Biosphere-Atmosphere Exchange of Pollutants and Trace Substances

A novel fast-response ozone sensor was developed for various environmental applications. The low weight and low power consumption open up applications which are not possible with conventional ozone analysers. The ozone sensor was successfully used in various environmental applications such as the on-line measurement of the vertical turbulent ozone flux onto various ecosystems (vegetation, soil, desert, sea water), high-resolution measurements of vertical ozone profiles and other airborne ozone measurements.

The recent development of a non-intrusive photothermal deflection (PD) instrument allows the quantitative detection of local ammonia concentrations in air to be measured continuously, sensitively and fast. Ammonia is vibrationally excited by an infrared CO₂ laser in an intracavity configuration. A HeNe beam passes over the CO₂ laser beam in a multipass arrangement and is deflected by a thermal, CO₂ laser induced, refractive index gradient. This deflection, due to the mirage effect, is a measure for the local ammonia concentration. In ambient air the detection limit is 0.5 ppb and the spatial resolution is 2.5 cm times π (0.282 mm2). The time resolution is 0.1 s for single line operation and 15 s for multi-line operation. The system is fully automated and suited for measuring periods of at least one week. Results are compared with those of a continuous-flow denuder system.

Automated denuder systems have been developed for measurement of concentrations and exchange fluxes of ammonia and for simultaneous measurement of concentrations of nitric acid, nitrous acid, hydrochloric acid and sulfur dioxide. A new technique was established for sampling and analysis of aerosol, based on growth of particles under conditions of supersaturation and collection of the droplets in a cyclone. Surface exchange of ammonia has been studied over wheat, heathland and coniferous forest. Deposition to leaf surfaces is very efficient under humid conditions. Emission of ammonia from semi-natural vegetation was observed under dry conditions during the day. The dry deposition velocity of different size classes of particles was measured over a coniferous forest. The deposition velocity for sulfates showed a near-linear increase with u_* and was twice as high as previously inferred, 1.2–1.5 cm s−1 on average.

It is now well established that the ozone molecules present in the lower troposphere generally undergo a downward flux through the surface layer, as a consequence of a dry deposition process at the surface. This fact was first described by Regener [1]. Rich et al. [2] studied the removal of ozone from the air by bean leaves under controlled conditions and found that its rate follows a pattern that is very similar to that of evaporation. Using the resistance analogy, they inferred from these observations that both processes are controlled by the stomatal aperture. Galbally [3] and Galbally and Roy [4] measured ozone deposition rates using enclosure techniques on various surface types, both bare and vegetated. Turner et al. [5] stated that both bare soils and vegetation are targets for ozone deposition.

The deposition flux of ammonia was determined over a coniferous forest. An uptake rate significantly higher than can be explained by stomatal uptake was observed. On the basis of these measurements the annual averaged deposition velocity of ammonia to forest was estimated to lie between 20 and 33 mm s−1 the estimated flux lies between 21 and 44 kg N ha−1 yr−1.

Since November 1992 vertical concentration gradients of SO₂, NO₂ and NH₃ as well as relevant meteorological parameters have been continuously measured at a Douglas fir forest site in the Netherlands. It was found that for SO₂ and NH₃ routine application of the gradient method is possible and annual average fluxes can be estimated. Routine application for NO₂ is not yet possible because of frequent instrumental failure and because the influence of chemical reactions on the fluxes can not yet be quantified. Eddy correlation measurements of u* and H at 30 m height by TNO and those at 36 m height were compared to evaluate the constant flux layer assumption. It is obvious from these measurements that during this month the constant flux layer assumption for momentum and sensible heat fluxes was valid. Flux profile relationships for heat and ozone were derived from simultaneous eddy correlation and gradient measurements by TNO. Conventional relationships can be applied, provided a correction is introduced. From the results of the hourly average measurements, dry deposition parameters have been derived. Surface resistances for SO₂ and NH₃ show strong diurnal variations during dry conditions. Then lowest values occur during the daytime. When the leaf surface is wet, R _c values are generally low and independent on the time of the day. An R _c parameterization for SO₂ derived from analogous measurements over a heathland was tested at the forest site. Good agreement between measured and modelled dry deposition velocities was obtained with no systematic deviations. A first order surface resistance parameterization for NH₃ has been derived from the measurements at Speulder forest. When the NH₃ flux is directed towards the vegetation, there is reasonable agreement between deposition velocities obtained from the parameterisation and from the measurements. However, when NH₃ is emitted (~ 20 % of the time), modelled and ‘measured’ V_d can differ considerably.

Fluxes of gases including NO, NO₂, O₃ CO₂ and water vapour were studied with the enclosure technique in three different coniferous forest ecosystems with different air pollution loads and in the laboratory on seedlings, also with different exposures of air pollutants. The results have been expressed in the form of resistances (conductances) which will serve as a basis for a more general estimate to any given forest ecosystem in northern Europe.The flux of NO at the forest floor (both up and down) is insignificant and can be neglected in Sweden and probably also at similar northern locations. The flux of NO to or from forest canopy is also small and can be neglected in comparison with the flux of NO₂. The flux of NO₂ to the forest canopy passes only through stomata and is, in southern Sweden, determined by stomatal resistance no additional internal resistance. This is valid for the whole concentration range investigated, from about 0.2 to more than 10 ppb. At a northern location this is true only at NO₂ concentrations above about 5 ppb. At this location and at lower NO₂ concentrations, an additional (internal) resistance is found. In model calculations, this additional resistance can be expressed as a compensation point, usually between 0.3 and 0.7 ppb, although the real mechanism can also be an internal resistance with no emission. Since the ambient summertime concentration of NO₂ in the region is about 1 ppb, the rate of uptake by the northern canopy is quite small. It is not yet well understood if the higher resistance is a geographical feature related to the lower pollution load, although there is some indirect evidence that this is the case. Deposition of NO₂ to the forest floor is important. Compared to the deposition to the canopy it is lower in southern Sweden but substantially higher at the northern site. The flux of O₃ to the canopy of Scots pine and Norway spruce can be described by stomatal conductance with no internal resistance, plus an external deposition.The conductance for the external deposition is linearly related to the PAR flux with a coefficient of 0.0015 (mm s−1) / (μE m−2 s−1), valid for projected needle area and was determined up to 2000 μE m−2 s−1 in laboratory experiments.The flux of HNO₃ and NO₂ (p) to a coniferous forest was measured with a foliar extraction method using branches distributed throughout the canopy. The deposition to individual branches was fund to be highly variable depending on their locations in the canopy. Comparison with simultaneously measured air concentrations will give an estimate of the deposition velocity.

The Environmental Protection Agency Timisoara is responsible for the environmental protection activity in Timis county, the biggest in the country, situated in western Romania. The complexity of the environmental problems is enhanced because of this border position. As a characteristic of the Timis county is the presence of a large livestock sector.

The surface exchange of nitric oxide (NO) and nitrogen dioxide (NO₂) was investigated over three European ecosytems (grassland, farmland, heathland) during extended field campaigns in 1989–1991. The applied flux measuring method (dynamic chamber technique) was successfully intercompared to micrometeorological and other techniques. In general, NO emission (from soils) and NO₂ deposition (to vegetation) was found for the different ecosystems which could be quantified for different time scales. Influencing factors of both, NO emission and NO₂ deposition were identified and studied by field as well as by laboratory experiments.

Uludag station is situated in the northwest part of the Turkey at longitude 40° 101 N, and latitude 20° 081 E (Fig. 9.7.1). The station is operated by the Middle East Technical University and Uludag University. Based on the fact that fluctuation of ambient concentration results in the changes in the vertical fluxes, first ambient concentration data were obtained. Starting in March 1993, measurements of O₃, NO_x, SO₂ and TSP concentration at ground level have been made continuously. Meteorological parameters wind direction, wind speed, temperature and humidity were also measured continuously. Hard winter conditions at the Uludag mountain site (average snow-fall is about 1.5m) resulted in some technical problems with the instruments, which in turn resulted in some gaps in the data.

From July 1992 until July 1994 continuous measurements of the exchange of ammonia between surface and atmosphere over cultivated pasture were carried out at Zegveld in the west of the Netherlands. Fluxes were determined according to the profile method, using thermodenuders and continuous flow denuders for NH₃ measurements. Concentrations showed a dominating diurnal trend of high values at night and lower values during the day, due to differences in mixing intensity with clean air from higher layers of the atmosphere. In the grazed season average concentrations were higher than in the ungrazed season, due to grazing, fertilizer and manure application and plant activity in the region. In the grazed season emission dominated during daytime, while at night and also during daytime in the ungrazed season deposition was usually observed.Exchange at this location can be roughly described with a model combining bidirectional exchange with the stomata, and deposition to (water layers on) the leaf surface.Additional SO₂ measurements showed efficient deposition to the outer leaf surface, probably due to uptake in water layers on this surface, according to a correlation with humidity. No clear evidence of co-deposition with NH₃ was found, although the slightly lower average SO₂ deposition in winter compared to summer could indicate the effect of a lower NH₃/SO₂ ratio.

Conflicting results in NO_x and O₃ dry deposition literature necessitates the requirement for understanding the effects of chemistry on their flux profiles. This is particularly important over forests where the micrometeorology and plant physiology is more complex than for other vegetation.In 1993 research groups from the United Kingdom (UMIST), Holland (TNO) and Denmark (RISØ) co-ordinated in a field campaign which was designed specifically to measure the flux of O₃, NO₂ and NO over coniferous forest. The experimental site was located in a mature Douglas fir stand (22 m high) at Speulderbos in central Holland. Flux measurements were carried out from a 37 m tower at more than one height above the forest canopy (25 m and 35 m), in order to determine the flux divergence between observational levels. A further measurement height beneath the canopy (7 m) was included to gain a fuller understanding of processes involved in gas-surface exchange and the important role which the canopy plays in these processes. Measurements were made on a continuous basis using both the eddy correlation and gradient techniques from the 19th June to 2nd July 1993. Soil emission and deposition was measured during the final period of the experiment using the chamber technique. Detailed data analysis of the experimental results has been carried out at UMIST. The main results are presented below along with a tabulated summary of two case study periods.A computer deposition model was devised in collaboration with TNO to reproduce the experimental results and describe some of the chemical and micrometeorological mechanisms involved.The basis for the model was also used by Baldocchi [1] for the estimation of SO₂ dry deposition to a deciduous oak forest and adapted for a coniferous forest by Deinum [2]. The revised model considers NO_x, O₃ and hydrocarbon reactions both above, within and below a coniferous forest canopy. It also accounts for an NO soil emission flux at the forest floor. Data form the Speulderbos experiment was used as direct model input to try and reproduce the observed NO_x and O₃ fluxes both above and below the canopy. Model results along with a sensitivity study were used to interpret the observed measurements made at Speulderbos and obtain a clearer picture of the chemistry and processes involved at the atmosphere-forest interface.

Our agency situated in the western part of Romania, was founded in August 1990. The activity area of the agency in Timis county is based on the following data: total area 869700 hectares; population 726000; 82 cities and villages; forest covered area 108184 hectares; 9 natural reservations. A characteristic of the Timis county is the great number of livestock (ca. 2000000 animals) which induce a high pollution level. Thus NH₃ is one of the most important air pollutants in this area. The complexity of the problems is also seen on the legislative side, since Timis county is close to the border and the agency has to take into account the international laws and conventions.Precipitation measurements, performed periodically, include: volume, pH and conductivity measurements. The atmospheric pollutant concentrations are measured by means of a stationary installation, while the wet and dry depositions are determined by photo-colorimetry and gravimetry, in several places. Heavy metal depositions were monitored in collaboration. Unfortunately the anthropogenic and biogenic emissions are not evaluated.

Results of the field campaigns carried out in Hungary in different seasons above different types of surfaces show that deposition velocities are higher in instable cases and for forest in contrast with stable cases and grass surfaces. There is no regular variation of deposition velocities as a function of season, very probably other meteorological and chemical parameters determine the rate of deposition rather than the seasonal variation.

In order to investigate the exchange of trace gases above a forest canopy in the densely settled area of Frankfurt, a tower of 51 m height was erected and a constant flux model for the determination of trace gas fluxes has been developed. The reliability of the gradient model was examined under different meteorological conditions by means of two fast response systems. These investigations showed us that the constant flux assumption is fulfilled if the wind passes over the fetch. Furthermore an intercomparison between the gradient model and the eddy correlation method was performed with respect to O_3·3.Beside ozone the concentrations of NO and NO₂ are also continuously measured and the appropriate fluxes subsequently calculated. The availability of detection methods for COS, DMS, H₂S, PAN, SO₂ and VOCs enables an application of the gradient method to be made to these species. A simultaneous measurement of all these trace gases was performed during a measuring campaign in August 1994. The goal of these efforts is to obtain information on the amount of pollutant uptake in the mixed forest ecosystem investigated for the measured species.In order to observe the influence of air pollutants on the monitored forest, their concentration together with the meteorological parameters are compared with the CO₂ deposition. Investigations made so far seem to indicate that the daytime CO₂ uptake by the forest canopy is diminished during periods of high air pollution. An intercomparison of CO₂ fluxes measured with those calculated by a global biosphere model, which actually does not consider the influence of pollutants and biotic interferences, is accomplished at present.

In the course of our studies we developed a new sampling method for reduced sulfur compounds and used this method for investigations on the role of trees, crops, algae, lichens and leaf litter within the sulfur cycle. These studies were done under field conditions in conjunction with micrometeorological measurements as well as leaf or branch cuvettes. Laboratory studies gave insight into some basic physiological processes and regulation mechanisms.

We have investigated the exchange of short chain organic acids between the biosphere and the atmosphere. Emission from different tree species under laboratory as well as field conditions was studied in order to get an estimate of the contribution of the release of organic acids into the atmosphere by the vegetation. Additionally we gathered information about the correlation of the emission with several plant physiological parameters such as photosynthesis, respiration, transpiration and stomatal conductance. Furthermore, a one year study (1994) supported by micrometeorological measurements over an agricultural site will give some insight into the fluctuations of acids in the air as well as the emission/deposition processes. Investigations on the role of leaf litter were done to discuss the role of litter layers in forests, and first experiments on the determination of the apoplastic concentration of acids and ions will help to discuss production, transport and release phenomena within and from plant tissues.

Sampling and analytical techniques were developed for the simultaneous measurement of isoprene and terpenes emitted by vegetation. An evaluation of the hydrocarbon emission rates by common tree and shrub species in Portugal was performed applying a semi-quantitative method, (∼ 50 different species were studied). Field and laboratory studies with the bag enclosure method were used to quantify the emission rate of monoterpenes and/or isoprene by Pinus pinaster and Eucalyptus globulus. In a pine patch forest tracer and bag enclosure methods were employed together to better access emission rates. Measurements were also done in the field during the year, for terpenes emitted as result of resin-tapping practices.

Biogenic hydrocarbons are known to act as important precursors in tropospheric photochemical ozone formation. Large uncertainties exist about the composition of the mix of volatile organic compounds emitted by various plant species, and the respective emission rates. A screening study of important woody and crop plants revealed that the portion of isoprene emitting species within deciduous tree species is small. But in contrast to the general assumption, all investigated deciduous tree species emitted traces of monoterpenes. As expected, oaks are strong isoprene emitters, however, some Mediterranean oak species emitted no detectable amounts. Isoprene emission is strongly dependent on temperature and insolation. Wheat and rape seed plants emitted only traces of isoprenoids. The emission and deposition behavior of wheat plants, as far as C₂ to C₉ hydrocarbons (NMHCs), formaldehyde, and acetaldehyde are concerned, was studied both in the field (BIATEX experimental site, Manndorf, Bavaria) and in the laboratory. Ambient temperature appears to control the fluxes of ethane, ethene, propane and propene, whereas acetaldehyde emission by wheat plants as well as by Norway spruce is controlled by light. Gradient measurements of NMHCs over a grassland area in the Rhine valley (TRACT Erlenhof 17, Brühl, Germany) showed that meadows have a negligible impact on ambient air mixing ratios of NMHCs. Over a spruce canopy (BIATEX experimental site Schachtenau Bayerischer Wald National Park, Germany) the twelve most abundant non-terpenoid NMHC exhibited no distinct diurnal cycle, and only small differences in mixing ratios were detected between the heights 31 and 51 m indicating that the influence of the canopy on the abundance of the non-terpenoid NMHCs present in the air above the canopy was small. Aldehyde mixing ratios above a spruce canopy, however, may significantly be influenced either by direct emission of aldehydes from spruce or by the production of aldehydes during photochemical degradation of precursors. Investigations of NMHC fluxes in strongly polluted forest ecosystems (BIATEX, Frankfurter Stadtwald, Frankfurt, Germany; BEMA, Castelporziano, Rome, Italy) revealed a complex flux pattern for selected biogenic and anthropogenic NMHCs, which is possibly a result of the heterogeneity of vegetation in the main wind direction.

The isoprenoids isoprene and monoterpenes are hydrocarbons with important ecological functions. In conifers monoterpenes are synthesised mainly for protection against insects. In a ca. 90 year old Norway spruce stand in the Bavarian Forest National Park, Germany, and on the Wank mountain near Garmisch-Partenkirchen, Germany, the formation of isoprene and monoterpenes in needles of Picea abies [L.] Karst., the emission into the atmosphere, the diurnal and seasonal cycles of emission and immission, the horizontal and vertical distribution of the source strength in and above the canopy and the fate of isoprenoids in the atmosphere were studied. The work was performed together with meteorologists and air chemists. Controlling factors in both, isoprene emissions and monoterpene emissions from intact needles are leaf temperature and photosynthetic active radiation (PAR). On the basis of these two parameters emission factors and emission algorithms were calculated. The sun crown of the trees is the dominating source for isoprenoids in the forest during the physiologically active period of the year. In 1991, 88% of the monoterpens were emitted in July and August. The contribution of the trunk compartment to total forest emission ranges from 1% to 64% and is not clear yet. The contribution of the soil is negligible (3%). The emission pattern and the emission rates of monoterpenes dramatically change after injuring of needles and trunks by opening the resin ducts. The α-pinene fluxes of the forest calculated using the gradient approach and the enclosure technique are similar and indicate that the sun crown of the forest is the main source of the isoprenoids transported in upper air layers. Around noon, however, the gradient approach resulted in deposition of α-pinene, although this compound is emitted from the twigs in large amounts due to high temperature and high solar insulation. At the site the ozone forming potential through photochemical isoprenoid degradation should be of minor importance, since the terpene/NO_x ratio in ambient air is mostly < 1 during the day. The interdisciplinary investigations of reactive trace gases released into the atmosphere by Norway spruce ecosystems demonstrate the complexity of the interactions soil, tree, canopy, atmosphere, man and animal.

During 1989–1994 the emission rates of VOCs (primarily isoprene and monoterpenes) from various plant species were assessed by different sampling methods, including the use of excised plant material (leaf disc method) and a dynamic flowthrough branch enclosure, developed during the course of this work. Both stressed and unstressed plants have been studied.Various tree (conifer and broadleaf), crop and herbaceous UK species were screened (leaf disc method), as were 19 typical Mediterranean species (enclosure method). All plants sampled with the enclosure method have been screened for their tendency to emit oxygenated compounds (BOVOCs). For example, maize emitted negligible amounts of isoprene and monoterpenes but various aldehydes and ketones.The effects of both elevated ozone (O₃) and carbon dioxide (CO₂) have been investigated. Elevated ozone levels had an inconclusive effect on emission rates. On the other hand, exposure to elevated levels of CO₂ resulted in increased emissions of VOCs.Relationships between environmental conditions and emission rates were studied. Of the UK species, emissions of isoprene from Sitka spruce, gorse and oak, and of monoterpenes from gorse, demonstrated a relationship with temperature. Relationships to the PAR regime were inferred, but not directly quantified, due to the uncontrolled nature of field measurements made under ambient conditions. Monoterpene emissions from the Mediterranean species Quercus ilex demonstrated a better relationship with light than temperature. Emissions from Eucalyptus globulus were exponentially related to temperature.Comparisons between laboratory samples, field samples and different ages of vegetation, have revealed differences in chemical speciation and absolute emission rates, which may be a result of their different growth environments and physiology. It was concluded that extrapolations from a few samples, or those based on laboratory measurements, must be treated with caution.

We have developed an automated analytical device, equipped with a preconcentration/thermo-desorption module, in order to monitor natural hydrocarbons that are emitted into the atmosphere.This on-line device was employed at various forested sites during the course of several measurement campaigns. Comparative tests revealed a satisfactory agreement between monoterpenes (MT) analyses carried out both in the field and in the laboratory by several groups.Daily cycles of atmospheric concentrations in terpenes were found to exhibit minimum values at daytime and maximum ones at night. By contrast, the emission rates and fluxes exhibit maximum diurnal values and minimum nocturnal ones.The emission rates and fluxes of terpenic compounds were correlated with the micrometeorological and physiological parameters of the vegetation.

The dry deposition velocity of submicronic aerosol particles has been measured during three campaigns corresponding to different ground cover. The eddy correlation method cannot be computed blindly, because of low frequency fluctuations. After proper filtration, the dry deposition velocity can be obtained and parametrised. During unstable situations, the dry deposition velocity gives high values.

Our contribution to the research covers field experiments and theoretical considerations. The fluxes of NO, NO₂, O₃, CO₂ and H₂O were determined by the eddy correlation method and the fluxes of particles by the gradient method. The details of materials and methods are given in the references [1–7]. The following gives an overview of the results.

Several sampling campaigns over flat terrain and forested areas were carried out from 1992 to 1994. The campaigns were aimed at obtaining preliminary information on the ambient concentrations of acidic aerosols and particle size and at analysing the daily cycles in order to examine their possible reactivity.Using filter packs, sulfate dry deposition and velocities were determined over flat vegetation and in a forested area by means of the gradient method. Micrometeorological parameters were evaluated using both the gradient method and the eddy correlation technique. The average values of the dry deposition velocities were 0.95 and 0.17 cm s−1 over low vegetation and in the forested area respectively. A strong influence of the Monin Obukhov length on the deposition velocity was observed at both sites and a satisfactory dependence on (−1/L)2/3 was found.

The biosphere, and soils in particular, interact importantly with the atmosphere. It was found that fertilisation with inorganic nitrogen (NH₄+, NO₃− and urea) contributes to the emission of NH₃, NO and N₂O. Field measurements of NO emission showed a clear diurnal fluctuation. The CaCO₃ content and pH of the soil are important soil parameters controlling NH₃, as well as NO and N₂O emission. However, incorporation of the fertilisers into the soil instead of surface application can substantially reduce the NH₃ emission.If upland soils are water saturated and contain enough organic matter they can emit CH₄. On the other hand, they can be a potential sink for atmospheric CH₄ if the methane diffusion into the soil is sufficient. The sorption is substantially reduced by the N turnover of NH₄+ or organic matter residues with low C/N ratios.On a global scale wetlands are the largest source of atmospheric methane. From our study it was clearly shown that wetlands have a large spatial and seasonal emission variation, mainly controlled by water status and temperature interactions. These variations are important in assessing flux estimates. Landfills, on the other hand, are an important anthropogenic source of methane. However, a good management of the landfills can reduce the emission substantially. A good and cheap managing practice is covering the landfill with an effective methane oxidising soil layer.

Potted spruce trees (Picea abies [L.] Karst.) were fumigated with NO₂ or 15NO₂ under controlled conditions in constant light. Current and previous flush needles, the bark and the fine roots were analysed for total 15N content and incorporation of 15N into the α-amino nitrogen of free amino acids. In addition, in vitro nitrate reductase activity and stomatal conductance of the needles were measured. In fumigated trees, nitrate reductase activity increased after a lag period of about four hours to levels which were five times higher than in controls exposed to filtered air. After a fumigation period of 24 hours nitrate reductase activity was linearly dependant on the NO₂ concentration used for fumigation both in previous and current flush. With an average of 9.1% 15N, free glutamate was the pool with the most label. Taking into account the time-course of the labelling of this pool, this figure can be taken as an estimate of the minimum contribution of NO₂ to the N nutrition of the needles. 15N labelled amino acids were also detected in the bark and the roots, indicating export from the needles. There were linear correlations between the stomatal conductance and the 15N uptake and between the 15N uptake and nitrate reductase activity.

The flux of NO between soil and atmosphere was demonstrated to be a function of simultaneous production and consumption reactions which are both due to microorganisms. Measurement of production and consumption rates in the laboratory allowed us to predict the NO flux in-situ by using a simple diffusion model. Both production and consumption of NO are controlled by soil variables such as temperature, pH, moisture and nutrients. NO is produced by many different microorganisms of which nitrifiers and denitrifiers seem to dominate. Their relative contribution to NO production is affected by soil conditions, e.g. moisture. NO is reduced to N₂O by denitrifiers, but is also oxidised to nitrate by aerobic bacteria. The flux of COS between soil and atmosphere is also due to simultaneous production and consumption reactions which are at least partially due to biological processes. NO₂ turnover in soil is a purely chemical process.

The TRACT experiment and the BIVOTEX studies at the Bavarian Forest site offered the possibility of investigating the composition and deposition behaviour of aerosol particles and gaseous compounds above and within tall vegetation (spruce forest with 14% beech) and in an agricultural area in the Rhine valley. During the intensive field campaign of TRACT a study was also performed on the thermodynamic behaviour of the chemically coupled system HNO₃−NH₃−NH₄NO₃ with respect to the phase transition and dry deposition or/and emission.

The exchange of inorganic and organic trace compounds in the soil-vegetation-atmosphere system is influenced by plant physiological, and physical and chemical processes in the soil and the atmosphere. A number of studies (called BIVOTEX I–V) to understand these complex interactions were performed at the Schachtenau experimental research site in the Bavarian Forest National Park, Germany [1–3]. The forest (about 86% spruce (Picea abies [L.] Karst), 14 % beech (Fagus sylvatica [L.]) has an average height of trees ≈ 28 m with LAI ≈ 5.5, crown between 14 m to 28 m, age 80–100 years, and a stand density of 721 trees/ha [1,2].

The exchange of momentum, heat and matter (e.g. H₂O, CO₂, SO₂, NO, NO₂, O₃, NH₃, HNO₃, VOC) in the soil-vegetation-atmosphere system over tall canopies is influenced by non-local transfer processes and the presence of distributed sources and sinks [1–4]. Depending on their water status tall plant canopies therefore have a microweather and a microclimate which significantly differ from those above neighbouring areas covered with bare soil or low vegetation. Thus also their nutrient budgets and especially the fluxes of trace substances between the atmosphere and the plant-soil system can be altered with respect to time and space. A number of studies suggest that chemical cycles inside the canopy can have a different behaviour compared to those from outside and can be coupled or decoupled from cycles in the overlying atmosphere by the various influences of dynamic and thermodynamic processes (e.g. [5–7]). Therefore a study was performed by the joint MIMMOL group to understand better the interaction between transport processes, trace gas emissions and deposition and chemical reactions within and above tall vegetation.

Based on forest experiments, we found a compensation point of 0.4 ppb for the exchange of ammonia with spruces at the Schachtenau site in June 1991. Thus, emission of ammonia from spruces was observed, when the atmospheric concentration was below 0.4 ppb, and uptake was found under conditions with higher mixing ratios. In general, a spruce site like the Schachtenau site acts as a sink for ammonia emitted from other ecosystems (e.g. agricultural sites). This is further supported by gradient measurements in correlation with cuvette data under higher atmospheric mixing ratios in September 1990. For a wheat field site at Manndorf we report a strong diurnal variation of net NH₃ fluxes showing emission over the day and deposition during the night in close correlation to CO₂ fluxes induced by the assimilation and respiration of the wheat field. This behaviour points to the plant physiological control of the NH₃ exchange.

Photolysis of ozone at wavelengths below 320 nm is the main source of tropospheric OH radicals: (1)$${O_3} + hv \to O{(^1}{\text{D}}) + {O_2}$$(2)$$O{(^1}{\text{D}}) + {H_2}O \to 2\;OH$$ Photolysis of nitrogen dioxide occurs in the wavelength range below 420 nm and is the basic supposition of tropospheric ozone production: (3)$$N{O_2} + hv \to NO + O{(^3}{\text{P}})$$(4)$${O_2} + O{(^3}{\text{P}}) \to {O_3}$$ The reaction rate constants of the photolytic processes (1) and (3) are given by the photolysis frequencies J(O₃) and J(NO₂). An accurate knowledge of these quantities is the basis for numerical modelling of photochemical reaction cycles.

ENEL SpA/ CRAM, Environment and Material Research Centre of the Italian Electricity Company, in collaboration with the Environmental Department of CISE, has been developing a long term project since 1992, in order to evaluate the contribution of deposition phenomena in the environmental problems related to atmospheric acidity. This activity is carried out at the meteorological station of San Pietro Capofiume, located in the Po river plain, a rural region in northern Italy affected by anthropogenic pollution events. This project also aims to study dry deposition over a long period in the Mediterranean area. Direct measurements of meteorological and chemical parameters were used to run inferential technique models, in order to quantify dry deposition. The big-leaf and multi-layer approaches were both tested. Concentrations of SO₂, NO₂, NO and O₃ were measured with half an hour resolution, using automatic analysers. During the vegetative season, concentration of gaseous species as SO₂, NO₂, HNO₂, HNO₃, NH₃ and fine particulate as SO₄2− NO₃− and NH₄+ were measured weekly, by an annular denuder system. In 1993, during the vegetative season, an intercomparison campaign was performed at S. Pietro Capofiume to test an inferential technique against eddy correlation ozone fluxes. The results of this comparison and the annual dry deposition fluxes are presented.

The throughfall technique cannot be used to estimate the dry deposition of ammonium and nitrate when the dry deposition is small because the two ions are lost in the canopy. A new phenomena partly explaining this loss is the conversion of inorganic nitrogen to organic nitrogen in the canopy and in the throughfall collectors if they are not kept dark. It is difficult to analyse the throughfall deposition of organic nitrogen because it is bound in particles and can also stick to the walls of the collector. The sampling technique will be investigated further using the addition of known amounts of inorganic nitrogen to the canopy.

Uptake of NO₂ and ozone via stomata and their elimination in the mesophyll were investigated using mechanistic mathematical models. The common hypotheses, postulating the disproportionation reaction of NO₂ being the predominant path for the transformation of NO₂ into the cellular nitrate and nitrite pools, could be falsified because it fails to reproduce the sorption characteristics as determined experimentally. If, however, the reduction of the pro-oxidative NO₂ by apoplastic ascorbate is taken into account the calculated uptake rates and the effects of changes in immission concentration and vegetation characteristics fit well within corresponding experimental results. The onset of acute injury in case of a fall in apoplastic ascorbate concentration below a critical level is postulated.In the case of ozone the results of the model calculations corroborate the hypotheses that apoplastic ascorbate is an essential factor controlling the uptake of this chemically reactive trace gas, too. The model developed contains furthermore a comprehensive mathematical description of the dynamic distribution of ascorbic acid/ascorbate between the different plant cell compartments.

Since 1988, our contribution to BIATEX research has been focused on modelling and measuring the deposition fluxes of different pollutants, such as SO₂, O₃ and NH₃, under unfertilised dry grassland, which is typical of large parts of inland Spain, as a typical representative of at least part of the Mediterranean area. Results contribute significantly to the knowledge of the deposition velocities and fluxes in a part of Europe where lack of data was very significant. Several deposition field experiments were performed in the Castilla — Leon region and in the final stage in the Madrid area. We investigated the deposition velocities and fluxes of SO₂, O₃ and NH₃. We used the Bowen ratio, profile and eddy correlation techniques to measure the deposition fluxes.