EXPERIMENTS AND LONG-TERM OBSERVATIONS
New technology and understanding … can have a devastating effect on long-term projects: there is always a loss of interest in continuing with a project set up using outmoded techniques, recorded on outmoded software (such as paper cards!), and set up to answer problems long since considered irrelevant.
JONATHAN SPENCER, 19951
This chapter is addressed to professional ecologists, teachers devising exercises for students, conservationists in charge of nature reserves and amateur naturalists or historians with a long-term interest in a wood or group of woods.
As all conservationists know, it is easier to start a new project than to sustain an existing one. Funds for new activities are much easier to raise. Volunteers love cutting down trees, but are most reluctant to return, even next year, to record the consequences. There is thus surprisingly little information about the effects of conservation activity. Have woods belonging to conservation organisations fared better or worse than those remaining in private ownership? The answer seemed obvious during the Locust Years. In the 25 years since grubbing and coniferisation ceased to be widespread the question has seldom been asked, but one day it will be, and conservation bodies ought to have an answer.
Investigators may persuade themselves that time sequences can be short-circuited: that comparing plots in 2005 that were felled in 1996, 1999 and 2002 is equivalent to following the same plot through three, six and nine years after felling. In practice this is seldom true: similar, even adjacent, plots have an infuriating habit of going through different trajectories in different years. There is no substitute for real time sequences.2
The anthropology of ecological science
Ecology has been forced into a cultural mould determined by other sciences: a world of full-time research, large grants of money to be spent within a limited time, PhDs to be gained within three years, careers to be pursued. Long-term, cheap, interdisciplinary, or exploratory studies are marginalised and tend to take place, if at all, outside the academic establishment.
These constraints limit the range of problems that can be studied, and create a bias towards certain kinds of interpretation. Change tends to be emphasised over stability: students – and referees of learned journals – are excited by change, but discouraged on discovering only stability. Gradualist explanations are favoured over catastrophist. The effect of ordinary rains can be studied with one year spent getting the apparatus together, one year of observation and one year writing up; but what research supervisor will advise a student to study the effect of a deluge that is unlikely to occur within the middle year?
Ecologists, taking ‘hard’ sciences like physics for their model, have striven towards a reductionist world where everything can be explained in terms of a small number of quantitative laws; in which observations can be repeated and notebooks thrown away when the work has been published. Even a century on, they have not fully come to terms with a real world in which the laws of nature are complex and semi-quantitative and have exceptions; in which observations are time-limited, and the primary data are often of more lasting importance than the theories. The Law of Evolution is not quite a ‘law’ in the sense of the Law of Gravitation.
Ecologists have become reluctant to investigate the unexpected or inexplicable. Discovery is out of fashion. A scientist recently complained to the British Ecological Society that only 43 per cent of ecological investigations were driven by the desire to confirm or disconfirm a hypothesis. This, for him, was not enough, even though some grant-giving bodies dislike projects that do not test a preconceived hypothesis.3 The time draws nearer when an investigator has to know (or pretend to know) the approximate answer to a question before applying for a grant to study it. I leave the reader to speculate where new hypotheses will then come from.
Woodland and the teaching of historical ecology
Any ancient wood provides abundant material for demonstrations and student exercises in the long-term functioning of a wood. By maps or transects one can in a few hours investigate the difference between the vegetation on a woodbank and that in the rest of the wood, or the occurrence of different humus types under oak, hornbeam or ash, or the spread of plants of ancient woodland into an extension to the original wood. A more ambitious student project might investigate the correlation between tree species and different types of ground vegetation. But other projects are well worth studying, though one would hesitate to recommend them even as a subject for a PhD, such as how the buried seeds of coppicing plants know that the wood has been felled and it is time to germinate.
LONG-TERM OBSERVATIONS
Some of my own studies, written up in Ancient Woodland (2003), continue photographic and other records begun by my predecessors B.A. Abeywickrama and D.E. Coombe in the 1940s and 1950s. Here I summarise two others.
Gamlingay Wood over 90 years
In 1911 R.S. Adamson wrote an account of Gamlingay Wood near Cambridge: one of the first ecological descriptions of a natural wood in Britain. His object seems to have been to introduce the American idea of plant-associations to this country. He made detailed maps of trees and shrubs and of the ground vegetation in about one-third of the wood. He drew attention to what I would now call a ‘sandlens’: a patch of sandy soil contrasting with the boulder-clay underlying the rest of the wood.4
Adamson was unaware that this was (as far as is known at present) the best-documented historic wood in the country. For 650 years before his time the wood, or part of it, had belonged to Merton College, Oxford, which kept excellent archives. Something is known of its ecology in the Middle Ages.
Time rolled on, and Gamlingay Wood passed through various vicissitudes. In 1938 Harry Godwin mentioned ‘felling and burning’ but alas gave no details. In the 1950s the college sold the wood, and parts of it were replanted in different ways and with varying success. It was bought by the Wildlife Trust for Bedfordshire and Cambridgeshire in 1991, which revived the coppicing and has removed many of the planted trees. Since then Gamlingay Wood has escaped the worst deer damage, but muntjac have become abundant and the wood now looks distinctly deer-bitten.
In 1991 and 2002 I was able to map the vegetation again, and in particular to repeat Adamson’s observations (Fig. 209). Table 25 is a summary of the findings.
Hybrid poplars (a fashionable tree crop in the 1950s) were inserted into small gaps in the otherwise intact wood. They had no measurable effect on the native trees or the ground vegetation. The effects of the other planted trees were not always what would be expected. Areas where ash was planted did not necessarily show an increase in ash. The most notable change in trees was birch, which had not been planted. In 1911 birch was confined to the sand-lens, which would traditionally be thought its proper habitat. By 1991 it had declined on the sand-lens, but become abundant everywhere else, especially in some of the planted areas. This is how birch increased generally in the twentieth century, taking advantage of the disturbance created by planting other trees. Planted trees (other than poplar) were particularly antagonistic to aspen, which although clonal is sensitive to competition.
Among herbaceous plants some differences indicate changes in waterlogging. In coniferised areas the effective winter rainfall would have been reduced, by needles intercepting and evaporating some of the rain; this would work in favour of bluebell and mercury and against oxlip and meadowsweet. Other changes in intact areas may result from runs of wetter or drier seasons. Global warming has not resulted in a decline of bluebell (though it may have of primrose). The decline of oxlip, as in many other woods, may be an effect of browsing. I cannot imagine why meadowsweet declined in both intact and replanted areas. This is peculiar to Gamlingay Wood; in nearby woods meadowsweet remains abundant.
Adamson’s record furnishes a rare if not unique opportunity to test how stable plant distributions and communities have been, and to measure the effect of different types of replanting. It confirms that planting damages a wood’s individuality, promoting commonplace plant communities, such as a monoculture of dog’s-mercury, at the expense of rare ones. However, in forestry terms this was barely a successful replanting. If it had yielded a reasonable crop of planted oak or pine, little of the ground vegetation, let alone the native trees, would have survived.
Lady Park Wood over 60 years
In a corner next to Gloucestershire and Herefordshire, Lady Park Wood climbs the northeast-facing crags of the Wye Gorge just inside Wales. It comprises some 90 acres (36 ha) (projected on to the horizontal) and adjoins other, more extensive, plantations and woods. It has been so called since 1608; who the Lady and the Park were is forgotten. The wood, on hard limestone, clays and scree, is most varied in its soils and vegetation. Ash, beech, lime, hazel and wych-elm predominate; sessile oak occurs on the acidic plateau above the cliffs. There are many other trees and shrubs. Ground vegetation ranges from wood-sage to Luzula sylvatica and dog’s-mercury.
After centuries as a coppice, the wood was last cut partly in 1902 and partly in 1943. Surrounding woods were coniferised gradually, beginning in the nineteenth century. This difficult terrain escaped, and in 1944 was set aside as a non-intervention nature reserve by the Forestry Commission, British Ecological Society and Oxford University Department of Forestry. It has been maintained, watched over and recorded by E.W. Jones, G.F. Peterken and E.J. Mountford.
By the 1970s there had been quite big changes. Beech increased in dominance; shade increased and understorey trees such as maple were killed; hazel was largely eliminated. ‘A perpetual rain of mature trees’ fell off cliffs and steep slopes, the limes remaining alive. Then came Elm Disease, which killed most of the elms to the ground. Then came the great drought of 1976, which killed some of the beeches and set back the others, to the benefit of ash and lime.5
By 1987 the investigators concluded that ‘succession is seen as an unpredictable process without a definite outcome’. Changes were driven by outside events, which varied from one part of the wood to another. Some, such as cessation of coppicing and Elm Disease, applied to many woods. Others, such as storms and trees falling from cliffs, were more or less local. Beech is unusually susceptible to drought, and is a ‘catastrophist’ tree, coming and going in response to unusual events.
Further changes have occurred in the last 20 years. Fallow deer had been present in modest numbers possibly since the time of the park; they have now proliferated to an all-too-familiar degree – they probably learnt to take refuge here from culling in neighbouring woods. They have eliminated ground vegetation, especially brambles, which survive inside exclosures. Grey squirrel damage is unusually severe, and probably affects the longevity of the trees. Sycamore threatens to invade. Felling of adjacent plantations increases wind-blow on the edges of the reserve. Pollution, which used to emanate from the mines and factories of South Wales, now takes the form of nitrogen compounds from nearby farms.6
Lady Park Wood is probably much more unstable than most ‘old-growth’ woods that result from non-intervention after coppicing. Of the factors promoting change, one – trees falling off cliffs – is rare, and another – the unstable behaviour of beech – is confined to beechwoods, but the others are widespread. However, it is unusual to find such a conjunction of instabilities in one site. Hayley Wood from 1965 to 2005, or Madingley Wood from 1950 to 1990, have been more stable than the first 40 years of observations were in Lady Park.
Thirty years in a hundred woods: the Kirby–Bunce survey
In 1971 Bob Bunce organised a detailed survey of 1,686 sample plots, each of 200 square metres, in 103 woods, located all over Great Britain, except for East Anglia, Essex, Kent, Cornwall, the Pennines and south Scotland. About two-thirds of the woods are now thought to have been ancient. Within each wood the plots were located at random, some including (for example) woodland grassland, others not. In 2000–3 the surviving 1,648 plots were relocated and re-surveyed by different people, and the results of the two surveys compared.7
At the time of writing the comparison was not fully published, and it would be premature to summarise or discuss it in detail. However, I have mentioned some of the findings in other chapters. There was a general change towards greater shade, greater fertility (in woods surrounded by farmland) and increasing deer. In general, plots had fewer species in c.2001 than in 1971, except those ‘damaged’ by the great storms of 1987 or 1990. The survivors tended to be plants tolerating shade, those of more fertile soils and those tolerant of browsing. Other changes were related to the lengthening growing season.
EXPERIMENTS
Although historical ecology will always be a mainly observational science like astronomy, there are opportunities for experimentation: for example Norman Moore’s new wood, or investigations of how waterlogging affects plant communities. A widely applicable type of experiment is to investigate effects of browsing by fencing various animals out of sample areas, with deer-, rabbit-, or mouse-fences (Fig. 158). One should beware of side effects, for example by the fence intercepting fallen leaves, or zinc dissolved from galvanised wire poisoning plants directly under the fence.
Any wood used for research is littered with remains of experiments that students of yesteryear abandoned to take up careers far away. Amateurs may have more opportunities for continuity. Long-term observations and experiments run the risk of destruction of the site, as happened to Poolthorn Covert. Management plans for woodland should always take account of the existence of previous investigations, even if they seem not to be still ongoing.
Long-term experiments incur another anthropological difficulty. The question that they were set up to answer, or the hypothesis that they might have confirmed, may turn out to be unimportant or merely unfashionable. Ecology, like many other sciences, is a creature of fashion: in its 150-year history it has been dominated successively by evolution, succession, energy flow, classification, nutrient flow, and recently again by evolution. (Is not a journal called Trends in Ecology and Evolution?) Any experiment lasting (by design or accident) for more than 20 years will have been through periods of being deeply un-trendy and unlikely to be continued for its original objective.
My own experiments and observations, set up in Hayley Wood to study the effects of coppicing on herbaceous plants, a few years later became of more interest in relation to the effects of deer. In recent years, with deer excluded from most of the wood, they have gained interest for the recovery of ground vegetation from browsing in relation to whether or not deer had damaged the underwood canopy.
Peterken has related the practical difficulties of maintaining observations over 60 years at Lady Park Wood. Money is less of a problem than lack of continuity of observers and equipment. Even the simplest experiments, using nothing more complex than measuring tapes and permanent markers for plots, are surprisingly difficult to compare with the work of previous observers using different kinds of tape and different methods of locating plots.
Experiments need to be simple and cheap to maintain and their objectives not too narrowly specified; otherwise they can deal only with short-term phenomena. Permanently marked plots that can be recorded or photographed at regular (or irregular) intervals are more robust than apparatus that needs constant maintenance and periodic replacement.
FIXED-POINT PHOTOGRAPHS
Successive photographs from the same point record change and stability as few other methods can do with so little expense or effort. Anyone can repeat a photograph every year or every ten years, beginning (if possible) by relocating a pre-existing photograph or historic painting.
It is important to repeat the exact scene (Figs 160 & 161). By careful observation of sightlines it should be possible in woodland to identify the viewpoint within 8 inches (20 centimetres) or so in three dimensions, even without a marker. For future use it is helpful to take the original picture with one’s back to a tree that seems likely to stay in place. A zoom lens is almost essential in order to reproduce the focal length of the original lens. It is not always easy to repeat the pictures at the same equivalent time of year: if anemones are in bloom one year along with their leaves and a bryophyte cover, the next year when the anemones are out the bryophytes may be covered by a layer of fallen leaves that earthworms have not finished eating. One tries to take successive pictures in roughly the same weather and at the same time of day.
An annotated sketch of the scene may be useful for identifying plants in the pictures. The difference between primrose and lesser celandine may not always be apparent in a photograph.
Very often it happens that an original viewpoint is now in the middle of a bramble thicket or otherwise obscured. This, of course, is a useful observation in itself, but it then becomes necessary to take another shot from as near as possible to avoid the obstruction (Figs 162 & 163). I normally think it proper to bend obtruding branches aside.
ECOLOGICAL ARCHIVES
Long-term projects require archives and an institutional memory: without them, the record will remain lost, or its existence will remain unknown. In the case of Lady Park Wood … the [Nature Conservancy Council] almost lost the lot when [X’s] papers were weeded following a promotion, but were saved by a secretary who thought they might be interesting.
G.F. PETERKEN, 2005
Ecology is time dependent. Stability and change need to be established over periods much longer than the life of a research project, longer than the life of a scientist or the life expectancy of an institution like English Nature. One can still (just) observe what elmwoods are like unaffected by Elm Disease, but to learn what a wood was like before the multiplication of deer depends on the chance of someone having taken and preserved observations.
Present observations will surely be of similar importance to ecologists a century hence. This is not limited to published material. Few ecologists want to publish all their data, even if editors and referees would let them. Some are so perfectionist that they take their knowledge with them to the grave. Notebooks and photographs belonging to deceased ecologists are part of the nation’s archives, and need to be properly housed and catalogued. As yet this is not systematically provided for, although some Biological Records Centres have made a valiant beginning.
Electronic storage
There is no remembrance of former things; neither shall there be any remembrance of things that are to come with those that shall come after.
ECCLESIASTES 1: 11
Electronic storage seems attractive, especially with the increasing habit of using Geographical Position Systems. Its limitations are illustrated by a well-known cautionary tale. To commemorate the 900th anniversary of Domesday Book a new project was set up in emulation of the original. A million people took part and promptly forgot about it. Within 15 years, although the disks survived, the equipment needed to read them had passed from ‘state of the art’ to obsolescence. The data were rescued, just in time, by a difficult recovery project.8 Had the equipment survived, the disks might not have lasted much longer. Meanwhile, William the Conqueror’s original text is still in excellent condition and can immediately be read by anyone who has mastered the abbreviations. The same has happened to other, less famous, electronic databases that were forgotten for 20 years or so.
Curating electronic archives is still in its infancy. Their great advantage in looking up data and saving space is opposed by their great disadvantage in needing frequent, expensive rewriting. Important data should still be preserved with printed copies on acid-free paper as a backup against a future when institutions stop spending money on material that is rarely consulted. Neglect is death to electronic archives in a matter of decades, but does not greatly matter to paper copies, which need only to be kept dry and can be scanned back into electronic form.