TL: Tree Survey of Southern England SO: Greenpeace Environmental Trust UK, (GP) DT: 1988 Keywords: atmosphere acid rain forests effects terrec uk greenpeace reports gp europe / Andrew Tickle Air Pollution Report no. 4 GREENPEACE ENVIRONMENTAL TRUST 1988 Summary . The health of three native tree species, beech, oak and yew, was assessed in southern England using parameters such as crown density, leaf discoloration, needle retention and the presence of abnormal growth patterns. ù The survey results show that nearly 40% of beech, over 50% of oak and nearly 75% of yew were found to be moderately or severely damaged, on the basis of crown density. Only 5% of beech, 10% of oak and 4% of yew were found to retain the full crowns indicative of entirely healthy trees. ù These results give cause for concern as the species examined in this survey were all found within their natural range. In addition, a large number of the trees surveyed were on sites of conservation ally valuable woodland. ù Although clear, causal proof that air pollution is damaging these trees is lacking, the balance of current evidence seems to suggest that the widespread and severe damage reported in this survey cannot be explained by the influence of other abiotic or biotic stresses alone. Contents Introduction Method Results Conclusions References Appendix 1 - Beech, Oak and Yew Damage Appendix 2 - Tree Damage by County Appendix 3 - Table 3: Beech Roloff and Chlorosis Results Site List Andrew Tickle is an independent scientist who has four years experience of investigating air pollution effects on native plant species, in both the field and laboratory. He has just returned from Czechoslovakia, where, during a two month research visit, he was involved in surveys of national forest health using standard European [UNECE] evaluation techniques. Introduction Surveys of forest health undertaken in the United Kingdom since 1984 (1,2,3,4) have revealed a seemingly rapid and marked deterioration in the health of our forest and woodland trees. Over the four years in which forest survey figures have been available for the UK (1984-88), the deteriorating health of our trees has quickly come to parallel, or exceed, the situation for comparable trees in Western Europe. The serious decline taking place in European forests is acknowledged to be causally linked to air pollution (5). In this country however the Forestry Commission "stands alone in its refusal to accept a nexus between air pollution and forest damage" (6), despite the Commission's own survey figures for 1987 (4) revealing that only 8% of Britain's oak and beech trees could be regarded as being completely healthy. This survey sets out to provide an independent assessment of the health of three native tree species and to give some indication of the spatial variation in tree damage over ten counties in the south of England. The species surveyed, beech (Fagus sylvatica), oak (Quercus petraea and Quercus robur) and the English yew (Taxus baccata) were chosen because of their national importance in the English countryside and heritage. Thus the survey differs from that undertaken by the Forestry Commission, which concentrates on the status of their commercial stock - mostly non-native conifers. Method The survey of southern England was undertaken over a 14 day period from mid-August to September 1988. Late summer is regarded as optimal for the evaluation of tree health, as the tree is still in full foliage prior to autumnal senescence and leaf fall. The area was arbitrarily split into ten counties (Kent, Sussex, Surrey, Hampshire, Berkshire, Wiltshire, Dorset, Avon, Gloucestershire and Buckinghamshire) and an approximately similar number of trees of each species was surveyed from within these counties. Sites for beech and oak were selected from available data on conservationally valuable woodlands, Forestry Commission plantations and other natural wood ants that were known to fall within the targeted age range of 70-140 years, and were larger than half a hectare. On-site requirements included lack of recent disturbance (felling or storm damage), good crown visibility and ease of access. Occasionally where there were insufficient sites or they turned out to be unsuitable, others were selected on an ad hoc basis. These fulfilled the same basic criteria of selection but tended to be non-woodland sites. Wherever possible for beech and oak, 24 trees per site were sampled and, site-permitting, these were recorded in groups of six from four sub-sample areas. Assessment of the trees was carried out using standard internationally recognised techniques and indices. Beech and oak were both assessed for crown density on the West German five-class system, in conjunction with reference photographs (7,8). Any leaf discoloration, necroses or abnormalities were noted as were instances of defoliation due to insect or fungal attack. In addition beech was scored for patterns of abnormal branch and twig growth which often contribute to the break up of the canopy and subsequent loss in crown density. This four-point 'Roloff' index (9) is a widely-used indicator of beech decline throughout Europe. Records of soil-type and the predominant ground flora were made wherever possible. All the yew sites were selected on an ad hoc basis and were mostly in old churchyards; consequently the majority of the trees fell into a much older and wider age range than that stipulated for beech and oak. Yews were assessed for crown density (classes 0-4 as above), needle retention (in years), needle discoloration or necrosis, and the presence of abnormal growth characteristics ('fear twigs or tin selling). Results In total over 1,100 trees were surveyed at more than 60 sites across southern England (site details are given in Appendix 3). The main parameter used throughout the survey, crown density, is the most widely used index of forest damage, both in the UK and internationally. However, discussion does exist as to the significance of the crown density categories. It is accepted that losses of crown density up to 25% can occur naturally without the tree having been damaged, although trees in the crown density category 1 (10-25% loss) are usually regarded as being at a 'warning' stage. Upwards of 25% foliage loss (crown density categories 2 and 3/4), significant damage can be said to be occurring and it is these categories that should receive the most serious attention. The full results for crown density in Southern England are presented in Table 1. Damage in the crown density categories 2 and 3/4 (26 -100% defoliation) has been further broken down into counties and is shown in Figure 1 over leaf. Full crown density results by counties and species are given in Appendix 2. Table 1. The percentage of beech, oak and yew in each crown density category, summed for the sites surveyed over ten counties of southern England. (Omitted .. unscannable) The overall crown density figures presented in Table 1 should provide a reliable indication of the current status of tree damage in southern England, the confidence of the estimate being based upon the relatively large sample sets for each species combined with a good spatial distribution of survey sites. However, extrapolations from a small number of locations assessed within a purely arbitrary area such as a county, cannot hope to estimate the damage status of that given area as accurately. Beech (Fagus sylvatica) On the basis of the results for crown density (Table 1), almost 40% of the beech surveyed in southern England were either moderately or severely damaged. A further 56% were found to be at the 'warning' stage (crown density category 1) and only 5% of the trees were found to be entirely healthy (fully foliated). Beech was also assessed for patterns of abnormal growth using the Roloff indices and the results are presented in Table 3 (Appendix 3). Over the south of England as a whole, only a small proportion of the trees surveyed (7%) had entirely healthy patterns of growth; another 45% are described as weakened, a stage also warning of further potential growth degeneration. Within the remaining 48% of beech the signs of serious decline were apparent: growth had become stagnant and a small proportion of the trees (6%) were actually dying. Although the results for these two indices of beech health are broadly similar, the Roloff system of assessment placed roughly 10% more trees in the higher categories of damage. The difference between the two figures probably results from the fact that a proportion of the beech surveyed which were well foliated in the lower areas of the canopy (contributing to a reasonable overall crown density score), often possessed thin upper crowns with abnormal branching patterns. This should largely explain the discrepancy in the results but the question still remains as to which method of assessment, crown density or Roloff, provides the most accurate indication of beech health status in Britain. Crown density as a parameter is subject to quite large variations on a yearly basis, being more easily influenced by short-term periods of stress, perhaps associated with climatic factors (eg. drought, frost) or outbreaks of biotic pathogens (eg insects, fungal disease). However, the abnormal branching patterns inherent to the Roloff analysis take many years to develop and fluctuate little from year to year. Thus the Roloff index is thought to be more reliable for the detection of longer term (chronic) stress and consequent decline in health status, such as might be associated with pollution. Symptoms of damage usually attributable to short-term (acute) climatic or biotic stresses (widespread browning, necrosis or defoliation of leaves) were limited to a very small number of trees. However, leaf chlorosis, in the form of significant yellowing, was observed repeatedly throughout the survey (for results see Table 3, Appendix 3). This is a symptom frequently noted in beech surveys but very little evidence yet exists to link its onset, occurrence or severity with the health status of beech in Britain. Although 'iron induced chlorosis' (a nutritional deficiency associated with highly calcareous soils) may be a contributing factor, the yellowing observed in the survey was not solely restricted to sites with calcareous soils. The yellowing is also often described as unusually early autumn coloration, and whatever the significance of this may be, on average nearly 30% of the beech surveyed across southern England were displaying the symptom. Oak (Quercus robur/petraea) The results for crown density (table 1) show that broadly half (52%) of the oaks surveyed were moderately or severely damaged, having suffered more than 25% reduction in crown density. Another 38% of oak were in the 'warning' category and the remaining 10% of the trees surveyed were healthy. Other damage symptoms such as leaf chlorosis and necrosis were noted but seldom contributed significant damage to the trees. Although a clear decline symptomology (comparable to the Roloff indices for beech) is not described for oak, increasing severity of damage did tend to follow a pattern. This began as a progressive loss of fine branch structure with leafiness then gradually being confined to epicormic shoots around the main branches. This generally results in the tree becoming 'stagheaded' - a feature normally associated with old age in oak. Loss of fine branches and the resulting reduction in crown density may, in some areas, have been influenced by wind damage. Whilst every effort was made to avoid trees obviously damaged by storm, the inclusion of a large proportion of non-woodland oak in the survey would tend to increase any bias in the results due to exposure. Yew (Taxus baccata) The results for crown density indicate that nearly three quarters (74%) of the yew surveyed had losses in crown density of between 26% and 60% (crown density category 2); 10% of the trees were severely damaged or dying and only 4% were found to be healthy. Although the diagnostic value of the other damage parameters examined is largely unknown their results can be summarised briefly. Needle retention in healthy yews was usually up to four years (including the current year's growth); in damaged trees this was reduced to about two years. Another damage symptom, the presence of large numbers of adventitious shoots ('fear twigs') was widespread but often tended to give compensatory density to a tree. However it does break up the normal branching pattern of the yew to give it a 'mossy' or clumped appearance. Lastly, the 'tinsel syndrome' - drooping, dying sub-branches - was also common in damaged trees. Several factors may be contributing to the severity of the damage recorded. Firstly, the upper age limit of the trees sampled could not readily be controlled, although little is known as to the age at which yews start to senesce naturally. Secondly, a large number of yew were examined at semi-urban or roadside sites which are more likely to be exposed to disturbance or purely local stress factors. Lastly, and perhaps most importantly, very little is known as to what constitutes a 'healthy' yew tree. Yews found in their native habitat, the understorey of mixed deciduous ancient woodlands, are naturally thinner than their churchyard counterparts. Thus, although the survey included both woodland and churchyard types, a scale based on a densely foliated churchyard specimen is likely to overestimate the damage in the population sampled. The overall results are intended to be comparable with those found in other large scale inventories of tree health published recently for the UK. Thus from the outset of the survey the sampling design for beech, and to a lesser extent for oak, was based on those employed in the Forestry Commission 1987 Main Survey (4) of beech and oak, and the detailed Imperial College study of beech on conservation sites in southern Britain (10). Despite these similarities in sampling design, detailed comparisons between the results of the three surveys should be made cautiously, bearing in mind that the surveys were completed over differing areas of the country at different times. This point should be made especially when considering the Forestry Commission's figures for 1987, which seem to show a higher proportion of beech and oak in the crown density categories 2 and 3/4 (Table 2). However the Commission's own analysis of the data (11) suggests that those categories are being strongly influenced by increasing losses of crown density towards the north and west of the country, areas outside the present survey. Table 2. Summary of the most recent surveys of native tree health undertaken in the UK to date. Figures are presented for the percentage of trees with crown density losses in excess of 25% (crown density categories 2 and 3/4). species BEECH OAK YEW survey Greenpeace, 1988 39 52 74 Southern England Forestry Commission 57 66 * Main Survey, 1987 (4) Imperial College (10) 23 * * Southern Britain 1987-88 Friends of the Earth 22# * 29# Dieback Survey, 1985 (12)- * not examined # figures adjusted for 10 counties of this survey - roughly corresponding to crown density categories 2 and 3/4 The Imperial College beech study was conducted over a much more comparable area but conversely its results estimate a smaller percentage of trees in crown density categories 2 and 3/4. This difference is difficult to explain but possible contributory factors could include differences in observer perception and an apparent trend towards increased beech health in the southwest of the country - an area examined by the Imperial College study but not in this survey. The only comparative data for yew health in the UK comes from the 1985 Friends of the Earth survey (12), although this employed a four-point dieback scale as a method of assessment, rather than crown density. The FOE figures (adjusted for the ten counties examined in this surveys showed over a quarter (29%) of yew in the categories roughly corresponding to crown density losses of over 25%, compared to nearly three-quarters (74%) found in the present survey. The use of differing survey techniques and volunteer observers by Friends of the Earth seems insufficient to explain such a large discrepancy in the two results. The possibility does remain, therefore, of a worsening in the health of yew in the south of England since the Friends of the Earth survey was conducted. No attempt has been made to relate the spatial extent of damage described for beech, oak and yew (Figure 1 a, b and c) to patterns of environmental variables, including that of pollution. Such an approach is deemed unsuitable as the use of a county as a variable would be statistically questionable and currently available information for environmental variables (especially the major air pollutants - ozone, nitrogen oxides and sulphur dioxide) cannot be extrapolated with enough confidence to reasonably predict conditions or concentrations prevailing in a particular area. Finally, note was made throughout the survey of the almost uniformly poor condition of both woodland and non-woodland trees adjacent to major trunk roads and motorways. Local stress factors such as soil disturbance, salt damage and particulate and gaseous vehicle emissions are all likely to have a strong adverse influence on the health of these trees. This still remains a matter of some concern. Conclusions Although some emphasis was placed on reporting the spatial extent of tree damage over southern England, the survey was not designed to provide information as to causal or correlatory relationships between amounts of damage and levels of pollution stress. Such an approach, if undertaken, is only likely to lead to dubious and possibly misleading associations, especially in areas where data for tree health and/or pollutant concentrations is sparse or completely lacking. However, the number of localities included in the study was sufficient to produce a relatively precise statement on the health of beech, oak and yew in southern Britain. The survey results show that significant proportions of the three native tree species (beech, oak and yew) examined throughout southern England are damaged, judged on the basis of crown thinness (the symptom most widely used to describe forest damage, both in the UK and abroad). Nearly 40% of beech, over 50% of oak and nearly 75% of yew were found to be moderately or severely damaged, having lost over 25% of their crown density. Only 5% of beech, 10% of oak and 4% of yew were found to retain the full crowns indicative of entirely healthy trees. These figures, although similar to those found by the Forestry Commission's Main Survey of 1987 (4), give cause for concern as the species examined in this survey were all within their natural range. In addition, a large number of the trees surveyed were from conservation ally valuable woodland. A range of secondary symptoms of decline such as abnormal growth patterns and early yellowing in beech, and reduced needle life in yew, were also widespread. Of these, the patterns of abnormal growth seen in beech are probably the most significant as they take several years to develop and thus indicate a medium or long term reduction in the vigour of growth. An estimate of beech health using this parameter (the Roloff index) shows that almost 50% are in serious decline. This survey has established that damage symptoms, comparable to those seen in forest declines already occurring in areas of continental Europe, are now widespread in beech, oak and yew over southern England. This only confirms the damage already reported (4) on a national scale by the Forestry Commission for beech and oak. The severity of damage to these broad leaved species at least equals that reported for Germany and Switzerland where research results so far indicate that air pollution is an essential, causal factor in the destabilization of forest ecosystems (5). Furthermore, on the basis of current concentrations of sulphur dioxide, nitrogen oxides and ozone,it is suggested (13) that a broadly comparable 'pollution climate' exists in both northern Germany and in central and southern England. Although clear, causal proof that air pollution is damaging our native trees is lacking, the balance of evidence seems to suggest that the widespread and severe damage reported in this survey cannot be explained by the influence of other abiotic or biotic stresses alone. References 1. Binns W.O., Redfern, D.B., Rennolls K. & Betts A.J.A. (1985) Forest Health and Air Pollution. 1984 Survey. Research and Development Paper no. 142. Forestry Commission, Edinburgh. 2. Binns, W.O., Redfern, D.B., Boswell, R. & Betts, AJ.A. (1986) Forest Health and Air Pollution. 1985 Survey. Research and Development Paper no. 147. Forestry Commission, Edinburgh. 3. Innes,J.L., Boswell, R., Binns, W.O. & Redfern, D.B. (1986) Forest Health and Air Pollution. 1986 Survey. Research and Development Paper no.150. Forestry Commission, Edinburgh. 4. Innes, J. I. & Boswell, R.C. (1987) Forest Health Surveys 1987. Part 1: Results. Forestry Commission Bulletin no. 74. HMSO, London. 5. International Co-operative Programme for Assessment and Monitoring of Air Pollution Effects on Forests (1987). Results of the 1986 Forest Damage Survey in Europe. In Executive Summary, Third Programme Task Force Meeting, Usti nad Labem (Czechoslovakia). May 1987. 6. House of Commons Environment Committee (1988) Acid Rain. Volume 1. HMSO, London. 7. EEC (1985) Diagnosis and classification of new types of damage affecting forests. EEC, Brussels. 8. Boss hard, W. (ed.) (1986) Kronenbilder, Eidgenossische, Birmensdorf, Anstalt fur tag Forstliche Versuchswesen. 9. Roloff A. (1985) Schadstufen bei der Buche. per Forst und Holzwirt 40, 131 - 134. 10. Ling, K., Power, S.A. & Ashmore, M.R. (in press) A survey of the health of Fagus silvatica in Southern Britain. in Air Pollution and Forest Decline. IUFRO Conference, Interlaken, September 1988. 11. Innes, J.L. & Boswell, R.C. (1988) Forest Health Surveys 1987. Part 2: Analysis and Interpretation. Forestry Commission Bulletin no. 79. HMSO, London . 12. Rose, C. & Neville, M. (1985) Tree Dieback Survey: Final Report. Friends of the Earth Ltd, London. 13 Ling, K.A. & Ashmore, M.R. (1987) Acid rain and trees. Focus on Nature Conservation no. 19. Nature Conservancy Council, Peterborough. Acknowledgements: The author thanks Katy Ling and Sally Power of Imperial College for their readily-given and invaluable help with the selection of suitable survey sites. (NOTE: This report contains 14 pages of colour photographs, tables and charts which were omitted here .. unscannable.)