Agrivoltaics over orchards from an investor's perspective: support, cost per hectare and what the trial plants show

Apple orchards are expensive to build. Many stand under hail nets, some under foil against rain, and frost is fought with sprinkler irrigation. Orchard PV starts exactly there: instead of a net or foil, the frame above the tree rows carries solar modules that let light through, keep out rain and hail and produce electricity on the side. Research plants have been running in Germany since 2021, and larger plants are now being built on commercial fruit farms.

For investors, orchard PV has so far been a niche. It turns up more and more often in conversations and exposés, usually with a reference to dual land use and to its support as a special solar installation. Both are true. Whether that makes a good site depends on other questions: how large the surcharge really is, what the tall frame costs, who receives the saved net costs and what happens when the fruit farm converts its orchard after some years. We do not currently offer an orchard PV project ourselves; this article is an assessment of the concept.

What is agrivoltaics over orchards, and what does the solar roof take over?

Agrivoltaics over orchards is a high-mounted solar plant whose modules form a roof above the tree rows. Fruit growing continues underneath with the same machinery as before. Most plants use semi-transparent modules that let through around 40 to 50 per cent of the light, depending on the design, on fixed frames or single-axis trackers. Fraunhofer ISE states that 60 to 70 per cent of the available light is enough for full yields in some apple varieties, so the design is always a compromise between electricity and fruit.

The roof is meant to take over three jobs that are solved separately today: hail protection, otherwise provided by a net; rain protection, since rain favours fungal diseases such as scab and otherwise needs foil; and protection from strong sun, which burns the fruit. On top comes the electricity, which the farm can use itself or feed into the grid. Technically, orchard PV belongs to category I of DIN SPEC 91434, the plants with farming underneath the modules. How agrivoltaics works in general and which designs exist is covered in Investing in agrivoltaics: dual land use, revenues and tax leverage at a glance.

The replacement argument does not hold everywhere to the same degree. According to a survey of fruit-growing advisers evaluated by the research service of the German Bundestag, around 15 per cent of Germany's apple area was under hail nets in 2015. In Baden-Württemberg it was 39 per cent, in Lower Saxony, the second-largest growing region including the Altes Land near Hamburg, only half a per cent. Where there is no net today, the solar roof does not save one.

Where does orchard PV sit in the EEG, and what support is there today?

Orchard PV is a special solar installation under § 37 (1) no. 3 (b) EEG: a plant on land that is not peat soil, with permanent or multi-year crops grown on the same land at the same time. Besides fruit this includes vines, berries, hops and tree nurseries. Under the Bundesnetzagentur's determination (Festlegung), permanent crops are crops outside the crop rotation that stay on the land for at least five years and deliver recurring harvests. Plants above one megawatt bid in the ordinary tender for first-segment solar plants; smaller ones receive the statutory value under § 48 (1) sentence 1 no. 5 (b) EEG.

The advantage in the tender is a surcharge of 0.5 cents per kilowatt hour on the award value if the plant is mounted with a clearance of at least 2.10 metres. It comes from the version of the EEG in force in May 2024 and applies to awards in 2026 to 2028. That this older version applies is due to § 101 (1) EEG: the more generous rules of the 2024 solar package, namely a separate quota for special solar installations, a cap of up to 9.5 cents and a bonus for plants outside the tender, may only be applied once the European Commission has approved them under state-aid law. According to the Bundesnetzagentur's notes on the 1 December 2026 bid date, that approval is still outstanding. Small plants outside the tender therefore receive no surcharge today.

The surcharge comes with obligations. The Bundesnetzagentur's determination of 1 October 2021, which continues to apply to permanent crops under § 85c (2) EEG, requires the plant and the fruit growing to meet the state of the art over the entire support period and refers to DIN SPEC 91434 for that. The operator submits an expert report to the grid operator at commissioning and then, in every third year, an expert confirmation that cultivation has continued. Aerial images, photographs or extracts from the fruit farm's field records (Schlagkartei) are sufficient. If the evidence is missing, the reference value drops by 2.5 cents per kilowatt hour (§ 54 (3) sentence 4 EEG) until it is supplied. That is five times the surcharge.

What does the EEG 2027 draft change for orchard PV?

The rate stays, the hope for more goes, and the penalty shrinks. The government draft of 7 September 2026 (Bundestag printed paper 21/7867) deletes the never-applied quota for special solar installations and the 9.5-cent cap. Instead, a new § 38a provides a fixed surcharge of 0.5 cents per kilowatt hour for agrivoltaics on arable land, permanent crops and grassland if the plant has at least 2.10 metres of clearance, or 0.80 metres for vertical plants. The same surcharge is to apply to smaller plants outside the tender (new § 48 (1b)). If the cultivation evidence is missing, the value would drop by only 0.5 instead of 2.5 cents under the draft.

In its opinion, the Bundesrat asked for 1.5 cents and a separate segment; the federal government rejected this in its reply (printed paper 21/8298). At the date of this article, the Bundestag has held the first reading and a committee hearing; the final vote is still pending, and the law is meant to enter into force on 1 January 2027. What else EEG 2027 changes for solar parks is described in EEG 2027: what the refinancing contribution means for a solar park direct investment.

How much agrivoltaics over permanent crops is actually being built?

Little, but more recently. The Bundesnetzagentur's tender statistics record the site type of every bid, including a separate column for special solar installations on permanent crops. We evaluated it for every bid date in which this type appears. From December 2023 to July 2026, 759 megawatts were bid on permanent crops and 214 megawatts awarded. That is 1.2 per cent of all awards in those rounds, at an award rate of 28 per cent. More than three quarters of the volume falls on the December 2025 and March 2026 rounds.

For comparison: agrivoltaics on arable land reached 978 megawatts of awards across all rounds, on grassland 137. How much of the permanent-crop volume stands over fruit cannot be read from the statistics. At around 600 kilowatts per hectare, the entire awarded permanent-crop capacity would correspond to roughly 360 hectares. That is just over one per cent of Germany's apple area of about 33,100 hectares (Destatis, tree fruit survey 2022), and only part of it is over fruit. Which land the other solar parks use is shown in Eligible land: where can a solar park be built in Germany?.

What do the trial plants show about harvest and fruit quality?

Less crop protection and less sunburn are consistent results, and so is a lower yield in some varieties. The most robust figures in Germany come from the Lake Constance fruit-growing competence centre (KOB) in Bavendorf, which compares four apple varieties under semi-transparent modules with 49 per cent light transmission and 3.5 metres clearance against a control under grey hail net. After two harvests, the yield under the modules was 63 to 95 per cent of the control: 95 per cent for Delcored, 78 per cent for Topaz on the fixed system and 89 per cent on the tracker, 63 per cent for Freya on the fixed system. Harvest came one to two weeks later, and the fruit was smaller and less coloured. Scab and sooty blotch occurred less often.

In Kressbronn on Lake Constance, a plant on a commercial farm with 239 kilowatts on 0.4 hectares, Fraunhofer ISE reported after two years 70 per cent less plant protection product, 50 per cent less irrigation and more than 20 per cent more electricity than simulated. In Bavendorf, by contrast, the trees under the modules needed more water than under the net. On water, the two sites contradict each other; conclusions should wait until more years are available.

Results from neighbouring countries point the same way. KU Leuven measured 24 per cent less light and 14 per cent less yield over three years for Conference pears, with smaller fruit. In southern France, Golden Delicious apples under modules were 17 per cent smaller in the first year and no longer smaller in the two following years. And a trial plant in Styria, Austria, did not keep the hail out in 2023 and 2024: 70 modules were damaged in 2023, and the 2024 harvest was almost completely lost. A solar roof is therefore not automatically a tight hail net; it depends on module glass, gaps and side protection.

What does one hectare of orchard PV cost, and what does it earn compared with a ground-mounted site?

One hectare of orchard PV carries less capacity and costs more per kilowatt. Fraunhofer ISE gives 500 to 800 kilowatts per hectare for high-mounted agrivoltaics, and the trial plants in Baden-Württemberg range from about 450 to 820. We calculate with 600 kilowatts per hectare against 1,000 on an ordinary ground-mounted site. For the substructure of high-mounted plants below four metres, ISE gives 243 to 306 euros per kilowatt, against 76 euros for a ground-mounted plant; that is 167 to 230 euros of extra cost for the frame alone. The Rhineland-Palatinate energy agency puts agrivoltaics over fruit and vines at four metres at 1,000 to 1,200 euros per kilowatt in total, against 450 to 600 euros for a ground-mounted plant. We therefore calculate three cases: 167, 230 and 550 euros of extra cost per kilowatt.

On the revenue side, we combined hourly profiles for the three largest apple regions with 2025 exchange prices. A fixed, south-tilted solar roof delivers around 950 kilowatt hours per kilowatt in the Altes Land, around 1,030 in the Rhineland and around 1,100 at Lake Constance; the trial plant in Gelsdorf in the Rhineland reached about 1,070 in its first year. Around a quarter of that fell into hours with negative prices in 2025, in which a plant in direct marketing receives nothing under § 51 EEG. Why that is so is explained in Negative electricity prices: what they mean for solar and storage investors.

Per hectare, orchard PV earns around two thirds of the electricity revenue of a ground-mounted site in this calculation, while keeping the fruit harvest. For the investor, what counts is the calculation per kilowatt. Converting the extra cost into an annuity at 5 per cent over 25 years and spreading it over every kilowatt hour produced gives 1.2 cents in the Rhineland (frame only, lower range), 1.6 cents (frame only, upper range) and 3.8 cents (whole system). The surcharge is 0.5 cents, and it is only paid on the kilowatt hours fed in outside negative-price hours.

Against this stands the saving on hail protection. KTBL, the German board for agricultural technology, put a gable-type hail net over apples at about 32,000 euros per hectare at 2021 prices, with a useful life of 20 years for the frame and 10 years for the net; including opening, closing and repairs this gives about 3,460 euros per hectare and year before interest. A simpler single-row net costs about 9,200 euros per hectare and 840 to 1,080 euros a year according to a trial report from Saxony; we use 1,000 euros for it.

Without discounting, simply adding up 25 annual values, the surcharge and the saved gable net even cover the frame cost in full. Discounted, a gap remains in every case, and a large one for the whole system. Values for better fruit quality, fewer sprayings or saved foil are not included, nor is a possible lower fruit yield. Both accrue to the fruit farm.

Who keeps the saving on the hail net?

The fruit farm, unless the contract says otherwise. Net, foil, sprays and part of the irrigation are costs of fruit growing. The frame and the modules are costs of the plant operator. If the fruit grower builds the plant himself or herself, both sides sit in one hand, and the calculation above applies as a whole. If an investor buys a park on someone else's orchard, the investor carries the extra cost while the saving stays with the landowner.

The benefit can be passed back in three ways: through a lower lease than on an ordinary ground-mounted site, through a construction contribution from the fruit farm towards the frame, or through an ongoing usage fee for the weather protection. Which route was chosen is in the lease or usage agreement, not in the exposé. What a solar park lease should regulate in general is covered in The land lease of a solar park from the investor's side: why it is a Mietvertrag, what the rent costs per kilowatt and how you step into the contract.

For orchard PV this includes some points an ordinary ground-mounted site does not have: who cleans the modules when spray settles on them; who is liable when a module breaks and spoils the harvest underneath; what driving clearance and row spacing the farm needs for its machinery; and who pays if hail comes through the roof. The plant's property insurance covers modules, not apples; more on that in Insuring a solar park or battery storage asset: which policies belong to a direct investment, and who holds them.

What happens when the apple orchard is grubbed up or converted?

Support continues if the new use again meets the requirements. The Bundesnetzagentur's determination expressly allows a change of crop, including between permanent crops and arable farming. The timeline is the harder part: a solar plant is designed for 25 to 30 years, while an apple orchard is grubbed up and replanted after a number of years, often with other varieties. Young trees bear little in their first years. Whether a young orchard without a meaningful harvest counts as continued cultivation in the three-yearly evidence is not expressly settled by either the statute or the determination.

For the investor this raises two questions: how old is the orchard at construction, and what does the contract provide for grubbing up? The final report of an Austrian trial plant also advises against building over an existing orchard because construction machinery compacts the soil. Building frame and orchard together is the recommended sequence there. And because a new variety needs different light, a conversion can change the yield under the modules even though the plant stays the same.

What applies to tax and farm subsidies?

The solar plant follows the same rules as any other park: it is a commercial business, and the investment deduction (IAB) and special depreciation under § 7g EStG are available for the movable parts if the conditions are met; they are set out in Investitionsabzugsbetrag: all §7g EStG requirements, and who can use it. Whether a tall frame anchored in the ground over an orchard is to be treated differently for tax purposes from the racking of a ground-mounted plant has not been expressly decided by the tax authorities or the courts; it belongs in the purchase price allocation of the specific project.

For the fruit farm, farm subsidies matter. Under § 12 (4) no. 6 of the German CAP direct payments ordinance (GAPDZV), land with solar plants is in principle not eligible unless the farm proves that the plant is agrivoltaics. For that, the land must still be workable with usual machinery, and the usable area may shrink by no more than 15 per cent under DIN SPEC 91434 (§ 12 (5)). Under planning law, an agrivoltaics plant at the farm is privileged in the outer zone up to 25,000 square metres of ground area, one plant per farmstead (§ 35 (1) no. 9 BauGB); larger ones need a development plan.

Which questions should investors ask about an orchard PV project?

• Is the plant registered in the award as a special solar installation under § 37 (1) no. 3 (b) EEG, and is the 2.10-metre clearance confirmed in the commissioning report?
• Which agrivoltaics surcharge does the model assume? Anything above 0.5 cents rests on a rule that has never been applied.
• Who prepares the three-yearly evidence, who pays for it, and what obligation does the fruit farm have to supply field records and photographs?
• Which varieties grow under the modules, how much light do the modules let through, and are there yield data for this combination?
• How does the saving on hail net and foil flow back to the operator: via the lease, a construction contribution or a usage fee?
• How old is the orchard, and what does the contract provide for grubbing up, replanting and changing variety?
• Who is liable for crop damage from broken modules, and who cleans the modules after spraying?
• Does the yield assessment use the semi-transparent modules actually installed and their output per module, or standard modules?
• Which extra frame costs are disclosed, and what share is covered by state funding?

How a yield assessment is structured and how to recognise its assumptions is described in Reading a yield assessment: what P50 and P90 mean for a solar park direct investment. General checks for any provider are listed in How to tell a trustworthy provider of energy direct investments.

How we assess new concepts such as orchard PV

We do not currently offer orchard PV. If such a project came to us, we would treat it like any special solar installation: we model the surcharge as a separate revenue line with its own risk, with one case with and one without the surcharge, and set the difference against the disclosed extra cost. We read the contract with the fruit farm as carefully as the grid connection, because the evidence, the return of the saving and the grubbing-up scenario all depend on it. We do not accept yield assessments that use standard modules instead of the semi-transparent modules actually installed.

We follow new designs even before we offer them. For peatland and floating PV we have done this in Peatland PV and floating PV as a direct investment: support, EEG 2027 and the evidence the revenue depends on. Several research projects on orchard PV end this year and next; once robust results are available, we will update this article.