Civil Law and Wind Farm Output Forecasting Error Liability Claims in Europe

1. Introduction

Wind-farm output forecasting is fundamental to the operation of European electricity markets. Unlike conventional generation, wind generation depends upon variable meteorological conditions. A wind-farm operator may therefore forecast that a project will produce, for example, 100 MWh during a particular settlement period, while actual generation may be only 75 MWh or may reach 125 MWh.

That difference can produce an imbalance between contracted/scheduled electricity and actual physical delivery. European electricity markets generally allocate financial consequences for such imbalances to the relevant market participant or balancing-responsible party. Research on European markets confirms that forecast errors can create substantial imbalance costs, and that the allocation of balancing responsibility differs between European jurisdictions. (IDEAS/RePEc)

This creates an important civil-law question:

When a wind-farm output forecast is materially wrong, who should bear the resulting financial loss?

Possible defendants include:

wind-farm operators;

turbine manufacturers;

engineering and construction contractors;

meteorological forecasting companies;

energy traders;

balancing-service providers;

power-purchase counterparties;

asset managers;

consultants;

technical advisers; and

in some circumstances, transmission or regulatory authorities.

The legal basis may be:

breach of contract;

negligent professional services;

negligent misrepresentation;

fraudulent misrepresentation;

breach of warranty;

failure to meet an energy-yield guarantee;

breach of a power-purchase agreement;

breach of balancing obligations;

regulatory compensation;

unjust enrichment;

tort/delict; or

non-contractual liability of a public authority.

Important qualification: reported European judgments dealing specifically with liability for a wind-forecasting algorithm's forecast error are still relatively scarce. Consequently, the strongest legal analysis combines the few directly relevant wind/renewable-energy disputes with closely analogous European cases concerning PPAs, balancing costs, energy-output guarantees, damages, causation and regulatory compensation.

2. What Is a Wind-Farm Output Forecast?

A wind forecast attempts to predict:

the quantity of electricity that a wind farm will actually produce during a specified future period.

Forecasts may be generated:

several days ahead;

one day ahead;

several hours ahead;

minutes ahead; or

continuously during real-time operation.

Forecasting models can use:

numerical weather prediction;

wind-speed measurements;

turbine SCADA data;

historical production;

atmospheric models;

satellite data;

machine learning;

artificial intelligence;

statistical models; and

ensemble weather forecasting.

The closer the market settlement period becomes to real time, the more accurate forecasts generally become.

3. Why Forecast Errors Create Civil Liability Issues

Suppose a wind farm forecasts:

50 MWh

but actually produces:

30 MWh.

There is a:

20 MWh shortfall.

If the operator already sold 50 MWh in the day-ahead or intraday market, it may have to acquire replacement electricity or pay an imbalance charge.

Conversely, if it forecasts:

50 MWh

but actually generates:

80 MWh,

there may be an excess imbalance and a different financial consequence.

Thus:

Forecast error

↓

Schedule/contract mismatch

↓

Imbalance

↓

Balancing-market settlement

↓

Financial loss

↓

Potential civil claim

European research specifically confirms that wind and solar forecast errors increase imbalance volumes and can affect electricity prices. (DOI)

4. The Central Legal Question

A forecasting error does not automatically constitute a legal wrong.

The fundamental distinction is:

Ordinary forecasting uncertainty

"The forecast was wrong because wind conditions were inherently unpredictable."

versus

Actionable professional failure

"The forecast was materially inaccurate because the forecasting company failed to use the agreed methodology, ignored available data, used defective software, failed to update the model, or knowingly supplied unreliable information."

This distinction is central to liability.

5. Major Categories of Claims

5.1 Contractual forecasting obligation

A forecasting agreement may require:

specified forecast accuracy;

specified confidence intervals;

specified update frequency;

specified data sources;

specified delivery times;

specified statistical methodology;

availability guarantees;

reporting obligations.

Failure to satisfy these obligations may constitute breach of contract.

5.2 Energy-yield guarantee

The relevant contract may not expressly guarantee the forecast but may guarantee:

minimum annual energy production.

This is particularly important in wind-turbine supply and EPC contracts.

A production guarantee is legally different from a forecast.

Forecast

"Expected annual production: 300 GWh."

Guarantee

"Contractor guarantees 285 GWh under specified conditions."

The second creates substantially stronger liability.

6. Case Law 1 — Glenfiddich Wind Ltd v Dorenell Windfarm Ltd [2025] CSOH 62

This is one of the most directly relevant recent European wind-energy decisions.

The Scottish Court of Session considered the contractual treatment of revenues associated with a wind farm, including:

PPA revenue;

constraint benefits;

loss-of-revenue claims;

turbine availability;

energy production;

power-output warranties; and

balancing/constraint arrangements.

The wind farm had a PPA under which electricity was traded in advance of generation. The litigation required the court to interpret how various electricity-related receipts and losses affected rent calculations. (BAILII)

The court also considered the relevance of the System Imbalance Price to quantification of a potential claim and emphasised the evidential difficulties where a party had not produced evidence enabling the court to calculate the alleged loss. (BAILII)

Importance

This case demonstrates an essential proposition:

Wind-energy losses must be established through the contractual mechanism and supported by reliable evidence of quantum.

For a forecasting-error claim, this means the claimant must normally establish:

what the forecast should have been;

what the actual output was;

what contractual position was created by the forecast;

what imbalance resulted;

what financial charge followed; and

why the defendant is legally responsible.

7. Case Law 2 — ACF Renewable Energy Ltd v Republic of Bulgaria, Award of 5 January 2024

Although this was an investment arbitration rather than ordinary civil litigation, it is highly relevant to the economic treatment of renewable-energy balancing costs.

The dispute concerned changes to Bulgarian electricity trading rules that exposed renewable-energy producers to substantially greater balancing costs.

The claimant argued that the regulatory system had previously included protections such as:

a renewable-energy balancing group;

deviation tolerance;

discounted balancing costs; and

arrangements recognising the forecasting difficulties associated with renewable energy.

The tribunal considered the relationship between renewable generation, forecasting difficulties and balancing costs. (Italaw)

Importance

The case demonstrates that:

Forecasting uncertainty can be economically significant enough to become a major component of the financial risk allocation surrounding a renewable-energy project.

It also illustrates the distinction between:

an inherent commercial risk of renewable generation; and

a legal entitlement arising from contractual or regulatory arrangements.

That distinction is critical in civil litigation.

8. Case Law 3 — Mainstream Renewable Power Ltd and Others v Germany, Award of 13 May 2026

The 2026 Mainstream Renewable Power arbitration concerned German regulatory measures affecting planned offshore wind projects.

The claimants alleged substantial economic losses resulting from changes to the German regulatory framework, including claims under the Energy Charter Treaty.

The dispute involved more than €353 million in claimed compensation. (Italaw)

Relevance to forecasting-error disputes

This is not a direct forecast-negligence case.

Its relevance lies in the broader principle of allocation of regulatory and commercial risk.

A wind developer's financial model inevitably incorporates assumptions concerning:

expected production;

electricity prices;

balancing costs;

support schemes;

grid conditions;

curtailment;

market rules.

A civil court or arbitral tribunal therefore needs to identify:

Which party contractually assumed the risk that the underlying assumptions would change?

That principle is directly applicable when a forecast turns out to be materially inaccurate.

9. Case Law 4 — Aquind v ACER, Case T-342/23, General Court, 11 June 2025

In Aquind v ACER, the General Court dealt with a claim for EU non-contractual liability concerning an electricity interconnector.

The Court reiterated the fundamental requirements for EU non-contractual damages:

unlawfulness;

actual damage; and

causal connection.

Failure to establish one cumulative condition is sufficient to defeat the damages action. (EUR-Lex)

Importance for wind forecasting claims

The same analytical structure is extremely useful.

A wind-farm owner cannot simply say:

"The forecast was inaccurate and we lost money."

It must establish:

Wrongdoing

The forecasting provider breached a contractual or legal duty.

Damage

The claimant actually incurred an identifiable loss.

Causation

The forecasting error caused that particular loss.

This prevents speculative damages claims based solely on a comparison between predicted and actual revenue.

10. Case Law 5 — Energia Group and Others v Commission for Regulation of Utilities, Case C-36/25

This is a particularly important current European energy dispute, but it must be treated carefully because the CJEU proceedings are not yet a final judgment.

The Irish Supreme Court referred questions concerning renewable generators and non-market-based downward redispatching under Article 13(7) of Regulation 2019/943.

The questions include:

how financial compensation should be calculated;

whether compensation should include foregone financial support;

whether certain generators may be excluded;

whether compensation can be paid to suppliers rather than generators; and

whether CPPA payments should be included in compensation. (EUR-Lex)

The Advocate General has proposed an interpretation under which qualifying renewable generators subjected to non-market-based downward redispatching may be entitled to compensation approximating the hypothetical revenue they would have obtained under market conditions. (EUR-Lex)

Importance

Although this is not a forecasting-error case, it demonstrates an important principle:

European electricity law increasingly treats renewable-generator financial losses through specific allocation and compensation mechanisms rather than leaving every loss to ordinary contract law.

It is therefore important to determine whether the loss is:

an imbalance cost caused by the producer's forecast;

a curtailment/redispatch loss caused by the system operator; or

a contractual loss caused by a counterparty.

These are legally different categories.

11. Case Law 6 — Ålands Vindkraft AB v Energimyndigheten, C-573/12

The CJEU's decision in Ålands Vindkraft concerned Swedish renewable-energy support and the compatibility of national renewable-electricity certificate arrangements with EU internal-market principles.

The case is not a damages claim.

Nevertheless, it is significant because it demonstrates that renewable-energy generation is subject to a complex interaction between:

national support mechanisms;

EU internal-market law;

electricity-market regulation; and

environmental policy.

Relevance to forecasting claims

A claimant calculating damages must therefore identify the relevant revenue stream.

For example:

Forecast error → lower electricity sales

is different from:

Forecast error → loss of renewable certificate revenue

which is different again from:

Forecast error → imbalance settlement charge.

Each may require a different contractual and regulatory analysis.

12. Case Law 7 — PreussenElektra AG v Schleswag AG, C-379/98

PreussenElektra is a foundational CJEU renewable-electricity case.

The Court considered the German renewable-electricity purchasing system and the obligation concerning electricity generated from renewable sources.

Although not a forecasting-liability case, it is important for understanding the special regulatory environment surrounding renewable electricity.

Relevance

Wind-farm revenue may arise through several legal mechanisms:

ordinary market sales;

PPAs;

feed-in arrangements;

certificates;

subsidies;

contracts for difference;

balancing payments.

Therefore, a forecast-error damages claim must identify exactly which legal revenue stream was allegedly lost.

13. Case Law 8 — Vattenfall / renewable-support litigation

European renewable-energy disputes have also generated substantial litigation concerning changes to renewable-support mechanisms.

The broader jurisprudence demonstrates that investors and operators cannot automatically treat projected renewable-energy revenue as legally guaranteed.

The legal question frequently becomes:

Was the projected revenue merely a commercial expectation, or was it protected by a binding contractual/statutory entitlement?

That distinction is essential when a wind forecast forms part of an investment model.

14. General European Contract-Damages Authorities

Because direct wind-forecasting cases remain relatively limited, general European contract authorities become particularly important.

14.1 Hadley v Baxendale

This foundational English contract case establishes the classic principle of remoteness of damages.

Losses must generally be sufficiently connected to the breach and within the relevant contemplation of the parties.

Application to wind forecasting

Suppose a forecasting company provides a defective forecast.

The wind farm claims:

€2 million lost from imbalance charges.

The defendant may argue:

"Those losses were too remote and were not within the contractual allocation of risk."

The answer depends on:

the contract;

the parties' knowledge;

the purpose of the forecasting service;

the pricing of the service;

liability exclusions;

caps;

and the nature of the imbalance mechanism.

15. The Achilleas — Transfield Shipping Inc v Mercator Shipping Inc [2008] UKHL 48

This is an important modern authority on contractual remoteness.

The case refined the conventional approach by focusing upon the scope of the contractual responsibility assumed by the defendant.

Application to wind forecasts

Suppose a forecasting company knows that:

the wind farm uses the forecast to make day-ahead bids;

forecast errors generate imbalance charges; and

the forecast is specifically purchased to minimise those charges.

The claimant can argue that imbalance losses fall squarely within the purpose and scope of the contract.

That argument becomes stronger if the agreement expressly states:

"The forecast is to be used for market scheduling and imbalance management."

16. Arnold v Britton [2015] UKSC 36

This Supreme Court case is important for contractual interpretation.

Wind-energy agreements frequently contain complex clauses concerning:

revenue;

imbalance;

constraints;

availability;

production;

forecast accuracy;

liquidated damages;

liability caps.

Courts generally begin with the contractual language and interpret it in its commercial context.

Application

If a PPA states:

"Seller shall be responsible for all imbalance charges arising from deviations between scheduled and actual production",

the precise wording becomes crucial.

The court must determine:

Does "deviation" include errors caused by the forecasting contractor?

The answer depends upon the contract structure and any subcontracting or indemnity provisions.

17. Cavendish Square Holding BV v Makdessi [2015] UKSC 67

This case concerns contractual penalties and liquidated damages.

It becomes relevant where a wind-energy agreement provides:

€X for every MWh of forecast deviation.

The court may need to determine whether the provision is:

genuine risk allocation;

agreed compensation;

or an unenforceable penalty.

This is particularly important where imbalance charges are calculated through contractual formulas.

18. Nature of a Forecasting Obligation

A key legal distinction is whether the forecasting service is:

A. Best-efforts obligation

The forecaster promises to use reasonable professional skill and care.

B. Accuracy warranty

The forecaster guarantees that the forecast will remain within a defined error range.

C. Performance guarantee

The forecaster guarantees a particular economic result.

The legal consequences differ dramatically.

19. Negligent Forecasting

A tort/delict claim may arise where the forecasting provider owes an independent duty of care.

Potential allegations include:

failure to calibrate the model;

use of obsolete weather data;

failure to correct systematic bias;

failure to account for turbine availability;

failure to incorporate curtailment data;

failure to update weather inputs;

incorrect aggregation;

coding errors;

data corruption;

failure to communicate known model limitations.

But merely showing:

"The forecast was wrong"

is generally insufficient.

Wind forecasts necessarily involve uncertainty.

20. Standard of Reasonable Professional Skill

The standard may be determined by:

contract terms;

industry standards;

professional qualifications;

forecasting methodology;

historical accuracy;

available technology;

known weather conditions;

data quality;

accepted forecasting practices.

Expert evidence will often be necessary.

An expert may compare:

Actual forecasting model

with

Reasonably competent industry methodology.

21. AI-Based Forecasting

Modern wind forecasts increasingly employ machine learning.

This creates additional liability questions.

Suppose an AI model systematically under-predicts production during high-wind conditions.

Potential issues include:

inadequate training data;

model drift;

defective validation;

inappropriate feature selection;

failure to retrain;

unexplained model changes;

data poisoning;

software defects;

lack of monitoring.

The central legal question remains:

Was the AI system operated in accordance with the contractual and professional standard expected of the provider?

22. Liability of the Wind-Turbine Manufacturer

A turbine manufacturer may face liability where the forecast depended upon technical assumptions supplied by the manufacturer.

For example:

Manufacturer states expected production = 400 GWh/year.

Actual production:

310 GWh/year.

Possible claims include:

breach of production warranty;

misrepresentation;

defective engineering;

breach of specification;

failure to achieve guaranteed performance.

This is different from a pure weather forecast.

23. Energy-Yield Forecasting Versus Meteorological Forecasting

The two should not be confused.

Meteorological forecast

Expected wind speed at location.

Power forecast

Expected electrical production.

A sophisticated wind-power forecast must account for:

wind speed;

wind direction;

air density;

turbine power curve;

wake losses;

availability;

electrical losses;

curtailment;

icing;

maintenance;

grid restrictions.

Therefore, liability can arise because the energy-conversion model was defective even when the weather forecast was accurate.

24. Causation

Causation is usually one of the most difficult issues.

Assume:

Forecast predicted 100 MWh.

Actual output:

70 MWh.

But the wind farm also experienced:

10 MWh turbine downtime;

5 MWh grid curtailment;

5 MWh icing;

10 MWh genuinely low wind.

The entire 30 MWh difference cannot necessarily be attributed to forecasting error.

A proper damages analysis may therefore require:

Forecast error → technical causation → market consequence → financial loss.

25. Loss Calculation

A sophisticated damages model might be:

Forecast error

× imbalance volume

× applicable imbalance price

replacement energy cost

transaction costs

− avoided costs

− revenues actually received

= net loss

The calculation should be performed at the relevant settlement interval rather than simply by comparing annual forecast and actual production.

European electricity balancing operates through settlement mechanisms designed to allocate financial responsibility for deviations from balanced positions. (EUR-Lex)

26. Why Annual Forecast Accuracy May Be Legally Misleading

Consider:

Annual or daily net error may appear small.

But imbalance prices can vary substantially between settlement periods.

Therefore:

A small annual forecasting error can produce a large financial loss.

This is one reason courts need expert market evidence.

27. Forecast Error and PPA Liability

A PPA may specify:

fixed price;

minimum delivery;

nomination procedures;

balancing responsibility;

force majeure;

curtailment;

deemed generation;

shortfall compensation;

liquidated damages.

A forecasting dispute may therefore become a PPA interpretation dispute.

The key question becomes:

Who contractually bears the consequences of deviation between nominated and actual output?

28. Forecasting Company as Subcontractor

Suppose:

Wind farm → energy trader → forecasting company.

The forecasting company provides the forecast to the trader.

The wind farm suffers €1 million in imbalance charges.

Can the wind farm sue the forecasting company?

Not necessarily.

The first question is:

Is there contractual privity?

If the wind farm has no contract with the forecasting provider, the claim may need to rely upon:

tort/delict;

negligent misstatement;

third-party rights;

assignment;

agency;

or another national-law doctrine.

29. Contractual Liability Chain

A typical project may contain:

Wind farm owner

↓

Asset manager

↓

Energy trader / BRP

↓

Forecasting provider

↓

Meteorological data provider

↓

Weather-data source

Each contract may allocate a different portion of forecasting risk.

Consequently, the final liability may involve:

contribution or indemnity claims between several defendants.

30. Force Majeure

The defendant may argue that the forecast error resulted from extraordinary weather.

But force majeure generally concerns an impediment to contractual performance, not simply an inaccurate prediction.

The parties may nevertheless contractually provide for:

exceptional weather;

extreme weather events;

weather-model failure;

data outages;

cyber incidents.

The exact clause is therefore crucial.

31. Exclusion and Limitation Clauses

Forecasting agreements commonly contain:

liability caps;

exclusions of consequential loss;

exclusions of lost profits;

exclusions for market losses;

disclaimers of forecast accuracy;

force-majeure clauses.

A clause might state:

"Provider's maximum aggregate liability shall not exceed the fees paid during the previous 12 months."

That could be extremely important if the actual imbalance loss is €5 million while annual forecasting fees are €100,000.

Whether such clauses are enforceable depends on applicable national law and the circumstances.

32. Fraudulent Forecasting

The situation is substantially different where the claimant proves intentional deception.

For example:

Forecasting company knows the model systematically underestimates output but advertises it as having 95% accuracy.

Potential consequences include:

fraudulent misrepresentation;

intentional breach;

aggravated damages where recognised;

invalidity/rescission;

broader damages;

possible regulatory consequences.

Fraud can also affect the enforceability of contractual exclusions under national law.

33. Professional Negligence

Professional negligence may arise where the forecasting provider:

fails to validate the model;

uses inappropriate weather stations;

fails to detect obvious anomalies;

ignores turbine-specific characteristics;

fails to incorporate known outages;

uses incorrect coordinates;

supplies stale forecasts;

fails to update forecasts when required.

Expert evidence is likely to be decisive.

34. Burden of Proof

The claimant generally needs to establish:

1. Duty

The defendant owed a contractual or legal obligation.

2. Breach

The forecasting methodology or service fell below the required standard.

3. Causation

The breach caused the relevant deviation.

4. Damage

The claimant suffered an identifiable economic loss.

5. Quantum

The amount can be established with reasonable certainty.

The Aquind decision is useful by analogy because EU non-contractual liability similarly treats unlawfulness, damage and causation as cumulative requirements. (EUR-Lex)

35. Statistical Evidence

Forecasting cases are unusually dependent upon statistical analysis.

A claimant may examine:

mean absolute error;

root mean square error;

bias;

confidence intervals;

forecast skill;

error distribution;

persistence;

seasonal effects;

extreme-event performance.

A court should avoid concluding:

"Forecast was wrong 20% of the time, therefore negligent."

A professional forecast can legitimately have significant uncertainty.

The relevant question is:

Was the error outside the range reasonably expected from a competent forecasting methodology?

36. Counterfactual Damages

The court may need to construct a counterfactual:

What would have happened if the forecast had been prepared competently?

For example:

Actual forecast

100 MWh

Competent forecast

90 MWh

Actual production

85 MWh

The claimant's loss is not necessarily based on the entire 15 MWh difference.

The relevant counterfactual may be the imbalance cost that would have resulted from the reasonable forecast, not a theoretically perfect forecast.

37. Mitigation of Loss

The claimant must generally consider reasonable mitigation.

Possible mitigation measures include:

intraday market trading;

updating nominations;

portfolio aggregation;

battery storage;

reserve procurement;

curtailment;

revised bidding;

alternative forecasting services.

If the wind farm could reasonably have reduced its imbalance exposure but failed to do so, the defendant may argue that the damages should be reduced.

38. Portfolio Forecasting

Large energy companies rarely rely upon one wind farm.

They may aggregate:

50 wind farms + solar + hydro + storage.

Portfolio diversification reduces forecasting error.

Therefore, a claimant may have to establish:

Was the alleged forecasting error actually attributable to the defendant, or did portfolio-level management create or amplify the loss?

This becomes particularly important where the defendant supplied forecasts to an energy trader rather than directly to the wind farm.

39. Regulatory Versus Contractual Imbalance Liability

This distinction is fundamental.

Regulatory imbalance

The wind farm pays the TSO/BRP under electricity-market rules.

Contractual imbalance

The wind farm pays its PPA or trading counterparty under contractual terms.

Professional negligence

The wind farm seeks recovery from the forecasting provider.

These three relationships can coexist.

40. Redispatch and Curtailment Must Not Be Confused with Forecast Error

A wind farm may generate less electricity because:

the system operator orders it to reduce output.

That is curtailment/redispatch, not necessarily forecast error.

The distinction matters because EU electricity law can provide compensation for certain forms of non-market-based redispatch.

The pending Energia Group litigation is particularly relevant because Article 13(7) of Regulation 2019/943 is being examined in relation to financial compensation for renewable generators subject to qualifying downward redispatch. (EUR-Lex)

Thus:

Low output

does not automatically mean:

forecasting failure.

41. Practical Hypothetical

Consider a 200 MW offshore wind farm.

The forecasting company predicts:

400 MWh

for a particular four-hour period.

The farm actually produces:

300 MWh.

The wind farm has therefore nominated 100 MWh too much.

The balancing charge is:

€180/MWh.

Potential imbalance cost:

100 × €180 = €18,000

The farm claims €18,000 from the forecasting company.

Legal analysis

Step 1 — Contract

Was the forecast contractually guaranteed?

Step 2 — Standard

What accuracy standard was promised?

Step 3 — Data

Did the forecaster receive correct turbine and meteorological data?

Step 4 — Error

Was the 100 MWh deviation within normal statistical uncertainty?

Step 5 — Causation

Would a competent forecast have avoided the €18,000 loss?

Step 6 — Market mitigation

Could the farm have corrected the nomination through the intraday market?

Step 7 — Contractual limitation

Is liability capped?

Step 8 — Quantum

Was €18,000 the actual net loss after replacement transactions and other revenues?

42. A More Complex Example

Suppose:

forecast error = 50 MWh;

imbalance charge = €200/MWh;

direct imbalance loss = €10,000;

intraday correction could have reduced loss to €4,000;

claimant failed to make the correction.

The defendant may argue:

€6,000 was avoidable.

The court may therefore award only the legally recoverable amount.

43. Importance of Glenfiddich for Quantum

Glenfiddich Wind v Dorenell Windfarm is particularly useful because the Scottish court addressed the evidential problem of quantifying losses linked to electricity imbalance and noted the absence of evidence sufficient to perform the necessary calculation. (BAILII)

The lesson is straightforward:

A wind-energy damages claim must be capable of being calculated from actual contractual and market data.

44. Forecasting Error and Misrepresentation

A project developer may claim:

"The seller represented that the project would generate 350 GWh annually."

If that representation was knowingly or negligently false, liability may arise independently of the later forecasting agreement.

Possible evidence includes:

original wind-resource assessment;

bankability report;

meteorological measurements;

turbine power curves;

P50/P75/P90 energy assessments;

internal emails;

financial models;

independent engineer reports.

45. P50, P75 and P90 Forecasts

Wind projects frequently use probabilistic energy estimates.

For example:

P50

Expected production with approximately 50% probability of exceeding the estimate.

P90

A more conservative estimate with a higher probability of actual production exceeding the estimate.

This has major legal implications.

A developer cannot automatically argue:

"The P50 forecast was not achieved, therefore somebody was negligent."

A P50 figure is inherently probabilistic.

The court must ask:

Was the probability model itself defective?

That is a much more sophisticated question.

46. Bank Financing and Forecasting Liability

Banks may rely upon:

energy-yield assessments;

wind-resource studies;

production forecasts.

A forecasting error can therefore cause:

debt-service problems;

covenant breaches;

refinancing costs;

reduced project valuation;

default interest;

restructuring costs.

Whether these consequential losses are recoverable depends heavily upon:

contractual wording;

remoteness;

foreseeability;

causation;

limitation clauses.

47. Investment Damages

Suppose an investor says:

"I invested €100 million because the forecast predicted 450 GWh."

Actual production:

300 GWh.

The investor cannot necessarily recover the entire €100 million.

The court may need to determine:

What would the investor's position have been if the true information had been known?

This is a counterfactual valuation exercise.

48. Expert Evidence

Expert evidence may be required from:

Meteorological experts

To evaluate wind conditions.

Renewable-energy engineers

To evaluate turbine performance.

Forecasting specialists

To assess model accuracy.

Energy-market economists

To quantify imbalance costs.

Financial experts

To calculate lost revenue.

Valuation experts

To calculate project-level diminution in value.

49. Documentary Evidence Checklist

A claimant should preserve:

forecasting agreement;

PPA;

balancing agreement;

BRP agreement;

forecast files;

actual SCADA data;

meteorological data;

turbine availability records;

grid instructions;

curtailment records;

market nominations;

intraday transactions;

imbalance invoices;

algorithm/version records;

correspondence concerning forecast accuracy;

technical reports;

financial models; and

expert reports.

50. Comparative Case-Law Table

51. Key Legal Principles Emerging from the Authorities

Principle 1 — Forecast error is not automatically negligence

Wind production is inherently uncertain.

Principle 2 — Contract wording is critical

The parties can allocate:

forecasting risk;

balancing risk;

market risk;

curtailment risk;

availability risk.

Principle 3 — Actual financial loss must be proved

A forecast difference alone is insufficient.

Aquind reinforces the importance of actual damage and causation in damages claims. (EUR-Lex)

Principle 4 — Imbalance costs are economically real

European electricity systems financially settle deviations between expected and actual positions, meaning forecast errors can have direct economic consequences. (IDEAS/RePEc)

Principle 5 — Forecasting liability is different from curtailment liability

A producer may generate less because of:

bad forecasting;

low wind;

turbine failure;

grid failure;

curtailment;

redispatch.

Each requires separate legal analysis.

Principle 6 — Evidence of causation is crucial

A claimant must separate:

forecasting loss

from:

losses caused by other operational or market factors.

52. Recommended Litigation Framework

For a European wind-forecasting liability claim, the court should proceed approximately as follows:

Step 1 — Identify the contractual relationship

Who hired the forecaster?

Step 2 — Identify the obligation

Was there:

a forecast;

a warranty;

a reasonable-care obligation;

an accuracy guarantee?

Step 3 — Establish the standard

What would a competent forecasting provider have done?

Step 4 — Analyse the forecast

Was it statistically and technically reasonable?

Step 5 — Compare actual production

Determine the true deviation.

Step 6 — Identify other causes

Exclude:

turbine failure;

curtailment;

grid outages;

extreme weather;

data failure.

Step 7 — Establish market consequences

Calculate the actual imbalance.

Step 8 — Apply the contractual mechanism

Determine the relevant PPA/BRP/balancing provisions.

Step 9 — Calculate damages

Use actual settlement data.

Step 10 — Apply mitigation

Deduct losses that could reasonably have been avoided.

Step 11 — Apply liability limits

Consider:

caps;

exclusions;

liquidated damages;

consequential-loss clauses.

Step 12 — Determine final liability

Only then should the court calculate compensation.

53. Conclusion

Wind-farm output forecasting error liability is an emerging and technically complex area of European civil and energy law.

The central legal difficulty is that forecasting uncertainty is inherent in wind generation. Consequently, an inaccurate forecast is not by itself evidence of negligence or breach.

Liability becomes significantly stronger where the claimant can demonstrate that:

the defendant owed a defined forecasting obligation + the forecasting methodology fell below the required contractual/professional standard + the error caused a measurable imbalance + the imbalance caused an actual financial loss.

The most directly useful recent European authority is Glenfiddich Wind Ltd v Dorenell Windfarm Ltd [2025] CSOH 62, because it illustrates how courts deal with contractual revenue, wind-farm output, constraint arrangements, imbalance pricing and the evidentiary requirements for quantifying losses. (BAILII)

The renewable-energy balancing analysis in ACF Renewable Energy v Bulgaria demonstrates the economic significance of forecasting and imbalance costs, while Aquind v ACER reinforces the cumulative importance of unlawfulness, actual damage and causation in damages litigation. (Italaw)

The pending Energia Group, C-36/25 proceedings are also significant for the future European legal framework because they concern financial compensation for renewable generators, redispatching and the treatment of CPPA-related revenue. They should presently be treated as pending proceedings rather than final CJEU case law. (EUR-Lex)

For a civil-law examination or research paper, the most useful conceptual formula is:

Forecasting obligation → professional/contractual standard → forecast error → causation → imbalance → actual financial loss → mitigation → contractual limitation → compensation.

The key distinction is therefore not simply “Was the wind forecast wrong?”, but:

“Was the forecast legally defective, did that defect cause the particular market loss claimed, and had the defendant contractually or legally assumed responsibility for that risk?”