Surface/atmosphere exchange and chemical interactions of reactive nitrogen compounds above a manured grassland

Abstract

Manure application to managed grassland is a common agricultural practice. There are, however, limited studies looking at the fluxes and interactions of reactive N compounds and aerosols following fertilisation with manure. In this study, state-of-the-art chemical analysers (GRAEGOR, QCLAS, PTRMS) were used to investigate concentrations, fluxes and chemical interactions of reactive nitrogen containing trace gases (NH₃, HNO₃, HONO) and aerosols (NO₃⁻) above a grassland fertilised with 164 kg N ha⁻¹ of cattle slurry. Emissions of NH₃ peaked at >67 μg m⁻² s⁻¹, based on a 30 min average. The estimated overall loss of total ammoniacal nitrogen (TAN) from the applied slurry through NH₃ emissions in the first 5 days was 33.5%. The average trimethylamine flux in the first 31 h following the first slurry application was 40 ng m⁻² s⁻¹ and amounted to 0.38% of the NH₃-N emissions. Apparent nitrate aerosol emissions were observed following the slurry application peaking at 13.0 ng m⁻² s⁻¹. This suggests formation of NH₄NO₃ from reaction of the emitted NH₃ with atmospheric HNO₃, consistent with the observation of gaseous concentration products exceeding the dissociation constants of ammonium nitrate. Fluxes of total nitrate (HNO₃ + NO₃⁻) were bi-directional and positive during the mid-day period after fertilisation, suggesting that the slurry acted as a net source for these compounds. There is evidence of small HONO emission following fertilisation (up to 1 ng m⁻² s⁻¹), although the production process is currently not identified. By contrast, all compounds showed deposition to the adjacent unfertilised grassland.

Introduction

Fertilisation with manure, a common agricultural practice throughout the world, causes the emission of a number of nitrogen (N) containing trace gases, which are of environmental importance and take part in atmospheric chemistry. One of the dominant and most studied compounds emitted is NH₃, where it is estimated that land fertilisation with manure contributes approximately 10% of the total NH₃ emissions in Europe (ECETOC, 1994), with detrimental environmental effects. The extent of NH₃ impact on the environment is highlighted in a recent document for the Economic Commission for Europe (UNECE), which estimated that by 2020, NH₃ will be the primary source of eutrophication, acid precipitation and secondary aerosol formation in Europe (Sutton, 2009). There is increasing evidence of eutrophication of ecosystems in the UK and elsewhere in Europe, as indicated by a decline in nitrogen (N) sensitive species, with a corresponding increase in N tolerant species across the UK (Hornung et al., 2002). Through the formation of ammonium sulphates and ammonium nitrates, NH₃ is an important precursor of aerosols, with implications for human health and the climate system. While NH₃ emissions have been studied in some detail, manured land is also likely to be a source of a range of organic N compounds and potentially, nitrous acid (HONO). Nitric acid, another important N containing trace compound not only reacts with NH₃ to form NH₄NO₃, but it also makes an important contribution to total nitrogen deposition of nitrogen (ROTAP report, 2009).

Further understanding of the nature and magnitude of fluxes of trace gases and aerosols allows us to develop a more detailed picture of the impact that anthropogenic sources are having on the environment. This is vital in order to optimise the abatement strategies, as well as to achieve and set realistic limits in legislation to control emissions. Measurement of reactive nitrogen fluxes and budgets are, however, restricted by availability of instrumentation and the interference of gas–particle interactions.

Ammonia is the primary gas traditionally associated with manure trace gas emissions. There have been a number of studies investigating the mechanisms controlling NH₃ emissions and the effect of abatement strategies. Several factors have been found to contribute to the emission rate from slurry, including the composition of the slurry, including total ammoniacal nitrogen (TAN) content and dry matter content, as well as the environmental conditions when application occurred, such as surface wetness and wind speed (Huijsmans et al., 2003, Sommer et al., 2003, Vogel et al., 2003, Misselbrook et al., 2005, Thompson et al., 1990a, Thompson et al., 1990b). Studies of fluxes of other N containing trace gases and aerosols such as HNO₃, NO₃⁻ and volatile organic N compounds (VONCs), however, are far fewer than those of NH₃ over fields fertilised with manure.

In general, HNO₃ is thought to be formed in the atmosphere as a secondary air pollutant, mainly as a result of photochemical reactions of nitrogen oxides, though other chemical pathways are known too (Zimmerling and Dammgen, 1999). Nitric acid has no known biological emissions to the atmosphere and is highly sticky (absorptive to surfaces). As a result the general consensus is that HNO₃ is continually deposited, at fast deposition rates as it is thought to have a zero surface resistance (R_c). There is, however, increasing evidence of possible nitric acid (HNO₃) emissions and R_c > 0 s m⁻¹ over different ecosystems (Farmer et al., 2006, Neftel et al., 1996, Nemitz et al., 2004, Sutton et al., 1997, Wolff et al., 2009), which contradicts conventional wisdom of source and flux mechanisms generally accepted for HNO₃. In particular, Sutton et al. (1997) presented evidence of increased HNO₃ concentrations downwind of slurry application.

Like HNO₃, nitrous acid (HONO) is a secondary reactive oxidised N compound, with the principal source attributed to NO from vehicular emissions. Photolysis of HONO is recognised as a significant source of hydroxyl radicals in the environment:

$$\text{HONO}\stackrel{h\nu }{⟶}\text{NO}+\text{OH}$$

In general, it is thought that HONO is formed principally during the night, whereas its lifetime with respect to photolysis is very short during the day. However, recent literature has reported HONO concentration increases during the day, implying that HONO is also an important OH source during daytime. It is likely that HONO is produced through photolytic surface reactions, with the main source postulated to be the reaction of NO₂ with H₂O (Harris et al., 1982, Harrison et al., 1996, Kleffmann, 2007, Lammel and Cape, 1990), though there are other suggested mechanisms of formation which have been widely debated (Lammel and Cape, 1990). For example, HONO recently has been identified to be formed from organic sources such as the photochemical reduction of NO₂ on humic acid (Stemmler et al., 2006). Overall the mechanism of day time formation of HONO still remains unclear (George et al., 2005, Kleffmann et al., 2003, Vogel et al., 2003). There are few flux measurements of HONO above vegetation, but most of these include periods during which emission was observed (Neftel et al., 1996, Trebs et al., 2006). In addition, there have been suggestions of a compensation point for HONO by Schimang et al. (2006) during laboratory studies of the uptake of HONO by various plant species.

Emission of organic compounds dominates the odours associated with manure application and can provide a nuisance to the public. One group of organic compounds, the amines, that has been associated with manure emissions (Schade and Crutzen, 1995), has recently received interest from the atmospheric chemistry community. Following the identification of amines as a considerable constituent of ultrafine (<40 nm) aerosol (Barsanti et al., 2009, Kurten et al., 2008), amines have been incorporated into revised particle nucleation schemes as their low vapour pressure has a significantly larger effect on promoting particle nucleation than ammonia.

Interactions between trace gases and aerosols can complicate the quantification of the measured exchange rates, unless the chemical interactions are explicitly quantified. In cases where the chemical time-scale for transformation (τ_c), is comparable or faster than the turbulent diffusion vertical transport (τ_d) of the component, assumptions of a constant flux with height will be violated and micrometeorological flux measurements need to be modified in order to derive the correct surface flux. Several studies have suggested that interactions of NH₃ with acidic compounds, such as HNO₃ and SO₂, result in such errors in the flux measurements which would therefore have an impact on earlier quantifications of the emissions of NH₃ from agriculture and fluxes of total N to semi-natural vegetation (Brost et al., 1988, Nemitz and Sutton, 2004, Nemitz et al., 2000b). In particular, above agricultural vegetation with high NH₃ emissions, the interaction between NH₃ and HNO₃ can result in the observation of HNO₃ deposition rates that exceed the theoretical deposition rate based on a zero surface resistance, due to the additional sink provided by HNO₃ reaction with NH₃. Sutton et al. (1993b), for example, on the reanalysis of work by Harrison et al. (1989), found that the deposition velocity of HNO₃ exceeded the maximum theoretical deposition velocity, based on a zero surface resistance, under the presence of large concentrations of NH₃ following urea fertilisation, and a similar observation was made by Dollard et al. (1987). The apparent large values presented were explained by Sutton et al. (1993b) to be the product of flux divergence due to the consumption of HNO₃ close to the surface to form NH₄NO₃ in the presence of high NH₃ emissions. Indeed, Nemitz et al. (2009) recently demonstrated that apparent upward fluxes of aerosols could be explained by NH₃ emissions resulting in growth of NH₄NO₃ containing aerosol in the size-range of the particles detected.

Such chemical conversion from fast depositing gaseous species (such as HNO₃) and species undergoing bi-directional exchange (NH₃) to slowly depositing aerosol species (e.g. NH₄⁺, NO₃⁻) can modify the total N flux at the site (Nemitz et al., 2004). By contrast, Nemitz et al. (2000b) found that above an oilseed rape canopy the chemical time scale of formation was too slow to violate the measurement of the NH₃ flux above the canopy. Within the canopy of oilseed rape, however, there were suggestions that the rate of formation of NH₄⁺ was sufficiently fast to modify the NH₃ flux. On a physical level, this would have further stimulated NH₃ emission (by lowering the NH₃ concentrations), while, on a technical level, it would have led to an underestimation of the true NH₃ flux from the plants because some of it would have shown up in the NH₄⁺ flux.

The opposite process has also been observed with apparent HNO₃ emissions or reduced HNO₃ deposition rates, being the result of the dissociation of NH₄NO₃ aerosol above semi-natural vegetation (Zhang et al., 1995; Nemitz and Sutton, 2004). Under these conditions, large apparent deposition rates for aerosol NO₃⁻ or NH₄⁺ have been measured (Nemitz and Sutton, 2004). This process converts slowly depositing aerosol species into rapidly depositing gaseous compounds, thereby enhancing total N deposition. Most former studies have suggested that the relative effect of chemical conversion is often small for NH₃, which often dominates the N flux. However, the relative effect can be large for HNO₃, NH₄⁺ and NO₃ ⁻ aerosol and even cause flux reversal. To understand the actual deposition of these compounds it is necessary to measure the deposition of all constituents of the NH₃–HNO₃–NH₄NO₃ triad (Wolff et al., 2009).

The work in this paper is one of the first integrated studies into the emission and interactions of a number of nitrogen compounds over manured grassland. The 2005 Easter Bush experiment took place in April 2005, with fertilisation of an intensively managed grassland over a period of three days using 6 month old cattle slurry. Flux measurements of NH₃, HNO₃, HONO and their aerosol counterparts (NH₄⁺ and NO₃⁻) were carried out, as well as concentration measurements of amines. The aim was to study the fore mentioned reactive N dynamics of an intensively managed grassland following fertilisation with cattle slurry. The specific objectives were:

• To quantify NH₃ concentrations and fluxes measured using both optical and wet chemistry methods based on aerodynamic gradient methods (AGM) and eddy covariance (EC) methods;

• To measure the concentrations and fluxes of acidic gases (HNO₃ and HONO) and aerosol species (NH₄⁺ and NO₃⁻), with emphasis on investigating the possibility of HNO₃ and HONO emissions following slurry application;

• To quantify the concentration of trimethlyamines following slurry application and to calculate fluxes using the concentration ratio technique;

• To determine possible interactions and relationships of NH₃ with other trace gases following slurry application, alongside environmental factors affecting the direction and magnitude of the fluxes.