Earthworks update.

By Andy Williams.

It hasn’t rained. It hasn’t rained in a very long time. We’ve had one brief shower since early May, roughly eight weeks ago. Now normally I’d be loving the opportunity to get so much done out on the croft, but we have a lot of bare earth baking in the sun, waiting for rain to sprout the meadow seed mix we’ve spread everywhere. The existing pasture is coping with the dry conditions well, but everything we’ve planted is struggling. In late winter we planted well over 700 bare root trees. They’ve done well, but a few weeks ago trees that were healthy started suddenly dying, When we pulled back the mulch around the base of a random tree, we found that the soil had shrunk so badly that it had left a huge crack that took two spadefuls of soil to fill. Every tree was like this, apart from the willow.

We’ve had to barrow soil to every tree and shrub we’ve planted, remove the stones holding the mulch down, remove the mulches, fill the trenches, reverse the process, then water them all. It’s taken some time. While we were at it, we added a scoop of slow release fertiliser to the fill around each tree. It’s not something we plan on doing in the long term but these trees have had a very rough start and at this point their survival has become the priority. Under the circumstances, I’m comfortable bending the rules occasionally to accelerate the establishment of the system. This is extreme weather, and once established the systems should be resilient enough to cope with such events easily, but only if we can get it there. The earthwork systems we’ve installed should make droughts like these practically irrelevant, but only after they’ve actually had some rainfall to harvest. Is it any wonder I’m frustrated?

One silver lining to it being so dry for so long is we’ve been able to see some of the wildlife benefits of the earthworks. Even when they’re bone dry, hard baked soil with no plant cover, the birds love the earthworks. The crest of each swale berm has a healthy sprinkling of bird droppings. Many bird species won’t search for food very far from a perch, and for some species fence posts just won’t cut it. Instead of an unbroken field of grasses and scrubby plants it’s now got high and low points, sheltered and exposed spots. The diversity of birds in the field has already started to increase as a result, and now occasionally at dusk we see a barn owl hunting. I often find evidence of bird kills on the very highest points of the berms, though I couldn’t say what species is doing it. The berms will eventually be planted to productive tree and shrub species, so will really benefit from the enrichment.

This week, after much sulking  thought, I worked out I could just about get a hose to the lowest earthwork. This is a stepped pond, with a curved berm behind it, shaped to catch and reflect the southern sun. It has a contour ditch to one side that collects overland flow and directs it into the pond, and should the pond fill completely during a major rain event the ditch will act as a level sill spillway so the berm won’t blow out from the weight of the water. There are several level platforms cut into the sides of the pond, the largest being the highest, just a couple of inches lower than the height of the level sill spillway.  In the event of the pond becoming overfull, first it’ll soak the top platform, then start to soak into the berm. Only then will the spillway kick in, so the water can’t get any deeper. It means that should we choose to, we can release water from further up in the landscape, flood the pond and soak the berm to water the whole area. The back of the berm has been planted with willow cuttings and then the whole area has been seeded with a meadow flower mix. It’s a small pond by agricultural standards, but even after months without rain it has a couple of feet of water in the bottom. Once it’s greened out the plant roots will knit the soil together and it’ll be a source of permanent water as well as being one of the most sheltered places on the croft. When the willow on the back of the berm has grown, it’ll be woven together to make a dense screen that will really shelter the whole earthwork from the wind. I dug a lot of big stone from this pond, so I’d planned on adding them to the berm to act as heat stores. If you’re going to add piles of stones to an earthwork though, you might as well make them comfortable to sit on. I’m going to rebuild the first of the stone seats I put together, but it’s hardly a priority.

With the hose just able to reach the end of the water collection ditch I was able to finally test it all. You can spend all day double checking the levels, but until you see water in it there’s always a niggling doubt in the back of the mind. This is the only truly finished part of the earthworks. By the time I’d finished this one, the sun had baked the soil so hard that trying to work it with spades is just brutal. I’ll post the other parts of the system as they become finished, because in the raw form they really don’t look like much. Have a look at these two photos for the contrast.

I don’t expect the pond to be full very often, and almost never during summer. If it’ll hold a couple of feet of water when it hasn’t rained in months however I’m sure it’ll be significantly higher for most of the year. Most new ponds seep a lot more water than they do when they’re older, clay particles washed into the pond are drawn into the tiny seep spaces and partially block them, making the pond to hold water better. You can buy preparations to give a pond a head start, or to fix a leaky pond. Other techniques use ducks to manure the water, because duck poo has particularly fine particles. We don’t have ducks yet but we do have clay. I’ve mixed up a tub of clay and water into the consistency of double cream. I’ve watered a third of it down and added it to the pond while it’s full. The pond already seeps very, very slowly, so any improvement should be significant. When the pond is full it’s over 8 feet deep in the middle. It’ll be interesting seeing what sort of depth it settles at. The next phase of the earthworks isn’t going to be built for at least a couple of years, but I’m so pleased by how this pond turned out that I’m really looking forward to it.

It’s the suspense that gets to you.

By Andy Williams.

It’s not often you find yourself waiting impatiently for rain when you live in the UK, but that’s what I’ve found myself doing for the last 36 hours. Here’s the thing. I’ve read all the books, researched online, then read the books some more. But until I see my earthworks dealing with a real rain event, there’s going to be a niggling doubt there. And it’s not rained in a week. It’s meant to rain tonight and tomorrow so I’m keeping everything crossed while muttering at the dark clouds on the horizon.

Today , after finishing tidying it up with spades, we’ve planted the berm of the swale up a little. We’ve put in the only proper apple tree we have, that a friend kindly gave us. I’ve carved it a little niche in the berm and lined it with stone to protect it while it’s little. We also put three autumn olive bushes around it. They’re a nitrogen fixing fruiting shrub that should do well here. We’ve put in 40 hybrid willow as a nursery crop, along with a few sycamore we found growing in the old garage. The plan is to replace them gradually as we buy the fruit and nut varieties we want. They can be chipped to be turned into a mulch that will encourage the fungally dominated soil we want to develop on the berm. We’ve put in four black locust on the end of the berm nearest the veg garden. They only produce a thin canopy even when they have leaves and will be coppiced routinely once they’re established so won’t cast any shade. We’ve added several buddleia for pollinators and even a few marigolds.

This is a very new swale, and it’s all bare earth. Covering that fast is vital, to prevent erosion of topsoil and stop invasives from moving into the open niche. We have a bee-friendly meadow seed mix ready to go on tomorrow, assuming it gives us some rain, but in the meantime we’ve seeded the berm with vegetable seeds. you know all the half empty veg seed packets that build up in your tin? All those saved varieties, bits shoved in unlabelled envelopes because ‘of course I’ll remember’? Wild seeds collected on autumn walks? We put the lot in a tub, added some sunflower and rocket seeds, gave it a good mix and spread the whole berm with it. Who knows, maybe we’ll even get some veg from it.

Under a foot of topsoil the ground is hard clay here. It’s hard work getting into it even with a mechanical digger. This means the bottom of the swale ditch after a week of wind and sun is like concrete. I plan on broadforking it (you know, assuming it ever rains) to break it up a bit and planting daikon radish fairly thickly to decompact the subsoil. In the meantime, I’ve put a sandbag across the pond entrance to keep the ditch from emptying into it. Hopefully a little soak should soften it up.


The pond…. I really had meant to dig a very small one. It had to be small, I’d already decided where the berm was going to go! But then I hit water and had to reconsider. I’m delighted with hitting reliable ground water, but it has made us evolve the plan. We’re going to enlarge the pond, to make better use of the water. Basically, the digger is coming back for a few days.

Still no rain. You know, in case you’re wondering.

Well, well.

By Andy Williams.

I went for a walk along the new earthworks this morning. We’ve yet to do the fine work with the spades so the pond in particular is still a little rough, so I wanted to see how much digging I’m in for this weekend. What I found was almost two feet of water. Superb. The top of our field is the ridge in the landscape, so it’s from the water table. This area is covered with wells and springs, but you can’t just sink a hole anywhere and hit water. We have a deeper, older ditch that was originally dug to take the water from the barn and it’s occasionally wet at the bottom (oo-er!) The new pond has water barely a metre down, and it’s not rain water. I noticed when I was using the digger that at one point I had water flow, but this is a lot more than I was anticipating. If this is reliable through summer it’s a huge deal. Over time we’ve always planned to take the croft off grid, so reliable water for livestock makes life a lot easier. What’s stunned me is that this water already has a pond skater. This pond is less than 24 hours old and someone’s moved in.

There’s a spot right at the top of the field that’s reliably wet. If we can develop that into a functional spring, we could gravity feed that water to anywhere on site. It’s starting to look like the site design has acquired another pond.

That’s no ordinary ditch…

By Andy Williams.

I know, I know, it’s been ages. I promise we’ve been very busy though. One of the frustrating things about having this blog is that when we’ve been doing interesting stuff we’re generally too busy to spend long blogging. We’ve had a couple of weeks of good weather, so have cracked on.

The front garden is almost finished. We’re waiting on a load of wood chip to be delivered. Once that’s spread over a load of cardboard it should take out the ground elder and few flowered leek. We’ve kept some of the leek in a pot, for something a bit different in the herb garden, but it can get away from you in the ground.

We finished marking the swales in the top field. We’ve bought a load of reclaimed timber to build a decent shed and some other reclaimed bits and pieces for the back garden. I’m sure I’ll blog about them when we get to it.

Best of all however is we’ve made a start on the earthworks! We hired a mini digger to level the shed site and booked an extra day to get a few other jobs done quicker. I must admit, I love it. It just gets things done so much faster and easier, it’s worth every penny. I’ve dug the lower swale and made a start on the integrated pond. I’ve designed it to work a little differently from a conventional swale. Rather than hold water until it infiltrates into the ground, this one drains into the pond at the end. It can be turned very easily into an infiltration swale, giving us the option to regulate the hydrology of the site throughout the year. The digger goes back after tomorrow so I’ll not manage to dig the other two this time, but it’s encouraging to see some of the main system components going in.

It still needs a little tidying with spades to finish, and I’ll need to go out during the next major rain event to tickle the spillway from the pond, but you get the idea. That digger will be back to visit again very soon I suspect.

Meet bucket.

By Andy Williams.

Bucket is wise. Bucket can tell us things. Bucket can be a surprisingly illuminating diagnostic tool. Hang on and I’ll tell you all about it.


I dug this hole back in March, so I could use the soil as a mound for the hazel tree we planted on it. We have many such holes around the croft that were dug at the same time, and they’ve been surprisingly interesting. We knew the ground here gets saturated very quickly when it rains very soon after buying it. As soon as we started planting we learned that we had roughly 18 inches of good topsoil over clay subsoil. Since we intend to dig ponds eventually this was excellent news, but it affects the hydrology of the site. At least one of us walks the fence line every day, checking the trees, so it’s easy to keep an eye on these holes. Immediately after rain they have up to four inches of water in them, draining to an inch or so within three days, and being no more than moist within four to five days. I put this down to a combination of water being infiltrated into the ground and the drying effect of the wind. Soon after we planted the trees I found the bucket jammed between some rushes. I shoved it into the hole so it wouldn’t blow away, intending to pick it up later, and promptly forgot about it until every time I passed the tree.

It added another dimension to water table observation. I’d read about using a bucket to monitor rainfall in a book I read years ago, but the author hadn’t thought of sticking it in a hole to get two sets of data. I really wish I could claim credit for it properly.

Typically, you stick a bucket somewhere it won’t blow away and watch it over a year. It’s as simple as that. If it always has water in it you have a wet climate. If it often dries completely you have a dry climate. All pretty obvious so far yes? Where it’s useful is in determining unusually dry spells, when pasture may need extra water to stay alive. Unlike proper rain gauges the bucket factors in moisture lost to the wind. It won’t however factor in moisture lost to ground infiltration, but my hole does (seriously, stop sniggering). Rain in Caithness is regular but not very often too heavy. In terms of inches of precipitation it’s similar to Kent, apparently. Windier though. The bucket has so far not gone beyond half full, but hasn’t gone lower than four inches deep either. The sun and wind evaporate the moisture fairly consistently, so will be having a similar effect on the pasture. Our soil isn’t waterlogged, it’s intermittently wet because moisture is slow to percolate through the clay subsoil. And that can be fixed easily through sensitive earthwork construction. It’s a game changer for our understanding of this site. And that, I think you’ll agree, isn’t a bad result from a forgotten bucket.

Shelterbelt design.

By Andy Williams.

There’s no denying it, Caithness can be windy. Most of the croft houses in the area seem to have trees planted around them, but when we bought our croft it was bare fields. The only trees were in the front garden, and that was a jungle. The trees were close enough to the house for the roots to be affecting the foundations so unfortunately most of them had to go. The ones closest to the house have already gone, with the others earmarked for after the trees we planted have grown big enough to give us some shelter.

High winds can cause many problems for agriculture. It makes it harder for livestock to put on weight so increases feed costs. For growing plants, the sheer physical damage to plants can be an issue. It makes irrigation tricky, because it can dry out soil too quickly. There are advantages however. Sheep farming here can be easier because the wind keeps fly strike from becoming such a problem. After heavy rain, wind dries pasture faster than still conditions will. Overall however, we could certainly do with a fair bit less than we currently receive. We don’t have a huge amount of land, a shade over five acres, so need to use it very efficiently. A belt of sheltering trees around a field needn’t cost anything in yield however. Estimates vary depending on which authority you’re reading, but typically it’s estimated that 10% of the perimeter of your field planted to the right species can increase the effective yield inside the shelterbelt by over 20%, without even factoring in any yield from the shelterbelt itself. Yes please, we’ll have some of that!

Back when we knew we were going to be buying a croft but hadn’t zeroed in on one yet, I had certain preconceptions when visualising what ‘our’ croft would look like. The biggest of these was that it would of course be a south-facing slope. Designing a system on a south facing slope is a doddle. You can put your huge nut trees along the north edge of your land where they won’t cast any shade over the rest of your system. You can plant lines of full-sized trees on contour because the slope will allow every row access to the light. I knew exactly what it would look like, and there’s no denying that it affected my research into how to design agricultural systems. When we bought a northeast facing slope I had to go back to the drawing board in a lot of areas. I spent many hours thinking about the peculiarities of the site. The main ones were:

  1. That loch view. Yes, it would be beneficial to plant dense, high, evergreen trees and shrubs to the northeast to block those cold winter winds but that’s the view that made us fall in love with the croft in the first place.
  2. The southwest edge of the site is the natural ridge. The height of the trees there would determine the amount of solar gain the whole croft receives in winter. Full height trees would make the whole site colder, particularly evergreens.
  3. When it’s windy, it’s *really* windy. While it would be lovely to plant a wonderfully productive shelterbelt entirely of fruit and nut varieties, until they grew big enough to provide at least a little shelter for each other we’d suffer heavy losses of expensive trees.
  4. There is a large barn in the southwest corner of the croft. We don’t own it, but it does affect the site. It provides completely impermeable shelter from winds coming from the south and west, for the bit of land in its lee at least. It causes shade that varies according to the time of day and the season.
  5. The site is wet. We hope to change the way water flows through the fields, but you can’t promise a tree it’ll be happier next year when it’s drowning. After rain the fields squelch underfoot, but after a couple of days of dry weather the water table can drop by over six inches.

I’ve approached the shelterbelt development in two phases. The first is the work we’ve already done, and the second phase will be implemented over the next few years as the plantings mature. One of the first things we did was remove the fence between the top and bottom fields. It was in a terrible condition but more importantly it’s more difficult visualising a piece of land as one system when it’s divided up by someone else’s straight lines. Nevertheless, because the design is different in what was the lower field because of the loch view we want to retain, I’ll stick to ‘top field’ and ‘bottom field’ for the time being. ‘The land formerly known as top field’ is a bit of a mouthful after all.

Northwest edge, top field.

This stretches from the house up to the barn in the southwest corner of the property. It receives the water running off the barn as surface water so has thick rush cover at the moment and is the wettest area of the whole site. Last year we stuck 147 willow cuttings along this edge after felling a tree in the front garden. At the time the pasture was knee-high in creeping buttercup so using the dilapidated fence as a guide has given us a somewhat wavy line. They’re just starting to put out shoots, so I’m going to remove them in the next couple of weeks because doing so later may damage the roots of the trees we’ve planted next to them since.

We’ve put in a double row of Italian alder, planted in a zig zag pattern, as the outer rows here. The outer row is four feet from the fence and planted four feet from each other. The inner row of alder is slightly inside this line, each planted between the trees in the outer line but still four feet from each of its pair of neighbours. Alders thrive in wet conditions, but as we’re going to divert the overland flow from the barn during the planned summer earthworks we needed to use a variety that can cope with drier conditions also. Italian alder can cope with occasional droughts as well as water logging, so should do fine here. It’s fast growing, putting on six feet in its first year, so should give us height quickly. It holds its leaves well into the autumn so should give plenty of protection from winds from the west for fruit trees that are finishing late. Like all alders it fixes atmospheric nitrogen as nodules on its roots via a symbiotic relationship with frankia bacteria, and these trees arrived ready inoculated. This nitrogen should encourage fast growth in the trees we’ve planted inside the alder. Italian alder has edible leaves. I’m rather curious to see how they taste to be honest.


Inside the alder we’ve planted a polyculture of bird cherry, crab apple, cherry plum and elder, all fruiting varieties. The area in the shade of the barn is elder-heavy because they crop well even in shade. There are also three small-leaved lime trees, which are reputed to have extremely tasty leaves.  They respond well to coppicing, so we can cut them in turn and always have one low enough to be easy to harvest. We’ve also planted some hazel on mounds to try to keep their roots out of the water-logging. Between and under the trees we’ve planted blackcurrant, redcurrant and sea buckthorn. The buckthorn doesn’t like waterlogged soils and some of them are struggling, but the earthworks should help that once they go in. Under all the fruiting species we’ve planted comfrey root cuttings. These will be divided and multiplied every year until they form a rhizome barrier to competing grass species and can be cut several times a year and left in place to feed the fruit trees and shrubs. No barrowing of fertilisers around the site, just a few minutes with a scythe. Between the trees and shrubs we’ve planted thousands of snowdrop and daffodil bulbs. They should have done their thing for the year before the trees and shrubs flower. We plan on having grafted, named cultivars of hazel, plums and apple inside the protection of the shelterbelt and all benefit from ‘wild’ pollen from seed grown trees. The mix of species should give a good nectar flow for bees over a sustained period as well as be attractive to humans. Who says beauty can’t be a yield too? An unexpected benefit of this area being covered with rushes is they’re providing excellent shelter for trees and shrubs nearby. Those planted near or among large clumps of rushes are leafing out significantly earlier than those in more exposed places. The Alder is the only species here that’s been planted in a straight line, in a double row. As vertical a face as possible to the wind has the best wind buffering effect, which is why we’ve planted the outer tree species in tight rows throughout the shelterbelt. Inside the protection of these we’ve intentionally kept things more random to look more natural and, hopefully, behave more like a natural ecosystem. We’ve left the areas of rushes bare of trees deliberately. After the water from the barn is diverted we’ll take the rushes out, and these areas will leave empty spaces for planting other species at a later stage.

Southwest edge, top field.

This is the driest part of the site, apart from the southernmost point where there is a dip in the ground full of rushes. It might lend itself to a pond at some point, but I’m not making that decision yet. Most of the outer row is hazel, four feet apart, but towards the wet corner we’ve used blackthorn. Inside this is a row of mixed blackthorn, bird cherry, elder, cherry plum, crab apple and rowan. Next to the barn is a single yew tree we grew from a seedling. Yew can be shade tolerant so should do well here. Under and between the trees we have more blackcurrant and redcurrant. All fruit species have comfrey root cuttings planted beneath them, that will need dividing next year to multiply. The hazel is more a multi-stemmed shrub than a tree, so should spread nicely to give good cover without getting too tall. Being deciduous it will allow some light through in winter when the sun is lower. It copes well with exposure and will itself produce a yield in time. We have deliberately left gaps in the inner rows of trees, for other species during the next phase of planting. We’ve planted many daffodils and snowdrops between the trees here also.

Northeast edge, top field.

The outer edge of this system is Bowles hybrid willow at fourteen inch spacings, four feet in from the fence. This hybrid is often grown as a biomass crop and grows incredibly fast. Ten feet in its first year, up to twenty-four feet at full height. It then spreads from the base to form a small clump. It leafs out early so should give excellent protection  from cold winds from the east for early blossoming species planted inside the shelterbelt, and is an early pollen source for bees. It also has a reputation for being very windfast, so I’m very curious to see how it performs. If it works as well as it should, the field should be much more sheltered by the end of this summer. Inside the willow we have a polyculture of bird cherry, blackthorn, cherry plum, rowan, sea buckthorn, blackcurrant, redcurrant, a couple of small-leaved lime, a couple of honey locust and a few radiata pine. The radiata are reputed to be one of the most windfast trees in the UK, a real pioneer species. They don’t like waterlogged roots, so we planted them on small mounds we dug for them. They’re suffering from severe wind burn at the moment, but of all the trees to suffer it could be worse. They were only ever intended to be for system establishment and eventually cut down, but I’m hoping some of them pull through. It’s nice seeing a little green in winter. This system hasn’t had comfrey planted for ground cover yet, but will early next year. It’s been the most exposed edge, so the trees are lagging behind the others. It seems likely that will change once the willow does its fast growth trick however. This part of the top field has been planted to have an inner edge that’s irregularly scalloped, to tie in with the approximate positions of the planned earthwork swales. The settling ponds dug inside these embrasures will create extremely protected microclimates that will hopefully allow us to grow things that would otherwise struggle here.


Northwest edge, top field.

This is the fence that we have removed. It originally divided the croft just uphill from the house and barn. The area between the house and the barn is being developed into a protected veg garden and will eventually have a greenhouse running along the approximate former fence line, but angled to face true south for maximum solar gain in winter. We’ve planted willow cuttings on six-inch spacing here as a temporary wind screen. The area between the barn and the northeast boundary will probably eventually become a perennial veg system, with the potential to develop some aquatics systems. We will plant a hybrid willow screen along the northeast edge of this section, between the barn and the southeast boundary. It will be several rows deep, so we can coppice a row each year for producing wood chip without reducing the wind buffering effect of the willow. We’re currently awaiting delivery of 200 cuttings for the initial planting.


View from the top field looking north past the house and barn.


Bottom field.

We’ve deliberately kept the trees here to a minimum, to keep the loch view open. The northwest edge has the phone line running above it, so we’ve limited it to a polyculture of smaller species. We have rowan, blackthorn, cherry plum, black locust, blackcurrant and redcurrant here. The southeast and northeast edges have a sparse planting of blackthorn and hawthorn with a few bird cherry and cherry plum. We’ve planted a couple of black locust with a couple of larger hawthorn behind them very specifically to screen our view from our house of the one house between us and the loch. We also have a radiata pine, a single spruce that was a freebie from a toilet paper promotion, and a single Korean pine that’s a retired living Christmas tree we’ve kept for years. We intend to add another couple of Korean pine eventually, they produce excellent nuts, but it’s one of the few species we intend to buy as older trees. They don’t start bearing nuts until they’re 20 years or so old, so it’s worth it for the head start. The Korean pine has a few blueberry planted under it, mostly to see how they develop.

Phase two.

Most of the species we’ve used in the shelterbelt have been relatively cheap to buy as cuttings or bare root trees. Other species that we intend to use however are expensive at £15 and upwards each. Because we want to use them in quantity they’re far beyond our budget. Instead we’ve bought bulk seed. We intend to establish a seedling tree and shrub area to grow our own for a fraction of the cost. A major advantage of this approach is we can grow enough plants to choose the ones that handle the local conditions the best rather than named cultivars. I’d rather hardy, high yielding plants than expensive ones that will struggle here.

Black locust. I love black locust. It’s a legume tree so fixes nitrogen. It’s a very attractive tree, with flowers that bees absolutely love. It has relatively open foliage so planted on pasture the grass still receives enough light to grow well. It burns green and coppices readily. Best of all though is how long its timber lasts. It resists rotting for anything from 500 to 2000 years depending on the environment. We plan on using it a lot. For example, I like wooden hurdles for fencing but don’t fancy having to remake them every few years. Using black locust I’ll only have to make them once. We’ll be planting plenty to go on the swales, but will also be using it to fill in odd gaps because it’s so useful.

Autumn olive. Another nitrogen fixer, this shrub also produces useable berries. We’ll use it to fill in some of the gaps in the top field shelterbelt, in particular the southeastern edge where there’s a lack of nitrogen-fixing support species so far. Particularly high yielding plants may be moved to the swale understorey planting.

Siberian pea tree. Again, a nitrogen-fixing legume. This is a bush that is extremely wind hardy and produces seeds that can be used much like lentils. It’s often used over and around chicken yards for the birds to self forage the seeds. We plan on using it a lot in the lower field, where we’ll be planting forage crops for a pig and chicken system, as well as filling some of the gaps in the understorey of the top field shelterbelt.

Juniper. This needs no introduction of course, we just haven’t found a supplier of cheap trees that won’t hammer us on the postage. I’ve observed natural polycultures in Scotland of pine over juniper over raspberry over herbaceous ground cover. We plan on using it in a similar role here and it should do well.

General design elements.

We’ve deliberately not used hawthorn in the top field because it can cause problems for pear trees when they’re planted too closely. There are hawthorn hedges up wind in neighbouring fields so we may struggle with pear here anyway, but it makes sense to at least maximise our chances.

The leaf fall from nitrogen-fixing species is extremely high in Nitrogen. Some of the leaf litter will remain below the trees that grew it but much of it will be wind-driven. Leaves tend to collect in swale ditches, so once they’re installed they’ll passively collect fertility from the shelterbelt using the remaining breeze to do the work. We’re deliberately going light on nitrogen fixing trees in the southwest shelterbelt. Nitrogen encourages fast, sappy growth in trees and we want them to grow slowly until the swales are installed and planted in their lee.

New trees typically have guards around them for a couple of years. They’re not cheap at £3 to £4 each (particularly when the trees have cost around 60p each) and nobody would ever describe them as attractive, but they’re a necessary evil to keep the trees from being browsed. Deer are often the biggest problem, a few deer can wreak havoc among a young woodland in very little time. Here however we’re surrounded by agricultural land. Five miles to the south we’d need deer fencing around the whole property but here they’re rarely a problem. Another major issue is usually rabbits. They ring bark young trees and kill them. When we were planning the system I realised we’d not seen many at all, so I hit the internet. A virus came through in 2016 and wiped out rabbits in the area. They’ll return of course, but at the moment we don’t have to worry about them. The only thing that’s even a possibility here is rodent damage. So far we’ve had no damage whatsoever, and considering the cost of replacing a couple of trees versus the cost of guarding them we’re taking a calculated gamble and leaving them unguarded.

The instructions you receive from tree nurseries all recommend mulching around new trees to a metre all around the trunk. When you’re planting over 650 trees however it’s not so simple, that much mulch would cost as much as the trees. We started with cardboard as mulch. We’ve had to order a lot of things online, and that means boxes. The trouble with cardboard mulches is they need weighing down with stones, and the cardboard under the stones rots very quickly. We soon got tired of adding to mulch constantly. We recently picked up over 500 stone slates that were going cheap locally. The bulk of them are earmarked for slabs and roof repairs, but the most damaged ones are being used for mulch around the trees and shrubs. Not even ceeeping buttercup can grow through Caithness stone.


We’ve tried to design the shelterbelt system to stack functions as much as possible. The inner sections are mostly composed of bee forage species, and the outer rows too as much as possible. We will get diverse yields of fruit and nuts, as well as pole timber from some of the hazel, locust and willow. It will have huge benefits to wildlife, which will generally benefit the whole site. The design divides the site into four different climates. The veg garden will be the most protected, with buildings completely enclosing it eventually. It will be the most labour intensive system on the croft, requiring the most inputs. This is the main reason it lies directly out of the back door. Almost as sheltered will be the area between the barn and the southeast boundary. It’s bigger than the veg garden, and a little further from the house but requires less labour and fewer inputs. Next in terms of shelter is the top field, which will be the swale system with rotational grazing between. This is the biggest part of the croft and will be the main food production systems. The lower field will be the least sheltered, but is easily accessed from the house and veg garden so will hold the pig and chicken forage systems. The only moderately buffered wind should help to keep insect populations at bay, but the wind hardy shrubs should provide localised shelter.

That’s it: our shelterbelt design. Writing it has been a challenge, producing over 3000 words on shelterbelt without mentioning breaking wind just once isn’t easy as I’m sure you can appreciate.

Earthworks design.

By Andy Williams.

From the very beginning of planning this site, I knew we’d need to do some serious earth moving. The site is intermittently wet because the subsoil is clay, so when it rains heavily there’s nowhere for the water to go once the topsoil is saturated. We’ll be improving this by increasing the organic matter in the soil by planned grazing, but it can only achieve so much. We plan to plant fruit and nut trees, which struggle with consistently waterlogged roots for the most part. The solution is a permaculture classic, almost a cliche really: swales.

A swale is a ditch dug along the line of contour, with the removed topsoil forming a berm on the downhill side of the ditch. Their purpose is usually to catch overland water flow and hold it, so it can be absorbed into the ground instead of running off the property. In dry climates they’re fantastic, and allow people to grow tree varieties without irrigation that would be impossible otherwise. Of course nobody would describe the north of Scotland as a dry climate, but here they’ll serve another purpose. The berm will allow us to plant trees in soil that’s higher above the water table, giving them a better chance. We’ll be able to plant other, smaller, water loving shrubs like wax myrtle lower down the berm where the soil is wetter and other productive species like ramsons in the bottom of the swale ditch. We’ll be able to create really interesting, productive polycultures because of the different niches on different parts of the swale with differing amounts of moisture and shade. An often overlooked benefit of swales is that they effectively increase the acreage of land. If a piece of A4 paper is representative of an acre of land, imagine a sheet of corrugated card the same size. Now imagine pulling that corrugated card from each end to flatten it out. Free land! Of course swales aren’t dug at the same concentration as the ridges in corrugated card, but the principle still holds. The swales will of course infiltrate a great deal of moisture into the ground during rain, after all it’s what they’re generally used for, so even during dry summers we shouldn’t get water stressed trees and shrubs. Not bad for a glorified ditch eh?

Ponds, even small ones, have a positive effect on natural ecosystems. The same is true within diverse agricultural systems. They have a moderating effect on microclimate because of their thermal mass. They encourage natural biodiversity, for example frogs and toads, which feed on slugs. Who’s ever grown food in a damp climate without wishing there were fewer slugs? Ponds can also be used more directly for food production of meat such as ducks and geese, plants such as watercress and large ponds can be used for aquaculture of fish. Our swales will be designed to fill relatively small ponds, but they’ll still be useful. I have a niggling little plan in the back of my mind that integrates water culture with a greenhouse that should be interesting but that’s for a different blog, far in the future.

Swales, as I’ve already said, need to be dug perfectly on contour. Water shouldn’t flow along them but remain static until infiltrated. Careful levels need to be marked out before any digging begins. Earthworks permanently change the hydrology of any landscape and need to be carefully thought out. In the case of our top field, we need to establish the way the ground slopes before finalising any design. Just looking at a field is never accurate enough. We know it slopes generally to the north of course, but is that north with a bit of west, or east? The human eye just isn’t good at estimating slight changes in slope.

There are many ways of setting levels within landscape. A laser level is beautifully accurate, but is expensive. I tend to buy good quality tools but only if they’ll be used regularly, and once we’ve marked out the contours in the field we’re not likely to need one often enough to justify the cost. One cheap alternative is an ‘A’ frame level. It’s simple to make and reasonably accurate, but doesn’t work well going through undergrowth. At the moment the top field is bare, but we may well need to set some levels after the swales have been put in. For us, the best option is the bunyip or hose level.

At its most basic, it’s a length of clear hose clamped or taped to a wooden stake at each end. With the hose mostly filled with water, when the stakes are put side by side on a flat and level surface the water level in each end of the hose will be exactly the same. If that level is marked on both stakes and one of the stakes is moved up, the water will remain level but will now be below the mark on the stake. It will of course be above the level on the other stake, because that’s how water always behaves. If the hose is a long one and two people have a stake each, they can ascertain whether the stakes are at the same height even if they can’t see each other just by telling each other whether the water level in the hose is above or below the line. It’s beautifully simple, and cheap.

I built ours a little fancier. The hose must have no airlocks in it, or it’s just not accurate. Airlocks are difficult to eliminate in a thin diameter hose, so I used 25mm braided hose. The thicker diameter also eliminates a lot of drag on the inner surface of the pipe which also should increase accuracy. Instead of marks drawn on the stakes I cannibalised a pair of cheap tape measures and screwed a tape to each stake. Hose is somewhat flexible, so if it’s squashed at all it changes the internal volume. This could lead to two people trying to describe just how much above the line their water level is. With tape measures on the stakes you can use figures that actually mean something to both people, nice and accurate.


We’ve had a bit of a play with it this afternoon, and it really does seem to work as well as the literature suggests. Tomorrow we’re going to survey the slopes in the top field and work out where the swales will run and the retention ponds will logically go. After years of study and planning, this is the start of the exciting stuff. As a bonus, there’s nothing quite like dragging a pipe full of water stuck to a couple of tape measures around a field, sticking in bamboo canes and waving your arms around, for entertaining your neighbouring farmers. Who knows, I might even crack out the new video camera!