Somewhat overshadowed by the recent and momentous Cabinet Reshuffle – and the news that Larry, the Number 10 Downing Street Chief Mouser, remains in post – was a little reported item that the ever eager and cherubic Michael Gove has worked a blinder. He has announced that millions will be spent in creating a new Northern Forest along the M62 corridor. So we will have lots more trees planted and everything in the garden will be lovely. Doubtless the rewilders will be anxious to append the label “#rewilding” to this activity, but the reality is that planting trees is just another land management option that happens to have a longer cropping interval than, say, wheat or potatoes.
In principle, I have no difficulty with lots more tree planting. Like most people, I like trees. But I also like to see open spaces and skies and the bare bones of the land as well. So my general view of large scale tree planting in Britain is that it should follow at least four rules:
- The trees should be native species.
- They should be native species of genuine British provenance, to reduce the risk of importing yet more exotic tree diseases.
- The species planted should respect the local species mix (let us call this the Rackham Rule ).
- They should be planted in areas which are appropriate to trees and not used to cover historically open landscape.
It is the last point which I will explore in this article.
The route for this exploration is a roundabout one and begins with an earlier debate with Dr Steve Carver of Leeds University, in which there was an exchange of views about the sizes of areas and how this influences ecosystems within them. My point was that the boreal rainforests of British Columbia are much bigger than a grouse moor in the UK and that ecological lessons in one are not always immediately transferable to the other. Dr Carver argued that size did not really make a difference and so the two areas are roughly comparable. He wrote: “Scale. Yes, size really does matter! But as a Geographer and a landscape ecologist I know that many, indeed possibly all, processes scale. This is true from the physical (e.g. drainage patterns) to the ecological (e.g. predator-prey relationships).” At this point, I was unsure what Steve Carver meant when he said “scale”, but he went on to say: “..the thing about area to edge ratio is geometrically interesting since a circle of 18,000 km2 has the same proportion of edge as a circle of only 18 km2 and even one as small as 1.8 m2.”
This caused me to take a sharp intake of breath. I tweeted to Steve that he might like to check his mathematics on the latter statement, but he responded that there was no need – that for any given geometric figure, the ratio of perimeter to area is constant regardless of its area.
So who am I – a mere knuckle-dragging prole – to question the authority of a senior lecturer at a Russell Group University?
Except that echoes of ‘A’ Level Biology lessons taken 45 years ago reverberated inside my head. Furthermore, before becoming a farmer, I spent 20 years as a quantity surveyor in civil engineering. Most days were spent measuring lengths, areas and volumes. And – you know how it is – when you’ve spent a while earning a living in a profession, you tend to get a bit of a feel for the principles of the thing.
So, like engineers, scientists and surveyors the world over, I started to write on the back of an envelope. And this is what I wrote:
Figure 1 – Calculation of circumference:area ratios
At this point, we are left to wonder if the Geography Department at Leeds University runs on a different kind of mathematics from everyone else, or whether Dr Carver is like the beaver in Lewis Carroll’s The hunting of the Snark. In this extract the butcher and the beaver find themselves thrown together when walking down an increasingly steep and narrow ravine, when the air is split by a dreadful shriek. The butcher recovers his nerve and says:
If I have difficulty with Dr Carver’s mathematics, I have similar difficulty when he says that symbiosis and predator-prey relationships are the same thing, because “they are all part of a continuum”. Well, red and violet are at opposite ends of the visible part of the light spectrum (or “continuum”), but most people consider them to be completely different colours. Also, Dr Carver’s repeated contention that managed grouse moorland, in ecological terms, is MAMBA (Miles and Miles of Bugger All) is equally open to question. I will leave it to the reader to decide whether “Landscape ecology” is a real subject, or whether it just involves making stuff up as you go along.
All this talk of ratios, perimeters and areas may leave many readers rather cold: and the debate which gives rise to it, to be little more than an argument between an academic and a cynical, grumpy old fart on the internet. But if you are about to plant millions of trees and alter the use of large tracts of land, an understanding of these principles becomes very important – so bear with me a little further.
Imagine a large wood of beech trees. The core of the wood is composed of many tall, straight trees which reach up to the sunlight. The canopy is far above your head. There are very few shrub layer plants such as Hazel; and the woodland floor is almost bare except for bluebells which grow in densely packed profusion. As the leaf mosaic of beech trees is extremely efficient at trapping light, the lack of light within the wood during the summer explains the lack of vegetation at lower levels. However, Beech is very late to come into leaf in the spring and so this is the only time when light reaches the woodland floor at the same time that temperatures are rising. Bluebells are one of the few plants which grow and flower at this early part of the year, and so that is why they grow in such profusion.
At the edge of the wood, the canopy slopes down towards the ground, but is nothing like as complete and shading as it is in the core of the wood. Light is now able to reach the woodland floor at all times of the year, because it comes in from the sides. Here, Hazel, Hawthorn and perhaps Elder grow, and under them might be brambles and so on. The Bluebells are shaded out by the shrubs and more competitive field layer. This outer part of the wood is a manifestation of the edge effect.
Now imagine a much smaller beech wood – a little spinney perhaps. Here the Beech canopy cannot close over enough of the ground to produce sufficient shade. The dense carpet of Bluebells seen in the large wood is missing – there may be some Bluebells, but not as many or as densely packed as in the large wood. In this case, the woodland floor is much the same as the edges of our large wood – it contains a shrub and field layer with much greater variety because light can penetrate all round. The ‘edge effect’ has eliminated the core or centre part of the wood. This is shown diagrammatically in Figure 2 below:
Figure 2 – Plans and cross-sections of idealised large and small Beechwoods, showing the increased proportional influence of the edges of the wood, as the area of the wood gets smaller.
This idea is of considerable importance in ecology, as the following quote from Sutherland (2000)  suggests:
“Small areas have a relatively large edge with a high proportion of individuals occurring near the edge, and this may lead to extinction. The climate may be different along the a habitat edge….or there may be increased herbivory or predation or reduced seed germination…..”
We can also see from Figure 2 that the core of the beechwood (or any other type of habitat) diminishes to zero as the habitat gets smaller. The core is eventually overwhelmed by the edge. Species which are dependent upon the habitat core become increasingly imperilled. This is taken up by Sutherland again here:
“Thomas and Hanski (1997) showed that the emigration of butterflies occurred simply due to individuals wandering off the habitat patch and then being unable to return. This emigration was greater in small sites and could then be too great for the population to sustain. Similarly, for various carnivore species, the mortality is considerably greater once they step outside the park… so that population survival is dependent upon the protected area being sufficiently large. Across species, the extinction probabilities are correlated with the home range, so that nomadic species like Wild Dogs Lycaon pictus can survive only in huge reserves while the comparatively sedentary Jaguar Panthera onca often persists in protected areas under 100Km2. Thus, for both butterflies and carnivores there is a size below which the species is unlikely to persist due to a loss of individuals along the edge. However, the scale varies enormously: Wild Dogs require minimum reserves of 3,600 Km2, while Silver-spotted Skippers Plebejus argus require an area of only 0.0005 Km2.”
So we can conclude that as habitats become smaller in area, they are increasingly and disproportionately influenced by their edge effect. This decrease in area renders those species which are dependent upon the conditions in the core of the habitat to be increasingly vulnerable, reduced and perhaps even eliminated.
This all considers conditions as they vary inside the boundaries of a given habitat – we might call this the internal edge effect. But many habitats also influence events beyond their own boundaries. They may provide shading, or alter the chemistry or flows of watercourses for example. Or they may harbour predators which range beyond the habitat and affect species that dwell outside. Let us call this the external edge effect.
To examine what might happen if we start to plant or rewild patches of uplands with trees, let us consider the case of the Lapwing – a bird whose decline nationally makes its upland strongholds even more important.
Figure 2 – Avoidance of woodland for nesting by Lapwing (Source: Graham Appleton; wadertales.wordpress.com)
Here, we are given a figure of 500 metres away from a clump of trees as an exclusion zone for nesting Lapwing. Lapwing do not like trees because they harbour predators such as Carrion Crows, Buzzards and Foxes etc so they nest as far away as they can. This is one reason why they like to nest in fields of sugar beet in the Fens – there are few trees and the growing crop provides suitable cover.
So imagine an idealised piece of moorland, completely devoid of trees, which is circular in plan and 2 kilometres in diameter (or 1000 metres radius). The total area of that moor will be 314 hectares. Let us also imagine that it is a haven for Lapwing and that there are 150 pairs that regularly nest upon it. This gives an average of 2.09 hectares per pair. In fact, Lapwing will nest in loose colonies or associations with territory sizes between 0.4 and 0.8 hectares, but we must acknowledge that the reality of habitat variation will mean that some places on our imaginary moor will not be suitable for them. Nevertheless, a rough figure of two hectares each looks reasonable.
Now imagine that we wish to plant (or rewild) a small circular piece of woodland on that moor of six hectares (approx 15 acres). By calculating the radius of that wood and then adding the additional 500 metres, we arrive at an area which we can call a “Lapwing Exclusion Zone” or LEZ. Alternatively, we could opt for a single smaller wood of two hectares; and there might be a further option of three small woods of two hectares each. We can then calculate the effect upon the numbers of nesting Lapwing for each option:
Figure 3 – Calculation of numbers of Lapwing on a 2 km diameter (314 ha) moor: with no woods, and combinations of two and six hectare woods.
The single new wood of six hectares reduces the Lapwing nesting from 150 down to 89 pairs. This is a huge cut, but at least there are some left. The two hectare wood gives a slight improvement and allows 100 pairs to nest. But this is the point of the principle of perimeter to area ratio effects being disproportionately high for smaller areas. If the ratios were constant, as believed by Dr Carver, then we would expect the 2 ha wood to exert an effect of one-third of the six hectare wood. In other words, we would expect there to be approximately 130 pairs of Lapwing. Instead, a single 2 ha wood allows only 100 pairs.
The effect of splitting the original six hectare wood into three equal plots and spreading them evenly over the moor is disastrous for the Lapwing and they are eliminated completely. Following my own particular interests of upland birds, I suggest that the exclusion zones for Curlew, Golden Plover, Merlin, Red Grouse, Common Sandpiper, Dunlin, and Greenshank will be of a similar order of magnitude as the Lapwing.
In theory, this kind of calculation could be carried out for many plants and animals. The distances of effect reaching beyond the new wood(s) could be calculated in the same way. As indicated by the quote from Sutherland above, the critical size of habitat or micro-habitat varies enormously depending upon which species is under consideration. Sadly, we do not know enough about the hundreds of species involved to do this exercise. But the above method is the kind of consideration that should be given by conservation or environmental scientists when considering changes of use of the uplands (or even other farmland) when large scale tree planting or rewilding is being considered. In other words, there should be a systematic environmental impact assessment for such changes.
It should not be taken as incontestable that more trees are necessarily always good. Back in the 1960s and on to the 1990s, there was a rush by the Forestry Commission to plant commercial woods of exotic conifers all over the country. Then, as now, the uplands of Britain (and the heathlands of lowland southern England) were considered to be largely empty and of little commercial use. The result was a rash of woodland that is now reaching harvestable age, but where the the timber produced is often of dubious quality and difficult to extract without considerable environmental damage. The impact on our upland birds by this planting has never, to my knowledge, been calculated. And what applies to the planting of exotic conifers, also applies to rewilding by native species. Carrion Crows and Buzzards will use both sorts of trees from which to watch and plunder the nests of ground-nesting birds. The quotation below from the late, great Oliver Rackham  illustrates another problem with large scale planting using the wrong species in inappropriate places.
An example of the grand design of post-war planning in forestry will be found on Bodmin Moor in Cornwall. The construction of Collyford reservoir, farming improvement, forestry planting and china clay extraction have all nibbled away at the moor. Johnson and Rose  record that 2,700 hectares have of moorland have been lost since 1946, 714ha to forestry. Bodmin Moor is estimated to have shrunk to a mere 41% of its size since 1800.
Today, a walk upon the moor will reveal a landscape which is full of ancient archeology. It is almost impossible to walk more than a few yards without crossing or stumbling over lines of stones, field boundaries, dwelling boundaries and other artefacts, which are Neolithic, Bronze Age, Early Medieval and later. This is a landscape which has been open, and essentially treeless, for thousands of years. It was shaped originally by the underlying dome of granite; and then grazed and modified and utilised by humans for 4000 years or more. Unlike the hills and mountains of the north of England, Wales and Scotland, Dartmoor and Bodmin were not glaciated during the last ice age. Their ecology has developed along with their human occupation and use. That occupation and use is still visible today, where it has not been planted over with alien clumps of conifers. Modern farmers continue to graze the moor with sheep, ponies and cattle in much the same way that they have done for millennia.
Figure 4 – A small block of exotic conifers near Brown Willy, Bodmin Moor. A haven for predators and a disaster for ground-nesting birds. (Source: the author)
The Curlew, Golden Plover, Snipe and Dunlin which once bred on Bodmin Moor have all long since departed as breeding species. Most of the Lapwing have also gone. Neighbouring Dartmoor, which is a similar but much bigger lump of granite, still has a tiny number of Dunlin – the most southerly breeding population in the world of this trans-Arctic species – but they are hanging on by a thread. Most of the other species are in a similar precarious position, except for the Golden Plover which have gone .
The usual explanation provided by environmentalists for this declining or local extinction of breeding species is that it is the result of “agricultural intensification”. But this is as lazy as it is gratuitous. A glance at the map of Dartmoor reveals an area of 954 km2 which is roughly circular and intersected by a couple of minor roads, Princetown, Postbridge and the Dart valley. Large blocks of commercial coniferous plantations are at Fernworthy Reservoir, Bellever Tor and Burrator Reservoir. But this leaves two large areas of contiguous moorland, each roughly eight by twelve kilometres across, which are otherwise uninterrupted by forestry or small plantations.
By contrast, Bodmin Moor has an area of 208 km2 (Dartmoor is 4½ times bigger). The southern one third is split from the rest by the A30 – the major arterial route into Cornwall. The surroundings to Colliford Reservoir are not afforested, but nearby Smallacoombe Downs are surrounded by two large blocks of forestry. The smaller Crowdy Reservoir to the north has a large block of forestry nearby. But the body of the moor has 17 small blocks of plantation – Figure 4 shows one of them – which splits the moor in a north-south string. The tops of Brown Willy, Cathole Tor, Leskernick Hill, Buttern Hill and Rough Tor enclose an area which is a mere two to three kilometres across of uninterrupted moor.
By comparing Dartmoor with Bodmin Moor, we can see the two effects of internal and external edge effects at work. On the one hand, Bodmin is much smaller than Dartmoor and so its core is also smaller and greatly influenced by its outer edges. This core is the part in which the upland birds would breed, given the chance. This might be considered to be the internal edge effect. Unfortunately, this already small area has been further broken up by the string of small plantations. These contain the predators which ensure that ground-nesting birds are unable to cope with. And this is where the external edge effects of the plantations manifest themselves.
In many ways, it really does not matter if the trees in the new Northern Forest are to be native hardwoods of local provenance, or if they are to be dark and alien conifers. As far as Curlew and Lapwing are concerned, lots of new trees in the wrong places will be a threat and a disincentive to breed. At a time when suitable nesting ground, and successful brood-rearing, are at an all-time low, we should be very careful before committing to grand political gestures – leaving our stamp upon the landscape – without first looking at what we might lose.
It is likely that there is not a square inch of land in the British Isles that has not been touched by human hand. Even the remotest corners of these islands can only be described as “semi-natural”. Most of it has been touched, felt, burnt, turned over, tilled, fertilised, grazed, planted, cropped and built upon many times over. There is a current fashion amongst environmentalists to condemn this human activity as being always destructive, and this is one of the driving forces behind the rewilding movement. It is as if these people really hate their fellow humanity and view any activity, such as farming, as somehow abhorrent; and they look upon farmers and others with complete disdain.
But I disagree with this unpleasant nihilism. Human activity in the landscape is not always something to be deplored. Indeed, much of it is to be celebrated. Humanity will always bring changes to biodiversity and the landscape, but often this can be beneficial. Much of our landscape has been modified by and for the many animals that live with us – and many others who find our activities useful to them. The Lapwing and Curlew of the moorland would not perhaps be there at all if it were not for the activities of the farmer and gamekeeper. And those things which we bring with us are also the things which we love. It would take a stoney heart indeed to listen to the thrilling, bubbling call of the Curlew on a misty spring moor, without marvelling at its beauty.
Only if we cover the landscape in concrete and tarmac will nature be excluded. As we have already built the M62, then a new forest around it will help to mitigate some of the noise and pollution and perhaps some of the losses of farmland. But, despite the exhortations of the environmentalists, trees are not a panacea. In some places they are a positive menace to some of our best loved animals and plants, as this article has pointed out. The lessons of the dismal forestry of the 1960s which swamped fen and moor have not yet been learnt.
So be careful, Mr Gove, to remember to ask your eager civil servants this simple question: “Where, precisely, do you intend to plant trees; and what will be the losses of your doing so?”
 Sutherland W, (2000): The conservation handbook: research, management and policy. Blackwell Publishing, Oxford.
 The ‘Rackham Rule’ (H/T David Lovelace @boscinet)
 Johnson N and Rose P (2008): Bodmin Moor: An archeological survey, Vol. 1: The human landscape to 1800 (page 4). English Heritage, Swindon.
 Balmer DE, Gillings S, Caffrey BJ, Swann RL, Downie IS & Fuller RJ (2013): Bird Atlas 2007 – 11: The breeding and wintering birds of Britain and Ireland. BTO Books, Thetford.
Thanks David… Just back from a week away. I don’t really have time for this at present, but just quickly what I actually said and what I meant about scaling was this…
“but the thing about area to edge ratio is geometrically interesting since a circle of 18,000 km2 has the same *uniform relationship between* edge as a circle of only 18 km2 and even one as small as 1.8 m2.”
Perhaps I wasn’t as clear as I could be about scaling by calling it proportionally the same when what I meant was if you plot radius versus perimeter or area vs perimeter you’d see that they increase proportionally in a uniform and monotonic manner; area vs radius plots as a straight line, and area vs perimeter plots as a uniform curve. Your own figures show this very nicely although you don’t plot them up for your readers to see. However, I apologise for the lack of clarity in my expression.
What matters here for ecology, however, is the functional relationship between area and perimeter, which if you look at these plots will show that the geometric relationship “scales” up very nicely as one might expect, whilst the functional relationship remains constant… a circle, is a circle, is a circle regardless of its absolute size. Of course, this gets more complex with the myriad of other geometric shapes and the non-uniform patches commonly found in landscape mosaics but the general principles remain the same.
In absolute terms the size of the habitat patch or landscape matters because with increasing area there is a greater the likelihood of finding more and more different species and with it the diversity of species-species interactions. This is the “species-area relationship” that I explain in my reply to your previous article. Crucially it also matters for ecological relationships between habitat geometry and species mix because habitat requirements vary between species. This was why I was suggesting that our grouse moors may be the territorial equivalents of the mountains of BC when considering the predator-prey relationships between stoats/weasels and ground-nesting birds, and those between wolves and moose/caribou respectively. That this is then implicated in the effects on “spill-over predation” is just a theory of mine but one that seems to be increasingly borne out in conversations here and there. More researched is need in other words.
I look forward to studying the rest of your article in more detail and will no doubt reply and perhaps draw some plots for you.
Thank you for your correction of your statement.
In many ways, the species-area relationship and the edge effect are similar expressions of the same phenomenon – that the bigger a habitat is, the greater the number of species adapted to that habitat can be supported. In particular, those species which require large areas can thrive only in an area which is the same size or bigger than, say, their territory size. The quote from Sutherland above which points out that Jaguars require smaller territories than hunting dogs, is an expression of the kind of constraints which influence the species richness of a habitat of any given area. The edge effect which I have detailed will limit the “effective” area of that habitat in the way that I have illustrated in my comparison of Bodmin Moor and Dartmoor.
So I think we can now agree that the bigger the habitat is, the more species it will support. Furthermore, I am sure we can also agree that if that habitat is broken up by human disturbance such as recreation, development, alien patches of forestry, or even 10m by 10m exclosures which encourage tree growth within them, then that will diminish the size and therefore species carrying capacity of that habitat. Once again, you can see this on Bodmin Moor rather starkly.
I’ll have to make the clarification in my article (with thanks).
I think this still leaves us with the question as to what “wildlife” we want and where we want it, and the differences between biodiversity and wildness/naturalness. I agree with you that the bigger the habitat patches, the better for species that are adapted to those conditions (viz your ground-nesting birds) and the greater likelihood of finding healthy populations of said species, as well as a few others that can survive under such conditions. However, the species-area relationship kicks in fully when there is a mixed mosaic of habitat types providing territories for a greater range of species. This may be to the detriment of some due to the effects of competition, reduction in habitat and predation. I will muse further on your “model” of Lapwing Exclusion Zones (LEZs… you’re beginning to sound like Sir John) but I suspect much hinges on spatial configuration of habitat mosaics, patch sizes, and the true nature of edge effects.
Thanks Steve. I look forward to it.