Dairy farms and solar: rooftop, ground-mount and grazing income compared

A dairy unit faces one of the more interesting solar decisions in UK agriculture. That’s because it usually has three separate options on the same holding.

First, a parlour and shed roof could carry a self-supply array. Second, spare land could host a ground-mounted system. Third, a developer may already have been in touch. If so, there may be an offer to lease that land outright for someone else’s scheme.

Very few farm types have all three options live at once. The right answer depends less on which option is “best” in the abstract. In particular, three things matter more. They are the herd’s own electricity draw, the condition of the buildings, and what the local grid will actually allow.

This piece works through that comparison for a working dairy operation. It builds on the wider picture in our South West England region guide. There, dairy and livestock farming is the dominant land use. That’s in contrast to the flat arable blocks that suit the biggest ground-mount schemes.

The parlour and shed load profile isn’t a simple daytime match

The general pitch for commercial solar is “generate when you use it,” and dairy farming only partly fits that pattern. Milking typically happens in two rounds a day, commonly early morning and mid-to-late afternoon. That schedule straddles daylight hours rather than sitting neatly inside the solar generation window.

AHDB’s guidance on dairy electricity use identifies the most energy-intensive processes on a typical unit. They are milk cooling, water heating and vacuum pumping. It’s worth separating those by how well they line up with solar output:

• Bulk tank refrigeration runs continuously, day and night, to keep milk at temperature between collections. This load matches solar generation best. It draws power right through the middle of the day, regardless of what else is happening in the parlour.
• Vacuum pumps and milk pumping run in short, heavy bursts during each milking round. That’s a good match if a round falls inside daylight hours, and a poor one if it doesn’t.
• Water heating for plant washdown is a large, somewhat schedulable load. It’s one of the easier loads to shift toward the middle of the day if you’re trying to raise self-consumption.

The practical upshot: self-consumption tends to be strong but rarely total. A well-sized array covers the refrigeration load and a good share of daytime washdown almost automatically. However, an early or late milking round still draws from the grid unless it’s paired with battery storage.

That’s why modelling matters. A proper feasibility study models that split against the farm’s actual half-hourly consumption, not a generic livestock-sector average.

Is the barn roof actually ready?

Cubicle sheds, dutch barns and grain stores often look like an obvious solar canvas. They’re usually large, south-facing or close to it, and rarely shaded on an open farmstead. Two things are worth checking before that assumption turns into a spec:

1. Roof material and age. A meaningful share of older UK farm buildings have asbestos cement roofs. This is especially true for those put up between the 1960s and the 1980s. That material falls under the site’s ongoing duty to manage asbestos-containing materials under the Control of Asbestos Regulations 2012.

   In practice, most installers won’t fix panel mounts directly into it. In particular, the routine solution is a combined re-roof-and-solar project. A competent contractor handles the existing sheeting under the regulations, and then new sheeting and panels go up together.

   That’s not a reason to rule a shed out. It is a reason to survey the roof before costing the array, since a shed that needs re-roofing anyway changes the combined project’s economics.

2. Structural capacity and system scale. MCS’s small-scale standard, MIS 3002, tops out at 50kWp. Installers must specify, install and commission anything larger outside that standard. That means a genuine structural assessment of purlins, trusses and existing roof loading — not a generic per-panel assumption.

Get both right, and a shed roof is usually the cheapest, least disruptive place to start. Get either wrong, however, and it becomes the single most common reason a promising-looking roof turns out not to be viable.

Rooftop, ground-mount and grazed land-lease, compared

	Rooftop self-supply	Ground-mount, self-supply/export	Land-lease to a developer	Agrivoltaic grazing
Who pays for the kit	The farm	The farm	The developer	Usually the developer
Capital outlay for the farm	Moderate — sized to shed roof	Higher — new mounting, cabling, fencing	None	None (farm typically leases the land)
Income mechanism	Saved electricity cost + SEG export	Saved electricity cost + SEG export	Fixed, index-linked rent	Fixed, index-linked rent
Effect on farmed land	None — roof only	Takes land out of grazing/cropping	Takes land out of farm’s own use	Land stays in grazing use, typically for sheep or youngstock rather than the milking herd
Grid dependency	Lower — sized to farm’s own demand	Higher — export capacity matters	Developer’s problem to solve	Developer’s problem to solve
Tax/ownership position	Farm owns the asset, claims capital allowances	Farm owns the asset, claims capital allowances	No asset owned; rental income; land may lose Agricultural Property Relief eligibility	Same as land-lease — ownership sits with the developer

The rooftop and ground-mount self-supply routes keep the upside — and the risk — with the farm business. Sized right, they cut the electricity bill on load the farm was always going to draw anyway.

The land-lease and grazed routes work differently. They hand the capital risk to a developer in exchange for a fixed, largely passive rent. That rent comes at a cost, however — the land’s own productive use, and often its inheritance-tax treatment.

Why grazing income usually means the flock, not the herd

Sheep grazing beneath and between panel rows is now well established on UK ground-mount sites. Sheep don’t climb or lean on the structures the way cattle can, and their grazing height suits standard panel clearances. Some schemes are trialling cattle grazing under panels, while fencing, panel height and ground-pressure requirements are more demanding.

As a result, it’s rarely the lactating milking herd that ends up on a grazed solar site in practice. A herd on a twice-daily parlour rotation needs to move reliably and often, in all weather. That doesn’t sit alongside a fixed array layout.

Where dairy farms do take a land-lease or agrivoltaic route, the grazing that continues is rarely the milking herd itself. In particular, it’s far more commonly youngstock, dry cows or a sheep flock run alongside the dairy enterprise.

That distinction matters for the business case. If the milking herd needs that land for its own grazing rotation, a land-lease becomes a genuine trade-off. It works against the farm’s own stocking capacity, not as a free income stream layered on top of it.

The grid and the land question sit together

None of this happens without headroom on the local network. Any array above the smallest domestic scale connects under Engineering Recommendation G99. That means an enquiry to the local Distribution Network Operator, then a formal application. For larger systems, a technical assessment follows before the DNO issues a connection offer.

Parts of the South West saw an earlier wave of wind and solar connections. That has used up much of the local headroom there, so some substations now carry only limited spare export capacity. As a result, it’s worth checking headroom for the specific site before a roof spec or a lease negotiation goes much further.

If a developer’s land-lease offer is also on the table, read it alongside the self-supply numbers rather than in isolation. Additionally, our land lease vs on-site solar guide sets out that trade-off in full.

It includes the point most offer letters don’t mention. Specifically, land let for solar generally stops qualifying as agricultural property for inheritance-tax relief purposes. Our glossary also covers terms like ALC grade, G99 and agrivoltaics in plain English.

Getting the sequence right

For a working dairy operation, the sensible order is usually this: check the roof first (material, structural capacity, orientation). Next, model the parlour and shed load against a real self-supply array. Then weigh any land-lease or grazing offer against what that land is actually worth to the herd’s own grazing plan.

Reversing that order causes problems. Signing a lease before costing the roof is one common mistake. So is costing the roof before checking what the grid allows.

That’s how farms end up with a decision that looks good on paper. It doesn’t hold up once all three options are compared properly. Our feasibility studies service exists to run exactly that comparison before any capital or land commitment.