Weather delays happen on every construction project. Rain stops earthworks. Frost prevents concrete pours. Snow shuts sites entirely. But under NEC4, a weather delay only becomes a compensation event if the weather measurement exceeds a specific, objective threshold. Subjective claims about “exceptionally bad weather” have no contractual standing. The clause is deliberately mechanical: either the recorded data exceeds the one-in-ten-year value, or it does not.
This makes weather one of the most precisely defined compensation events in the contract, and one of the most frequently mishandled. Contractors who understand the mechanism recover time and cost. Those who do not lose entitlement to weather events that were genuinely exceptional, simply because they failed to compare the right numbers at the right time.
What Clause 60.1(13) Says
Clause 60.1(13) states that a compensation event occurs when a weather measurement is recorded within a calendar month at the place stated in the Contract Data, and by comparison with the weather data, the measurement is shown to occur on average less frequently than once in ten years.
Three conditions must all be satisfied for the clause to apply:
- A weather measurement is recorded. The measurement must be one of the types stated in the Contract Data. There must be an actual recorded value, not an estimate or recollection.
- The measurement is within a calendar month. Each calendar month is assessed independently. January is compared to the January historical data, February to the February data, and so on. Conditions spanning two months are not aggregated.
- The measurement exceeds the one-in-ten-year value. The recorded value is compared against the weather data identified in the Contract Data. If it exceeds the value that occurs on average less frequently than once in ten years for that calendar month, it is a compensation event.
This is a supervening event under Category 3 of the compensation event classification. The Contractor notifies under clause 61.3, and the eight-week time bar applies.
Importantly, clause 60.1(12) for physical conditions explicitly excludes weather: it covers conditions within the Site “which are not weather.” This means weather can only be claimed under 60.1(13) with its specific measurement test. If weather is severe but does not exceed the one-in-ten-year threshold, there is no alternative clause and no fallback. The Contractor bears that weather risk entirely.
Why NEC4 uses an objective test
Other standard forms use subjective tests like “exceptionally adverse weather conditions” (JCT) or “unforeseeable” conditions (FIDIC). These invite argument about what counts as exceptional. NEC4 removes the argument entirely. The weather data either shows the measurement exceeds the once-in-ten-year threshold, or it does not. The test is binary and verifiable by anyone with access to the same data.
The Four Weather Measurements
The Contract Data in NEC4 typically specifies four weather measurements to be recorded and assessed each calendar month. These are the standard measurements used on most UK contracts:
| Measurement | Unit | What it captures |
|---|---|---|
| Cumulative rainfall | mm | Total rainfall recorded during the calendar month |
| Days with rainfall > 5mm | Number of days | Days where daily rainfall exceeds 5mm, indicating working days lost to heavy rain |
| Days with minimum air temperature < 0°C | Number of days | Frost days affecting concrete, earthworks, and exposed operations |
| Days with lying snow at 0900 GMT | Number of days | Days with snow cover at the specified observation time, affecting site access and operations |
Each measurement is assessed independently. If cumulative rainfall exceeds the threshold but frost days do not, only rainfall triggers a compensation event. Multiple measurements can trigger separate compensation events within the same month if each independently exceeds its threshold. However, the delay assessment must not double-count: if 3 excess rain days and 2 excess frost days overlap on the same dates, the programme impact is assessed once for the combined effect, not as 5 separate lost days.
The four measurements are not exhaustive. The Client can amend or add measurements in Contract Data Part 1 to reflect site-specific conditions. For example, wind speed may be added on exposed coastal or elevated sites where high winds prevent crane operations.
What the Contract Data Must State
Clause 60.1(13) only works if the Contract Data provides three pieces of information. If any of these is missing or incomplete, the clause becomes unworkable and the Contractor may have no weather entitlement at all.
1. The place where weather is to be recorded. This is typically the nearest Met Office weather station to the Site. It must be a location with reliable recording equipment and historical data. Using on-site measurements is possible but creates practical difficulties because there is no historical baseline for comparison.
2. The weather measurements to be assessed. These are the four measurements listed above, or alternative measurements chosen by the Client. Each measurement must be defined precisely so that recorded values can be compared against historical data without ambiguity.
3. The weather data. This is the historical dataset against which actual measurements are compared. It is typically ten years of Met Office records for the stated weather station. The dataset must include monthly values for each measurement so that the one-in-ten-year threshold can be derived for each calendar month.
Check the Contract Data before you start
On mobilisation, verify that all three items are present in Contract Data Part 1. If the weather station is too far from the Site, or the historical data period is too short, or measurements do not reflect the operations being carried out, raise it immediately. Once the contract is signed, these become the contractual reference points regardless of whether they are appropriate.
The One-in-Ten-Year Threshold
The one-in-ten-year threshold is derived from the weather data stated in the Contract Data. For each calendar month, the historical values for each measurement are ranked from lowest to highest. With ten years of data, the highest recorded value for that calendar month is the value that occurred once in the dataset. Any actual measurement exceeding this value occurs less frequently than once in ten years, satisfying the clause 60.1(13) test.
For cumulative rainfall, this means finding the rainfall total for each November over the ten-year period, ranking them, and identifying the highest. For frost days, the same process is applied to the count of days with minimum temperature below 0°C for each month. In practice, the weather data provided in the Contract Data may include pre-calculated return period values from the Met Office, which avoids the parties having to derive the threshold from raw data.
Deriving the one-in-ten-year threshold: historical data ranked per calendar month, actual measurement compared
Scroll to see full diagram
The threshold is specific to each calendar month. January may have a threshold of 120mm cumulative rainfall while July’s threshold may be 95mm. A measurement that triggers a compensation event in July would not trigger one in January. Always compare the actual value to the threshold for the correct calendar month.
Assessment Methodology: Only the Excess Counts
When a weather measurement exceeds the one-in-ten-year threshold, the compensation event is not assessed on the full impact of all weather that month. It is assessed only on the impact of the excess above the threshold. This is the most important principle in weather CE assessment and the one most frequently misunderstood.
The logic is straightforward. The Contractor is expected to have allowed in their programme and pricing for weather conditions up to and including the one-in-ten-year level. The contract only compensates for the additional impact beyond that threshold. While the trigger test is entirely objective (the measurement either exceeds the threshold or it does not), the assessment stage requires judgement: identifying which specific days caused delay, demonstrating causation to the critical path, and quantifying the cost impact of the excess weather.
How to calculate the excess
For each measurement that exceeds the threshold:
- Cumulative rainfall: Subtract the threshold from the actual total. If the threshold is 100mm and the actual rainfall is 142mm, the excess is 42mm.
- Days with rainfall > 5mm: Subtract the threshold number of days from the actual. If the threshold is 5 days and 9 days were recorded, the excess is 4 days.
- Frost days: Same approach. If the threshold is 8 days and 13 were recorded, the excess is 5 days.
- Snow days: Same approach. If the threshold is 2 days and 5 were recorded, the excess is 3 days.
The Contractor then assesses the delay and additional cost attributable only to the excess days or excess rainfall, not the full month’s weather impact. This is where the practical difficulty lies: identifying which specific days caused the delay when the total exceeds the statistical threshold.
Worked Example: Rainfall on a Highway Scheme
A contractor is delivering a highway improvement scheme under NEC4 Option C. The Contract Data states weather is recorded at the nearest Met Office station, 4km from the Site. The weather data covers 2014 to 2023. Earthworks are programmed through November.
Worked ExampleThe ten November cumulative rainfall values from the weather data are ranked lowest to highest:
| Rank | Year | Nov. Rainfall (mm) |
|---|---|---|
| 1 (lowest) | 2017 | 52 |
| 2 | 2022 | 61 |
| 3 | 2014 | 68 |
| 4 | 2020 | 74 |
| 5 | 2016 | 79 |
| 6 | 2019 | 85 |
| 7 | 2021 | 91 |
| 8 | 2023 | 98 |
| 9 | 2018 | 106 |
| 10 (highest) | 2015 | 118 |
The highest November cumulative rainfall in the ten-year dataset is 118mm (recorded in 2015). This is the value that occurred once in the ten-year period. Any actual measurement above 118mm occurs less frequently than once in ten years. The one-in-ten-year threshold for November cumulative rainfall is 118mm.
In November 2025, the Met Office station records cumulative rainfall of 138mm.
138mm exceeds the 118mm threshold. This is a compensation event under clause 60.1(13). The excess rainfall is 138mm − 118mm = 20mm.
The commercial team reviews daily rainfall records to identify the heaviest rain days that represent conditions beyond the threshold. Three working days (18, 22, and 27 November) recorded the most intense rainfall and are the days when earthworks operations were prevented by saturated ground conditions. These 3 days represent the delay attributable to the excess weather:
| Element | Defined Cost |
|---|---|
| Standing plant (excavators, dumpers) — 3 days | £11,400 |
| Labour on site unable to work — 3 days | £6,600 |
| Preliminaries (site management, welfare) — 3 days | £4,800 |
| Additional drying-out and re-compaction | £2,200 |
| Total Defined Cost | £25,000 |
| Fee (7%) | £1,750 |
| Total change to Prices | £26,750 |
The quotation includes a revised programme showing 3 days of delay to planned Completion. The earthworks are on the critical path, so the Contractor also claims a 3-day extension to the Completion Date.
The remaining rain days during November that fell within the 118mm threshold are not compensable. The Contractor was expected to have allowed for normal November weather in their programme and pricing. This is the critical distinction: the contract compensates for weather beyond what should have been anticipated, not for all weather-related disruption.
Notification and the Time Bar
Weather is a Category 3 supervening event. The Contractor must notify under clause 61.3 within eight weeks of becoming aware that the event has occurred. For weather, awareness typically arises when the month’s data becomes available from the weather station and can be compared against the threshold.
The practical challenge is timing. Met Office monthly data summaries are usually available within the first two weeks of the following month. If November data becomes available on 10 December, the eight-week clock starts on 10 December, giving the Contractor until 4 February to notify.
Do not wait for the Met Office data. If site records show rainfall is tracking well above the threshold during the month, prepare the notification in advance. Submit it as soon as the official monthly figure confirms the threshold has been exceeded. Waiting unnecessarily shortens the window for quotation preparation.
The assessment typically uses actual Defined Cost rather than a forecast. Because a weather CE can only be notified after the calendar month ends and the data is available, the work affected has usually already been done by the time the quotation is submitted. The dividing date will have passed, meaning the assessment is based on records of what actually happened rather than forecasts of what might happen.
Clause 63.7 also applies to weather. If the Contractor saw rainfall tracking towards the threshold mid-month but did not give an early warning under clause 15, the PM can assess the CE as if early warning had been given. The assessment would then reflect whatever mitigation an early warning meeting could have achieved, such as rescheduling weather-sensitive operations. Give early warning as soon as site conditions suggest the threshold may be exceeded.
Gather’s QS AI Agent cross-references daily site diary weather entries against Met Office station data in real time. When monthly rainfall, frost days, or snow days track towards the one-in-ten-year threshold, commercial teams are alerted before the month ends. The comparison happens automatically because every site diary entry already records weather conditions. No separate weather log. No end-of-month scramble to compile data.
Five Common Mistakes with Weather Compensation Events
1. Claiming for all weather, not the excess. The compensation event covers only the impact above the one-in-ten-year threshold. Submitting a quotation for every rain day in the month will be rejected and may delay the assessment of a legitimate claim.
2. Using site measurements instead of the stated weather station. The clause requires the measurement to be recorded at the place stated in the Contract Data. If the contract specifies a Met Office station 4km away, on-site rain gauges are supporting evidence at best. The contractual comparison must use the stated station’s data.
3. Aggregating weather across calendar months. Each month is assessed independently. A wet October followed by a wet November does not combine into one claim. Each month must independently exceed its own threshold. Two below-threshold months cannot be aggregated to create a compensation event.
4. Not linking weather to actual delay. Exceeding the threshold creates the compensation event, but the assessment must demonstrate that the excess weather actually caused delay to operations on the critical path. Rain on a day when only internal fit-out was planned has no programme impact. The quotation must include a revised programme showing the specific delay caused by the excess weather, supported by contemporaneous site records.
5. Missing the notification deadline. Weather claims feel less urgent than PM instructions because the weather has already happened. But the eight-week time bar runs from awareness, and awareness arises when the monthly data is available. Commercial teams that do not track weather data monthly can lose entitlement to genuinely exceptional weather simply by missing the notification window.
.webp)
