Chapter 12. Deposition of air pollutants

Table of Contents

12.1. Dry deposition of trace gases
12.1.1. Field measurements
12.1.2. Deposition/exchange models
12.1.3. Some results
12.2. Dry deposition of aerosol particles
12.2.1. Field measurements
12.2.2. Modelling the dry deposition of particles
12.3. Modelling of wet deposition
12.3.1. Wet deposition in Europe

Deposition of air pollutants is an important loss of gases and aerosol particles from the atmosphere. At the same time, deposition processes of different air pollutants can cause various harmful effects both on ecosystems and built environment. Deposition of an air pollutant affects its atmospheric concentration as well as the state of the environment or human health (see Chapter 13). Therefore, it is an important factor in different types of atmospheric chemical-transport models, and surface exchange models.

The removal of gases and particulates from the atmosphere can occur by dry or wet forms (Figure 12.1). The importance of each deposition form varies with species and locations. Dry deposition is a continuous process, while wet removal can be realized only in the presence of precipitation. Therefore, despite of the dry process is slower than wet deposition, the accumulated removal quantity of a pollutant could be more important in case of dry deposition. Both the dry and wet depositions depend on the properties of the gases or particles and removal processes are governed by several environmental factors (Table 12.1).

Dry and wet deposition processes in the atmosphere

Figure 12.1: Dry and wet deposition processes in the atmosphere

Table 12.1: Factors affecting dry and wet deposition of gases and particles

 

dry deposition

wet deposition

gases

near-surface concentration

physical and chemical properties of tracer

weather condition

soil, surface, vegetation properties

chemical reactions

gas-specific parameters

cloud parameters

precipitation

aerosol particles

particle properties

near-surface concentration

weather condition

particle properties

cloud parameters

precipitation

12.1. Dry deposition of trace gases

Exchange of trace gases between surface and the atmosphere is mainly governed by the turbulent diffusion. During this process gas transfer can be bi-direction depending on concentration profile of tracer in the near surface layer. If the concentration of a given species increases with height, the turbulent flux of the trace gas is directed from the atmosphere into the surface (deposition). Otherwise, an upward flux is generated (emission). The dry flux of trace gases is generally described by the following form:

(12.1)

where w’ and c’ are the fluctuation of vertical wind components and the concentration of pollutant. The minus sign denotes that downward flux is negative. Here we assume stationarity and horizontal homogeneity of wind, air temperature, humidity and concentration of pollutant. Additionally, equation (12.1) is valid only if chemical sources and sinks are negligible. This means that the characteristic transport time (τt – the time, while the pollutant is transported from a zr reference level to the surface) is much shorter than characteristic reaction time of the given pollutant (Foken et al., 1995). The layer, when this condition is satisfied, sometimes called constant flux layer.

Several micrometeorological measuring techniques are available for measuring the flux of tracers in this layer. Next to the flux measurements, detailed deposition models are also widely used to estimate the exchange of gases. In these models, the deposition flux of gases is calculated from the product of the concentration gradient of trace gas and the so-called deposition velocity.

12.1.1. Field measurements

In the last few decades, several micrometeorological methods to quantifying the trace gas exchange (deposition and emission) in the soil-surface-vegetation-atmosphere system have been developed (for example, eddy covariance method, relaxed eddy-accumulation method, aerodynamic profile method). For a review about these measuring methods see e.g. Foken et al. (1995) or Grünhage et al. (2000).

The eddy covariance method is an accurate direct method, which requires sophisticated, fast response sensors to measure simultaneously the fluctuations of wind velocity, air temperature, water vapour and the trace gas concentration. Eddy covariance technique is widely used for the determination of H2O, CO2, SO2, O3, NO2 fluxes in the turbulent surface layer. However, for some gases (e.g. NH3), this technique is not available therefore other micrometeorological measuring methods must be applied. Flux data obtained from eddy covariance measurements by different fast response sensors requires further corrections (e.g. correction of density fluctuation, sensor separation correction, etc.).

The relaxed eddy accumulation method (REA) was proposed by Businger and Oncley (1990). This method based on a conditional sampling of pollutants in the function of vertical wind velocity. The air samples are taken with a constant flow rate into two separate reservoirs for updraft and downdraft air sample, and the flux can be calculated by the measurement of pollutant’s concentration in each reservoir. This method is generally used for gases (e.g. for ammonia – NH3), when fast response sensor is not available (see e.g. Hensen et al., 2009).

The fluxes of gases can also be calculated by the measured vertical profiles of concentration, wind, temperature and humidity. This aerodynamic profile method is based on the Monin–Obukhov similarity theory (see e.g. Foken, 2006), which describe the turbulent transfer of heat, momentum and scalar quantities in the surface layer. Parameters used in this theory can be derived from the measurement of wind speed, temperature, humidity and trace gas concentration at different levels over the surface (Figure 12.2). A simplified version of this measuring technique is the gradient method, when measurements are carried out at only two levels above the surface.

A schematic picture of a micrometeorological measuring system in a pine forest

Figure 12.2: A schematic picture of a micrometeorological measuring system in a pine forest, in Mátra Mountain, Hungary, used for the calculation of trace gas fluxes by the aerodynamic profile method.

12.1.2. Deposition/exchange models

Gaseous dry deposition models require various types of input data for the estimation of the dry flux of a trace gas. For flux calculation, the concentration of tracer must be known. Airborne concentration of pollutants a few meters above the ground or over the vegetation can be obtained from both field measurements or from the results of a dispersion model. Other input data depends on the complexity of the model, but generally contain several meteorological elements, physical and chemical properties of pollutant, soil-, surface- and vegetation-specific parameters.

The dry deposition process can be interpreted as an analogy of Ohm’s law. Based on this approach, the flux is proportional to the concentration gradients. This proportionality can be expressed by the deposition velocity:

,

(12.2)

where vd is the deposition velocity, c(zref) and c(0) are the concentration of trace gas at a reference level above the surface and at the surface, respectively. This latter term is zero for a few gases, which have no emission from the surface, such as ozone (O3), sulphur-dioxide (SO2) or nitric acid (HNO3).

The deposition can be estimated by the widely used „big-leaf” model, in which the deposition velocity is defined as the inverse of the sum of the atmospheric and surface resistances:

vd = (Ra + Rb + Rc)–1,

(12.3)

where Ra, Rb, and Rc are the aerodynamic resistance, the quasi-laminar boundary layer resistance, and the canopy resistance, respectively. Each term are given by less or more detailed parameterization in different models.

The aerodynamic resistance (Ra) describe the turbulent processes over the surface and is independent on the species. The boundary-layer resistance (Rb) affects the deposition in a very thin layer of air contact with the surface elements, where the transfer occurs by molecular diffusivity. These terms can be calculated for example using the Monin–Obukhov similarity theory taking into account atmospheric stability (details can be found e.g. in Nemitz et al., 2009).

In the parameterization of the deposition over vegetated surface, the canopy resistance (Rc) is the most complex part of this resistance network. It is generally contains different resistances, which represent the different pathways of pollutants in the vegetation:

,

(12.4)

where Rst,Rm, Rcut, Rac and Rsoil are the stomatal, mesophyll, cuticular, in-canopy aerodynamic and soil resistances, respectively. The stomatal resistance is a key parameter in deposition modelling, which is affected in different degree by both the weather conditions and several plant and soil characteristics. The scheme of this simple resistance network can be seen in Figure 12.3.

Considering, that the flux is constant between the reference height and the top of the canopy, the total ozone flux can be written as follows:

,

(12.5)

where Cr is the concentration at the measuring height, and Cc is the concentration at the top of the canopy, defined as a level, where the flux divides into stomatal (Fst) and non-stomatal (Fns) part (Cieslik, 2004):

,

(12.6)

where Rst is the stomatal resistance and Rns is the non-stomatal resistance covering all deposition pathways but stomatal. According to Equations (12.5) and (12.6), the stomatal flux is calculated separately:

,

(12.7)

therefore:

.

(12.8)

Figure 12.3: A widely used simple resistance network in dry deposition models

The main limitation of these deposition models lies in the uncertainty and variability of the model input data, such as the time- and species-dependent parameters. Therefore, these parameters may give rise to significant uncertainties in the simulation results, and it is very important to know the effect of the individual input parameters on model output. Nonlinear models, such as most of the deposition models, can magnify the uncertainties of some parameters and damp others. In many cases, the models may over- or underestimate the fluxes through the calculation of deposition velocity (Mészáros et al., 2009a).

12.1.3. Some results

A typical diurnal course of surface ozone flux obtained from eddy covariance measurements over grassland can be seen in Figure 12.4. (the negative flux means deposition). A pronounced difference appears between night-time and daytime fluxes. At night-time, due to the stable stratification in the near surface layer, the turbulence is weak, therefore no large flux can be observed. Night-time flux is generally no change significantly in time. After sunrise, the stratification becomes unstable, resulting increased flux. Over vegetated surfaces, daytime flux can be more explicit, when environmental conditions are optimal for exchange processes through the plants. At the same time, several environmental factors (e.g. high water stress, low turbulence etc.) can reduce the daytime deposition flux.

Typical daily course of the dry deposition flux of ozone

Figure 12.4: Typical daily course of the dry deposition flux of ozone (O3). Results are obtained from eddy covariance measurements over managed grassland (Braunschweig, Germany), on 7 June, 2000.

Beside weather conditions, the agricultural activities (e.g. cut, fertilization) can also significantly influenced the O3 fluxes. During a measuring campaign over grassland (Mészáros et al., 2009b), three different phases of vegetation were covered to describe the ozone flux under different conditions: tall grass canopy before cut, short grass after cut, and re-growing vegetation after fertilization. Results of ozone flux measurements indicated that daytime ozone flux decreased after cut, but to a smaller extent than would be expected due to the drastic reduction of Leaf area index (LAI), which decreased both stomatal and cuticular uptake of the ozone. However, at the same time, with decreasing vegetation height and LAI, the importance of ground flux and chemical reactions are increased. After fertilization, nitric oxide (NO) emissions are assumed to have increased (inferred by greatly increased soil nitrate levels) providing a further (chemical) sink for ozone thereby affecting the deposition. These complex, highly nonlinear effects reveal the importance of canopy structure and non-stomatal pathways on ozone fluxes.

Similar diurnal pattern of ozone flux was found over pine forest (Figure 12.5). The dry flux of the ozone was measured during a field campaign by eddy covariance method in May, 1998 at Nyírjes station (Mátra Mountain) and the dry deposition velocity of ozone (Figure 12.6) was derived from these measurements. Based on the measurements the average values of deposition velocity for ozone ranges between 0.1 and 0.3 cm s-1 in daytime, and was lower than 0.1 cm s-1 at night-time.

Next to the micrometeorological measurements, detailed model simulations were also performed for the estimation of the dry deposition of ozone. Based on the big-leaf models a one-dimensional ozone dry deposition model was developed for continental climate area. In this model, analogously to Ohm's law, the deposition flux is inferred by multiplying the concentration and the deposition velocity of ozone. The deposition velocity is related to different resistances in the soil-plant-atmosphere system (see Chapter 12.1.2). First investigations were performed for a pine forest (Figure 12.5 and 12.6) in May, 1998.

Average daily courses of measured and modelled dry deposition flux of ozone

Figure 12.5. Average daily courses of measured and modelled dry deposition flux of ozone (O3). Results are obtained from eddy covariance measurements over pine forest (Nyírjes, Hungary) and model simulations.

Average daily courses of measured and modelled dry deposition velocity of ozone

Figure 12.6: Average daily courses of measured and modelled dry deposition velocity of ozone (O3). Results are obtained from eddy covariance measurements over pine forest (Nyírjes, Hungary) and model simulations.

The dry deposition is strongly depending on the season and the type of the surface. Therefore, model simulations were extended to a longer period and for grassland, too. Figure 12.7 and 12.8 show monthly averages of daytime and night-time deposition velocity for ozone over pine forest and grassland, respectively. Average daytime values are higher (between about 0.25 and 0.35 cm s−1) over forest during the vegetation period. In winter, under mainly stable stratification the typical daytime deposition velocities are between 0.1 to 0.2 cm s−1. Above grassland, during the vegetation period, the daytime dry deposition velocities vary between 0.05 and 0.15 cm s−1. Night-time deposition is lower, than daytime values in all cases.

Monthly averages of daytime and night-time deposition velocity for ozone over pine forest

Figure 12.7: Monthly averages of daytime and night-time deposition velocity for ozone (O3) in 1998 over pine forest (Nyírjes, Mátra Mountain). Results are obtained from model simulations.

Monthly averages of daytime and night-time deposition velocity for ozone over grassland

Figure 12.8: Monthly averages of daytime and night-time deposition velocity for ozone (O3) in 2000 over grassland (Hortobágy). Results are obtained from model simulations.

Due to the higher leaf area index of the vegetation, as well as the stronger turbulence over the canopy, the deposition velocity generally higher in case of forest than for grass.

Dry deposition of ozone shows a typical temporal pattern. Both daily and yearly time series experienced a dual maximum. During the year, the greater values generally appear in late spring and in the beginning of autumn, while during the day in the morning, and afternoon. In winter and in night-time too, the stable stratification blocks the deposition. In summer-time period in the year, and in noon-time period in the day the higher water vapour deficit can also decrease the fluxes.

Deposition velocity of ozone has also been estimated by a deposition model using a regular grid (0.025 × 0.0375 degree resolution) over Hungary. The input meteorological datasets (air temperature, relative humidity, cloudiness, wind speed and air pressure) were taken from the ALADIN meso-scale limited area numerical weather prediction model used by the Hungarian Meteorological Service. Spatial distribution of deposition velocity and total dry flux of the ozone can be seen in Figure 12.9 and 12.10, respectively for July, 1998. The land surface was categorised into 11 land uses, and 5 soil types. Each land use category was associated with a seasonal pattern of vegetation parameters, stomatal response characteristics and other resistance parameters. In Hungary, in case of vegetation-covered surface, the ozone fluxes mainly depend on soil water content (Mészáros, 2009a). This impact prevails in day-time, when ozone transferred through the stomata, and it is negligible in night-time.

Average daytime deposition velocity for ozone in July 1998 over Hungary

Figure 12.9: Average daytime deposition velocity for ozone (O3) in July 1998 over Hungary. Results are obtained from model simulations.

Average daytime deposition flux of the ozone in July 1998 over Hungary

Figure 12.10: Average daytime deposition flux of the ozone (O3) in July 1998 over Hungary. Results are obtained from model simulations. (Negative sign denotes that the flux is downward.)

Monthly averaged deposition velocities with standard deviations at 12 UTC over different vegetation types

Figure 12.11: Monthly averaged deposition velocities with standard deviations at 12 UTC over different vegetation types (grass, orchard, agricultural land, agricultural land+forest, deciduous forest, coniferous forest, mixed forest) over the model grid.

Temporal variability of daytime ozone deposition velocity during vegetation period is presented in Figure 12.11 for different vegetation types (grass, orchard, agricultural land, agricultural land+forest, deciduous forest, coniferous forest, mixed forest). According to different plant physiology, and characteristics, there are significant differences among deposition velocities in each month even as over each surface type. Due to the plant growth (higher leaf area index), and the optimal environmental conditions for vegetation (higher temperature together with sufficient soil water content), generally higher values occur in June. However, for coniferous forest, higher deposition velocity values were detected in spring than in summer. In this case no significant changes in leaf area index between each period, at the same time the lower temperature in April and in May is more favourable for the stomatal uptake of this type of vegetation. These physiological effects can also be observable less in case of mixed forest, and mixed agricultural land and forest. Decreasing soil water content (due to the warmer period of the year without precipitation) in July and August in 2007 decreased the deposition velocities in all cases. In contrast of this, in September, the values were raised because the soil water content was increased again (for more details see Czender et al., 2009).