III Gaseous Pollutants 5-6

Literature related to higher plants (i.e. herbs, grasses, trees and shrubs) and air pollution is vast. There is a growing concern of the effects of air pollutants, particularly ozone (O₃) on crops and forests and literature in these areas are particularly extensive. Bennett (1996) prepared a floristic summary of plant species in the air pollution literature derived from the BIOLEFF database (Bennett and Buchen, 1995). Smidt (1996) summarised the type of assessments used in the evaluation of air pollution stress on forest ecosystems.

This section is directed towards published literature focusing on monitoring of gaseous air pollutants and no attempt has been made to cover the extensive publications on the effects of air pollutants on higher plants. Bioindicator plants have been used to indicate the relative air quality of specific areas and regions and can provide unique information as to the ambient air quality within a particular area.

The most common method of bioindication is visible injury. Prior to the establishment of the Environment Agency in England and Wales, UK, a guidance manual on the diagnosis of air pollution injury to vegetation was produced by HM Industrial Air Pollution Inspectorate (HMAPI) (Taylor et al., 1990). This is a comprehensive volume, which successfully attempted to provide a guide to aid personnel in identifying the range of factors, which may lead to particular symptoms in the field, particularly in association with industrial sources. The text covers most major pollutants and a wide range of plant types and plant parts. This manual is unique in that it extends beyond reporting pollutant effects on vegetation in that it is structured to enable the user to detect the pollution (i.e. bioindication).

Generally, visible injuries are typically non-specific and may indicate various stresses on the plants. Increasingly physiological, structural and biochemical effects are being used in biomonitoring studies. These responses not only occur prior to visible injury and therefore represent early detectors but also are regarded as more precise and objective parameters.

Turcsanyi (1992) reviewed the use of plant cells and tissues as indicators of environmental pollution. The effects of acid gases such as NO₂, SO₂, and HF on cells and tissues were discussed. Saxe (1996) provided an excellent overview of bioindicative methods involving biochemical, physiological and ultrastructural processes. This was a sequel to a previous review by the author (Saxe, 1991). The extensive 1996 review focused primarily on forest decline and air pollution injury and considered laboratory and field based conditions. The mechanisms discussed included photosynthesis and stomatal conductance, leaf pigmentation, chlorophyll fluorescence, element content, metabolite content, enzyme activity, morphological analyses, ultrastructure and histopathology and genetical analysis. This literature is used extensively in the succeeding sections. References

5.2 Critical loads/levels

The concept of critical loads and levels is gaining profile world-wide. Critical loads and levels are not directly related to biomonitoring using plants as such but do warrant recognition in this report. Bioindication using sensitive species plays a role in establishing critical levels and loads.

The basis of the critical levels/loads approach is that that once the sensitivity of an ecosystem to acid deposition or to gaseous pollutants is established a critical load of wet deposition or a critical level of gases is calculated and set. Below this value the ecosystem will not be significantly damaged. These values are then compared with actual or estimated concentrations at the locations under study. Load and level exceedances can then be established and emission reductions and controls can be initiated. This is regarded as a more scientific, quantifiable method of assessing effects and planning emission controls compared with the previous arbitrary percentage reduction approach. Critical loads and levels are being mapped for countries in Europe and North America. To date, critical levels have been set for the effects of SO₂, O₃, NO_x, and NH₃ on crops, trees and natural vegetation by UN ECE (1993). Values were based on literature of the effects of these pollutants on vegetation. The next stage is to map and/or model critical level exceedance.

It is beyond the scope of the report to appraise methods, models and approaches to setting and determining exceedance of critical levels and loads. Instead the reader is referred to the following useful pieces of literature. Bull (1991) reviewed some different approaches to the calculation of critical loads and levels. Sanders et al. (1995) discussed specifically how critical levels for the effects of air pollution on crops, trees and natural vegetation. Exceedance of critical levels of O₃ in Europe was treated by Benton et al. (1995). The use of critical load maps in estimating the impact of air pollution on environmentally valuable sites in the UK was discussed by Brown et al. (1995). More recently, results of exceedance mapping of O₃ for crops and trees in the Netherlands were presented by de Leeuw and van Zantvoort (1997). Setting air quality standards, particularly O₃ to protect vegetation in the US were discussed by Lefohn and Runeckles (1987), Lefohn and Foley (1993) and Larsen et al. (1993). References

5.3 Monitoring design

Ideally plants which show specific responses to specific pollutants are most suitable as bioindicators. Sensitive species are more useful as bioindicators and tolerant species are more appropriate as accumulative indicators. Agrawal et al. (1991) used the air pollution index (APTI) to examine the susceptibility of a number of plant species growing in an urban/industrial region of India. APTI considered levels of total chlorophyll, ascorbic acid, leaf pH and relative water content. Plants were classified into three categories of sensitivity:

The authors demonstrated that plants classed as sensitive to SO₂ under laboratory conditions were also ranked as the most sensitive species by APTI calculation under field conditions. Kalyani and Charya (1995) calculated the APTI of 54 plant species under natural conditions in Warangal City to establish which species were the most suitable for biomonitoring purposes.

Species selection primarily involves choosing herbs and/or grasses or tree species. References

Kovács (1992b) and Taylor et al. (1990) listed sensitive and accumulative plant indicators of gaseous air pollutants. Seedlings, semi-mature and mature plants have been used in air monitoring studies. The age of the leaves used in bioindication is important. Apparently young leaves of cultivars of the genus Petunia are particularly effective indicators of SO₂ pollution. Gladiolus gandavenis is an excellent bioindicator by showing marked responses to fluoride pollution. Kozuharov (1986) reviewed the appropriateness of plants as bioindicators. The author suggested that in annual plants, younger specimens were more sensitive than older ones. Such a distinction in the different stages of perennials is not as obvious. References

Air pollution studies in relation to trees may involve tree parts, individual trees, tree stands or entire forests. Tree foliage reflects changes in pollution conditions within a relatively short time (two or three years). By contrast, air pollution impacts on entire tree stands are detected in the much longer term. Tree stands are commonly used in the assessment of forest decline of which air pollution is believed to be an important contributing factor.

Suitable tree bioindicators of air pollutants were listed in Kovács (1992c) and Taylor et al. (1990). Coniferous trees are regarded as more sensitive indicators of air pollutants than deciduous trees. This can be attributed to their continual exposure to air pollution all year round, their longevity and their response to low levels of contaminants. References

Site selection will depend on the extent and purpose of the survey. For the EC and UN ECE programme on the assessment of forest condition in Europe, the transnational survey is based on a 16 x 16 km transnational grid of sample plots. The national surveys are conducted on a national grid basis.

During biomonitoring programmes using higher plants it is often important to consider other environmental conditions at the study site.

Sampling procedures should be consistent throughout the study period and area. Jones et al. (1991) emphasised that if a bioindicator is used in different locations for comparison of responses, then the same variety and seed source should be used. In addition only plants at similar stages of plant development should be compared. Plant parts of the same age should be used consistently throughout a survey.

Sample frequency will depend on the plant under investigation. Some plant species may be restricted by time of year or even time of day. Generally the greater the sample frequency the better. However, this is in many cases limited by resources. In addition the collection of extensive amounts of data is fruitless without effective programme planning and design in the offset. It is pointless to concentrate efforts on collating large amounts of random data, which provide us with little more information than a smaller, focused dataset.

Some countries have adopted standard, national methods for sampling. For example, in a Finnish study of S content in pine needles sampling, preparation and analyses were undertaken in accordance with Finnish standard method SFS-5669 (Haapala et al. 1996). References

5.4 Sulphur dioxide (SO₂)

Many factors affect the sensitivity of plants/trees to SO₂ (Taylor et al. 1990). Young fully expanded leaves/needles are more sensitive than older needles and generally seedlings are more sensitive than older plants. Drought, low temperature, N, S, and P deficiency all reduce SO₂ sensitivity whereas high relative humidity, wind, K and Ca deficiency increase sensitivity. References

Changes in species distribution of higher plants are rarely used in biomonitoring in comparison to lower plants. Many established methods are available to assess species diversity, cover etc. but these are generally used as conservation tools not as biomonitoring tools. Cape (1989) suggested that changes in composition of sensitive herbaceous species would be appropriate bioindication of air pollution in forest ecosystems. Similar decreases in the floristic composition of the herbaceous layer were observed in zones of increasing pollution around Shaktinager Power Plant in India (Agrawal et al. 1991).

Sweden undertakes extensive vegetation surveying at national reference areas as part of their national monitoring programme (PMK). Vegetation is surveyed every twenty years at transects situated in one square kilometre catchments (Bernes, 1990). More frequent sampling is taken in circular subplots of radius ten metres along the transects. The development of trees and tree stands (e.g. their status, trunk diameter, fallen trees) are recorded. Within these circular plots more intensive surveying of all vegetation is carried out in smaller 'tree plots'. In the vicinity of the tree plots more intensive sampling of tree, shrub, field and ground layers is undertaken. This includes estimations of cover and fertility. Even more thorough monitoring of ground and field layers takes place in 'intensive square plots' containing 0.5 x 0.5 m quadrats.

In 1996 the Swedish Environmental Protection Agency produced a report on the impacts of air pollutants on processes in small catchments which forms part of Sweden's integrated monitoring network. (Brakenhielm, 1996). Results from intensive plot monitoring were reported. Estimates of percentage cover, species diversity and the composition of indicator groups of lichens, mosses and vascular plants were recorded. Species diversity was calculated by the Shannon-Weiner (H') index according to the following formula:

The composition of the indicator groups was expressed by two indices, the Acid Tolerance Index (ATI) and the Nitrogen Demand Index (NDI) where:

These were calculated for each site and were based on the values of sensitivity of each species to acidification and eutrophication. The data was subjected to Principal Component Analysis (PCA) and regression analysis. The NDI appeared to be the best parameter for detecting N deposition-induced impact on forest understorey vegetation. References

Measurements of plant injury can be very uncomplicated (e.g. injured vs. non-injured) or more elaborate such as the application of numerical values or indices as measures of severity. Both systems are discussed below. Plant injury symptoms generally indicate acute damage to plants. References

Most multi-national monitoring programmes are concerned with forests and tree stands. One of the most comprehensive ongoing programmes in Europe is the monitoring of forest condition. According to Lorenz et al. (1997) and Lorenz (1995) 'In response to growing concern about forest damage caused by air pollution, in 1985 the United Nations Economic Commission for Europe (UN ECE) under its Convention on Long-range Transboundary Air Pollution (CLRTAP) established the International Co-operative Programme on the Assessment and Monitoring of Air Pollution Effects on Forests (ICP Forests). In 1986 the Member States of the European Union (EU) agreed upon the European Union Scheme on the Protection of Forests against Atmospheric Pollution. Since then ICP forests and the European Commission (EC) have been monitoring forest condition in close co-operation.'

This programme is not aimed exclusively at SO₂ but includes air pollution in general (e.g. O₃, NO_x). For convenience general air pollution studies will be included in this section.

Defoliation and discolouration are the major assessment parameters of crown condition in the transnational survey. Defoliation is assessed by 5 to 10% stages in comparison to a standard reference tree. This has replaced the traditional five-class system given in Table 3.5. Discolouration is also evaluated by a five-class system (Table 3.5). Additional parameters reported include country, plot number, plot co-ordinates, altitude, aspect, water availability, humus type, soil type, mean age of dominant storey, tree numbers, tree species, observations of easily identifiable damage. Crown condition is not a specific indication of air pollution damage since many other conditions such as climate, drought or disease may cause similar symptoms. However, by undertaking correlation analysis with the aforementioned parameters some factors, which show no correlation, may be eliminated. Analyses of results indicate that crown condition has deteriorated over the 11 year study period, particularly in central Europe. Although some discrepancies in sampling work have occurred between countries, this is an example of an effectively co-ordinated, large-scale, biomonitoring programme. The collection of a long-term dataset will allow the determination of spatial and temporal patterns of forest ecosystems and individual tree species. This integrated approach will also aid in the identification of more specific cause-effect relationships. References

The UN ECE programme mentioned above incorporates national surveying and reporting currently from 33 countries (Lorenz et al. 1997). The national surveys vary in intensity depending on country. For example, in Finland five variables are used in the assessment of tree vitality: defoliation, crown degeneration, needle and leaf discolouration, cone yield and male flowering on conifers and abiotic and biotic damage. Whereas other countries employ the transnational 5% class method alone.

The most recent synopsis of tree health as a consequence of air pollution in the UK was printed in 1993 (DoE, 1993). The results indicated localised areas of decline in crown condition in the UK. References

In response to increased acid deposition from S and N in south-east Norway, a local monitoring network of Norway spruce stands growing on soils of poor neutralising capacity (Solberg and Tørseth, 1997). By using crown density as an indicator of tree health, surveyors found no evidence to support the hypothesis that S and N deposition are having deleterious effects on crown condition. The authors stipulated that other environmental factors might be concealing any limiting effects that deposition may be exerting.

An earlier study on the effects of air pollutants (mainly S and N) on ecological forest condition in the eastern part of the Gulf of Finland was conducted by Haapala et al. (1996). Bioindication methods were based specifically on collected needles and more generally on tree specimens.

Twigs were removed from the sample plots and the number of needle age classes counted in addition to the number of needles in each age class. A sample of 100 needles from each plot were analysed as follows:

This index of damage based on a qualitative five-point scale has been used in other studies. For example in the Czech Republic (Tichy, 1996).

Needle damage parameters did not show great statistical variation between sampling areas but some localised acute damage to needles was recorded. Needle age structure differed between areas, with general lowest needle longevity in the polluted regions. Defoliation was not regarded as severe and no significant differences between sites were obtained. By contrast epiphytic lichen distribution was seriously affected by pollution. The authors concluded that a whole array of abiotic and biotic factors affect tree stands and the effects of air pollution stress were not as apparent as observed for lichens.

The Canadian Forest Service have been monitoring the condition of Canadian boreal forests since 1984 under the auspices of the Acid Rain National Early Warning System (Hall, 1995). Two classification systems are applied: conifer tree crown and hardwood tree crown classification systems. The former is based on the amount (expressed in nine percentage classes) of defoliation in the crown. Hardwood crowns are classified into 13 classes according to the visible parts of the outer crown, the quantity and quality of foliage in the outer crown and the percentage of the outer crown that contains bare twigs and dead branches. By noting any observed effects of climatic conditions, nutrient deficiencies, air pollution, insects and diseases, poor tree condition and mortality was not attributed to long range transport of air pollutants.

SO₂ fumigation of 41 seedlings of herbaceous species in England resulted in reductions in dry mass of plants, roots and shoots in many of the species (Ashenden et al., 1996). Antagonistic effects were more apparent on root growth than shoot growth. The former could possibly be used in bioindication of SO₂. References

Taylor et al. (1990) review of the symptoms of acute SO₂ injury in vegetation will be summarised in the succeeding paragraphs. The authors also present in tabular form, concentrations of SO₂ known to cause injury in a variety of plants. Dose-response concentrations will depend on exposure time. Short, high doses are probably more representative of industrial emissions and/or accidents. It is also important to note that many of these concentrations will be based on experimental conditions and more field evidence is required.

The symptoms of acute damage (SO₂ >1 ppm) according to Kovács (1992b) are necrosis on the upper and lower leaf surfaces, at the apices, margins and between the veins. The tissues around the stomata may also decompose. Taylor et al. (1990) also reported watersoaked appearances on leaves in many species. Specific coloured leaf necrosis is common. For example light brown necrosis in daffodil, grey necrosis in geranium and black necrosis in broad beans. Tip necrosis on sepals has been observed in marigold and gladiolus. Necrosis on awns of grasses and cereals has been reported. Barley, bracken and clover are regarded as very sensitive species to SO₂ exposure.

Observation of visible symptoms was used to assess air pollution burden in three locations in Egypt (Ali, 1993). The extent of chlorosis, necrosis, red pigmentation and growth parameters such as height, leaf area and stem diameter in clover and Egyptian mallow plants was generally related to pollution load.

To test the suitability of a variety of plant species as bioindicators, Agrawal et al. (1991) subjected the plants to two hours a day of 0.15 ppm SO₂ for a period of 30 days under controlled laboratory conditions. Most plants showed bifacial interveinal chlorosis and necrosis but the authors concluded that the use of foliar symptoms alone was not a specific enough method of bioindication unless one air pollutant dominated. References

Acute visible injury affecting various parts of coniferous trees has been observed in response to SO₂. Young needles show chlorosis and are poorly developed and stunted. Middle aged needles often show yellow then red/brown discoloration succeeded by necrosis. Necrosis commonly affects the tip first but in larch and spruce the tip is often not the most sensitive part of the needle. Abscission of older needles in fir trees is frequently an immediate response but generally occurs after several months SO₂ exposure in pine trees. SO₂ exposure may bleach stems of young shoots.

Larch is viewed as a particularly sensitive tree species to SO₂. Taylor et al. (1990) reported solely on leaf damage in broad-leaved trees and shrubs. Symptoms include interveinal chlorosis, irregular interveinal necrosis and abscission in many species. Necrosis characterised by brown/orange colouring has been revealed in lime, beech and hazel trees while black necrosis has been observed in pear trees. Distortion, puckering and curling of leaves has been detected in birch and maple. A scale to assess the extent of necrotic spots in birch (Betula pendula) was devised by Jäger (1980) as cited by Kovács (1992c). Observations of reduction in annual shoot growth in broad-leaved trees have been identified (Kovács, 1992a). References

Sulphate from external SO₂ can be accumulated in plant shoots, a portion of which may be transported to the root (Rennenberg et al., 1996). Miller (1989) reported on a few American studies of decreased S foliar tissue content with increasing distance to industrial works.

Manninen and Huttunen (1995) found very good correlations between S content in young Scots pine needles and SO₂ load. The authors concluded that needles under pollutant stress are extremely influenced by high short term SO₂ doses which have implications for the setting of critical levels to forest ecosystems. S concentration in Scots pine needles from forest of the Gulf of Finland correlated with S emissions in the region (Haapala et al., 1996). Regression analysis demonstrated pollution gradients in pine needle content with increasing distance from known pollutant sources. S content in two Ligustrum species at a variety of sites in Argentina was elevated in those sites with high traffic density (Carreras et al., 1996).

Foliar analysis of S and N was used to assess the air pollution burden in national parks and landscape protection areas in Slovakia (Mankovska, 1997). S content ranged from 0.72 to 6.77 g kg^-1 in hardwoods and from 0.98 to 4.3 g kg^-1 in softwoods. Results indicated cause for concern in some areas. References

Plant responses depend not only on the characteristics of a species but also on the stage of development, age and nutritional status of the plants (Kovács, 1992b). The biochemical/physiological and ultrastructural changes discussed in the following paragraphs represent chronic damage to plants. The following information summarises and supplements the extensive review in this area undertaken by Saxe (1996). Much of the work on the effects of pollutants on plants is under controlled laboratory or field conditions using fumigation chambers. There is paucity in natural field based research or even correlations between laboratory and fieldwork. Open top chambers have tried to overcome some of the pitfalls of laboratory experiments but results should still be interpreted with caution.

The lack of availability of simple methods is a potential drawback to using responses at cellular/molecular levels for the bioindication of air pollutants. However, these techniques may often provide additional information on causal factors.

Agrawal et al. (1991) showed that different plant species showed varied responses to different parameters. For example, a species which showed the highest reduction in relative water content in response to SO₂ did not show the greatest reduction chlorophyll content.

The choice of physiological/ biochemical/ultrastructural method will depend on the species. This section will focus on plant responses, which can be used in bioindication. Areas of activity such as the mechanisms of air pollution effects on plants, dose-responses and gas absorption are beyond the scope of this text. Hippeli and Elstner (1996) and Winner (1994) treat some of these aspects. References

Measurement of photosynthesis and stomatal conduction are common measures of gaseous air pollution damage in that they respond quickly to air pollutants and can be measured by non-destructive techniques (Winner, 1989). They are obvious measures of gaseous air pollution in that the primary function of stomata in plants is gas exchange.

Csintalan and Tuba (1992) and Saxe (1996) reviewed published experiments of the effects of SO₂ on photosynthesis, stomata functioning and transpiration. Depending on dose, exposure time and species and other abiotic and biotic factors, SO₂ can increase or decrease photosynthesis or open or close the stomata.

Disruption of chloroplast metabolism has been implicated in the inhibition of photosynthesis due to SO₂ exposure under controlled conditions (Veeranjaneyulu et al., 1990). Stomatal conductance in association with S and N foliar content was used to assess the air pollution burden in national parks and landscape protection areas in Slovakia (Mankovska, 1997). References

In their study of suitable biomonitoring parameters, Agrawal et al. (1991) suggested that determination of total chlorophyll level could be a good bioindicator of chronic SO₂ conditions. Total chlorophyll is measured as part of the APTI mentioned in Section 5.3.1.

Chlorophyll reduction directly relates to damage in plants (Heath, 1989). However it is often regarded, like chlorosis, as a non-specific indication of SO₂ stress. Chlorophyll content was lower in one year old needles of damaged spruce trees in comparison with healthy specimens when studied in three different sites in northern Germany (Godbold et al., 1993). No correlation with a specific pollutant was made. Total chlorophyll determination in potted plants transferred to three different locations in Egypt showed correlation between pollution burden at the sites (Ali, 1993). Both clover and Egyptian mallow indicated up to 29% reduction in total chlorophyll levels in plant leaves grown in the most polluted sites. Also under field conditions Pandey and Agrawal (1994) found reductions in chlorophyll content in leaves of three woody perennials in an urban area of India. Chlorophyll levels were correlated with ambient SO₂ concentrations.

The usefulness of chlorophyll fluorescence has been discussed in Section 2.3.3.2. This method is regarded as a more sensitive measure of photosynthetic activity than pigment content and more specific than the measurement of photosynthetic rate (Saxe, 1996). Chlorophyll fluorescence assessment is non-destructive and can be achieved with relatively little effort in the field (Saarinen and Liski, 1993). Chlorophyll fluorescence using a portable fluorometer in the field and pigment (chlorophyll a and carotenoid) content determined by laboratory analysis in Scots pine needles were mapped around the vicinity of an oil refinery in Finland by Saarinen and Liski (1993). References

Saxe (1996) provides several examples of metabolic changes to plants induced by air pollutants. These include changes in amino acid, polysaccharides and ATP/ADP ratios. A recent study by Julkunen-Titto and Lavola (1995) demonstrated changes in the production of phenolic secondary chemicals and soluble sugars by willow species in response to 0.11 ppm SO₂ fumigation for three weeks.

Enzyme activity has been used as a biochemical stress bioindicator of air pollutants. In Germany, acid phosphatase and peroxidase activity in needles of healthy Norway spruce trees was generally lower than in damaged trees (Godbold et al., 1993). However, the authors could not relate these effects to specific stress factors. Saxe's (1996) review summarised similar findings.

Saxe (1996) and Berg (1989) documented varying responses of leaf cuticles to pollution stress in the published literature and concluded it was not a very specific bioindication tool. By applying electron microscopy, Manninen and Huttunen (1995) observed that the epicuticular wax structure of Scots pine needles was very badly degenerated in trees in close vicinity to an oil refinery in southern Finland. The destruction rate of the needle surface wax decreased with decreasing S content in needles. References

Cook and Innes (1989) reviewed the value of tree ring analysis in assessing the impact of lower-level regional air pollution on forests. Tree ring patterns offer a long-term, baseline dataset whereby changes in growth (displayed as annual tree-ring increments) in response to pollution stresses can be detected. This method is complicated by the fact that air pollution effects on ring increments are not necessarily distinct and may be prone to misinterpretation. The potential of tree-ring analysis as a bioindication method in air pollution diagnosis is obvious. However, further research in this field is necessary to determine the credibility and future of the technique.

Currently, increments can be quantified non-destructively or destructively. Non-destructive analysis measures the radial ring widths from cores taken at breast height of a tree. Detailed stem analysis is destructive but provides a more accurate measure of annual volume increment and complete growth layer profiles.

A more recent review of the application of tree-ring analysis in air pollution studies is presented by Turcsanyi (1992). References

By effecting photosynthesis and translocation, gaseous pollutants may reduce carbon allocation to roots. This in turn may reduce root growth, turnover and capacity for water and nutrient uptake. Richards (1989) reviewed the techniques used to measure these responses in their application in the monitoring of gaseous air contaminants.

Any response method is complicated by the fact that several factors may cause the same or similar reaction. Specificity is enhanced if a number of these responses are observed together (Saxe, 1996; Miller, 1989). Combining several response parameters into multivariate indices is a promising tool recommended by Saxe (1996).

Four pollution zones around a power plant in India were observed by analysing changes in a selection of physiological and biochemical parameters in plant leaves (Agrawal et al., 1991). Correlations between ambient SO₂ concentrations and decreases in the levels of chlorophyll, ascorbic acid and specific leaf area and increased S content were obtained. The authors recommended this type of monitoring as an essential companion to chemical monitoring in India. References

Little published literature appears to exist in relation to transplantation of higher plants to polluted sites. A method known as grass cultures was discussed by Weinstein and Laurence (1989). This is a system predominately utilised in Germany whereby self-watering rye grass cultures are established and samples removed at specific intervals. Samples are analysed for SO₂, F or heavy metals. More details of the system are provided in Kovács (1992a).

Specimens of the tropical shrub Carissa carandas grown in controlled pot conditions were transplanted to a selection of sites of varying pollution load in India for a period of two years. At four monthly intervals plant growth and morphological parameters such as height, leaf number, extent of chlorosis and necrosis and basal diameter were observed. The aim of the study was not primarily for biomonitoring purposes but to establish the adaptational responses of the shrub. References

5.5 Fluoro-compounds

The most phytotoxic and best studied fluoro-compound is hydrogen fluoride gas. Other gases include silicon fluoride (SiF₄) and fluorine (F₂). In the following discussions fluoride as reported in the literature is thought to represent the total fluoro-compounds measured unless otherwise stated. Fluorides differ from other gaseous air pollutants in that they are readily accumulated in tissues. Therefore, chronic exposure over long periods of time can result in translocation to leaf tips and margins where visible symptoms are exhibited.

Many of the factors which affect plant sensitivity to SO₂ are applicable to fluorides. An interesting observation in relation to fluoride is that different cultivars of the same plant species react in very different ways to the pollutant. For example, there is a large variation in the sensitivity of different cultivars of gladiolus and tomato plants. The susceptibility of gladiolus varies with flower colour. References

The following paragraphs summarise the reviews by Kovács (1992b) and Taylor et al. (1990) on fluoro-compounds (measured as total fluoride content) and air pollution injury. The characteristic symptoms of fluoride damage are tip and marginal chlorosis which later extend to the inter-venial areas. This is succeeded by tip and marginal necrosis which gradually covers the whole leaf area. Loss of leaf may result. Ivory necrosis has been observed on tomato, wheat and oats, red/brown in St. Johns Wort and black in dahlia. Tip necrosis has been recorded on awns and bracts and marginal necrosis on sepals and petals.

Kovács (1992b) presented a scale devised by Arndt et al. (1984) which relates the magnitude of leaf necrosis in vine leaves to fluoride pollution (Table 3.6). References

Damage class	Symptoms


No recognizable damage	Slight apical and spot necrosis on every tenth leaf
	No difference in growth vigour

Slight damage	Recognizable necrosis on every fifth leaf
	Necrosis damage approx. 2 to 3 cm²
	Growth of axillary shoots normal

Medium damage	Every second leaf is necrotic
	Growth of entire plant reduced

Extensive damage	Assimilating surface of every leaf is reduced
	Entire marginal area is necrotic
	Tendrils are shorter

Most extensive damage	Each leaf damaged
	Abscission occurs
	Tendrils stunted

Total damage	>80% of leaves are entirely necrotic
	Axillary shoots and tendrils are missing
	Growth is greatly reduced

Gladiolus cv. White Friendship and Hemerocallis cv.References ed Moon were cultivated and exposed to twelve sites in the region of Cubato and the Serra do Mar at varying distances from hydrogen fluoride (HF) emission sources for 28 days (Klumpp and Klumpp, 1994). HF induced injury, estimated by measuring the area of tip and margin necroses on the sample plants, was most severe at sites nearest the fluoride-emitting fertiliser industries.

In coniferous trees, young chlorosis in young needles is a common symptom of fluoro-compound contamination, followed by red/brown discolouration. Tip burn may occur which usually results in necrosis of the whole needle. Frequently necrosis appears as red/purple bands distinctive from healthy tissue.

The leaves of deciduous trees and shrubs show marginal and tip necrosis which turn into sharply defined red/brown bands between necrotic and healthy tissue. Necrosis in fruiting parts of trees has also been observed. References

Miller (1989) reported on a few American studies of decreased fluoride foliar tissue content with increasing distance to industrial works. As part of an active biomonitoring scoping study in Brazil, Klumpp and Klumpp (1994) found that foliar fluoride concentrations in transplanted Gladiolus cv. White Friendship and Hemerocallis cv. Red Moon correlated with severity of fluoride-injury. Plants at close vicinity to fluoride-emitting fertiliser industries contained fluoride content of 80 to 120 µg g^-1 dry weight compared to content of >10 µg g^-1 dry weight at reference sites.

Chemical analysis of transplanted standardised grass cultures of Lolium showed that it was a much more effective bioaccumulator of fluoride than the two foregoing plant species.

Photosynthesis is inhibited at relatively low hydrogen fluoride (HF) concentrations and this may be a potential bioindicative response. Plant stomatal conductance probably shows the greatest variation in response to HF than any other gaseous air pollutants (Csintalan and Tuba, 1992). References

5.6 Nitrogen oxides and ammonia (NO_x and NH₃)

Nitrogen content in plants is regarded as a questionable bioindicator of NO_x and NH₃ because it so easily translocated throughout the plant (Saxe, 1996). However, some crop species such as bean, leak and pea are regarded as very sensitive to nitrogen oxides. Kovács (1992b) lists sensitive and accumulating plant indicators of nitrous gases. Young leaves and needles are more sensitive to NO_x than older ones. High relative humidity and N deficiency increase plant sensitivity to NO_x, whereas N excess and drought conditions decrease sensitivity. NH₃ has received less attention in the literature than NO_x.

Relatively higher concentrations of NO₂ are needed to produce acute symptoms on plants in comparison to SO₂. Different plant groups react differently to NO_x and NH₃ (Taylor et al., 1990). Concentrations of these two pollutants known to cause injury in a variety of plant species are found in Taylor et al. (1990). Most reports of acute pant injury in response to NH₃ are in relation to accidental release or spillage. References

Many species show a watersoaked appearance on the leaves followed by necrosis in response to acute NO_x exposure. Leaf glazing has been observed in annual poa, cabbage and spinach. Necrotic streaking and interveinal necrosis has been recorded in many narrow-leaved and broad-leaved species. Legumes among many other species show ivory necrosis while some species display yellow, orange or brown necrosis. Tip necrosis has been observed on other plant parts such as awns, bracts and sepals.

Yellow discolouration, watersoaked appearance, glazing and bleaching have been observed on leaves in response to NH₃. Red/purple discolouration of leaf upper surfaces in cereals is common. Ivory necrosis, reddish necrosis and brown/black necrosis are typical NH₃-injury symptoms in certain species.

Chlorosis of young needles is a common symptom in response to NO_x. Tip burn of older needles is often observed in coniferous species. Pine trees display bleaching followed by sharply defined red/brown bands between necrotic and healthy tissue in older needles. Immediate abscission of older needles occurs in spruce.

Red/yellow discolouration of young spruce needles has been observed. Many coniferous species display black discolouration and tip burn of older needles as NH₃-injury symptoms.

Herringbone necrosis in the older leaves of beech, hazel and apple trees have been observed. Ivory necrosis, red/brown necrosis and black necrosis have been recorded in certain species.

Yellow discolouration of leaves has been observed in sycamore species. Other symptoms recorded in many species include watersoaked appearance on leaves, intercostal necrosis and finally desiccation and abscission of damaged leaves. References

Published literature with respect to the effects of NO_x and NH₃ is limited and literature on the use of biochemical and physiological responses as bioindicators is even sparser.

Plant photosynthesis and stomatal conductance are thought to be relatively tolerant to NO_x exposure (Csintalan and Tuba, 1992).

Nitrate reductase activity in plants has been measured in response to nitrogen deposition. Addition of ammonium nitrate to experimental grassland plots in the UK increased nitrate reductase activity in vascular plants (Morecroft et al., 1994). Krywult et al. (1996) found higher nitrate reductase activity in ponderosa pine needles correlated with sites characterised by higher nitrate deposition to branches. However, the authors concluded that nitrate reductase activity was not a sensitive enough parameter to be used in bioindication studies because it is influenced by many other biotic and abiotic factors.

Other enzymes may have potential as useful response parameters in nitrogen deposition bioindication studies. The possible role of the enzyme glutamate dehydrogenase in the adaptation of plants to ammonia assimilation with respect to air pollution was discussed by Schlee et al. (1994). The activity of this enzyme may be used in bioindication of nitrogen deposition but substantial further work would be required. Glutathione reductase and ascorbate peroxidase activity in red spruce needles was enhanced by exposure to acidic mists (Chen et al., 1991).

Soares et al. (1995) proposed the use of multivariate analysis of physiological and biochemical parameters such as enzyme activity and total nitrogen content of woody and herbaceous plant species as a means of assessing plant susceptibility to acid rain. References

5.7 Ozone (O₃)

Currently tropospheric O₃ is probably the gaseous pollutant of most concern to plant scientists. Estimates suggest that tropospheric O₃ concentrations have increased substantially over the last few decades. The available literature on the adverse effects of O₃ on forest and crops has accelerated over the last 10 to 15 years. Much effort has been placed on finding suitable cultivars for bioindication purposes.

In the assessment of O₃ on plants de Leeuw and van Zanvoort (1997) claimed that the spatial distribution of O₃ over a particular area is of more interest than concentrations recorded at monitoring stations. Plant biomonitoring techniques are particularly useful in establishing the distribution of air pollutants in a study area.

Ozone is not bioaccumulated in plants and can only be detected by sensitive plants. The major effects of O₃ on terrestrial vegetation include visible foliar injury, reductions in growth and productivity, changes in crop quality and increased sensitivity to either abiotic or biotic stresses. References

With respect to O₃, most attention in the published literature has been placed on observations of visible injury in plants and trees. Much of the work reported here is from field based studies but fumigation studies are still necessary.

Taylor et al. (1990) presented typical acute O₃ injury symptoms. As before these can be divided into plant type. O₃ symptoms tend to be exhibited in plant species after a relatively short exposure time.

Most visible O₃ injury symptoms have been described for crop species. In general, white, fawn, tan, grey and brown necrotic streaks on the upper surface of leaves are typical of O₃ injury. Kovács (1992b) presented typical O₃ injury symptoms on certain crop species. For example, bean plants show foliar browning and chlorosis, cucumber plants display white stipple, onion plants show white flecks and tip dieback while spinach plants exhibit grey to white flecks on leaves.

Chlorotic flecks later becoming pink lesions followed by orange-red tip necrosis have been observed on current year pine needles in response to O₃.

A variety of O₃ injury symptoms have been observed in deciduous trees and shrubs. Ash and maple show dense purple or reddish stipple on upper leaf surface. Many species including lime and apple show leaf bronzing while some species such as birch display leaf bleaching. Leaf curling and tip drying have been observed in lilac. Fruits are often affected and may prematurely drop (e.g. citrus trees). References

At one time, elevated concentrations of O₃ were thought to be restricted to urban areas but currently larger scale biomonitoring surveys of O₃ are in the increase. Visible damage to vegetation and agricultural crops over wide areas is becoming an important problem. Medium and long-range transport of O₃ precursors and O₃ itself have resulted in increased concentrations in rural areas. Bioindicators have been used to evaluate the effects of O₃ pollution in a number of impact areas.

The United Nations Economic commission for Europe (UN ECE) Convention on Long-range Transboundary Air Pollution includes an international co-operative programme to evaluate the effects of air pollution on agricultural crops. This programme has three main aims:

An initial stage of this project was to select a single species which could be used as a convenient bioindicator of phytotoxic effects of pollutants. Jones et al. (1991) reported on the potential of radish (Raphanus sativus L. cv. cherry belle) as a bioindicator for this programme. Radish seedlings were subjected to one of three treatments – placed in unfiltered, open top chambers, placed in filtered chambers or treated with ethylenediurea (EDU protects plants from O₃). Results demonstrated a reduction in radish plant yield (measured as shoot and root dry weight and leaf area) in the unfiltered chambers in the month of July. The authors concluded that cherry Belle radish plants were sensitive enough bioindicators to be used in a co-ordinated programme to evaluate the effects of O₃ on crop plants. References

From the published literature, Tobacco plants (Nicotiana tabacum), in particular variety Bel-W3, are the most widely used bioindicators of O₃ for the following reasons (Kovács, 1992b; Koppel and Sild, 1995):

The drawbacks of using tobacco plants are their susceptibility to diseases and pests and it can only be used during a limited period of the year. A review of the use and origin of tobacco cultivars until 1990 is presented in Heggestad (1991). A variety of techniques using tobacco have been employed in O₃ biomonitoring on a regional context.

Schenone and Lorenzini (1992) used three techniques to study the effects of regional air pollution on crops in northern and central Italy. The extent of ivory interveinal flecks on Nicotiana tabacum cv. Bel-W3 correlated with ambient O₃ concentrations indicating that levels exceeded the threshold for phytotoxic effects. Ambient air pollution exclusion experiments with open-top chambers demonstrated air pollution effects on crop productivity. Furthermore in controlled SO₂ fumigation chambers plant growth and physiology was only affected at concentrations above those measured at rural sites in Italy. This collaborated work confirmed that O₃ was the most important phytotoxic gas at a regional level in northern and central Italy. Further work in this area has continued over the past five years.

By estimating the percentage increase in injured leaf area (necrosis) of cultured semi-mature Nicotiana tabacum cv. Bel-W3, Mignanego et al. (1992) calculated LIIs at 23 sites in northern Italy over a six month period. Leaf injury was most pronounced in the summer months. This is the time of year when photochemical smog (of which O₃ is a major constituent) is generally at its highest. LII monthly and seasonal averages were calculated. The authors were able to characterise sites depending on their seasonal indices (SI) values.

Lorenzini et al. (1995a) reported on a modification of the procedures undertaken by Mignanego et al. (1992). An easily transportable miniaturised kit utilising two week old ozone supersensitive tobacco (Bel-W3) seedlings was used to monitor O₃ at 27 sites of varying pollution burdens in Tuscany, Italy. Design details are discussed in detail by Lorenzini (1994). In summary, tissue culture plates were used and seedlings were grown in an ozone-free environment before being transplanted to the monitoring sites for one week. The severity of O₃ pollution was assessed by recording the percentage area of injured cotyledon and leaf compared to control plants. Previous pilot fumigations studies validated that visible O₃ foliar injury in younger leaves, characterised by grey water-soaked marks (flecking), was analogous to those observed in mature leaves. Cotyledon injury showed good correlation with a number of O₃ parameters. This method was also successfully used to map O₃ distribution across a wide geographical area over a land-sea transect in Italy (Lorenzini et al., 1995b). The authors recommended the application of this simple, cost effective method for biomonitoring in large areas in developing countries.

Another approach using tobacco to estimate the phytotoxic effects of existing O₃ levels in two regions over a three year period in Spain was reported by Gimeno et al. (1995a). Three tobacco cultivars, Bel-W3, Bel-C and Bel-B of increasing sensitivity to O₃ were grown in greenhouses to the fourth true leaf stage before being transferred to sampling locations spread over the two regions. Ozone induced visible injury was assessed and mapped according to the classifications described in Table 3.7. This study demonstrated the need to consider environmental conditions in O₃ biomonitoring and in the calculating critical levels. At coastal sites the effects of relative humidity appeared to enhance O₃ injury in the sample plants. The use of the foregoing three cultivars was recommended by Heggestad (1991). References

Table 3.7 Characterisation of O₃ phytotoxicity using three tobacco varieties (from Gimeno et al., 1995)

Two varieties of tobacco, BelW3 (indicator) and Samsun (control) were used to study O₃ contamination in four locations in Estonia over a two year period (Koppel and Sild, 1995). Plants were grown in chambers for approximately 3.5 months prior to transplantation to the sampling areas. By observing leaves >10 cm in length, necrotic leaf indices (NI) were estimated as the percentage of the leaf blade covered with necrotic flecks. The mean daily increment of the necrotic index for the period between two observations (typically 6 to 14 days) was calculated for each plant (NII_plant) and for the site (NII_site) where the later was used to indicate the variability of O₃ episodes during the growing period. Highest NII_sitevalues corresponded to the site closest to major thermal plants and cities in Estonia. As may be expected O₃ injury was greatest in the year with the warmest summer. Temperature data is therefore a further consideration in O₃ biomonitoring and the setting of critical levels.

Other crop species besides tobacco have been used in O₃ biomonitoring studies. Watermelon (Citrullus lanatus) was used as a passive bioindicator of O₃ contamination in eastern Spain (Gimeno et al., 1995b). This study differed from the foregoing investigations in that it did not involve transplantation of plants grown from seed. Commercial fields of watermelon were visited and O₃ injury was evaluated using the classification outlined in Table 3.8. O₃ injury was widespread throughout the study area but varied in intensity. The authors recommended that all the considerations mentioned in Section 5.3.3 above should be adhered to during field sampling. Furthermore, plant canopy structure and the agricultural practices that affect it should be considered. This passive method of biomonitoring is simpler than transplantation exercises. However a drawback of using this approach over different areas is that the sample plants may be affected by other conditions unknown to the observer such as contaminated soil or drought.

Measurements of foliar injury in trees and shrubs have been used in the evaluation of regional O₃ pollution. The incidence and severity of O₃ injury in black cherry (Prunus serotina) was used to assess the impact and extent of O₃ pollution in Great Smoky Mountains National Park (Chappelka et al., 1997). Incidence was defined as the number of individuals with visible foliar symptoms of O₃ injury. Severity was measured as the percentage of foliage injured per plant and as the percentage of leaf area injured for the injured leaves. Statistical analysis was used to determine significant injury levels within different park areas. Skelly et al. (1997) also used black cherry in a study of the Desierto de Los Leones National Park in Mexico City. A sample of mature trees were subjected to an intensive survey, where the extent of surface stipple, leaf reddening and premature senescence was recorded. A more general survey estimated the percentage of trees exhibiting O₃ induced injury. References

Miller (1989) reported the common visual symptoms used in ozone-injury detection in coniferous forests in the United States:

The same paper reviewed some pine tree injury indices caused by ozone. Crown density, extent of chlorotic mottle on each whorl, needle length and extent of needle retention appear to be useful parameters in the calculation of combined indices.

At present, most O₃ injury based investigations are concerned with crops and forests over wide areas and urban and industrial point source investigations are limited.

De Bauer and Krupa (1990) summarised observations of typical O₃ injury symptoms to assess the air quality burden on vegetation in Mexico city over the past twenty years. Allegrini et al. (1994) favoured integrated evaluation of tropospheric O₃ pollution in urban and semi-rural areas in Italy using physico-chemical and biological monitoring.References eduction in biomass and leaf area of radish plants (Raphanus sativus L. cv. Cherry Belle) correlated negatively with ambient O₃ concentrations. The sensitivity of this cultivar favours their use as early warner systems to indicate that injurious levels of O₃ had been reached.

Kovács (1992b) presented a table indicating potential dangers to humans from O₃ exposure by correlating effects with O₃-injury symptoms displayed by tobacco. This is duplicated as Table 3.9. References

Short-term O₃ concentration	Physiological effect on:
	humans	tobacco (Bel-W3)
0.13 - 0.23	Decreased performance in sports (the partial pressure of blood oxygen decreases)	At high temperature and humidity the first flecks appear on the leaves
0.4 - 0.6	The mouth and throat become dry during sport; chest pains develop leading to asthmatic symptoms	The extent of flecks amounts to 50%
> 0.7	Without physical effort respiratory problems occur	Leaf fleck >90%
1 - 1.2	Premature death of ill and aged persons	Decay of plants

Literature with respect to laboratory and field studies using different types of chambers under ambient and fumigation conditions is extensive. Such studies are essential in establishing plant thresholds and responses prior to adaptation to natural conditions. For example, open top chamber experiments have been used to determine the most sensitive plant species to be utilised in biomonitoring studies. Such studies have been useful in classifying quantitative responses of plants to actual O₃ exposure.

Gimeno et al. (1995b) classified watermelon injury into four classes depending on the number of diurnal hours O₃ concentrations > 40 ppb using open top chamber experiments. This classification was then applied in a passive biomonitoring programme to determine the intensity and extent of crop damage in eastern Spain.

The chemical ethylenediurea (EDU), when applied to plants as either a soil drench or to foliage, can protect the plant against elevated O₃ concentrations. Hence the response of EDU-treated and untreated plants can be compared. However, the effectiveness of EDU is variable and this procedure should be used with caution (Jones et al., 1991).

Most experimental studies are in relation to effects of O₃ on crops and in establishing dose-response relationships but the information generated is often useful in determining sensitive species and potential bioindicators. Table 3.10 summarises some recent experimental procedures used to assess the response of certain plant species and cultivars to ambient and elevated O₃ exposure. This table is purely an illustrative guide to some examples of the published literature in different countries and is by no means exhaustive. References

Reference	Location	Experimental procedure	Species and cultivar	Response
Brennan et al., 1990	USA	EDU treated and untreated plots	Soybean (Glycine max L. Merr.)	Foliar damage
Carey and Kelley, 1994	USA	Open top chamber using ambient air, carbon filtered air and non-filtered air plus extra O₃	Loblolly pine seedlings (Pinus taeda)	Growth parameters
Clarke et al., 1990	USA	EDU treated and untreated plots	White potato (Solanum tuberosum)	Foliar damage
Kasana, 1991	England	Closed-chamber fumigation	Chick pea (Cicer arietinum), Black-gram (Vigna mungo), fenugreek (Trigonella foenumgraecum)	Growth parameters
Keller, 1988	Switzer-land	Closed chamber fumigation	American aspen (Populus tremuloides Michx.)	Growth parameters
Kobayashi et al., 1994	Japan	Closed chamber fumigation	Rice (Oryza sativa L.)	Yield and growth parameters
Kraft et al., 1996	Germany	Open-top chamber using charcoal filtered air and charcoal filtered air supplied with O₃	Spring wheat (Triticum aestivum cv. Turbo), white clover (Trifolium repens cv. Karina), maize (Zea mays cv. Bonny)	Foliar damage symptoms and spectral reflectance measure-ments
Maggs et al., 1995	Pakistan	Open-top chamber using ambient and charcoal-filtered air	Wheat and rice	Yield parameters e.g. grain weight
Meier et al., 1990	USA	Closed chamber fumigation	Loblolly pine (Pinus taeda)	Growth parameters
Paakkonen et al., 1997	Finland	Open-top chamber using charcoal filtered air and charcoal filtered air supplied with O₃	Betula pendula, Betula pubescens	Foliar damage and growth parameters
Pleijel et al., 1991	Sweden	Open top chamber using ambient air, charcoal filtered air and non-filtered air plus extra O₃	Spring wheat (Triticum aestivum L., cv. Drabant)	Yield and grain quality parameters
Renaud et al., 1997	Canada	Open top chamber using ambient and filtered air	Alfalfa (Medicago sativa L.)	Yield and growth parameters
Runeckles et al., 1990	Canada	Zonal Air Pollution System (ZAPS) simulating various exposure regimes	Processing peas	Yield parameters
Soja and Soja, 1995	Austria	Closed-chamber fumigation	Winter wheat	Yield parameters

Stomatal response to O₃ varies between species and even cultivars. Some species show an increase in stomatal conductance, some show a decrease and some are unaffected. However at concentrations above 200 ppm O₃, most plants close their stomata (Csintalan and Tuba, 1992).

Under controlled conditions, loblolly pine seedlings exposed to 0.12 ppm O₃ for seven hours per day, five days a week for 12 weeks, displayed a 16% reduction in photosynthesis in comparison with plants exposed to charcoal-filtered air (Spence et al., 1990). The same plants failed to exhibit visible injury under the elevated O₃ conditions. Schenone et al. (1994) used open top chambers to compare field grown bean plants under ambient conditions and filtered air. Net photosynthesis was less in ambient air. Visible injury was not detected. These experiments demonstrate the potential of such parameters as early warning bioindication techniques.

A drawback of using photosynthesis, stomatal conductance and transpiration in biomonitoring of O₃ is their lack of specificity. References

Chlorophyll levels are a direct measure of leaf damage but again are not a specific measure of plant damage to pollutant type.

Tenga and Ormrod (1990) measured total chlorophyll concentration tomato plant leaves using a leaf greenness meter. Chlorophyll levels decreased with increased O₃ exposure at levels too low to cause visible damage. Smith et al. (1990) and Fernandez-Bayon et al. (1993) also reported reductions in chlorophyll content prior to the onset of visible injury.

Metabolite content has been recommended as a potential response parameter in air pollution biomonitoring (Saxe, 1996). Fumigation of Aleppo pine with O₃ resulted in delayed rate of ethene emissions, accumulation of total polyamines and increase pool sizes of reduced glutathione and ascorbate in current year needles (Wellburn et al., 1996). References

All major air pollutants effect enzymes and their activity (Saxe, 1996). O₃ fumigation reduced light regulation of Rubisco activity in barley leaves (Machler et al., 1995). Alternatively, Ranieri et al. (1994) demonstrated increased peroxidase-catalase detoxification in response to exposure to ambient air in open-top field chambers in Italy.

Turcsanyi (1992) reviewed plant cells and tissues as bioindicators of environmental pollution and found that ultrastructural responses to O₃ were not very specific. The effects of peroxy-acetyl-nitrate and O₃ are often indistinguishable.

Degeneration of needle wax layer in response to SO₂ has been demonstrated (Section 5.4.4.6) but Barnes et al. (1990) discovered that O₃ did not significantly alter the needle wax layer in Norway spruce.

5.8 Complex studies

In nature pollutants often occur in combination with other pollutants. It is therefore necessary to investigate the effects of pollutant mixtures on plants. Responses which allow the identification of pollutant mixtures would be extremely useful in bioindication. Integrated usually involve fumigations and open or closed chamber systems. It is often necessary to rank pollutants according to their degree of damage. Obviously, studies using most natural conditions and pollutant levels are more useful. The more information obtained from complex studies the easier it would be to ascribe different responses to different sources and emission types.

Generally, plants are more severely affected by mixtures of pollutants than individual pollutants. For example, the threshold limit of tobacco Bel-W3 to O₃ is reduced if low concentrations of SO₂ are present (Heggestad, 1991). Closure of stomata occurs at lower concentrations of O₃ if O₃ and SO₂ are present together (Csintalan and Tuba, 1992). Pollutant mixtures may produce synergistic, additive or antagonistic responses. A combination of O₃ and HF enhanced senescence in maize which was not apparent when similar concentrations of the pollutants on their own were applied (MacLean, 1990).

Heck (1989) summarised a number of studies involving pollutant mixtures including exposure information and plant responses. Most work has been in relation to mixtures of O₃ and SO₂ (for example, Fialho and Bucker, 1995; Mcleod, 1995). To a lesser extent SO₂ and NO₂ combinations have been investigated (Kasana and Lea, 1994) and sometimes a combination of all three are studied (for example, Bucker and Guderian, 1994).

Some studies have investigated metal and gaseous pollutant mixtures (Dueck et al., 1987; Keller and Matysseck, 1990; Edwards et al., 1992). Others have focused on the effects of gaseous pollutants and acid rain mixtures on vegetation (Blank et al., 1990; Blaschke and Weiss, 1990; Ashenden et al., 1996; Shan et al., 1996). References

Defoliation class	Needle/leaf loss	Degree of defoliation

0	up to 10%	None
1	>10-25%	Slight (warning stage)
2	>25-60%	Moderate
3	>60-<100%	Severe
4	100%	Dead

Discolouration class	Foliage discoloured	Degree of discolouration

0	up to 10%	None
1	>10-25%	Slight
2	>25-60%	Moderate
3	>60-<100%	Severe
4		Dead

Ozone phytotoxicity	Observation
Low	O₃ injury recorded on Bel-W3 cultivar only
Medium	O₃ injury recorded on Bel-W3 and Bel-C cultivars
High	O₃ injury recorded on Bel-W3, Bel-C and Bel-B cultivars

Class*	Injury description

I	< 10% of foliar area affected brownish-red spots diffuse.
II	between 10 and 40% of foliar area affected increase in number of spots.
III	between 40 and 80% of area affected most of upper surface of leaf covered by brownish-red spots
IV	> 80% injury white necrotic areas covering most of leaf