Warm Water — Clear or Murky? The Correlation Flips by Region, Measured Across 30,000 Dives

"Warm southern seas are blue and clear." Half true, half false. When we compute the correlation between water temperature and visibility from our measurements, the sign flips depending on the site. Places where "warm = clear" and places where "warm = murky" both exist within Japan.

The numbers: correlation runs both positive and negative

From observations recording both visibility and temperature, we computed the Pearson correlation coefficient (−1 to +1) per site.

Hirasawa's −0.42 and Echizen's +0.34 are opposite signs. Same two variables, yet one says "colder is clearer," the other "warmer is clearer." That reversal is the whole point.

Why it flips — Hirasawa (warm = murky)

The monthly breakdown explains it:

• February: 15.9°C, 11.4m visibility (cold and clear)
• August: 25.6°C, 7.1m visibility (warm and murky)

In Izu's sheltered bays, warming water coincides with plankton blooms — Kuroshio-borne nutrients plus sunlight drive phytoplankton that cloud the water. Winter cools the water, plankton fade, clarity returns. Hence "colder is clearer." This negative correlation is exactly why Izu divers prize winter visibility (though winter Izu demands a drysuit).

Why it flips — Echizen (warm = clear)

The Sea of Japan is the opposite:

• March: 11.4°C, 7.3m (cold and murky)
• Aug–Sep: 25–27°C, 10.3m (warm and clear)

On the Japan Sea coast, winter is the stormy season; rough seas stir bottom sediment and cloud the water. Summer brings the warm Tsushima Current and calm seas — warm and clear together. Here what governs clarity isn't temperature but whether the sea is rough, and that happens to follow the same seasonal rhythm as temperature, producing the correlation.

So: correlation is not causation

The single most important caveat. Temperature does not control visibility. Common underlying factors (season, plankton, waves, currents) move both at once. Therefore:

• In Izu, "it's warm today so it's probably murky" tends to hold — but because of plankton, not temperature
• Make the same guess in Echizen and you'll be wrong
• Carrying another site's rule of thumb to a new site inverts it — the classic mistake of the traveling diver

Practical: know your site's sign

• Memorize just the sign of your destination's correlation (table above) and you can roughly read visibility odds from the temperature forecast
• But the correlation is weak (±0.4 at most) — don't bet on it. Check recent measurements in the archive before you go (last week's reading is your best forecast)
• Chasing days when both are good is covered in our jackpot-rate article

Note: A correlation coefficient describes the direction and strength of a relationship, not causation or any single day's guarantee. Figures come from this site's measured distributions.

Data sources

• Visibility & temperature measurements: this site's database (2006–2026; observations recording both, per site, minimum 300 each), browsable in the archive