Guide

RONI vs ONI: what changes and why it matters

The Oceanic Niño Index (ONI) and the Relative Oceanic Niño Index (RONI) are both three-month averages of the Niño 3.4 sea surface temperature anomaly. The difference is the reference. ONI compares Niño 3.4 with its own past over a sliding 30-year base period. RONI compares Niño 3.4 with the rest of the tropics in the same month, using a fixed 1991–2020 baseline, and rescales the result. NOAA now uses RONI for its official ENSO monitoring and still publishes ONI. ENSO Tracker plots RONI and does not plot ONI.

Side by side

RONI and ONI compared, with the weekly Niño 3.4 series for contrast
Property	RONI	ONI	Weekly Niño 3.4 (a different product)
Averaging period	Three overlapping months, one new season per month	Three overlapping months, one new season per month	One week, published weekly
Definition	Relative: tropical-mean anomaly subtracted, then rescaled	Conventional: Niño 3.4 anomaly only	Published in both conventional and relative forms
Baseline	Fixed 1991–2020	Centred 30-year periods, updated every five years	Fixed 1991–2020
Source dataset	Relative ERSST v6 (monthly)	ERSST (monthly), version stated on NOAA’s ONI page	OISST v2.1 and Relative OISST v2.1 (weekly)
On ENSO Tracker	Yes, the seasonal chart	No, not ingested or plotted	Yes, the four regional readings and the weekly chart

Reference periods and the tropical adjustment

ONI was designed to answer a simple question: is Niño 3.4 warmer or cooler than it usually is at this time of year? To keep that question fair as the climate warms, NOAA calculates ONI against a 30-year base period centred on the decade of each value and shifts that period every five years. Recent values are therefore compared with recent climatology and older values with older climatology. The method works, but it means an ONI value from 1985 and one from 2025 are measured against different baselines, and the most recent years are compared with a base period that has to be extended forward until enough new data exist.

RONI takes a different route. It keeps one fixed baseline, 1991–2020, and instead subtracts the sea surface temperature anomaly averaged across the tropical oceans for the same month. If the whole tropical belt is 0.4 °C warmer than the baseline, that 0.4 °C is not an ENSO signal, so it is removed before Niño 3.4 is judged. Because subtracting the tropical mean also shrinks the spread of the index, NOAA then rescales the relative anomaly so that its variability is comparable with the conventional one and the familiar ±0.5 °C thresholds still apply. RONI is therefore not simply ONI minus a constant, and the difference between the two is not the same from one month to the next. The precise steps are set out in NOAA’s RONI information circular ↗.

NOAA’s monitoring transition

NOAA’s Climate Prediction Center announced in a public notification issued in February 2026 ↗ that it was moving its operational ENSO monitoring and its historical episode table to the relative index. The monthly ENSO Diagnostic Discussion, the El Niño and La Niña watches and advisories, and the classification of past episodes now follow RONI. ONI continues to be published as a separate record on NOAA’s ONI page ↗, so both series remain available. Check the notice and NOAA’s current pages for exact effective dates; this page describes the change rather than dating it.

Why historical comparisons can differ

The two indices tell nearly the same story in the middle of the twentieth century, when the tropical mean anomaly relative to the baseline was small. They diverge in recent decades. As the tropics have warmed, conventional Niño 3.4 anomalies have carried a growing warm component that RONI removes. Warm events therefore tend to look somewhat weaker in RONI than in ONI, and cool events somewhat stronger, with the gap widest in the most recent years. Under a relative definition, some borderline periods that ONI classified as neutral can meet the La Niña threshold, and some marginal El Niño seasons may not. This is why NOAA publishes a historical episode table based on RONI alongside the ONI table, and why rankings built on ONI should not be mixed with rankings built on RONI. When comparing a current value with a past event, use the same index, the same dataset version and the same season definition on both sides.

No ONI values appear on this site, so no current RONI-to-ONI difference is quoted here. The magnitude of the adjustment in any given month can be read by placing the two NOAA tables side by side; the values must come from the same dataset version to be comparable.

What this site displays

The seasonal chart and the “Seasonal RONI” reading on the homepage are RONI, taken from NOAA’s published file, with the reference lines at ±0.5 °C and shading by threshold. The regional readings and the weekly chart are the separate weekly Niño region products, in relative form by default and conventional form on request. Neither weekly series is ONI, and averaging weekly values does not create ONI or RONI. If you need ONI itself, NOAA’s table is the source; ENSO Tracker links to it rather than reproducing it. Historical shading in the chart follows the threshold rule, while NOAA’s episode classification additionally requires persistence over five consecutive overlapping seasons, so shaded cells and classified episodes are related but not identical.

Open the seasonal RONI record →