Rain Stops, The Ground Doesn't

It rained on Monday and the day that ends up in dispute is Tuesday, when the Bureau of Meteorology recorded nothing at all and the site still couldn't work.

Rainfall is the one part of a weather claim nobody has to argue about. The Bureau of Meteorology publishes it, both sides can look it up and neither party is going to dispute what fell out of the sky. If a claim were only about whether it rained, it would take about four minutes to settle.

What gets argued about is which days were lost. That's a different question entirely, with a different kind of evidence behind it.

First, check whether there's any entitlement at all

Whether rain earns an extension of time at all is a question for your contract and its Annexure. Under AS 4000-1997 the definition of a qualifying cause of delay is broad enough to cover inclement weather before the date for practical completion, while the same definition lets the parties strike causes out through Item 23, which is where weather often gets removed. Published commentary and the clause text don't always agree on this point, so the only reliable answer is the one in the contract in front of you.

Time and money are separate questions either way. Delay damages under AS 4000-1997 are payable for a compensable cause, which covers acts of the principal or superintendent plus anything the parties list in Item 26, so weather generally earns time without cost unless somebody put it there.

The contract is where the detail lives. Contracts frequently define what counts as inclement weather and some set a threshold, meaning a minimum period of delay before anything becomes claimable at all. Many require the event to be evidenced by published Bureau data, while some ask for rainfall to be reported monthly to the principal, which turns the record into a running obligation instead of something assembled when a claim comes up.

Worth reading your own contract for all three before the first wet week, since the definition, the threshold and the evidence requirement are the things that decide whether the days you lost were ever claimable.

The allowance runs out

Most programmes carry some allowance for wet weather, a handful of days built in on the assumption that a few will be lost. It's a sensible way to plan and it covers the ordinary case.

Once it's used up the arithmetic changes, since every further day lost to weather has to be claimed. A claim under AS 4000-1997 goes in within 28 days of when the contractor should reasonably have become aware of the causation, carrying the facts of causation and of the delay including its extent, and in practice a programme analysis is how the effect gets shown even though the clause doesn't demand one. Standards Australia published AS 4000:2025 in June 2025 and the numbering has moved, so check which edition your contract uses before quoting a clause at anybody. That's a serious amount of work and it falls due per event instead of once a season.

So the cost of a wet winter isn't only the days lost. Every day past the allowance generates a document, produced by the same people trying to recover the programme.

The station is not the site

Bureau data comes from a weather station, which might be some distance from your job. Rainfall is local in a way that annoys everybody, so a station reading that differs from what actually fell on your ground is an ordinary outcome, and it goes both ways just as often.

That matters because the contract often points at the station data as the evidence, which means the number used to test your claim isn't a measurement of your ground. Where a contract allows site rainfall records as well, keeping them is cheap and the comparison is the thing that makes the argument.

The days nobody can evidence

Here's the part that gets overlooked, which is that rain stops and the ground doesn't dry.

The Bureau shows no rainfall for Tuesday, while on Tuesday the excavation is holding water, the haul road has gone, the crane pad is too soft to stand a crane on and nobody is getting near the formwork. That day is as lost as the day it rained, with no weather data anywhere that says so.

Those days are often the expensive ones, since a single heavy day can take a week of ground out of use. They're also the ones a claim built on rainfall figures will quietly miss. The only evidence that exists for them is what somebody on site wrote down about the state of the ground and what it stopped.

It's also why a weather claim can't be assembled backwards. Rainfall can be pulled off the Bureau's site months later, while nobody can go back and reconstruct what the subgrade looked like on a Tuesday in June.

What a wet weather day needs recorded

A wet weather day needs the same things as any other delay event, with one addition. The activities that couldn't proceed, named so they can be matched against the programme. The times work stopped and restarted, since a half day and a full day are different claims. The reason recorded consistently, so the same cause reads the same way on every job. Who was stood down and what got moved to another area, since that's what shows the delay was managed rather than simply accepted.

Then the addition, which is the state of the ground itself, meaning standing water, access, whether plant could get on and whether the area was trafficable. A photograph with a date against it does more work here than anywhere else, since ground conditions are visible in a way that most delay causes aren't.

Weather data and a weather record

Weather data is free, public and completely neutral between the parties, telling you what fell and when, which is the part nobody was arguing about.

A weather record is what somebody on site wrote down about what that weather did, on the day, while standing in it. That's the part that decides the claim, it costs ten minutes a day to keep and it can't be bought, backdated or downloaded.