Measurement and Modelling of Ammonia Emissions at Waste Treatment Lagoon-Atmospheric Interface

Global emissions of ammonia are approximately 75 Tg N/yr (1 Tg =10¹²g). The major global source is excreta from domestic animals ( 32 Tg N ^-1yr^-1). Waste storage and treatment lagoonsare used to treat the excreta of hogs in North Carolina (NC). Proteins and nitrogen rich compounds in the lagoon are convertedto ammonia, through a series of biological and chemical transformations. The process of ammonia emission has been investigated using two different model approaches: (1) CoupledMass Transfer with Chemical Reaction Model (Model I), and (2)Mass Transport without Chemical Reaction Model (Model II). Asensitivity analysis is performed with the models, and the modelresults are compared with ammonia emission experiments at a swinewaste storage and treatment lagoon in NC using a dynamic emissionflux chamber.Results of model predictions of emission flux indicate an exponential increase in ammonia flux with increasing lagoontemperature and pH, a linear increase with increasing lagoontotal ammoniacal nitrogen (TAN), and a secondary degree increasewith the increasing wind speed. In addition, the fluxes predictedby Model I are consistently larger than fluxes predicted by Model II. Experimental values of flux agreed well with model predictions, with the experimental values lying in different positions between the two model predictions under different physical and chemical conditions. Further, when compared to diurnal and seasonal experimental flux values, Model I corroborates the results in calm meteorological conditions (windspeed U10 = 1.5 m s^-1). However, the observed results are better predicted by Model II during unstable conditions, when wind speeds are higher than 2.0 m s^-1 and physical transfer process functions dominate.     

Ozone Dry Deposition Velocities for Coastal Waters

Air-sea exchange rates for ozone were measured by the eddy correlation technique at a site on the north Norfolk coast in the UK. The average surface resistance to ozone uptake was found to be, r_s(O₃) = 1,000 ± 100 s m^-1. Micrometeorological measurements of trace gas fluxes to ocean surfaces are rare but a review of available measurements suggests that we can constrain sea water surface resistance for ozone to between 1,000 (Regener (1974), and this work) and 1,890 s m^-1 (Lenschow et al., 1982), yielding surface deposition velocities between 0.53 and 1.0 mm s^-1. These values are more than an order of magnitude greater than can be explained by laboratory determined mass accommodation coefficients for ozone to water. The importance of dry deposition with respect to process air-sea exchange models is highlighted. A trend in surface deposition velocity with wind speed was also observed supporting a surface chemical enhancement mechanism of ozone uptake which in turn is enhanced by near surface mixing processes.     

Winter and Spring Thaw Measurements of N2O,NO and NOx Fluxes using a Micrometeorological Method

The impact of nitrogen fertilizers on gaseous emissions duringwinter and spring-thaw is not well understood and was the objective of this research. Using a micrometeorological method,N₂O, NO and NO_x fluxes from ryegrass were measured from November 1997 to March 1998. Three different mineralfertilizers were applied in November: urea (U), slow-release urea(SRU) and ammonium nitrate (AN). N₂O emissions during the winter were small, increasing significantly in March. Total losses of N₂O-N were significantly higher from SRU and U plots, with winter N₂O emissions accounting for 50% of annual losses. Nitric oxide fluxes from all plots weresmall during the measurement period (<0.9 ng N m^-2 s^-1). The NO fluxes from U and AN fertilized plots were significantly higher than from SRU and control plots. NO₂ fluxes were always negative (–6 ng N m^-2 s^-1)indicating deposition, but decreased to –2 ng N m^-2s^-1 when snow was present on the soil surface. Our resultsindicate that the form of inorganic N applied has an effect on NO+ N₂O emissions but not on NO₂ fluxes.Sponsored by CAPES – Brasília, Brazil     

Dynamics of Ammonia Exchange in Response to Cutting and Fertilising in an Intensively-Managed Grassland

Continuous micrometorological measurements of ammonia (NH₃)exchange were made for a period of 19 months (May 1998–November 1999) over intensively managed grassland in southern Scotland. This study focused on the influence of management activities, such as cutting and fertilising, on vegetation-atmosphere exchange of NH₃. Measurements were conducted within the European project GRAMINAE (GRassland AMmonia INteractions Across Europe) within which the Scottish site forms one of 6 sites in an E–W transect across Europe. NH₃ emissions were enhanced (up to 300 ng m^-2 s^-1) after cutting followed by larger emissions after fertilising (up to 1400 ng m^-2 s^-1). Annual budget calculations show the intensive grassland acted as a net source (1.8 kg N ha^-1 yr¹) although fluxes were bi-directional with deposition dominating in the winter and emission in the summer. Initial modelling of the NH₃ exchange using a `canopy compensation point' model has been conducted for key periods. The dynamics of the fluxes during these key periods, such as before and after cutting and fertilising, may be reproduced by introducing different values of the apoplastic ratio, = [NH₄ ⁺]/[H⁺].     

Picocyanobacteria – Sensitive Bioindicators of Contaminant Stress in an Alpine Lake (Traunsee Austria)

The abundance and photosynthetic activity ofpicocyanobacteria in the oligotrophic alpine lake Traunseewere measured at a station located close to the outlet ofindustrial soda waste and at a mid-lake reference stationduring spring, 1999 through to autumn, 2000.Picocyanobacterial numbers measured by flow cytometry inTraunsee (0.7–13.2 × 10⁴ ml^-1) were comparable tothose of other oligotrophic lakes, and there was nosignificant difference between the contaminated and thereference sampling location. Picoplankton (<2 m)photosynthetic rates measured in vitro by the ¹⁴C-technique were significantly reduced at the contaminated siterelative to the reference station at low photosyntheticallyavailable radiation (10 E m^-2 s^-1), while nodifference between these two stations was found at moderatelyhigh light intensity (100 E m^-2 s^-1). Theinvestigation was complemented by laboratory experiments withcultured picocyanobacteria. Three Synechococcus spp.strains were exposed to water taken from either of the twoTraunsee stations and from a control station located inneighbouring Attersee. Cell-specific photosynthetic activitymeasured by 4-h in vitro incubations revealed no significantdifference among the three stations investigated. Growthrates of the same three Synechococcus spp. strains weremeasured by flow cytometry over several days in thelaboratory. One strain, in particular, was sensitive to watertaken from the contaminated site; growth rate of this strainwas significantly reduced, relative to when exposed to watertaken from the reference station. Taken together, our resultsdemonstrate that picocyanobacteria are highly sensitivebioindicators of contaminant stress. The overall impact ofthe emissions from the industrial outlet on thepicocyanobacteria was, however, relatively minor.     

Correlating the Spatial Distribution of Atmospheric Ammonia with δ15N Values at an Ammonia Release Site

A field ammonia (NH₃) release experiment and open top chambers containing moorland monoliths continuously fumigated with NH₃ or sprayed with NH₄Cl were used to assess the potential for using ¹⁵N values in determining the area of influence around a point NH₃ emission source. ¹⁵N values are being increasingly used as environmental tracers and we tested the hypothesis that the ¹⁵N signal from an NH₃ emission source is observable in nearby vegetation. Using modified monitoring devices, atmospheric NH₃ concentrations were found to decrease with distance from source, with ¹⁵N values also reflecting this trend, producing a signal shift with changing concentration. Open top chamber studies of ¹⁵N values of Calluna vulgaris (L.) Hull indicated a correlation with deposition treatments in current year shoots. Analysis of Calluna shoots from the NH₃ release showed a similar trend of ¹⁵N enrichment. Significant linear correlations between ¹⁵N and percent N in plant material were found, both in the controlled conditions of the open top chambers and at the NH₃ release site, illustrating the possible use of this technique in N deposition biomonitoring.     

Biodegradation of Low Molecular Weight Organic Acids in a Limed Forest Soil

The effect of liming (3.45 and 8.75 t ha^-1 dolomite; 16 yr after application) on the biodegradation of three low molecular weight organic acids (citrate, oxalate and propionate) in forest soils was investigated. The concentration of organic acids in the soil solution followed the series propionate > citrate > oxalate with liming having no significant impact on soil solution concentrations (mean organic acid concentration = 8.7 ± 2.3 M). Organic acid mineralization by the soil microbial community was rapid in surface organic horizons (mean half-life for citrate = 2–6 h), with biodegradation rate gradually declining with soil depth. Concentration-dependent biodegradation studies (0 to 350 M) showed that the mineralization kinetics generally conformed well to a single Michaelis–Menten equation with V_max values following the series oxalate > citrate > propionate (mean = 9.8 ± 1.0 nmol g^-1 h^-1) and K_M values following the series oxalate = citrate > propionate (mean 168 ± 25 M). The V_max values declined with soil depth, which was consistent with a general reduction in microbial activity down the soil profile. Liming induced a significant increase in V_max for citrate with no change for propionate and reduction in V_max for oxalate. The latter was probably due to adsorption and precipitation of Ca-oxalate making it unavailable for microbial uptake. The higher adsorption/precipitation capacity for oxalate in the limed soils was confirmed by adsorption isotherms. Generally, liming increased soil microbial activity by approximately 10 to 35% with calculations based on soil solution concentrations indicating that organic acid mineralization constituted approximately 3 to 15% of the total soil respiration.     

Air–Soil Exchange of NO,NO2 and O3 in Forests

The exchange of NO, NO₂ and O₃ at the soil surface wasmeasured with automatic dynamic chambers in a spruce forest and in abeech forest during periods of several months.NO was emitted from the soil at a rate of0–8 ng N m^-2 s^-1(spruce) and 0–15 ng Nm^-2 s^-1(beech), but there was no simple relationship between the flux andeither soil temperature or soil moisture. NO₂ and O₃ weredeposited at the soil surface. Deposition velocities forNO₂were on average 0.3 mm s^-1 (spruce) and 0.1 mms^-1(beech), and the deposition velocities of O₃ were on average 1.6 mm s^-1 (spruce) and 1.4 mm s^-1 (beech). The depositionvelocity of O₃ is fairly constant whereas the deposition velocityof NO₂ varies greatly, but the reasons remain to be investigated.     

Investigation of Biosurfactant Activity and Asphaltene Biodegradation by <Emphasis Type="Italic">Bacillus cereus</Emphasis>

In this research, a biosurfactant-producing bacterium with capability of asphaltene degradation was isolated from oil-contaminated soil samples, and identified as Bacillus cereus. This strain produced an effective biosurfactant in the presence of molasses and the surface tension was reduced to the level of 36.4 mN/m after 48 h under optimum conditions. The optimum values of carbon-to-nitrogen ratio (C:N), pH, and temperature for biosurfactant production were determined as 30:1, 7.3 and 29 °C, respectively, using response surface methodology. The maximum emulsification activity in the culture broth was 53 % after 48 h using kerosene at 25 °C. The goodness of fit of four growth kinetic models including Tessier, Contois, Logistic and Westerhoff was compared for the bacterial growth and molasses utilization of B. cereus in 5-L batch bioreactor during 120 h. Conducted kinetic study showed that biosurfactant production had a good fit with the Contois growth kinetic model (R² = 0.962) and the maximum specific growth rate (µ _max ), saturation constant (K _s ) and the yield of biomass per substrate (Y _x/s ) were determined to be 0.145 h^?1, 1.83 g/L and 0.428 g/g, respectively. The asphaltene biodegradation in flask was evaluated by FTIR analysis and quantified by a spectrophotometer. This bacterium was able to degrade up to 40 % of asphaltene as a sole carbon and energy source after 60 days at 28 °C. The resulting surface tension of 30.2 mN/m with the critical micelle concentration of 23.4 mg/L indicated good efficiency of the biosurfactant.     

The Effect of Nitrogen Enrichment on the Carbon Sink in Coniferous Forests: Uncertainty and Sensitivity Analyses of Three Ecosystem Models

Estimates of the global carbon sink induced by nitrogen enrichment range vary widely, from nearly zero to 2.3 Gt C year^-1. It is necessary to reduce this uncertainty if we are to make accurate predictions of the future magnitude of the terrestrial carbon sink. Here, we present a Monte Carlo approach to uncertainty and sensitivity analysis of three ecosystem models, Century,BGCand Hybrid. These models were applied to a coniferous forest ecosystem in Sweden. The best estimate of the change in total carbon content of the ecosystem with the cumulative change in nitrogen deposition over 100 years, C_total/N_deposition was 20.1 kg C (kg N)^-1 using the pooled mean, with a pooled standard deviation of 13.8 kg C (kg N)^-1. Variability in parameters accounted for 92% of the total uncertainty in C_total/N_deposition, and only 8% was attributable to differences between models. The most sensitive parameters were those which controlled the allocation of assimilate between leaves, roots and stem. In particular, an increase in allocation to fine roots led to a large reduction in C_total/N_deposition in all models, because the fine roots have a very high turnover rate, and extra carbon allocated there is soon lost through mortality and decomposition.     

On Sand and Duststorms and Associated Significant Dustfall Observed in Chongju-Chongwon,Korea during 1997–2000

Monitoring of the occurrence of sandstorms and dustfallphenomena from 1997 to 2000 was carried out. We refer tosandstorms, duststorms and blowing sand as the phenomenaoccurring in the source region, while significant dustfall is thephenomenon that occur in the sink area. It was noted that while the sandstorms in deserts and loess in NW-N China and Mongolia occurred any time during the year while severe sandstorms occurred frequently from March to May after the springthaw. The wind erosion in the source region usually occurred withwind speeds greater than 8 ms^-1. However, severe sandstorm phenomenon occurred with wind greater than 15 ms^-1 for at least 3 hr or more. In the source regions, sandstorms occurred over60–120 days per year. In Mongolia, the observation of thenumber of sandstorm days decreased due to the increase in rainy days.When a dust cloud from a desert reached the sink region,significant dustfall in the Korean Peninsula occurred due tothe lack of strong winds. We defined the episode ofsignificant dustfall (SD) with the following values: TSP250, PM₁₀ 190 g m^-3 and visibility 6.5 km or less. In Korea, 8–12 episodes of SD occurred per yearwith 12 to 22 SD days. The maximum concentrations of TSP recorded were 989–1396 g m^-3 with PM₁₀ values of 861–996 g m^-3 and with the lowestvisibility of 1.4–1.6 km.     

Hysteresis in Reversal of Central European Mountain Lakes from Atmospheric Acidification

Extremely high emissions of S and N compounds in Central Europe (both 280 mmol m^-2 yr^-1) declined by 70and 35%, respectively, during the last decade. Decreaseddeposition rates of SO₄ ^-2, NO₃ ^-, and NH₄ ⁺ in the region paralleled emission declines. The reduction in atmospheric inputs of S and N to mountain ecosystemshas resulted in a pronounced reversal of acidification in the Tatra Mountains and Bohemian Forest lakes. Between the 1987–1990and 1997–1999 periods, concentrations of SO₄ ^-2 and NO₃ ^- decreased (average ± standard deviation) by 22±7 and 12±7 mol L^-1, respectively, in theTatra Mountains, and by 19±7 and 15±10 mol L^-1, respectively, in the Bohemian Forest. Their decrease was compensated in part (1) by a decrease in Ca²⁺ + Mg²⁺ (17±7 mol L^-1) and H⁺ (4±6 mol L^-1), and an increase in HCO₃ ^-(10±10 mol L^-1) in the Tatra Mountains lakes, and (2) by a decrease in Al (7±4 mol L^-1), Ca²⁺ + Mg²⁺ (9±6 mol L^-1), and H⁺ (6±5 mol L^-1), in Bohemian Forest lakes. Despite the rapid decline in lake water concentrations of SO₄ ^-2 and NO₃ ^- in response to reduced S and N emissions, their present concentrations in some lakes are higher than predictionsbased on observed concentrations at comparable emission rates during development of acidification. This hysteresis in chemical reversal from acidification has delayed biological recovery of the lakes. The only unequivocal sign of biological recovery hasbeen observed in erné Lake (Bohemian Forest) where a cladoceran species Ceriodaphnia quadrangular has recentlyreached its pre-acidification abundance.     

Pools and Fluxes of Cations,Anions and Doc in Two Forest Soils Treated With Lime and Ash

The effect of liming and ash treatment on pools, fluxes and concentrations of major solutes was investigated at two forestedsites (Norway spruce) in S. Sweden. One site was treated 15 yrprior to sampling (Hasslöv-Hs; dolomite: 3.45 and 8.75 t ha^-1) and the other 4 yr before (Horröd-Hd; dolomite: 3.25 t ha^-1; wood ash: 4.28 t ha^-1). Effects of limingwere most pronounced in the O horizon solutions where higher pH,elevated Ca (120–700 M) and Mg (50–600 M) were observed as compared to control plots. The impact on the mineralsoil was more moderate. Soil solution concentrations were combined with modelled hydrological flow to calculate mass flows,which largely followed the trends of the solution composition. Liming also resulted in large increases of both exchangeable Caand Mg as well as effective cation exchange capacity (CEC_E;2–5 times the controls). The base saturation (BS%) was raised to 60–100% in the O horizon while in the mineral soil elevated values were only seen at the Hs site (20–60%; down to 10–15 cm depth for 8.75 t ha^-1). Ash treatment did notaffect either the soil solution nor the exchangeable pool to thesame extent as lime. In general, the impact at the Hd site was less pronounced especially in the mineral soil, which might be due to shorter treatment time (4 vs. 15 yr) and also differentthickness of the O horizon. Budget calculations for Ca and Mg originating from the lime showed that a major part of the Ca (40–100%) was retained in the top 30 cm of the soil, of which30–95% was present in the O horizon. The mobility of Mg wasgreater and it was estimated that a significant part had been leached from the profile (30 and 50 cm depth) after 15 yr. Increased mass flows of NO₃ ^- due to nitrification resulting from liming at the Hs site were calculated in the range120–350 mmol m^-2 yr^-1 (or 1.2–3.5 kmol ha^-1 yr^-1). There was significant leaching of Al (25–60 mmol m^-2 yr^-1), of which about 70% was inorganic, in thelower B horizon at both sites with no influence of liming.     

Trends in Chemical Composition of Wet-only Precipitation at Rural French Monitoring Stations Over the 1990–2003 Period

The long-term monitoring of precipitation and its chemical composition are important for identifying trends in rain quality and for assessing the effectiveness of pollution control strategies. A statistical test has been used to the atmospheric concentrations measured in the French rural monitoring network (MERA) in order to bring out spatio-temporal trends in precipitation quality in France over the period 1990–2003. The non-parametric Mann–Kendall test which has been developed for detecting and estimating monotonic trends in the time series was used and applied in our study at annual values of wet-only precipitation concentrations. The emission data suggest that SO₂ and NO _x emissions decreased (−3.3 and −2.0% year⁻¹, respectively) contrary to NH₃ emissions that increased slightly (+0.2% year⁻¹) over the period 1990–2002 in France. On the national scale, the pH values have a significant decreasing trend of −0.025 ± 0.02 unit pH year⁻¹. and concentrations in precipitation have a significant decreasing trend, −3.0 ± 1.6 and −3.3 ± 0.6% year⁻¹, respectively, corresponding with the downward trends in SO₂ emissions in France (−3.3% year⁻¹). A good correlation (R ² = 0.84) between SO₂ emissions and concentrations was obtained. The decreasing trend of was more significant (−5.4 ± 5.2% year⁻¹) than that of (−1.3 ± 2.4% year⁻¹). Globally, the concentration of the major ions showed a clear downward trend including marine and alkaline ions. In addition, the relative contribution of HNO₃ to acidity precipitation increased by 51% over the studied period.     

Impact of Lime and Ash Applications on Soil Solution Chemistry of an Acidified Podzolic Soil

Soil solution samples were taken from two sites (Horröd and Hasslöv) in the south part of Sweden to evaluate how soil solution chemistry responded to different treatmentswith dolomite and wood ash. At Horröd, samples were taken four years after application of wood ash, 4.28 ton ha^-1 and dolomite, 3.25 ton ha^-1. At Hasslöv dolomite, 3.45 ton ha^-1 and 8.75 ton ha^-1 was applied and samples were taken 15 yr later. It was found that treatment with dolomite at one site (Hasslöv) resulted in higher pH values (<2 pH units) and higher nitrification. It was also found at this site that the total Al and the inorganic Al concentrations decreased with dolomite treatment. The Ca, Mg, DOC, Fe, SO₄ ^2- and Cl^- concentrations, mainly in the topsoil, were found to be higher at both sites, following dolomite treatment; Ca and Mg concentrations were 2–8 times higher (<820 M) than in controls (<70 M). Wood ash was found to have less impact. The PO₄ concentration in the O2 horizon at Hasslöv decreased due to dolomite-treatment. ANOVA (Analyse of Variance) and PLS (Partial Least Square) were used to evaluate the data from the two sites.     

Physical Modelling of Urban Roughness using Arrays of Regular Roughness Elements

The aerodynamic behaviour of large urbanagglomerations must be represented in increasingly greaterdetail, as large-scale numerical weather prediction and airpollution dispersion models are refined. The present studyprovides detailed measurements of the flow field in regulararrays of obstacles to obtain representative data on meanflow and turbulence statistics in urban-type areas. Obstacle arrays consisting of simple cubes and flat plateroughness commonly used in boundary layer simulations wereplaced in a simulated atmospheric boundary layer flow in ahydraulic flume. The scale factor was about 1:200 based onthe obstacle height (50 mm). The results show noappreciable `constant stress' region in the internalboundary layer above the buildings, since in any finite-length test array the boundary layer is always developing.However, if the RMS turbulence components are scaled by thelocal values of the shear stress, then there seems to be auniversal scaling, with _u/u_* = 2.1,_v/u_* = 1.65 and _w/u_* = 1.2. This greatly simplifies the parameterization of the first orderturbulence statistics in obstacle arrays. It was alsoobserved during the experiments that, compared to results inthe cube arrays, the turbulence kinetic energy and theReynolds stresses were almost doubled in the flat plateroughness arrays.     

A Relaxed Eddy Accumulation System for the Automated Measurement of Atmospheric Ammonia Fluxes

A continuous relaxed eddy accumulation system ispresented for the automated measurement of surface/atmosphere exchange fluxes of atmospheric ammonia (NH₃) at a single height. The new system features sampling by parallel plate denuder inlets, online chemical analysis using the conductivity cell of a commercially available NH₃ analyzer and online flux calculation. The effective detection limit of the system for air concentrations is 0.2 g m^-3 and it is estimated to resolve fluxes > ± 20 ng m^-2 s^-1, depending on the NH₃ concentration and turbulence. The performance of the system was tested in two measurement campaigns above agricultural grassland, in which it was compared with a 3-point continuous gradient system. During the first campaign, after urea application of 47 kg N ha^-1 in autumn, the REA system derived fluxes which were on average twice as large as the gradient fluxes, while concentrations agreed closely (on average within 4%). Possible reasons include differences in the footprint and an over-correction of the gradient flux in stable conditions. Due to wet and cold conditions, only 0.3% of the fertilizer N was volatilized as NH₃ during the first week. Results from the deployment of an improved system are presented for a summer day, 6 days after fertilization with calcium ammonium nitrate. The agreement of both concentrations (on average within 13%) and fluxes (26%) was very encouraging and similar to the agreement found between two state-of-the-art gradient systems with online analysis.