ONI vs. RONI: what changed, and why
If you've followed ENSO for a while, you know the Oceanic Nino Index (ONI) as the number: three months of average sea-surface temperature anomaly in the Nino 3.4 region, compared against a fixed 30-year climatology, updated monthly by NOAA's Climate Prediction Center. On February 1, 2026, NOAA began publishing a second number alongside it — the Relative Oceanic Nino Index, or RONI — and made RONI its operational index for ENSO monitoring. ONI didn't go away. It's still published, still tracked, still the number in the history books. But RONI is now the one NOAA leans on for calling El Nino and La Nina in real time.
This page is the explainer we're proudest of on this site, because most ENSO coverage online either hasn't caught up to the change yet or glosses over the "why." We'll walk through both indices, the actual reasoning behind the switch, and — most importantly — what it means when the two numbers don't match for the same season. (They often won't, and that's not a bug.)
What ONI actually measures
ONI is a rolling 3-month average of sea-surface temperature (SST) anomaly in the Nino 3.4 region, a stretch of the tropical Pacific roughly between 5°N-5°S and 170°W-120°W. "Anomaly" means the departure from a fixed climatological baseline — NOAA periodically updates that 30-year base period, but at any given time it's a fixed reference, not a moving one. Each 3-month "season" (like DJF for December-January-February, or JJA for June-July-August) gets its own ONI value, and NOAA has published this series continuously back to 1950. It is, by a wide margin, the most widely cited ENSO index in the world, and it's what almost every "strongest El Nino ever" list — 1982-83, 1997-98, 2015-16, 2023-24 — is built from.
ONI's strength is exactly its simplicity: one region, one fixed baseline, one number per season, consistent methodology for 75+ years. Its weakness is the same fixed baseline. The tropical Pacific, like the rest of the ocean, has been warming over the long run. A fixed climatology from decades ago doesn't move with that warming — so a Nino 3.4 SST reading that would have counted as a moderate warm anomaly in 1985 can look like a smaller anomaly today, purely because the background ocean is warmer, not because the El Nino signal itself is stronger or weaker.
What RONI measures, and the formula
RONI is built to correct for exactly that drift. Instead of comparing Nino 3.4 SST only to a fixed historical baseline, RONI subtracts out the broader tropical ocean's own warming trend at the same time:
RONI = (Nino 3.4 seasonal SST anomaly) − (tropical-mean 20°N–20°S SST anomaly)
In plain terms: take the same Nino 3.4 anomaly ONI uses, then subtract the average anomaly across the whole tropical belt (20°N to 20°S) for that same period. What's left is how anomalously warm or cool the Nino 3.4 region is relative to the rest of the tropics right now — not relative to a climatology fixed decades in the past. That's the "Relative" in Relative Oceanic Nino Index.
Why NOAA made the switch
The motivating problem is basin-wide warming. Global sea-surface temperatures, including the broader tropical Pacific and Atlantic, have trended warmer over recent decades. Because ONI's baseline is fixed, that background warming quietly leaks into the ONI number: raw Nino 3.4 anomalies have been gradually inflated by the rising tide of the whole ocean warming underneath them, not just by El Nino/La Nina activity itself. Left uncorrected, that makes recent decades look more "El Nino-prone" when compared to earlier decades — a distortion that has nothing to do with how ENSO itself is behaving and everything to do with the fixed reference point.
RONI's subtraction of the tropical-mean anomaly is designed to net that background trend out. The result is meant to be a more stable, more era-to-era comparable index: a RONI value from 2026 and a RONI value from 1986 are measuring "how anomalous is Nino 3.4 relative to its own contemporaneous tropical ocean," rather than both being measured against the same aging yardstick. NOAA kept ONI in publication — it remains valuable for its long, unbroken, simple historical record — but adopted RONI operationally because it better isolates the ENSO signal itself from long-term ocean warming.
What this means for the numbers you see
Here's the practical part. Because the current era is a relatively warm one for the tropical oceans overall, RONI will often read lower than ONI for the exact same season — sometimes by several tenths of a degree Celsius. That is expected behavior, not a data error, and not a sign that one of the two indices is "wrong." They're answering slightly different questions:
- ONI asks: how warm is Nino 3.4 compared to a fixed historical baseline?
- RONI asks: how warm is Nino 3.4 compared to the rest of the tropical ocean right now?
In a warm era, the answer to the second question is naturally more modest than the answer to the first, because part of what looks like a warm Nino 3.4 anomaly is really just "the whole tropical ocean is warmer than the old baseline," which RONI subtracts back out. A season that reads as a solidly moderate event on ONI can land in a lower strength band on RONI — and that gap is exactly what RONI was built to reveal.
Reading the two side by side
The qualitative relationship, in recent warm-era seasons, tends to run like this:
| Situation | What tends to happen |
|---|---|
| Warm tropical-ocean era (recent years) | RONI often reads several tenths of a degree lower than ONI for the same season |
| Tropical ocean near its long-run average | ONI and RONI tend to sit close together |
| Cool tropical-ocean stretch | RONI can read slightly higher than ONI, for the same underlying reason in reverse |
We're deliberately not printing live index values in this table — those change every month and belong on the ONI/RONI index page, where both numbers are pulled fresh with their own "data as of" timestamps. This page is about the relationship between the two indices, not a snapshot of either one.
Why we show both, always
This is exactly why Nino34 displays ONI and RONI side by side rather than silently picking one and hiding the other. If we only showed RONI, longtime ENSO followers comparing today's number to 1997-98 or 2015-16 would see a value that looks "too low" with no explanation. If we only showed ONI, we'd be ignoring NOAA's own operational correction for basin-wide warming. Neither index is more "official" in the sense of being the only one that counts — NOAA publishes both, for good reason, and so do we. Check the current ONI and RONI readings to see both numbers for the latest season, with their sourcing and freshness clearly marked.
For how this site turns either index into a phase and strength classification — and why RONI now leads that classification per NOAA's operational convention — see our full methodology page.
Related: