A Serengeti at Stonehenge?

Last week, Mark Avery kindly hosted my guest blog in which I started to set out the case for trophic ‘rewilding’ at the Stonehenge World Heritage Site and adjacent Salisbury Plain Training Area. The next day, the government announced its intention to consult (yet again….) on putting the A303 at Stonehenge in a tunnel. As noted by Mark, consigning the A303 to a tunnel would finally unify the Stonehenge landscape and render the introduction of surrogate aurochs and wild horses more feasible. I’ll return to the theme of rewilding at the Greater Stonehenge Landscape in later posts. For now, I reproduce my guest blog text below:

Making space for wild nature in England’s wheat belt

Rewilding advocates searching the UK for places to host a pilot or two are unlikely to settle long upon the endless wheat deserts of Wiltshire. Indeed, advocates of land sparing will probably look at Wiltshire as the sort of place where very high-yield farming should continue, to allow land to be spared for nature elsewhere.

But examine central Wiltshire in Google Earth and one is struck by the existence of an extensive landscape of relatively flat grassy plains, sitting directly north of Stonehenge and the A303. Here, in one of southern England’s most productive arable landscapes, is one of the north-west Europe’s earliest, largest, inadvertent rewilding areas.

The hidden 28,000 hectares

At around 28,000ha, the core chalk grassland plains of the Salisbury Plain Training Area (SPTA) are the largest ‘surviving’, intact example of this habitat anywhere. ‘Surviving’ is probably not the correct term, because a fair proportion of what is now a large military training area has been cultivated in the past, only reverting to grassland upon acquisition by the state. The MoD continue to revert arable to grassland here, and parts of the adjacent Stonehenge World Heritage Site (WHS) are the subject of more recent arable reversion.

The first phase of rewilding on a grand scale?

If one includes the publicly-owned arable bordering the SPTA grassland, the nearby Porton Down, and the parts of the WHS in conservation ownership, we find around 40,000 hectares (400 square kilometres) of land potentially available for trophic rewilding.

Essentially, the first stage of inadvertent rewilding has already taken place, partly to protect sub-surface archaeological remains and create a more sympathetic setting for Stonehenge, and partly to serve the current needs of the military.

Not surprisingly, SPTA has retained extraordinary biodiversity value, and part of its value is the product of its sheer scale, and of disturbance.  Parts of what is now SPTA and the WHS have never been ploughed, and these old chalk grassland patches have provided a source of plants and animals to colonise areas reverted from arable. This process of grassland habitat defragmentation, consolidation and enlargement is on-going with the potential for even more grassland creation on publicly-owned arable controlled by the MoD.


Military training with a metallic-megafauna of tanks currently creates and maintains a patchy disturbance regime upon which so much biodiversity depends. Tanks churn the chalky substrate, creating germination patches for ‘arable’ flora and parched crumbly soil for invertebrates. Tank-track puddles support a thriving fairy shrimp population. The impact area has frequent fires, whilst very light, rotational grazing by domestic stock maintain a mosaic of sward heights and densities. This mix of metallic, pyric and domestic herbivory and disturbance greatly influences the patterns of settlement by ground-nesters like skylarks, patchy marsh fritillary distributions and patterns of species-rich chalk scrub expansion and recession.

Bringing back the aurochs

So, we have here something that’s quite difficult to achieve in the English lowlands – habitat at scale. Can we foresee a future in which, as one alights to view Stonehenge, one is also treated to the spectacle of a landscape rich in wild megafauna?

Imagine the ecotourism potential here! People drive from all across the country – and fly from across the world – to see Stonehenge. Imagine the added attraction of hikes across rewilded plains full of grazing and browsing megafauna and, if we can modestly increase the extent of chalk scrub and copses, their predators too. If we can accommodate domestic cattle at Stonehenge and the huge training area to the north, why not their wilder counterparts?

What sorts of wild megafauna might be appropriate? The status of European bison in the UK is passionately contested, and the European wild horse and aurochs are, of course, extinct. Efforts are being made to back-breed into extant cattle the traits which were assumed to characterise wild aurochs. Of course, we can’t know the precise ecological traits of aurochs, but we can make well-informed inferences, and groups are doing just that. We will very likely soon have available animals that are, ecologically, very much more like aurochs than the domestic ‘rare and hardy breeds’ we currently use for grazing conservation landscaped, including those at Stonehenge and on the Salisbury Plain Training Area and Porton Down. If we’re happy using domestic livestock, and celebrate the ecological benefits of metallic-megafauna (tanks), how could we logically object to things back-bred to behave more like aurochs? The same with horses: why would rewilded feral horses behaving much like their wild ancestors be so objectionable to anyone (other than the MoD!)?

Some argue in favour of Pleistocene rewilding – introducing what they assume to be wild ecological surrogates for extinct species like elephants and rhinos. I’ll leave that open, saying merely that it needs a lot more thought and debate, and it’s not something I’d advocate.

But what about the ecological risks? Surely the core chalk grasslands of SPTA are of substantial biodiversity value which could be damaged by a revved-up rewilding programme?

Well, yes, the SPTA core grassland is a SAC and SSSI, and this and a larger arable area is also an SPA and SSSI. But the SAC value is really a function of the fact that it’s massive and open, and its ecology is largely driven by disturbance and patch dynamics – it’s not the sort of place where the precise ecological attributes of every square metre can be usefully defined. I’d argue that the careful reintroduction of wild and wild-type megafauna will not adversely alter the ecology of the area. How about the SPA? Well, stone curlews are a key interest here, and much of the population nests on specially cultivated plots in grassland and arable, or other heavily disturbed areas, with adult birds foraging for food on the bare ground of tank tracks and tilled areas – much of the grassland is too tall in summer for feeding stone curlews. The mosaic of sward heights created by mixed grazing and browsing herbivores – aurochs and wild horse surrogates breaking up scrubby areas and thick upright brome swards and creating grazing lawns – could well benefit stone curlews. And where tanks create water-filled ruts and churned pits of loose soil, could aurochs create water-filled wallows and sandy bathing pits? Tanks plough through scrubby patches, just as aurochs surely once did.

More specifically – where to start?

Notwithstanding the need to protect archaeological remains, it’s not unreasonable to say that the Stonehenge World Heritage Site area north of the A303 is already available for megafauna reintroductions. This could be the first phase. One would then need to negotiate with the MoD to revert arable land between this NT-owned area and The Packway, the road dividing the WHS from the larger SPTA. This MoD arable is public land, all-be-it leased to tenants. Incorporation to create a grassy corridor to SPTA is therefore feasible. But what of The Packway itself? It’s not a particularly busy road: one could foresee a mix of cattle grids and traffic calming, creating a megafauna crossing stretch along say 100 m of the carriageway. Of course, any other State would fund a wildlife tunnel or green bridge. This would give megafauna access to a relatively quiet part of southern SPTA west of Larkfield. If even a small area of MoD grassland here is opened-up for megafauna – say 4 square kilometres – it would be a start.

The big constraint here is not land ownership – much of the 400 square kilometres is publicly owned; it’s not human population density – the core 28,000 ha is of course unoccupied; it’s not existing ecological value – wild herbivores and their predators would compliment existing drivers of biodiversity value here; it’s not cost, notwithstanding some infrastructure tweaks and arable reversion. The real constraint upon the bigger part of the area is of course current use for military training. Carefully managed domestic stock are completely different beasts to free-ranging herds of essentially wild cattle and horses. Could such populations be managed so as not to pose an insurmountable constraint on military training requirements?

If the government and others were to be persuaded to consider accommodating relatively free-ranging herds of wild bovids and equids at Stonehenge and parts of SPTA, we’d have the core of an impressive rewilding project in southern England, through which passes a major transport corridor, and at the core of which already exists the major global tourist attraction of Stonehenge.



Cirl bunting recovery: What next?

The English cirl bunting recovery programme has so far been a huge success. Sterling efforts by the RSPB and farmers in the south-west have seen the residual cirl population in the South Hams increase very dramatically, and the re-establishment, through translocation, of a population down in Cornwall appears to have gone to plan.

Here, I ask if it’s about time we moved to the next phase – a further translocation much further east, somewhere close to the English south coast. This has long been an objective of the UK Species Action Plan for the cirl bunting, and the RSPB has demonstrated that translocation of cirl buntings can work.

To recap, the UK cirl population was once very widespread across the south of the country. In the UK, cirl buntings are very much a farmland bird: it’s probable that their population expanded with the expansion of low-intensity mixed farming.

As agricultural technologies developed, farming systems began to specialise, with arable concentrating in the east, and pastoralism in the west. Cirl buntings love – well, in the UK they required – mixed farming. The loss of spring-sown cereals and over-winter stubbles, rich in seeding broadleaved weeds, denied cirl buntings access to seed-rich habitats critical to over-winter survival. A dramatic decline in the abundance of grasshoppers across farmed landscapes was one consequence of the application of chemical fertilisers to agricultural grasslands. Cirl buntings love to feed grasshoppers to their broods, particularly in late summer.

These key factors driving the collapse of the cirl bunting population were revealed by intensive research conducted primarily by Dr Andy Evans at the RSPB. Having worked out the causes, the RSPB worked with Natural England (or was it the NCC?!) and the Countryside Commission (CoCo) to build cirl-bunting-friendly options into the pilot Countryside Stewardship Scheme then being tested in the South Hams. A particular problem at the time was the conversion of small coastal arable fields, upon which the surviving cirl bunting population depended for winter seed, to grassland. Local RSPB staff – particularly Leigh Lock and Cath Jeffs – worked with CoCo and local land owners to ensure that valuable spring-tilled arable was retained and weedy stubbles left over winter, and that grasshopper-rich pasture was managed sensitively to provide and maintain tussocky grassland foraging habitat over summer. Tussocky grass margins were established around arable fields too, enabling grasshopper populations to build up along the bases of thick hedgerows in which cirl buntings like to nest.

Having demonstrated that an agri-environment scheme can be used to provide all that cirl buntings need, Leigh, Cath and their local RSPB advisory team then worked to expand advice to farmers across the remaining range of cirl buntings in south Devon. Thanks to their heroic advisory efforts, and sustained funding from NCC…English Nature...Natural England and the RSPB, the cirl bunting population was stabilised, and their numbers proceeded to increase.

The trouble was that, although the numbers of cirl buntings increased, their range didn’t – in fact, in some places they continued to decline. A key objective of the UK Biodiversity Action Plan for cirl buntings was to increase their range, and establish populations elsewhere in southern England, away from south Devon. One reason for this was that a population of birds confined to a very small part of the country – like south Devon – could be vulnerable to a chance event, such as a very cold and prolonged winter. If you have a second population some distance away, chances are one or other will not be affected by the same event.

Given that the cirl bunting population in Devon was (and still is) showing very little evidence of expanding out of its South Hams stronghold, plans were devised to translocate some birds in an attempt to establish a second population. Cornwall – an area reasonably close to the existing population and with increasingly suitable habitat (because Stewardship agreements were maturing) – was chosen as an area in which to test translocation techniques.

The Cornwall translocation – Phase 2 of the recovery project, if you like – appears to have been a great success too. A breeding population has been established and this appears to be increasing of its own accord, without the need to supplement the population with further translocations.

Now, having proven the concept of translocation, is it time to move to the next phase – to establish a population away from the south west, somewhere in southern or south-east England?

I’d argue Yes, absolutely. The south-west populations remain vulnerable; there remains very little sign that the South Devon population is expanding east of the Exe estuary, into east Devon, from which it would then continue east into Dorset. Maybe we’ll arrive at a point where the density of pairs in some parts of south Devon is so high that range expansion suddenly takes off, but that could be decades away, and the northern portion of the Devon population (around Dawlish, Exminster and Exeter) remains at a pretty low density and is increasingly threatened by built development, so it seems unlikely that this will provide a source at the leading edge of an expanding range. And presumably appreciable range expansion of what’s proving to be a very sedentary species would take tens of decades – will they ever reach Dorset?! Basically, new cirl bunting pairs appear to establish territories as close as possible to where they were raised – a few tens or hundreds of metres away. Range expansion could therefore be a very, very slow process indeed.

Where, then, might we attempt the next translocation? Somewhere right on the coast would seem sensible, because the cirl bunting population held on along the coast of the South Hams, so the mild coastal strip may be the best area to create a new source population. Mixed farming – arable and grassland on the same holdings – is essential. And, I’d suggest, a reasonable long history of high quality agri-environment schemes in place: features like grassy margins take time to mature, to build up good grasshopper populations, and benign arable flora populations take time to build up in arable rotations.

How about The Fleet hinterland in west Dorset, Purbeck, or the Isle of Wight? The Isle of Wight certainly retains some promising habitat, but it’s an island, and it would take some determination on the part of non-dispersive cirl buntings to cross the Solent from the Isle of Wight. Those that make it across will be confronted by the New Forest and the sprawl of Southampton and Portsmouth – neither of which appear to provide particularly attractive habitat for an expanding cirl bunting population. The Fleet? I’m not sure to be honest, but it looks pretty good to me.

Personally, I’d settle on Purbeck, an area that has a history of farmland bird focused land management advice, which retains ample nesting and summer foraging habitat, and where I assume we might readily expand suitable over-winter stubble and other seed-rich habitats.

There is another possibility, if we’re prepared to look further inland: Salisbury Plain. Salisbury Plain Training Area and Porton Down cover about 40,000ha, of which about 28,000 ha is grasshopper-rich calcareous grassland, all in public ownership. Fringing the chalk grassland core is public-owned arable, where there is great potential to work with tenant farmers to establish weed-rich stubble and patches of wild bird seed crops. Crossing the plain and extending beyond it are chalk river valleys, the steep margins of which support some cracking chalk grassland and scrubby slopes – further potential nesting habitat. So maybe a bunting rear-and-release facility could be established somewhere within the training area.

So, is it time to move to a new phase in the cirl bunting recovery project – a translocation to central southern England?



The Diversion of Land

Published back in 1991, long before ‘land sparing’ and ‘rewilding’ entered the public consciousness, this fantastic book, The Diversion of Land: Conservation in a Period of Farming Contraction‘, remains probably the best account of the potential for land sparing in the UK. Although much of its analysis remains fascinating, it is probably now correct but for the wrong reasons, and an update is long overdue:



Riverside Land Trusts (1): Overview

I observed previously that in the UK we don’t really have a fully-fledged and diversified community and civil society land trusts movement, in contrast to, say, the US and many countries in Latin America. I’m not sure why this is the case, but there’s great potential, and one area where this seems particularly the case is riverside land – i.e. the often rather narrow bands of land that sit either side of even the smallest streams and which get rather soggy when it rains.

You can see the areas I mean in England if you look at the Environment Agency online flood risk maps here:


Zoom in to an area and you’ll see flood zones following river networks. They form often narrow ribbons of blue (i.e. flood risk) that meander their way through farmland and towns and cities.

We often go to extraordinary lengths to keep these areas dry, dredging the streams that flow through them, keeping drainage ditches clear and, in towns, we encase the channels in concrete, building right up to their edge.

Because they are frequently waterlogged or, in towns, vulnerable during larger flood events, they tend not to be the best yielding parts of your farm, and not the best place to build houses or industry.

I wonder if these ribbons of riverside land might be the focus of a new riverside land trusts movement. With appropriate start-up funding, charitable land trusts could be established at county or district level to work in the countryside, and city riverside trusts focused on urban river corridor acquisition and management. Some larger dollops of public cash could support one-off land purchases, particularly in anticipation of longer-term savings to the public purse on account of relaxed management once river corridors are under land trust stewardship. In towns, trusts could work with planners to ensure that local plan policies and proposals map identify strips of riverside land from which re-development must be set back, opening up corridors, and secure developer support via planning gain.

Of course, bringing riverside land into relaxed management will not address all the ills facing our stream networks – far from it. So many problems afflicting streams and rivers have their source well away from riverside lands, in the wider catchment, and these problems need to be tackled as close to source as possible.

But freeing rivers from intensive channel management, often conducted to increase drainage of remarkably narrow pieces of riverside land, would at least enable us to start to tackle physical river and riverside habitat quality. And we could start to establish new national riverside path networks, set at the edge of the floodplain or directly at the river channel edge in places, and with well designed river access points where people can get down into the river itself. I recall as a child visiting a big river ‘riffle’ where crowds of people gathered to play and look for bullheads and minnows in the riffle and downstream plunge-pool.

Streams and riverside land should be seen as a national treasure – an asset to be cared for and nurtured – not simply a water, nitrate and silt disposal facility.



News from Rampisham Down

Great news via Miles King – no solar farm to be built on Rampisham Down SSSI, Dorset

a new nature blog

thumb_P1040207_1024 Rampisham Down radio mast © Miles King

Regular readers will recall the story of Rampisham Down and the plan to build a solar farm on a nationally important grassland Site of Special Scientific Interest (SSSI). I have written about this case many times over the past few years, and you can read the back story in those blog posts; the story starts 3 years ago here.  Just search on Rampisham to find all the other ones. The story has plenty of plot twists, bizarre and slightly unbelievable characters, and some unsung heroes.

I’m delighted to say, after all this time, We Won!

Yes West Dorset District Council quietly slipped out the news before Christmas, that they have given planning permission to a smaller solar farm across the road from the proposed Rampisham Down SSSI site. As part of the permission, the developers, British Solar Renewables, have agreed to withdraw their…

View original post 130 more words


So few and so skittish as to be invisible?

I’ve read in a couple of places recently some knocking of what’s described as a ‘safari park vision’ of visible big herbivores and their predators – a vision said to be espoused by Rewilding Europe. I’m not sure what the critics are saying, but I assume their contention is that, within Europe, large herbivore and predator populations will always remain at such low densities and be so skittish as to be basically invisible.

Predators, yep, but large herbivores?

Whilst I agree (and hope) that we’re never going to have Serengeti-style mass herds or migrations of megafauna across treeless plains in Europe, I don’t think we can say that we’ll never have impressive herds of European bison, visible in open habitats. My guess is that as numbers of bison build up at various re-introduction sites, they’ll become ever-more obvious, and, provided we avoid spooking them too much, they’ll be pretty relaxed having onlookers.

How about aurochs, or their anticipated ecological stand-ins, the Tauros? My guess is that the real aurochs formed herds too – I’ll go as far as to say obviously they did. And they’re more grazers than browsers so probably favoured more open regions. So, if the Tauros created by back-breeding really do exhibit the ecological traits of aurochs, I guess they’ll form herds that’ll venture into open habitats. Again, as long as they’re not hounded too much, they’ll probably be pretty visible.

Wild horses (or whatever ecological surrogates we come up with)? Their herding behaviour varies depending on the habitat, but they’re another grazer and will form visible herds in more open habitats.

Groups of cervids are quite happy to venture out from the woodland edge to nipple at low shrubby vegetation and will do so during the day in quieter areas.

If you put all this lot together, in protected areas with a mix of open and more wooded habitats, free from harassment by people, then surely they’ll remain pretty visible to discrete onlookers?



How and where to induce land sparing

Evidence of the benefits or otherwise of land sparing – concentrating agriculture in the best and most versatile areas, and thereby leaving other land (and marine) areas aside for other uses – is growing.

Much of the evidence, where researchers use the model framework devised by Green et al. (2005)*, appears to favour land sparing, at least in frontier landscapes, where sensitive species associated with unconverted habitats persist. I’m not sure there’s much evidence either way for long-settled, cultural landscapes such as those across north-west Europe, and people using the Green model in such areas will need to be careful in how they select baseline habitats (I’ll come back to this).

If ecologists convince us that land spring is, indeed, a way forward, two practical questions confront the conservation and land use planning communities: Where to induce land sparing? and How to induce land sparing?

I’ll be discussing both the Where and How questions over the next few months.  Here are a couple of delivery mechanisms that I’ll expand upon in future posts:

Spatial planning: this is kind of obvious, really. Take the oil palm industry. Conservationists correctly continue to put a huge amount of effort into the Roundtable on Sustainable Palm Oil (RSPO) certification principles and standards. But I think we’ve neglected spatial planning, which is the responsibility of governments at various levels, and which could steer oil palm expansion (certified or otherwise) into already-converted lands. It’s no good if RSPO-certified growers only go for concessions in largely converted areas, whilst growers not intending to gain RSPO-certification go for and are issued concessions in unconverted areas. The NGO community could work with government land-use planners to initiate some sort of roundtable on biodiversity (and terrestrial carbon) friendly spatial planning. This ought to focus in the tropics where rapidly expanding tropical commodities are a major driver of biodiversity loss. What role might spatial planning play in, say, the UK, where land sparing is more about expanding habitats, rather than preventing further conversion and habitat loss? Well, in England, District councils prepare proposals maps – a detailed spatial plan of their areas showing where things like new homes will be built. These maps tend to ignore what goes on away from areas subject to built development, but could define habitat creation areas much more clearly. And in a future where production is being withdrawn from marginal farmland, do we really need to maintain such a dense highway network? Maybe we do, but we decommission other sorts of obsolete infrastructure – such as coal-fired power stations and old nuclear – so why continue to maintain a dense web of low-usage lanes? Converting some of these to green lanes with access only for cycling and walking could help to de-fragment landscapes and render recovery of bigger extirpated species more feasible. And spatial planning could work to identify the best parts of the highway network to decommission.

Land trusts: what to do with all this ‘spared’ land? Someone or something will have to be responsible for its upkeep. Conservation and community land trusts are an increasingly important means through which land is ‘spared’ from conversion and managed long term. They range from very local community groups convened to manage a particular site, to country-wide non-governmental organisations, such as the Jocotoco Foundation in Ecuador, created to establish networks of reserves protecting (sparing) habitat critical to the survival of endemic species. Then there are regional organisations set up to support the activities of these local land trusts, such as Nature & Culture International which is focused on the tropical Andes, and international organisations, such as the UK-based World Land Trust and the US-based Rainforest Trust and, of course, The Nature Conservancy, which support projects across the global tropics. Here in the UK, we of course have the National Trust, and reserve-network creating NGOs like the RSPB and county Wildlife Trusts. I suggest we need a big push in the UK to create a more acquisitive land trust culture. We need riverside land trusts that specialise in acquiring entire river corridors (if entire floodplain corridors were in the ownership of community trusts who’s activities weren’t compromised by flooding, we could relax about how the river channels themselves are managed); we need land trusts focused on creating big landscapes subject to full-on trophic rewilding (i.e. where extant and surrogate wild herbivore and carnivore communities are re-established); we need urban-fringe community land trusts focused on creating big natural open spaces for people.


*Green, R.E., Cornell, S.J., Scharlemann, J.P.W. & Balmford, A. 2005. Farming and the fate of wild nature. Science 307: 550–555.   [At the time this paper was published, I was delivering advice to farmers on how to make their farms better for wildlife – and I absolutely hated the message being presented by Green et al. Now I see this as probably the most important paper in conservation science and land-use planning to appear in recent decades…… ]