately de scribe the exchange process, and an alternative model was applied, whi ch treats both the resistance for deposition to leaf cuticles, R-w, an d exchange with a ''stomatal compensation point,'' chi(s). This link b etween chi and direction of F-g may result in an important effect of c loud processing: depletion of chi by in-cloud reaction would be expect ed to favour NH3 emission from down-wind agricultural land and moorlan d, though emission from the hill itself during immersion in cloud is u nlikely. Comparison of two measurement techniques to determine air con centrations (batch wet rotating denuder, inlet 0.5 m height; continuou s wet denuder, inlets 0.3, 2 m heights) showed acceptable agreement, a lthough because vertical concentration gradients were large (small R-c ) the height of sampling had a substantial effect. Vertical gradients are also relevant to the use of the measured concentrations as estimat es of NH3 in the air mass passing over the hill, for modelling atmosph eric budgets. Where NH3 deposition occurs at the maximum rate, concent rations measured at Im require a 35% correction in neutral conditions when scaling to a reference height of 10 m. (C) 1997 Elsevier Science Ltd.