Adverse weather evidence
Weather evidence that survives scrutiny.
“It rained all week” is an assertion. A reading from a named gauge, captured on the day with its source and a checksum, is a fact the other side can check. Construction Metric keeps the weather half of the site record that way, every night, so when the delay is argued the record already exists and nobody has to reconstruct it.
In development with early customers · The site record is live today · UK sites
The contract context
Two tests, one record
JCT
Exceptionally adverse weather
The JCT forms list exceptionally adverse weather conditions as a Relevant Event. In the 2016 Standard Building Contract it sits in the clause 2.29 list (2.29.8, or 2.29.9 in some versions), in Design and Build 2016 at clause 2.26.8 and in the Intermediate form at clause 2.20.8. The contract sets no number. Whether the weather was exceptional is a judgement, and the judgement needs a record of what an ordinary month at that site looks like.
NEC
A 1-in-10 return value from recorded data
NEC3 and NEC4 treat weather as a compensation event under clause 60.1(13) when a measurement recorded within a calendar month, at the place stated in the Contract Data, is shown by comparison with the weather data to occur on average less frequently than once in ten years. Only the difference beyond that value is assessed. The comparison is against a return value worked out from recorded data, not against an average, and the record has to be able to make it.
Precedence
The executed contract decides
Editions differ, and amendments change the numbering, the measurements and sometimes the test itself. The executed contract, its edition and its amendments take precedence over anything on this page. Nor does the record decide entitlement: the contract administrator or the project manager does. What the record does is put measured, source-linked, contemporaneous figures in front of them.
What you get
What a weather record needs to survive scrutiny
Measured rainfall, with the station named
Rainfall from the nearest Environment Agency gauge, with the station and its distance from site stated on every day. A measurement from a named gauge can be checked by the other side. A grid-square average cannot.
Modelled conditions, kept separate
Temperature, wind and the rest come from weather-model data, and every line says so. A model estimate and a gauge reading are different kinds of evidence. A record that blends them is the one that gets picked apart.
The forecast that was available at the time
Each day keeps the forecast that existed when the decision to stand down, or to carry on, was made. Whether a decision was reasonable is judged on what was known then, not on what the gauge recorded afterwards.
Source, timestamp and checksum on every value
Every daily figure carries the source it came from, the time it was captured and a checksum. The record is captured as the work happens rather than reconstructed at submission, and a gap is flagged, never filled.
A standdown log linked to the diary
When weather stops work, the standdown is logged against the day: the operations affected, for how long, and who decided. It sits beside the diary entry, which is the link between weather and delay that a claim has to prove.
A monthly summary against the long-term record
Each month is set against the long-term record for the site, with the 1-in-10 return value worked out from the recorded years. A ten-year average is not a 1-in-10 return period, and the summary says which one you are reading.
How it works
From the gauge to the claim
Captured every night
For each site: measured rain from the named gauge, the modelled conditions, daylight, and the forecast that was in force. Nobody on site has to remember to do it.
Standdowns logged
When weather stops an operation, the standdown is recorded from the diary: what stopped, for how long, who decided. It sits next to the weather line for that day.
The month read
The monthly summary sets the site against the long-term record and shows the 1-in-10 return values beside the actuals, with the sources named and the years used stated.
The period handed over
Daily values with sources, timestamps and checksums, the standdown log and the summary, exported together to sit behind a notice, a claim or a delay analysis.
The record does not decide entitlement, and this page describes the design. The weather record and the site diary are live in Construction Metric today. The rest of what this page describes is being built with early customers, and the design will follow what they need. Under JCT the contract administrator decides whether the weather was exceptionally adverse and whether it delayed completion. Under NEC the project manager decides whether the 1-in-10 value was passed and what the difference is worth. The executed contract, its edition and its amendments take precedence over anything on this page. What Construction Metric does is keep the weather record organised, source-linked and gap-flagged from the day the work happened, so the decision is made on evidence rather than memory. For a high-value dispute a formal report from the Met Office may still be needed. A record kept this way is what makes that report straightforward to reconcile against.
Stop reconstructing the weather at submission.
Start on one site and the weather line is kept from the first night, beside everything else the site says and photographs.