BFS112m to 24m watercourse buffer strip on cultivated land

What it actually is

A grass strip between the crop and the water. It has to be at least 12 metres wide and it is paid up to 24 metres, averaged across each land parcel, running alongside a ditch, pond, river or stream. It stays in grass for the whole agreement and it does not move.

GOV.UK's stated aim is a grass buffer strip on land at risk of soil erosion or surface runoff which is alongside a watercourse or pond and has an intact sward throughout the year, without tracks, compacted areas or poaching. The page gives the purpose as protecting watercourses, reducing the risk of sediment, pesticides and nutrients being carried into them in surface runoff, providing habitat and links between habitats, and supporting an integrated pest management approach where the strip sits close to cropped areas.

There is a gate on the way in that most actions do not have. The land has to be identified by you as being at risk of soil erosion or surface runoff, and you have to keep written evidence that it is.

Source: GOV.UK, BFS1 action page, checked 05/08/2026.

What it pays, and what that comes to

The rate is in the panel at the top of this page. Multiplied out:

Source: arithmetic on the payment rate shown at the top of this page, which comes from Tilth's copy of the GOV.UK action catalogue and carries its own last-checked date. The hectarages use the action page's own measurement rule, which is length in metres times width in metres, divided by 10,000. The 1,000 metres is an illustration.

Put the other way round: at 12 metres, a hectare of strip is about 833 metres of watercourse. At 24 metres it is about 417 metres. Most farms find they have far less ditch than they assumed, so it is worth measuring the actual length off the map before working out what this is worth.

Source: arithmetic on the widths given on the GOV.UK BFS1 action page, checked 05/08/2026.

Is it worth it on your ground?

Step one: what it costs to put in. The action lets you establish the strip either by sowing a seed mix or by natural regeneration, so the seed cost is a choice rather than a given.

Source: ProGreen, own listed price and stated sowing rate for a 4/6 metre grass and buffer strip mix, looked up 31/07/2026. Price your own mix. Cultivation, drilling and any early cutting to encourage the grass to tiller are your own costs and are not in this. Natural regeneration, which the GOV.UK action page also permits, carries no seed cost at all.

Step two: what the ground would otherwise return. Gross margins per hectare for England, non-organic, in the 2023/24 crop year. Variable costs have already been taken off these figures, so there is nothing further to deduct.

Source: Farm Business Survey, Crop Production in England 2023/24, Rural Business Research, England, non-organic, 2023/24 column, checked 02/08/2026. The winter wheat figures come from a sample of 572 farms.

The agreement runs three years, so the comparison has to run three years too, and that ground will not be in wheat for all three of them. Three rotations, worked from the crop figures above:

Source: arithmetic on the crop gross margins in step two, from the Farm Business Survey, Crop Production in England 2023/24, checked 02/08/2026. These three are illustrations. Work your own out from the same crop figures.

Step three: the gap.

Source: the payment rate at the top of this page, less the £38 a hectare a year of seed from step one, set against the rotation averages above. The figures are subtraction. Cultivation and drilling are not in them. If you establish by natural regeneration there is no seed to take off, so each gap above narrows by £38.

So on average arable ground this is short, by somewhere around £95 to £167 a hectare a year if you sow, and £57 to £129 if you let it come back on its own. That is a narrower gap than it looks, because the ground this action goes on is not average ground.

Step four, which is the one that matters here: the gap is not the same at 12 metres as it is at 24.

What follows is Tilth's own reading rather than scheme guidance, and it adds no requirement beyond the GOV.UK action page.

The payment is a flat rate a hectare whatever width you go to. The ground is not flat at all. The twelve metres nearest a ditch tends to be the part of the field that already earns least: shaded on one side, turned on and compacted, the last to travel in a wet spring, and the strip an operator is most likely to miss or double. The next twelve metres, the ones you take on to get from the minimum width to the paid maximum, is ordinary field.

So widening the strip doubles the payment and doubles the area, but the second half of that area is worth more as crop than the first half was. The rotation figures above are field averages, which makes them roughly fair for the outer part of a 24 metre strip and probably generous for a 12 metre one.

That is the actual decision on this page. It is not really whether to do BFS1. It is how wide, and the honest answer to that is different on a ditch that runs along a wet shaded northern boundary from a ditch that runs through the middle of your best ground.

Two things to set against the crop figures above. Also Tilth's own reading.

The survey reports one winter wheat figure, not first and second wheat separately. A rotation with two wheats in it carries a second wheat at the same gross margin as the first, which is unlikely to be how it turns out on your ground. We have not found a published split to cite, so we have not applied a discount of our own invention. Read the rotation averages as the generous end for the crop.

It is one year, and not a strong one. The same report's 2022/23 column put winter wheat's gross margin at £1,570 a hectare against the £988 above for 2023/24. A three-year decision resting on a single crop year is a shaky thing, and this particular year was the lower of the two. Against that stronger year the gaps above widen and the action looks worse.

Where that leaves it. The money is close enough that national averages cannot settle it. On ground that crops like the field average, at the full 24 metres, this does not pay. On a narrow strip beside a ditch that has been costing you money for years, on land you have already judged to be at risk of runoff, it may well. The number that decides it is what that particular field edge yields, and that number is on your farm rather than in a national survey.

Does it work, and what makes it work better

This section is what we have found in published sources about the practice this action pays for. It is not scheme guidance, it is not Defra's or the RPA's view, and it adds no requirement beyond what the GOV.UK action page sets out. We have named every source and given the date we checked it. What it means for your ground is your decision.

None of the work below is about BFS1 by name. It is about planting grass buffer strips and margins around arable and pasture fields, and this action is one way of paying for that practice. Two things about the fit are worth saying before the findings rather than after them.

The first is width and position. Conservation Evidence describe the practice as planting field margins with a grass-rich seed mixture, not fertilised and spot-treated with herbicide only if necessary, which is close to what BFS1 asks for. The widths in the studies below are not. They run from under 1 metre to 56 metres, and the 6 metre Countryside Stewardship margin recurs throughout, which is half the width at which this action starts. Only two of the studies were set up beside water at all, one in Finland and one in the Netherlands, and neither of those measured what the water was carrying.

The second is the subject. GOV.UK gives BFS1's purpose as protecting watercourses and reducing the risk of sediment, pesticides and nutrients being carried into them in surface runoff, alongside providing habitat and links between habitats. Between them the two evidence syntheses below cover 84 studies, and not one of them measures sediment, nutrients or pesticides reaching water. They speak to the habitat half of that purpose and not at all to the water half. If what you want to know is whether a 12 metre strip keeps soil out of your ditch, this section does not answer it, and we have not found a source we would put on a page that does.

The graded assessment. Conservation Evidence, based at the Department of Zoology at the University of Cambridge, summarises the published studies behind individual conservation actions and has independent subject experts score each one. This practice is assessed twice, in two separate synopses, and the two assessments do not land in the same place.

In the Farmland Conservation synopsis it comes out in the top category, Beneficial, on 69 studies, with expert scores of 90 per cent for effectiveness and 65 per cent for certainty. Harms were not assessed. In the Bird Conservation synopsis the same practice, assessed for birds alone, comes out a category lower, Likely to be beneficial, on 15 studies, with 47 per cent for effectiveness, 54 per cent for certainty and 0 per cent for harms.

Effectiveness and certainty are not the same thing. Effectiveness scores only the size and direction of the effects that were reported. Certainty is the separate score for how many studies there are, how well built they are and how much ground they cover. The gap between the two assessments is mostly a gap in subject rather than a disagreement. The farmland synopsis counts beetles, spiders, bees, butterflies, plants and small mammals as well as birds, and it is on those groups that the large effects sit. Narrow it to birds and the picture gets more modest.

Source: Conservation Evidence, action 246, Plant grass buffer strips/margins around arable or pasture fields, Farmland Conservation synopsis, and action 191, the same practice for birds, Bird Conservation synopsis, both checked 06/08/2026. Cite as Dicks et al. (2020), Farmland Conservation, pages 283 to 321, and Williams et al. (2020), Bird Conservation, pages 137 to 281, both in What Works in Conservation 2020, Open Book Publishers.

What the studies found, both ways. The farmland synopsis groups its 69 studies by what was measured. Nineteen studies from Finland, the Netherlands, Sweden and the UK, seven of them replicated and controlled, found grass buffer strips increased the abundance, species richness and diversity of insects and spiders. Nine UK studies found benefits to birds, in numbers, densities, species richness and foraging time. Seven studies from the Netherlands and the UK found increased plant cover and plant species richness. Five studies from Finland and the UK found more small mammals and more small mammal activity. And six studies from the Netherlands and the UK, three of them replicated controlled trials, found no clear effect on insect numbers, bird numbers or invertebrate pest populations. A separate site comparison found sown grassy margins were not the best option for conserving rare arable plants.

That last group matters as much as the first four. A 90 per cent effectiveness score sits alongside six studies that found nothing, not instead of them.

Source: Conservation Evidence, action 246, key messages, checked 06/08/2026.

Assessed for birds on their own, the split runs like this. Nine studies from the UK and the USA found more positive population trends, higher populations or strong habitat associations for some or all species where there were grass margins. Three replicated UK studies found no positive effect on some or all of the species they looked at. Both of the studies that looked at habitat use found birds used margins more than other habitats. One replicated UK study found grey partridge had smaller broods in grass margins than in other habitat types.

Source: Conservation Evidence, action 191, key messages, checked 06/08/2026.

Some of the individual numbers, so the size of the effects is visible rather than described. On birds, Buckingham and others across 23 pastoral farms in the West Midlands recorded 16 times the winter density of seed-eating birds within 6 metres of boundaries on fields with grass margins, 1.1 birds a hectare against 0.1, and at the same time twice as many blackbirds near the boundaries of fields without margins, 1.8 against 0.9, with no statistical analysis performed. Davey and others, across 97 one kilometre squares in East Anglia in February 2008, found 19 of 24 farmland bird species responded positively to scheme margins, but only yellowhammer and possibly blackcap responded strongly. Against those, Marshall and others found no more birds and no more species on 42 southern England fields with 6 metre margins than without, 11 to 18 species a site against 11 to 15, and Kleijn and others found 6 metre margins had no effect on the number of birds or bird species breeding or foraging across seven pairs of fields in three parts of the UK.

On the other groups, Marshall and others found 5.2 grasshoppers and crickets of 1.8 species on fields with 6 metre margins against 0.9 individuals of 0.6 species on fields without. Pywell and others, across 32 ten kilometre squares in England, counted 6 to 8 foraging bumblebees of 1.3 to 1.4 species on a transect of sown grass margin against 0.2 bumblebees of 0.1 species on cropped margins. Smith and others in Cambridgeshire recorded 74 woodlice a square metre in sown margins and none in the winter wheat, with beetles at 80 against 41 and earthworms at 281 against 244. Shore and others in North Yorkshire trapped 26 to 38 bank voles an autumn period on 6 metre margins and 13 to 14 on 3 metre margins, against 0 to 1 on cropped controls. Huusela-Veistola in Finland caught roughly 1,200 to 3,000 spiders in 12 metre grass and clover strips against 400 to 900 in the crop.

And the studies that found nothing. Thomas and Marshall found no more invertebrates, no more overwintering invertebrates in the soil and no more ground beetles in 4 metre rye grass margins than in cropped controls, at 110 to 130 invertebrates a sample either way. Telfer and others found no significant difference in ground beetle numbers between 3 metre margins, 6 metre margins and cropped field edges. Carvell and others found 6 metre plots sown with a tussocky grass mix supported no more bumblebees than conventionally cropped margins. Van Alebeek and others, across 15 kilometres of sown margins in a 400 hectare block in the Netherlands, found no clear effect on pest populations.

Source: Conservation Evidence, actions 246 and 191, supporting evidence from individual studies, checked 06/08/2026.

Three things about those lists are worth knowing before you weigh them. Conservation Evidence say themselves that studies are not directly comparable or of equal value, and that study size, design, reported metrics and relevance to your own situation matter more than "simply counting the number of studies that support a particular interpretation". Several of the studies test more than one thing at once, so the cirl bunting recovery reported by Aebischer and others came from a package of overwinter stubbles, grass margins, managed hedges and set-aside rather than from margins on their own. And the individual studies run from 1994 to 2011, so every one of them predates this scheme and this action.

Source: Conservation Evidence, actions 246 and 191, checked 06/08/2026, including the note under key messages and the descriptions given in the individual study summaries.

What we have not found. Nothing measuring BFS1 itself. Nothing measuring sediment, nutrient or pesticide interception, which is the action's first stated purpose, in either synopsis, across all 84 studies. Nothing comparing 12 metres against 24 metres on the same ground. Two studies were carried out on riparian margins, Yletyinen and Norrdahl in Finland and Musters and others in the Netherlands, and both looked at what lived in the strip rather than at what the strip stopped.

What the sources say makes it work better. All of the following comes from the studies listed on those two Conservation Evidence pages, and it is a description of what was found rather than advice from us.

The clearest finding is about management rather than seed. Henderson and others ran a randomised replicated controlled trial of sown margins in eastern England from 2002 to 2006, setting three seed mixes against three spring treatments: cut to 15 cm, soil disturbed by scarification until 60 per cent of the area was bare ground, and treated with a grass-killing herbicide at half the recommended rate. They concluded that management affects bird use more than the seed mix does. In summer, bird densities were higher on the disturbed and herbicide-treated plots than on the cut ones, and bird densities tracked the densities of ground beetles. In winter it went the other way, with twice as many birds on cut margins as on uncut ones across ten farms.

Those same three treatments were tested on other groups, in that trial and in four more. Ogilvy and others found more birds on scarified than cut margins in July 2004, with greater plant diversity and more unsown plant species on the scarified plots, while cutting maintained plant diversity in the mixes that contained wildflowers. Woodcock and others found scarification raised the species richness of predatory beetles. Blake and others, on 6 metre strips created in 2004 on two arable farms, found the highest butterfly species richness where scarification and the grass-specific herbicide were used together. Pulling the other way, Smith and others found scarified plots had fewer woodlice, fewer species of woodlice and lower densities of predatory species than cut or herbicide-treated plots, and Ramsay and others found fewer planthoppers on scarified plots. Scarification is not uniformly good news across every group that was counted.

On weeds, three studies point the same way. Asteraki and others, on 38 plots around four Countryside Stewardship fields in Wiltshire, found sown plots had significantly lower cover of nettle, creeping thistle and blackgrass than plots left to regenerate naturally. Kleijn and others in the Netherlands measured creeping thistle at 8 g a square metre and couch at 9 in rye grass margins, against 33 and 28 in naturally regenerated ones. Moonen and Marshall found hedges with an adjacent sown strip carried fewer pernicious weeds than hedges without.

The trade-off is plants. Sown grass margins carry fewer plant species than the same ground left to come back on its own. Field and others recorded 20 plant species in sown 6 metre margins in 2003 against 35 in naturally regenerated 6 metre margins on the same farms, with richness declining in the sown margins from 1998 as the sown grasses took over. Critchley and others found sown grass margins averaged 4.4 plant species against 5.5 for naturally regenerated margins and 6.3 for uncropped cultivated margins, the lowest of the ten habitats they compared, and reported that sown grassy margins more than two years old ran to 87 to 95 per cent grass cover.

On width the evidence is real but mixed. Smith found meadow grasshopper density significantly higher in 6 metre margins than in 2 metre margins. Shore and others found more bank voles on 6 metre margins than on 3 metre ones. Merckx and others found more moths in fields with 6 metre margins than in fields with margins of about 1 metre, then in a second study found no such difference. Askew and others found the most small mammals in 2 metre margins, then 6 metre margins, then set-aside. Holland and others, comparing standard margins under 1 metre against wide ones of 5 to 6 metres, found more spiders in the wide-margin fields and a higher proportion of aphids parasitised. Conover and others in Mississippi, in the bird synopsis and the only study we have found that varied width across the range this action works in, recorded 35 birds a hectare on borders of 20 to 56 metres, 18 on borders of 6 to 12 metres, and 3 on fields with no border.

On seed mix, Woodcock and others found the plain grass Countryside Stewardship mix carried more beetles, and more beetle species, than a fine grass and wildflower mix, and about as many as a tussock grass and wildflower mix. Pywell and others found the reverse for pollinators, with 35 to 47 bumblebees on the wildflower mixes against 10 on grass alone. Ramsay and others found grass-only margins carried more planthoppers than the wildflower mixes. Field and others, surveying 100 farms across three English regions, found hedges alongside floristically enhanced grass margins held an estimated 0.4 yellowhammers a metre against 0.2 for hedges with no margin, while hedges alongside unenhanced margins showed no such difference. Which mix comes out best depends on what is being counted.

Source: Conservation Evidence, actions 246 and 191, supporting evidence from individual studies, checked 06/08/2026.

What follows is Tilth's own reading of how those findings sit against this action, and it adds no requirement beyond what the GOV.UK action page sets out.

Three of those findings land on decisions this page has already put in front of you.

The first is the seed decision in step one. Establishing by natural regeneration saves the seed cost, and the money section says so. The evidence now cuts both ways on that saving. Henderson's conclusion, that management affects bird use more than the seed mix does, points away from the mix being where the difference is made. But Asteraki, Kleijn and Moonen all found sown strips carried fewer pernicious weeds than ground left to itself, and blackgrass was named in one of them. Against that, Field and Critchley both found sown grass margins ended up with fewer plant species than naturally regenerated ones. If your worry about a 12 metre strip along a ditch is what colonises it, that is the trade you are making, and it is a trade rather than a right answer.

The second is width, which is where the money section said the real decision sits. The findings do not settle it. Two studies found more of something on the wider strip, one found the opposite, one found a difference and then failed to find it again. The only study that reached the widths this action works in was carried out in Mississippi with US species, and it found more birds on the wider borders. So the evidence leans towards wider being better for wildlife while the money gets worse as the strip gets wider, and neither side of that is strong enough to make the decision for you.

The third is the cut. The action requires you to cut part of the strip's width next to the crop in late summer, after the bird breeding season. Henderson found more birds on cut margins in winter and fewer on cut plots in summer. Askew found more small mammals in 6 metre margins cut every two or three years than in those cut every year, though only significantly so for common shrews in one of the two years. Gardiner and Hill watched grasshopper density in two Essex margins fall from 0.4 and 0.1 a square metre to 0.05 over seven years and put it down to the annual cut for hay. The requirement and the findings sit in roughly the same direction on birds, but the studies cut whole plots and the action cuts part of a width, so they are not measuring the same thing.

One more word on the scores. Harms were not assessed at all in the farmland synopsis, and scored 0 per cent in the bird one, while a study in that same bird synopsis reported smaller grey partridge broods where more of a site was under planted grass strips. A harms score is an expert assessment across the target group as a whole, not a promise that no study reported anything negative. If you are managing that ground for partridge in particular, that study is worth reading rather than taking the score on its own.

What you actually have to do

Establish and maintain a grass strip that is at least 12 metres wide and up to 24 metres wide, on average in each land parcel, alongside a ditch, pond, river or stream, to create a buffer between the edge of the crop or fallow land and the watercourse. It can be more than 24 metres wide, but you will only be paid for the 24 metre width.

You must make sure the grass strip is in addition to any regulatory requirements relating to buffer strips.

Establish the strip within the first 12 months of the action's duration and maintain it through each subsequent year. Once it is established, manage it in a way that can reasonably be expected to achieve the action's aim.

During late summer, after the bird breeding season, cut part of the strip's width next to the edge of the crop or fallow land, along as much of the strip's length as possible.

If the strip is in a flood risk area and a flood risk management authority such as an Internal Drainage Board carries out flood risk management works on it, you must repair any damage as soon as reasonably practicable.

Keep written evidence that the land entered into this action is at risk of soil erosion or surface runoff, and keep evidence of what you have done to complete the action, such as field operations at a land parcel level and associated invoices, photographs and other documentation.

Source: GOV.UK, BFS1 action page, checked 05/08/2026.

What you cannot do on it

On the established grass strip: no livestock access, no regular vehicular access, turning or storage such as storing bales or machinery, no fertilisers or manures, and no pesticides except herbicides used to weed wipe or spot treat injurious weeds, invasive non-native species, nettles or bracken.

Source: GOV.UK, BFS1 action page, checked 05/08/2026.

Where it can go, and any cap

Agricultural land below the moorland line, identified by you as being at risk of soil erosion or surface runoff, alongside a ditch, pond, river or stream. The eligible land types are arable land used to grow crops and temporary grassland and arable land lying fallow, all with an arable land cover. It can also go on eligible land in a flood risk area.

The action is static, so it stays in the same place each year of the agreement.

You can maintain an existing grass strip and be paid for it, provided it meets the action's requirements and is not already being paid for under another environmental land management scheme option.

Source: GOV.UK, BFS1 action page, checked 05/08/2026.

BFS1 is not one of the limited area actions under SFI26, so the 25 per cent of total agricultural area rule that applies to actions like CAHL2 and AHW9 does not apply to it. In practice the ceiling on this action is the length of watercourse you have.

Source: GOV.UK, SFI26 scheme rules and guidance, section 4.5, checked 05/08/2026.

Things worth watching

This section is Tilth's own reading of where this action is easy to get wrong. It is not scheme guidance, it is not sourced from Defra or the RPA, and none of it adds a requirement beyond what the GOV.UK action page sets out.

Find out what you are already obliged to leave before you measure anything. The action page says in plain terms that the grass strip must be in addition to any regulatory requirements relating to buffer strips. Whatever width you are already required to keep clear beside that watercourse is not what this pays for, and the paid strip sits on top of it. That changes the arithmetic at the front of the job rather than at the end of it, so it is worth settling before you decide a width.

The 12 metre minimum is a real threshold, not a starting suggestion. Other buffer strip actions begin at four metres. This one does not exist below twelve. If your ditch sides will not carry twelve metres on average across the parcel, this is not the action for that ground.

The width is an average across the land parcel. That helps where a boundary wanders, and it also means a stretch that narrows is only offset by a wider stretch somewhere else in the same parcel. Measure the whole parcel, not the best bit and not the worst.

At 24 metres you have given away the headland, not the margin. The prohibition on regular vehicular access, turning and storage applies across the whole strip. A four metre margin and a twenty-four metre one are very different propositions when the sprayer needs somewhere to turn. Work out where the machinery goes before you commit a width, because the answer is much harder to change afterwards.

It is static. There is no moving it for year two if the cropping plan changes. Whatever you pick sits there for the length of the agreement.

The flood risk clause has no number attached to it. If your ditch is under a drainage board and they come and do works on the strip, the repair is yours and it has to happen as soon as reasonably practicable. On a lot of low-lying arable ground that is not a remote possibility. There is no way to put a cost on it in advance, which is exactly why it is worth knowing about in advance.

The risk judgement is yours to make and yours to keep. The land has to be identified by you as at risk of soil erosion or surface runoff, and you have to hold written evidence of that. It is made before the agreement starts and it has to still be findable at the end of it.

Natural regeneration is cheaper on paper and slower in practice. It costs nothing in seed, but the strip still has to be established inside twelve months and still has to hold an intact sward with no tracks or bare patches. On clean, well-structured ground that may come right on its own. On a compacted ditch side it may not. The saving is £38 a hectare a year, which is not much against the cost of getting it wrong.

Keeping the evidence

Also Tilth's own reading. GOV.UK sets the requirements; what follows is a practical suggestion about record keeping.

This action asks for two different kinds of record, and they are made at different times.

The first is the evidence that the land was at risk of soil erosion or surface runoff. That is a judgement made before you apply, and there is no form for it. Whatever you looked at when you made it, the wet corners, the silt at the ditch bottom, the photographs from a bad autumn, the slope, is worth writing down at the time with a date on it. Three years later nobody remembers why a particular field edge was chosen.

The second is the ordinary record of doing the work. GOV.UK's own examples are field operations at land parcel level with their associated invoices, and photographs or other documentation. For this action that means the establishment method and date, the seed invoice if you sowed, the date of the late summer cut each year, and photographs of the strip through the year showing an intact sward.

The measurement is the third thing to get right, because the payment comes straight out of length times width, and the width is an average.

Tilth measures the strip off the field boundary you have already drawn, holds the establishment and cutting records against it, and keeps the risk note and its photographs with the same parcel rather than in a different folder.

The part only your own ground can answer

This page has argued itself down to one question, and it is a question about width. Twelve metres of ditch side and twenty-four metres of ditch side pay the same per hectare and cost you very different amounts of crop, and nothing published anywhere will tell you what those particular metres return, because national gross margins are field averages and this is a decision about field edges. If your fields, your yield maps and your NDVI history are in one place, you can look at that edge and answer it for yourself. Request an invite.