Review and parameterisation of bi-directional ammonia exchange between vegetation and the atmosphere
- 1Center for Ecology and Hydrology, Edinburgh Research Station, Bush Estate, Penicuik Midlothian EH26 0QB, UK
- *now at: INRA, AgroParisTech, UMR 1091 EGC, 78850 Thiverval Grignon, France
Abstract. Current deposition schemes used in atmospheric chemical transport models do not generally account for bi-directional exchange of ammonia (NH3). Bi-directional exchange schemes, which have so far been applied at the plot scale, can be included in transport models, but need to be parameterised with appropriate values of the ground layer compensation point (χg), stomatal compensation point (χs) and cuticular resistance (Rw). We review existing measurements of χg, χs as well as Rw and compile a comprehensive dataset from which we then propose generalised parameterisations. χs is related to Γs, the non-dimensional ratio of [NH4+]apo and [H+]apo in the apoplast, through the temperature dependence of the combined Henry and dissociation equilibrium. The meta-analysis suggests that the nitrogen (N) input is the main driver of the apoplastic and bulk leaf concentrations of ammonium (NH4 apo, NH4 bulk). For managed ecosystems, the main source of N is fertilisation which is reflected in a peak value of χs a few days following application, but also alters seasonal values of NH4 apo and NH4 bulk. We propose a parameterisation for χs which includes peak values as a function of amount and type of fertiliser application which gradually decreases to a background value. The background χs is based on total N input to the ecosystem as a yearly fertiliser application and N deposition (Ndep). For non-managed ecosystems, χs is parameterised based solely on the link with Ndep.
For Rw we propose a general parameterisation as a function of atmospheric relative humidity (RH), incorporating a minimum value (Rw(min)), which depends on the ratio of atmospheric acid concentrations (SO2, HNO3 and HCl) to NH3 concentrations. The parameterisations are based mainly on datasets from temperate locations in northern Europe making them most suitable for up-scaling in these regions (EMEP model for example). In principle, the parameterisations should be applicable to other climates, though there is a need for more underpinning data, with the uncertainties being especially large for tropical and subtropical conditions.