Prolepsia forecastYes / no

Will NOAA's Climate Prediction Center report a three-month Niño 3.4 index value (ONI or relative ONI, whichever CPC uses as its official index) for any season from SON 2026 to MAM 2027 higher than the peak of the 2015-16 El Niño on the same index?

As of 17 November 2026, Prolepsia puts the chance of yes at 84%.

84%chance of yes

Yes 84%No 16%
Prolepsia forecast made

How Prolepsia reasons

The report Prolepsia wrote with this forecast.

How forecasts are made · Track record

TL;DR

The forecast is 83.8% Yes and 16.2% No. The current historical benchmark is +2.3°C RONI, so a published +2.4°C in any eligible season would qualify; reaching +2.5°C is not required. CPC’s near-term outlook supports an exceedance, but no qualifying season has yet been published.

Key evidence

  • The threshold is lower than the headline discussion suggests. CPC’s current historical table places the maximum for JJA 2015–AMJ 2016 at +2.3°C, in NDJ 2015–16. The comparison must use the same official index and methodology for both events, not the familiar historical ONI peak (CPC RONI table).
  • The official outlook places the expected peak above the qualifying value. CPC’s November outlook gives median RONI values of +2.50°C for OND and +2.53°C for NDJ. Its approximately 60% chance of reaching +2.5°C concerns a harder threshold than the current requirement for this question (numerical outlook; diagnostic discussion).
  • Observed seasonal strengthening is clear, but does not yet establish Yes. Displayed RONI rose from +1.0°C in MJJ to +1.4°C in JJA, +1.7°C in JAS and +2.0°C in ASO. ASO falls outside the eligible window (official observations).
  • Recent weekly readings provide a caution. Relative Niño-3.4 declined from +2.5°C on October 21 and October 28 to +2.3°C on November 4 and +2.2°C on November 11. These weekly measurements are not the resolving seasonal index (weekly series).

Uncertainties

The main risk is an early peak followed by cooling strong enough to keep every eligible seasonal value at or below the historical maximum. The seven seasons overlap, so they are not seven independent chances to win.

Measurement differences also matter. October’s monthly relative OISST reading was +2.47°C, versus +2.16°C in ERSSTv6, which underlies official RONI (OISST series; ERSSTv6 series). The warmer reading cannot substitute for the official metric. Resolution uses the September 1, 2027 snapshot and published precision: a tie is No. Revisions or a methodological change could affect a close comparison.

Read the full report

TL;DR

The forecast is 83.8% Yes and 16.2% No. CPC’s current same-methodology historical benchmark is +2.3°C RONI, making a published +2.4°C sufficient to qualify (official table). The decisive evidence is CPC’s outlook for an OND–NDJ peak above that level; the main threat is sustained cooling before those seasonal averages are complete (November outlook).

Context

As of November 17, 2026, the question remains unresolved. The latest published seasonal observation is ASO 2026 at +2.0°C RONI, outside the eligible window. CPC’s November 12 discussion describes a strengthening El Niño, while its numerical outlook places the highest median seasonal values in OND and NDJ (historical observations; diagnostic discussion; outlook).

The question is a strict numerical comparison, not a judgment about impacts or whether the event is described as record-breaking. Only SON 2026, OND 2026, NDJ 2026, DJF 2027, JFM 2027, FMA 2027 and MAM 2027 count. Resolution uses the official table available on September 1, 2027, with the historical and current events measured using the same index, dataset and methodology. A reported tie is No.

Evidence

The historical record provides both the benchmark and a caution against assuming that a strong autumn guarantees a new peak. In CPC’s current RONI history, five autumns from 1950–2025 had displayed ASO values of at least +1.5°C: 1965, 1972, 1982, 1997 and 2015. Their subsequent SON–MAM peaks were +2.0°C, +2.0°C, +2.4°C, +2.3°C and +2.3°C, respectively. Only the 1982 event exceeded today’s benchmark. That is a 20% historical frequency, but it comes from a small, varied reference class. It is not a substitute for the current observations and forecast distribution (CPC historical table).

The benchmark itself is decisive. The complete comparison window for the older event contains the following displayed values, all in °C RONI (CPC historical table):

Historical season Displayed RONI
JJA 2015 +1.3
JAS 2015 +1.5
ASO 2015 +1.8
SON 2015 +2.0
OND 2015 +2.2
NDJ 2015–16 +2.3
DJF 2016 +2.1
JFM 2016 +1.8
FMA 2016 +1.3
MAM 2016 +0.7
AMJ 2016 +0.1

The threshold to beat is currently +2.3°C, not +2.5°C. A displayed +2.4°C qualifies if the benchmark remains unchanged. The historical file contains more precision, but that does not override the question’s requirement to compare values at the precision published in the resolving table (CPC table).

The distinction between RONI and traditional ONI explains why the benchmark can differ from familiar accounts of the older event. CPC adopted RONI as its official index effective February 1, 2026. RONI adjusts Niño-3.4 temperature anomalies for broader tropical ocean temperatures and applies a variance adjustment before three-month averaging. The current series uses ERSSTv6 and a 1991–2020 climatology. Comparing the current RONI event against an older ONI peak would answer a different question (implementation notice; RONI announcement).

Current seasonal observations show a steady rise: MJJ 2026 was +1.0°C, JJA +1.4°C, JAS +1.7°C and ASO +2.0°C at displayed precision. These are observations, not projections. They establish strengthening through the latest published season, but none is eligible to resolve Yes (official seasonal series).

The strongest forward-looking evidence is CPC’s numerical November outlook. Its forecast medians and outer percentile ranges are shown below. These are projected seasonal RONI values, not observations (CPC outlook).

Eligible forecast season 5th percentile, °C Median, °C 95th percentile, °C
OND 2026 +2.19 +2.50 +2.82
NDJ 2026–27 +2.00 +2.53 +3.07
DJF 2027 +1.59 +2.28 +2.97
JFM 2027 +1.08 +1.79 +2.50
FMA 2027 +0.60 +1.23 +1.86
MAM 2027 +0.18 +0.74 +1.31

This outlook concentrates the opportunity in OND and NDJ, with DJF providing another route. Later medians decline sharply. CPC’s approximately 60% probability of reaching +2.5°C addresses a higher threshold than the currently qualifying displayed +2.4°C. It therefore should not be read as the probability of this question resolving Yes (diagnostic discussion).

The main contrary signal is recent weekly cooling. CPC’s relative Niño-3.4 series recorded +2.5°C on October 21, +2.5°C on October 28, +2.3°C on November 4 and +2.2°C on November 11. That sequence supports concern about an early peak. It does not establish an equivalent decline in the official three-month ERSSTv6 index (weekly observations).

The measurement gap is visible in the monthly data. Both series strengthened from June through October, but October’s OISST reading rose much further than ERSSTv6. The full supplied comparison is below (relative OISST; relative ERSSTv6).

Month in 2026 Relative OISST, °C Relative ERSSTv6, °C
June +1.04 +1.07
July +1.38 +1.33
August +1.75 +1.67
September +2.09 +2.07
October +2.47 +2.16

October’s gap was +0.31°C. The inference is caution, not an automatic correction in either direction. The warmer analysis cannot replace official RONI, and there is no basis to assume ERSSTv6 must rise to meet it. Nor is there evidence that CPC’s explicitly RONI-based outlook ignored the distinction (CPC outlook; ERSSTv6 methods).

Physical signals are mixed rather than uniformly bearish. Subsurface diagnostics show cooling and western cold water, but substantial eastern subsurface warmth remains. Continued central-Pacific westerlies support the event, while the MJO outlook gives a less clear signal for further reinforcement. These observations argue against treating the weekly decline as proof that seasonal warming has ended (subsurface temperatures; ocean cross-section; MJO assessment). Alternative guidance also retains substantial warmth, but its model-relative monthly values do not substitute for CPC’s official seasonal metric (GFDL November outlook).

What's non-obvious

The largest informational edge is in the definition, not a novel physical signal. A reader focused on the widely quoted +2.5°C outlook can miss that the current same-index historical maximum is lower. The question asks for a strict exceedance of the older event’s RONI peak, not attainment of a headline intensity category. That makes the current qualifying value +2.4°C (historical table).

The second point cuts against overconfidence: seven eligible seasons do not provide seven independent attempts. Adjacent three-month seasons share months, ocean conditions persist, and forecast errors move together. A cooling episode can weaken several candidate seasons at once. The broad window helps, but the outlook shows that the outcome is concentrated around a single expected peak rather than spread evenly across the window (seasonal outlook).

Uncertainties

The central physical uncertainty is whether the weekly decline is a temporary pause or the start of sustained cooling. The lower tail of CPC’s near-term outlook matters more than the possibility of an exceptionally high peak. The evidence supplied does not establish a definitive arrival date for an upwelling oceanic wave or a calibrated numerical wind-stress forecast that settles this issue.

There is also no verified historical calibration here for the current outlook’s errors across overlapping seasons. Better data on joint seasonal forecast errors would clarify how much protection the broader window provides. Archived publication vintages would help distinguish ordinary revisions from differences between observing systems; current revised values alone cannot do that.

Finally, the comparison is close enough for publication precision and revisions to matter. CPC warns that recent RONI observations can be revised for up to two months. The September 1, 2027 snapshot reduces reliance on provisional observations, but the final benchmark must still come from the same official dataset as the new event. A successor index would require a consistent historical series, and a displayed tie remains No (CPC RONI documentation). These risks explain the meaningful No probability without overturning the strong evidence for Yes.

Forecast history

How Prolepsia’s forecast has movedOne forecast so far: 84% on 17 Nov 2026.
DateForecast
84% chance of yes

Sources Prolepsia read

  1. Climate Prediction Center - Relative Oceanic Niño Index (RONI)www.cpc.ncep.noaa.gov
  2. subseasonal to seasonal variations we see in the midlatitudes and over the United States.www.weather.gov
  3. ERSSTv6 methodsrepository.library.noaa.gov
  4. November 2026 El Niño Predictions – Geophysical Fluid Dynamics Laboratorywww.gfdl.noaa.gov
  5. Southern hemisphere monitoringwww.bom.gov.au

The question

As of November 17, 2026, NOAA's Climate Prediction Center (CPC) is monitoring a strengthening El Niño event. In its November 12, 2026 ENSO Diagnostic Discussion, CPC reported that a strong-to-very-strong El Niño was likely through January–March 2027, with a probability exceeding 79%. CPC also estimated approximately a 60% chance that the three-month Relative Oceanic Niño Index (RONI) would reach at least +2.5°C during October–December 2026 or November 2026–January 2027, potentially surpassing previous El Niño events dating back to 1950. ([cpc.ncep.noaa.gov](https://cpc.ncep.noaa.gov/products/analysis_monitoring/enso_advisory/ensodisc.html)) This question asks whether the peak three-month Niño 3.4 index value reported by CPC for the 2026–27 El Niño, within the specified window, will strictly exceed the peak value recorded during the 2015–16 El Niño, measured using the same official CPC index and methodology. The eligible seasons are September–October–November 2026 (SON 2026), October–November–December 2026 (OND 2026), November 2026–January 2027 (NDJ 2026), December 2026–February 2027 (DJF 2027), January–February–March 2027 (JFM 2027), February–March–April 2027 (FMA 2027), and March–April–May 2027 (MAM 2027). These are seven overlapping three-month periods, and the question resolves YES if any one of them exceeds the historical benchmark. CPC adopted the Relative Oceanic Niño Index (RONI) as its official ENSO classification metric in February 2026. RONI measures three-month average sea surface temperature anomalies in the Niño 3.4 region relative to broader tropical ocean temperatures, whereas the traditional Oceanic Niño Index (ONI) measures Niño 3.4 anomalies without that tropical adjustment. CPC currently publishes RONI values based on ERSST version 6, with historical values extending back to 1950. Accordingly, RONI is the expected resolution metric, although the question allows for a subsequent change in CPC's official index. ([cpc.ncep.noaa.gov](https://www.cpc.ncep.noaa.gov/products/GODAS/ocean_briefing_gif/global_ocean_monitoring_2026_03.pdf)) The historical benchmark will be the highest three-month value during the 2015–16 El Niño, calculated using exactly the same index definition, dataset version, climatological baseline, and processing methodology as the values used to evaluate the 2026–27 event. The comparison concerns the magnitude of the reported three-month index, not the duration of El Niño conditions, associated atmospheric impacts, or the highest weekly or monthly Niño 3.4 anomaly.

How it resolves

The question resolves YES if at least one of the seven eligible three-month seasons from SON 2026 through MAM 2027 has a CPC-published official Niño 3.4 index value strictly greater than the highest corresponding index value recorded during the 2015–16 El Niño. The question resolves NO if all seven eligible seasons have official published index values that are less than or equal to the historical 2015–16 peak. The primary resolution source will be NOAA CPC's official historical ENSO index tables, specifically the Relative Oceanic Niño Index (RONI) table at https://www.cpc.ncep.noaa.gov/products/analysis_monitoring/enso/roni/, provided RONI remains CPC's official index at the time of resolution. CPC's official monitoring documentation will determine which index is considered official. If CPC replaces RONI with another official three-month Niño 3.4 index before resolution, that successor index will be used for both the 2026–27 values and the 2015–16 benchmark, provided CPC publishes a consistent historical series covering both events. The 2015–16 benchmark will be established by taking the maximum published three-month value for seasons within the historical event, defined for this question as June–July–August 2015 (JJA 2015) through April–May–June 2016 (AMJ 2016), inclusive. The benchmark and the 2026–27 values must be taken from the same official historical dataset and methodological version. This avoids comparing an older ONI value against a newer RONI value or comparing values calculated using different sea surface temperature datasets. Resolution will use the official CPC historical table as available on September 1, 2027, allowing approximately three months after the final eligible season ends for the publication and stabilization of the MAM 2027 value. CPC notes that its current ERSSTv6-based RONI observations may be revised for up to two months after their initial publication because of the filtering applied to the underlying data. The September 1, 2027 snapshot is therefore intended to avoid relying on provisional estimates. ([cpc.ncep.noaa.gov](https://cpc.ncep.noaa.gov/products/analysis_monitoring/enso/roni/)) If CPC has published values for all seven eligible seasons and the historical benchmark by September 1, 2027, the question will resolve using those values, even if CPC subsequently revises the series. If CPC's official tables are temporarily unavailable, archived official CPC data or CPC's official seasonal index files may be used, provided the same methodology is applied to both periods. If CPC has not published all the necessary observations by September 1, 2027, resolution may be postponed until December 31, 2027. If no consistent official dataset covering both periods is available by that date, the question will be annulled rather than resolved using incompatible indices or unofficial reconstructions. Comparisons will use the numerical precision actually published in CPC's official historical table, ordinarily one decimal place in degrees Celsius. A reported tie does not count as exceeding the historical peak. No interpolation, external recalculation, or unreported additional precision will be used to break a tie.

Fine print

The historical comparison must use the same index for both events. In particular, the historical ONI peak commonly associated with the 2015–16 El Niño must not automatically be treated as the RONI benchmark, because the two indices measure different quantities. The seasons are identified by the calendar year in which their first month occurs, except where conventional CPC notation labels a season by its final calendar year. For avoidance of doubt, NDJ 2026 refers to November 2026–January 2027, and DJF 2027 refers to December 2026–February 2027. Only the seven explicitly identified overlapping seasons are eligible. An exceptional index value occurring in August–October 2026 (ASO 2026) or April–June 2027 (AMJ 2027) does not qualify, even if it belongs to the same broader El Niño event. The comparison is strictly greater than, not greater than or equal to. For example, if the historical benchmark in the selected official index is +2.5°C, a reported value of +2.5°C does not qualify, while +2.6°C does. The +2.5°C threshold discussed in CPC's November 2026 outlook is background context and is not automatically the resolution threshold; the actual benchmark is the historical maximum calculated from the selected CPC dataset. If CPC changes its official index after September 1, 2027, the outcome will not be reopened. Likewise, later historical revisions will not retroactively change an already resolved outcome. A CPC announcement describing the 2026–27 event as the strongest El Niño on record will not independently trigger a YES resolution unless the published index values satisfy the numerical comparison specified above.