Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
THE USE OF BIOMARKERS TO MONITOR FOREST DAMAGE IN EUROPE J. N. Cape Institute of Terrestrial Ecology Bush Estate, Penicuik EH26 OQB Midlothian, Scotland ABSTRACT Forest health is now routinely monitored in most Western European countries, using a common protocol based upon visual assessment of leaf loss, crown structure and discoloration. Problems of observer bias have been recognized, and training is now more thorough. There are also problems of interpretation; it is not clear what are "normal" patterns, and how they vary with latitude, maritime/continental climates and altitude. Many hypotheses have been formulated to explain observed patterns of damage. measurements, and early INTRODUCTION Attention is now being turned to alternative methods of assessment, aimed at providing objective quantitative diagnosis (i.e., pre-visible symptoms.) These techniques, still in the exploratory stages, rely on three approaches: (i) comparison of visibly damaged and undamaged trees at the same sites, (ii) comparison of trees at different sites exposed to different pollution climates, and (iii) direct controlled experiments (e.g., whole trees in open-top chambers). Specific biochemical/physiological tests have been applied using some or all of these three approaches. Collaborative research and further development are required before such tests can be applied in large-scale field surveys. The possibility that the long-range transport of air pollutants derived from fossil fuel combustion could cause damage to forest health was raised by Scandinavian scientists some 16 years ago (Anon., 1972~. However, it was not until this decade that the possibility became a reality in the public mind, following a widespread visible decline in forest health in West Germany. The scientific community remains to be convinced, perhaps not that air pollutants are involved, but that air pollutants are primarily responsible for the observed effects. MONITORING FOREST HEALTH One of the difficulties encountered has been the quantification of this 'new' type of forest decline, both in space and time. Systematic surveys of forest tree health started in W. Germany in the early 1 980s and are now routinely undertaken in most European 63
64 countries, under the auspices of the United Nations Economic Commission for Europe (UNECE) (Anon., 1 987a). Within the European Economic Community, such surveys, based by UNECE, are now mandatory L studies using their complemented by a 12 regions (Innes & on a uniform 16 x 16 km grid with protocols drawn up (Anon., 1 986a). In addition, several countries conduct more detailed own protocols. In Britain, for example, the grid-based survey is stratified sampling system for economically important tree species in Boswell, 1987~. Damage is scored visually in late summer each year by comparison with "standard" illustrations for each of the major tree species (e.g., Bosshard, 1986~. The scoring system is based upon crown structure, leaf loss and foliar discoloration (Table 1~. Table 1. Criteria used under the UNECE protocol for visual assessment of damage. · Grid 16 km x 16 km intersecting with woodland area > 0.5 ha. · Results reported for 5 most important broadleaved species and 5 most important conifer species. · Crown density (leaf loss) expressed in 5 categories: o 1 2 3+4 0 - 10% no damage 11 - 25% indicative 26 - 60% moderate damage ~ 60% severe damage and death · Discoloration expressed in 4 categories 0 0- 10% none 1 1 1 - 25% 26 - 60% > 60% indicative moderate severe It has become apparent that such survey techniques require thorough assessors. The documented biases in the earlier W. German surveys (Krause have also been observed in Britain (Innes et al., 1986~. component of international cooperation. Since efficient training of ~ et al., 1986) Training is now an important methods have been established, the apparent vitality of W. German conifer forests has changed little, but broadleaved species such as beech (Fagus sylvatica) and oak (Quercus spp.) are showing a clear, if slow, deterioration (Anon., 1 986b). It is too early yet to identify time trends from other national surveys. The debate on the efficacy of such surveys continues, despite the agreed protocols. Some simple sources of error, such as sampling with or against the light, are amenable to assessor training. The most serious point of contention is the use of "standard" reference photographs over a wide range of latitude, altitude and ecotype. It has been suggested (Kilz & Hanisch, 1988) that separate standards should be set up for different types of ecosystems, became crown form and crown density are known to be influenced strongly by environmental factors other than air pollution. This suggestion may overcome difficulties associated with crown form, but in the absence of good historical data it is impossible to define "normal" crown density at sites which are now subject to
65 air pollution stress, and the interpretation of the visible "damage" remains subjective. Other refinements have been suggested, such as surveying forests twice yearly. More frequent observations would enable the crown structure of deciduous trees to be evaluated better, and pathogenic influences to be identified. However, resources are unlikely to be available to progress from a simple inventory toward the identification of the possible causes of the observed symptoms. The visible "damage" symptoms (leaf loss and discoloration) used in such monitoring surveys of this type are non-specific, and may be produced by biotic, climatic or edaphic factors in the absence of air pollution. Visible symptoms alone give no information on the role of air pollutants. In general, data on the levels of air pollution are unavailable for the forest site,s where symptoms appear, so it is not even possible to look for spatial or temporal correlation between air pollution damage and symptoms. More specific markers are required to link forest decline with air pollution. THE USE OF BIOMARKERS A biomarker is any biological measure which may be used to monitor the effects of stress on plants, at every level from the plant cell to the ecosystem. The range of possible biomarkers may be subdivided into visible biomarkers (usually giving qualitative information) and non-visible, which may give qualitative and quantitative information. The objective of the study must be defined before an appropriate biomarker is selected. If one wishes to use plants as biological indicators of air pollution exposure, lichens, herbaceous and woody plants may be chosen to indicate critical levels of exposure to specific air pollutants (Arndt et al., 1987~. If one wishes to study subtle effects of air pollutants on forest ecosystems, then it may be more appropriate to use sensitive herbaceous species on the forest floor as bioindicators, or to study changes in species composition. There have been large changes in the species composition and production of fruiting bodies of ectomycorrhizal fungi over the past two decades in the Netherlands (Anon., 1987b), which are thought the result from excess deposition of nitrogen. The above are examples of the use of visible biomarkers, or bioindicators. In studying forest decline, however, most attention has been given to detailed investigations of trees, and non-visible biomarkers of tree health have been developed to extend observations beyond the visible symptoms of crown thinning and discoloration. Recent investigations of the vitality of forest trees in Europe have been reviewed by Cape (1988), and the methods used are summarized in Table 2. Table 2. Methods for investigating the effects of air pollutants on trees (after Cape, 1988~. 'whole leaf' - external surfaces - physiological - external surfaces cell ultrastructure electrolyte leakage element composition leaf Nettability amount and composition of surface wax Hartel turbidity test photosynthesis/respiration water relations biochemical
66 (non-destructive) - (destructive) - roots - reflected light (remote sensing) induced fluorescence electron paramagnetic resonance nuclear magnetic resonance emission of organic compounds biochemical enzymes and substrates/precursors pigments other metabolites buffer capacity elemental analysis mycorrhizal associations The ways in which such methods are used may be classified under three headings: i) comparison of visibly damaged and non-damaged trees at the same site. ii) comparison of trees (usually showing some damage symptoms) at different sites. iii) experimental manipulation of whole trees or parts of trees, to determine the types of measurement which give clear indications of specific stresses, and which may be useful for diagnosis under field conditions. In the first case, it may be assumed that climate and exposure to pollutants are similar across the site, and, more dubiously, that soils are homogeneous. Measurements may identify the limiting factors or impaired functions in individual trees showing visible symptoms, and may indicate possible causes. Markers of tolerance may then be identified as part of the genetic variability expressed in the visible symptoms. If a good correlation can be found between a biomarker (or biomarkers) and visible symptoms, then the biomarker may be used objectively to evaluate the degree of "risk" of the forest populations at that site, and in the selection of resistant genotypes for future breeding. However, as the precise cause of the differential response is not known, there are no guarantees that the same biomarker will be useful elsewhere. An example of such an approach may be found in the work of Villanueva et al. (1988~; they measured a wide range of biochemical constituents of individual trees showing different degrees of damage but growing at the same site. In the second case, there are many possible sources of variability when comparing different sites. To be useful, a biomarker should be less variable within sites than between sites, so that significant differences may be observed between sites. If biomarkers that show good spatial and temporal correlation with visible damage symptoms can be identified, then there is a basis for a more objective assessment of forest health, possibly even before the appearance of visible symptoms. One recent example is a pilot study, supported financially by the Commission of the European Communities, which involved four institutions in Britain and Germany (Cape et al., 1988~. In this study, samples were taken from three tree species growing at 12 sites in a transect from southwest Germany to northeast Scotland. This material was subjected to a wide range of potential diagnostic tests (Table 3), and significant differences were sought between "damaged" and "undamaged" trees within sites, and between sites. A unique feature of this study was the wide range of tests applied to the same plant material. Some interesting correlations occurred between apparently unrelated properties; for example, the emission of ethylene from spruce needles was strongly correlated with their buffering capacity after maceration. Although some of the tests were too variable between trees to be useful as biomarkers, others allowed discrimination between sites. Of the 42 measurements made on Norway spruce, 14 were shown to be significant in
67 terms of site discrimination. The straightforward determination of foliar nutrients ranked as the most important discriminating factor. The range of pollution climates (i.e. different mixtures of pollutants and climate) at the different sites also permitted the correlation of specific effects with exposure to different pollutants, ant! pointed to possible underlying mechanisms. Those tests which appeared to be useful are now being applied to experimentally-treated trees in an attempt to understand the genetic, climatic and dynamic factors affecting the response. Table 3. The range of tests of tree vitality applied to samples of foliage collected from 12 sites in Europe as part of a study on the early diagnosis of forest decline (Cape et al., 1988~. Visible symptoms Histology of leaf sections Leaf Nettability Amounts of surface wax and dust Hartel turbidity test Hydrocarbon emission Pigment analysis (chlorophylls + carotenoids) Anti-oxidants (alpha-tocopherol) Buffering capacity Modulated fluorescence Water relations Nutrient analysis (S. N. K, Ca, Mg) In the third case, experimental approaches have been of three types: the exclusion of supposed causes (for example, by growing trees in filtered or unfiltered air), the removal of symptoms (by adding fertilizer to nutrient deficient trees), and by the addition of supposed causal factors (fumigation or spraying with acid rain). Each of these approaches addresses a different aspect of the cause-and-effect relationship. In all cases a response is sought that is specific for a given treatment, and which may be used in the field to indicate the presence or absence of a particular pollutant stress. Exclusion of supposed causes may narrow the range of possible causal factors if an effect is observed, but it may not give sufficiently detailed information to indicate the appropriate strategy for pollution control. On the other hand, such experiments show the likely benefit of a given reduction in air pollutant exposure. The use of open-top chambers in studying effects of air filtration on mature trees has formed part of the CEC initiative for studying effects of air pollutants on crops and trees (CEC, 1987~. Work on mature trees is expensive, and the experimental design must permit optimum use of the facility. For example, at the open-top chamber site set up by Arndt and colleagues (Arndt, 1987) at Edelmannshof in the Swabian-Frankonian Forest of West Germany, research workers from other institutions make measurements and take samples of plant material. Not all of the studies are relevant to the development of biomarkers, but the facility provides an opportunity to test potential biomarkers under conditions similar to those of a mature forest stand. In the absence of a meaningful decrease in pollutant emissions, the use of ameliorative experiments may give little or no information on the causes of observed symptoms, but points to the feasibility of practical approaches in alleviating damage, at least in the short term. Zottl and co-workers (Zottl & Huttl, 1986) have shown, for
68 example, that discoloration caused by magnesium or potassium deficiency can be rapidly reversed by applying fertilizer, but the underlying cause of the nutrient deficiency is still unknown. Such studies may again be used to test the response of potential biomarkers. The direct testing of supposed causal agents, following the ideas enshrined in Koch's Postulates (e.g., Last et al., 1984), has been employed primarily under well- controlled conditions in the laboratory. The aim is to reproduce symptoms typical of field conditions, and thereby establish a direct cause-and-effect relationship. Such experiments have been used both to identify potential biomarkers, and to test biomarkers against a known stress. Although this appears to be a straightforward approach, a number of problems may arise. First, if the field symptoms or biomarker responses are non-specific, reproduction of symptoms or responses experimentally in the laboratory only indicates that an applied treatment can produce the effect, but it does not necessarily indicate that the treatment does produce the effect under field conditions. Second, there are subtleties of interaction between different pollutants, and between pollutants and other stresses (climatic or pathogenic), which may be untestable in the laboratory. Under these circumstances, Koch's Postulates may be unworkable. Third, at the most basic level, such studies may experience difficulties in reproducing appropriate experimental treatments. Nevertheless, the value of such experiments is that they (i) isolate the effects of individual pollutants at all levels in the plant, from biochemical processes through to the overall effect on growth, and (ii) suggest potential biomarkers for study in the field. Microscopic examination, for example, can now distinguish between the effects of ozone exposure, winter desiccation and magnesium deficiency, based on results obtained from laboratory experiments (Fink, 1988~. The best example of this experimental approach is the study based around the environmental chambers of the GSF near Munich, where complex control over pollutant exposure and climate can be achieved. A large number of measurements are being made on the exposed trees in a major collaborative experiment (see Payer et al., 1986~. However, such experiments cover only a minute fraction of the combinations and permutations available in the field. The widespread application of biomarkers will be determined by the practical constraints of logistics and finance. The benefits derived from the use of specific biomarkers (for example, a biochemical assay) will need to be seen as cost-effective relative to other techniques (for example, an increased frequency of scoring visible damage). However, biomarkers that measure the biochemical and physiological responses of trees to their environment provide the best means of quantifying non-visible injury and identifying causes. The attention given to tree health in relation to air pollution stress has highlighted the deficiencies in our knowledge of the fundamental biochemical and physiological mechanisms that operate in forest trees. CURRENT DEVELOPMENTS A recent CEC workshop on Forest Decline Symptomatology (CEC, 1988) concluded that the problem of forest decline cannot be solved by foresters or biochemists working in isolation, and that an intensive multi-disciplinary approach should be adopted at a few key sites. Specific biomarkers need to be identified which can be applied on a large scale in forests, and several research groups in Europe are working toward this end. The scale of the problem is encouraging a multi-disciplinary approach, but too often the research effort is fragmented, with intensive measurement of air pollutants at one site, and intensive research into tree physiology and biochemistry at another. The major funding agencies have attempted to bring together the different approaches, and publish regular reports. The French "DEFORPA" (Deperissement des Forets Attribue a la Pollution Atmospherique) programme (Landmann, 1988) and the West German "PEF"
69 (Projekt Europaisches Forschungszentrum fur Massnahmen zur Luftreinhaltung) project publish regular reports in conjunction with the CEC, which also publishes its own series of "Air Pollution Research Reports" (in English) as the proceedings of workshops held under its auspices. SUMMARY In Europe, national and international surveys of forest health have been developed in response to widespread visible symptoms of decline in both coniferous and broadleaved trees. These surveys cannot of themselves be used to establish cause-and-effect relationships, and many studies are now in progress on the use of tree biomarkers in detecting and quantifying damage attributable to air pollutants. There are, however, a number of questions that arise, and a number of statements that can be made. 1 ) In the visual assessment of tree vitality, can or should allowance be made for the influence of known factors such as climate or genotype? Does a "normal" tree exist, against which comparisons can be made? 2) Within sites, a range of damage symptoms may occur. Detailed studies of individual trees at the same site may give quantitative information, and may be used to identify markers of damage before visible symptoms appear. 3) 4) 5) Between sites, a biomarker used as a diagnostic test must show a correlation with visible symptoms. It is unlikely that cause-and-effect relationships can ever be established from field data. Plausible mechanisms of action of air pollutants or other factors may be derived from well-designed and controlled manipulative experiments on trees. Such experiments are more easily performed on small or young trees, and extrapolation to older trees in forests is unlikely to be straightforward. In field based studies there is a need to use as much information as possible in attempting to understand ecosystem responses to air pollution stress. To this end, research efforts should be concentrated at key sites to ensure that the different disciplines (atmospheric chemistry to plant population ecology) act in coordination. 6) The dynamics of forest responses to air pollutants are poorly understood, both on short time scales (minutes) and long (decades). The sequence of events in the development of damage is also not well known, and may indicate mechanisms of cause-and-effect. These issues were discussed at a recent workshop on forest decline symptomatology sponsored by the Commission of -the European Communities (CEC, 1988~. The conclusions will be used as the basis for the development of research programmer on forest decline in Europe. REFERENCES Anon. 1972. Sweden's Case Study for the UN Conference on the Human Environment. The Impact on the Environment of Sulfur in Air and Precipitation, Stockholm.
70 Anon. 1 986a. Council Regulation (EEC) No. 3528/86 on the protection of the Community's forests against atmospheric pollution. Official Journal of the European Communities, L326/2-4. Anon. 1986b. 1986 Forest Damage Survey. Federal Ministry of Food, Agriculture and Forestry, Bonn. Anon. 1987a. Manual on methodologies and criteria for harmonized sampling, assessment, monitoring and analysis of the effects of air pollution on forests. Convention on Long-range Transboundary Air Pollution, International Cooperative Programme on Assessment and Monitoring of Air Pollution Effects on Forests. UNECE, Geneva. Anon. 1987b. Qualitative and quantitative research on the relation between ectomycorrhiza of Pseudotsuga menziesii, vitality of the host and acid rain. Report 25-01 Agricultural University, Wageningen. Arndt, U. 1987. Open-top chamber projects in Hohenheim and Edelmannshof. Pp. 299-317 in Microclimate and plant growth in open-top chambers. CEC Air Pollution Research Report 5 (EUR 11257~. CEC, Brussels. Arndt, U., Nobel, W., & Schweizer, B. 1987. Bioindikatoren: M~glichkeiten, Grenzen und neue Erkenotnisse. Ulmer, Stuttgart. Bosshard, W. (ed). 1986. Kronenbilder. Eidgendssische Anstalt fur das Forstliche Versuchswesen, Birmensdorf, Switzerland. Cape, J.N. 1988. Recent developments in the diagnosis and quantification of forest decline. Pp. 262-305 in Air Pollution and Ecosystems (P. Mathy, ed.~. D. Reidel, Dordrecht. Cape, J.N., Paterson, I.S., Wellburn, A.R., Wolfenden, J., Mehlhorn, H., Freer-Smith, P.H., & Fink, S. 1988. Early Diagnosis of Forest Decline. Institute of Terrestrial Ecology, Grange-over-Sands, UK. CEC. 1987. Microclimate and plant growth in open-top chambers. Air Pollution Research Report 5. (EUR 1 1257), CEC, Brussels. CEC. 1988. Forest Decline Symptomatology. CEC, Brussels. Air Pollution Research Report (in press) Fink, S. 1988. Histological and cytological changes caused by air pollutants and other abiotic factors. Pp. 36-54 inInt. Symp. Air Pollution and Plant Metabolism, S. Schulte-Hostede, N.M. Darrall, L. Blank, and A.R. Wellburn (eds), Elsevier, London. Innes, J.L., Boswell, R., Binns, W.O. & RedEern, D.B. 1986. Forest health and air pollution: 1986 survey. Forestry Commission Research and Development Paper 150. Forestry Commission, Edinburgh. Innes, J.L., & Boswell, R.C. 1987. Forest Health Surveys 1987, Part 1: Results. Forestry Commission Bulletin 74, HMSO, London. Kilz, E., & Hanisch, B. 1988. Characterization of visible symptoms. In Forest Decline Symptomatology, CEC Air Pollution Research Report, Brussels (in press).
71 Krause, G.H.M., Arndt, U., Brandt, C.J., Bucher, I., Kink, G., & Matzner, E. 1986. Forest decline in Europe: development and possible causes. Water, Air, and Soil Polln. 31, 647-668. Landmann, G. 1988. Les recherches sur le deperissement des fordts en France: structure et principaux resultats du programme DEFORPA. Pp. 261-281 in Air Pollution & Ecosystems (P. Mathy, ecI.~. D. Reidel, Dordrecht. Last, F.T., Fowler, D., & Freer-Smith, P.H. 1984. Die Postulate von Koch und die Luftverschmutzung. Forst. Centralblatt, 103, 28-48. Payer, H.D., Bosch, C., Blank, L.W., Eisenmann, T., ~ Runkel, K.H. 1986. Beschreibung der Expositionskammern und der Versuchsbedingungen bei der Belastung von Pflanzen mit Luftschadstoffen und Klimastress. Forstw. Cbl. 105, 207-218 (see also other papers in this volume). Villanueva, V.R., Mardon, M., Moncelon, F., & Santerre, A. 1988. Biochemical markers in polluted Picea trees. Pp. S8-00 in 2nd Int. Symp. Air Pollution and Plant Metabolism, GSF-Bericht 9/87, GSF Munchen. Z0ttl, H.W., & Huttl, R.F. 1986. Nutrient supply and forest decline in southwest Germany. Water, Air, and Soil Poll., 31, 449-462.
