Satellites can tell you the viz in metres
Turning satellite water colour straight into metres doubled the error compared with quoting the site's usual viz. Added to the usual viz, it improved the metres only marginally.
For the VIZ group · November 2026
Visibility apps keep appearing, many of them promising the metres days ahead. Six years of real viz reports from our own VIZ group let us test what is actually possible in Sydney waters.

Preface
Today's forecasting is very good at waves, wind and rain several days out. Visibility is a harder problem. It depends on what is floating in the water at your entry: stirred-up sand, plankton, runoff, a dirty layer sitting on top of clean water. No forecast model or satellite measures that, at that spot, on the day.
Satellites see only the top few metres of the sea, from pixels kilometres wide, once a day, and only when there is no cloud. Global ocean models work on grids several kilometres across. Neither can see the reef you are about to drop onto.

So a single figure like 5.6 m promises a precision nobody has. Two divers in the same water on the same day often differ by 2 m. A one-number forecast will be off almost every day, and it hides by how much. A range is honest about the spread: it tells you what a normal day looks like here, and how good or bad it can get.
Typical miss from giving each site one fixed value that never changes: its usual viz.
Typical miss from estimating each day's viz in metres from satellite water colour, on the same reports. The fixed value had about half the error.
Often, simply quoting a site's never-changing usual viz is more accurate than trying to estimate the day. Any day-by-day forecast, ours included, has to beat that bar before it adds anything.
The landscape
Services are not named: this is about approaches, not brands. Each row summarises what we could find on that service's own public pages in mid November 2026. Services change, so their pages are the authority on what they offer today.
| Service | Kind of forecast | Diveability rating | Satellite | Weather and waves | Ocean model | Tuned on diver reports | Accuracy measured on dives | Area | Cost |
|---|---|---|---|---|---|---|---|---|---|
| VIZ Forecastthis app | Entry & exit forecast; visibility forecast: usual range per site and warnings; water temperature | ✓ | ✓ | ✓ | ✓ | ✓ | ✓right about 60% of the time: the reported viz fell inside its range (Sydney) | Sydney and Illawarra | Free |
| Service A | Viz range, today and tomorrow | – | ✓ | ✓ | – | ✓ | ✓right about 60% of the time (UK reports) | Several countries | Paid; limited free tier |
| Service B | Viz in metres, days ahead | ✓ | ✓ | ✓ | – | – | Figure claimed, method not disclosed | Global; Sydney as three regions | Paid; limited free tier |
| Service C | Visibility forecast, days ahead | – | ✓ | ✓ | ✓ | ✓ | Figure claimed, method not disclosed | Global | Paid; limited free tier |
| Service D | Viz estimate and a traffic-light rating | ✓ | ✓ | ✓ | – | ✓ | ✓typical miss of about 3 m (mostly Victorian sites) | Australia | Free account for the viz estimate; extras paid |
Key. ✓ the service says it offers or uses this (diveability: a go/no-go or conditions rating for the dive), or for the last column, says it measured its accuracy against real dives. – not mentioned on its public pages, which does not mean it is not used. "Figure claimed, method not disclosed": the service gives an accuracy figure, but its public pages do not say how, or against what, it was measured.
For scale. Simply quoting a Sydney site's usual viz every day misses by about 2.5 m, and a range that is right about 60% of the time is about as good as anyone has shown. Results from other countries or states are not directly comparable with Sydney.
The data columns look much alike: nearly everyone draws on the same satellites and weather models. The difference is the last column, whether anyone has checked the result against what divers actually saw.
Two more forecasting services were left out: one is invite-only, so members cannot use it, and one does not cover Sydney. We also looked at six services that show reports, satellite maps or a dive rating without forecasting viz. Most forecasting services offer a free tier that shows little; the full forecast sits behind plans from a few dollars a month to over A$100 a year. VIZ Forecast is free, with nothing behind a paywall.
Common ideas, tested
We tested the ideas behind visibility forecasting on Sydney data, scoring every test on years it had not seen. These verdicts are about the ideas, not about any service.


Satellites can tell you the viz in metres
Turning satellite water colour straight into metres doubled the error compared with quoting the site's usual viz. Added to the usual viz, it improved the metres only marginally.
Satellite images show murky water
They do, near the surface. On a site's murkiest satellite days, poor viz was reported about half as often again as on its clearest days. But the satellite sees only the top few metres of water.
Big swell stirs up the bottom
The strongest weather signal we found. Soon after a big swell, poor viz roughly doubles, and it held in every year we tested.
Rain ruins the viz
After a very wet week, shore divers reported poor viz more often. The effect is mostly in shallow water near the shore.
Currents and upwelling drive the viz
Modelled currents, layers and sea-bed temperature did not track viz consistently in Sydney reports.
Exact viz, hour by hour, a week out
We could not find a published test showing it, and the 2 m gap between divers on the same day sets a limit on any such number. Our best combination of everything beat the site's usual viz by only a couple of per cent.
Read it this way: even when a warning sign is present, about 70% of dives are fine. Warning signs shift the odds. None of them predicts the metres.
What works best
A recent report from the same site told us more about the next day's viz than any satellite or weather input we tested. A project overseas, built on tens of thousands of daily dive-shop logs, found the same thing.
Sydney has no daily dive-shop logs. It has you. On most sites a fresh report exists on only a small share of days, so every extra post makes the next call better for everyone.

What VIZ Forecast shows
Three example days in the app's own wording. The days are made up to show each case.
Saturday · Bare Island West
Viz: usual for here4–8 m · best viz 9am–1pm · 🌡 19 °C at 15 m
🔭 Usual viz Usually 4–8 m here, going by past VIZ reports.
An ordinary day: the site's usual range, plus the hours most likely to be clearer.
Sunday in July · The Steps, Kurnell
Viz: better odds than usualusually 5–8 m · clearest time of year · best viz 1pm–5pm
🗓 Season This is usually the clearest time of year here.
No warning out and the site's clearest months: the odds lean your way, still without a promised figure.
Tuesday · Bass Point
Viz may be reduced
Viz likely below usualusually 6–13 m · best viz 7am–11am
Why Recent conditions that, in past reports, often came before poorer viz. Most days with a warning are still fine.
A warning that shifts the odds. It does not name the metres.
Hall of fame
Ranked by posts that give a viz figure or verdict, counted from the group's own posts, each post once however many times it was shared or exported. Between them these ten posted 601 viz reports.
And thank you to the hundreds of other members whose posts are in the count. Every one of them is in the data behind this page.
Every result on this page rests on VIZ members who posted the viz, the time and the site, year after year. They let us test ideas honestly, instead of taking anyone's word for it, ours included.
To keep it going, post the viz in metres and the time you were in, like Viz 8 m · ITW 9:00am. In the app, open a site's Details and tap "Copy a VIZ post" for a ready-made post.
About this page
The landscape table summarises information each service published on its own website or app-store listing, as we found it in mid November 2026. Services are not named, and their accuracy figures are not reproduced or re-tested. "We could not find" means exactly that: a test may exist that is not public. Corrections are welcome.
The verdicts are about general approaches to visibility forecasting, tested on Sydney dive reports, and are our own opinion based on that testing. They are not a review of any product. Sydney shore entries are a hard case, and results elsewhere may differ.