Experimental and model-based comparison of wind tunnel and inverse dispersion model measurement of ammonia emission from field-applied animal slurry

Volatilization of ammonia from field-applied animal slurry is a significant problem. Accurate emission measurements are needed for inventories and research, but are not provided by all measurement methods. Wind tunnels may give emission values substantially above or below micrometeorological results, which have been shown to be accurate. This limitation reduces the utility of wind tunnel results, which make up a large fraction of available measurements. The present work focused on understanding wind tunnel measurement error by comparing micrometeorological and wind tunnel measurements with the aid of a semi-empirical model. Ammonia loss from digestate after field application was measured in high time resolution in three field trials using wind tunnels and the backward Lagrangian stochastic (bLS) dispersion technique simultaneously. Differences in measured emission were interpreted using the ALFAM2 model, and measurements were used to evaluate the model. Results showed that wind tunnel and bLS methods provided different cumulative emission estimates, although there were similarities in measured emission dynamics. The ALFAM2 model was generally able to reproduce emission dynamics for both measurement methods, but only when differences in mass transfer between the two methods were incorporated in an experimental parameter set. This important result suggests that: 1) the simple structure of the ALFAM2 model captures the essential physical and chemical processes controlling emission and 2) the two measurement methods differ only (or mainly) through mass transfer above the slurry/ soil surface and rain. Therefore, with careful selection of wind tunnel air flow it should be possible to approximately match emission that occurs under open-air conditions. But without temporal variation in air flow, actual emission dynamics cannot be captured. This work provides a template for integrating and comparing measurements from different methods, and suggests it is possible to use wind tunnel measurements for model evaluation and even parameter estimation.