GaWx

For the first time in awhile, Nino 3.4 didn’t warm per OISST. Instead, there was a notable cooling of 0.065C (see image below). After such an impressively strong warming since May 31st, this isn’t surprising. To me it is just El Niño taking a temporary breather of sorts. Ups and downs always happen. It’s still up at ~+1.0 RONI/+1.5 ONI equivalent snapshot. I expect the next round of notable warming to be later this month or in early July at the latest, especially if the -SOI persists, as that’s what ENSO models are suggesting.

I obviously agree with your last sentence as that’s essentially what my quoted 2026 sources said and it’s logical to me, regardless. Regarding the first part, I don’t see why RONI shouldn’t be the preferred way for historical rankings. As far as heat being released into the atmosphere: we know that heat released into atmosphere increases as SSTs rise from El Niño. But we know that the atmosphere also directly warms up from AGW, which adds to SST warming by what essentially is the equivalent of ONI less RONI more or less. In the upcoming case, a huge amount will be imparted by what has a good chance to be a record breaking RONI or at least close to a record. All ENSO regions are looking to be quite warm and thus will contribute a large amount of heat. That’s different from a severely E based that may not contribute as much heat because of the other ENSO regions not being as warm. *Edited to improve my wording about portion of SST increases due to AGW.

With the way you’re wording this, it seems to me like there’s double counting of sorts between the warming from AGW, itself, and El Niño.

Thanks, Chris. The following two sources are very recent (2026) and are in support of RONI over ONI: New NOAA El Niño-Southern Oscillation Index Supports Drought Early Warning RONI accounts for the long-term ocean temperature trends in a way that the traditional ONI does not, thus providing a better representation of the seasonal climate variability. Not accounting for the warming oceans means recent El Niño temperature anomalies look bigger than they really are, and La Niña anomalies look smaller. Consider the last six winters: Five of the last six winters saw a La Niña pattern, which is one of the reasons much of the southern half of the Nation experienced persistent and recurring drought throughout the first half of this decade. Using the traditional ONI, 2020-2023 experienced a “triple-dip” La Niña, but RONI classified this as one continuous, unbroken La Niña event that lasted three years. These years were also very dry across the Southwest and Southern Plains. The 2024-2025 and 2025-2026 winters were ENSO-neutral using the traditional ONI. Using RONI, these events would instead be classified as definitive La Niñas. Winter 2025-2026 brought Moderate to Severe Drought (D1-D2) or worse for a broad swath of the Southern U.S. from Arizona to Florida. The La Niña pattern in place since around August 2025 is a primary driver of this drought. This season mostly followed the traditional La Niña response, with a few exceptions https://www.drought.gov/news/new-noaa-el-nino-southern-oscillation-index-supports-drought-early-warning-2026-03-11 ———————————————— Relative Oceanic Niño Index (RONI) A clearer, more reliable way to track El Niño and La Niña More reliable in real-time: ONI depends heavily on the choice of a 30‑year average. As tropical ocean temperatures shift over time, that reference can lag behind current conditions. RONI reduces this sensitivity, leading to more consistent classifications. More stable ENSO classifications RONI is less sensitive to which 30‑year reference period is used, meaning that the classification of past El Niño, Neutral, and La Niña events is more stable. Clearer connection to impacts RONI better reflects when ENSO‑related atmospheric patterns are actually present, improving alignment with seasonal temperature and precipitation outlooks. How is this change to RONI going to benefit decision makers who rely on an accurate ENSO forecast? RONI is better able to capture the expected changes in the atmosphere inherent to ENSO, and El Niño/La Niña categorization is more stable even as the climatology changes. Because RONI is more accurately reflecting the ENSO state, users will be able to more accurately make decisions that rely on associated seasonal forecasts and risk assessments. https://www.cpc.ncep.noaa.gov/products/analysis_monitoring/enso/roni/pdf/roni-info-may-2026.pdf

Note that Nino 3 anomalies are stronger than Nino 1+2. So, this prog isn’t severely E based as of 2/1/27.

1877-8 followed by cold neutral and then weak La Nina 1888-9 followed by strong La Niña and then weak La Niña But in neither of these cases were they immediately preceded by La Niña like ‘26.

Thanks. Some of us have already been discussing ‘26-7 vs ‘97-8. On Saturday I posted that per the Euro progs, ‘26-7 is looking to evolve into a much less E based Nino than ‘97-8, which became the most E based Nino on record (back to 1951-2). When I refer to E based, I’m referring to the differences between 1+2 and 3 vs 3.4 and 4. This is what I posted: More on 97. Here were the monthlies: 1+2 3 4 3.4 1997 6 26.12 3.00 28.13 1.51 29.23 0.26 28.82 1.09 1997 7 25.59 3.63 27.90 2.09 29.37 0.47 28.86 1.56 1997 8 24.96 3.96 27.71 2.59 29.29 0.50 28.75 1.89 1997 9 24.69 3.96 27.74 2.84 29.44 0.68 28.85 2.13 1997 10 24.69 3.67 28.06 3.08 29.34 0.58 29.08 2.36 1997 11 26.12 4.47 28.37 3.27 29.39 0.69 29.12 2.41 1997 12 27.06 4.25 28.53 3.30 29.11 0.57 28.89 2.29 1998 1 28.12 3.55 28.74 3.08 28.95 0.64 28.93 2.38 1998 2 28.74 2.64 28.90 2.49 28.79 0.59 28.78 2.03 1+2 (peak 4.47) and 3 (peak 3.3) were way, way stronger than 4, which remained just wk (peak 0.69) and even way stronger than 3.4 (peak 2.41)! The forecast for ‘26 is much stronger in 3.4/4 and stronger in 3, but not as strong in 1+2: Per latest Euro, here are peak non-rel. monthlies: 1+2: 3.9 3: 3.9 3.4: 3.7 4: 2.4 That’s only a contrast of 1.5 from E to W compared to 3.8 in ‘97! And contrast from 1+2 to 3.4 is a mere 0.2 vs 2.1 in ‘97!! So, the prog is for ‘26 to be E based but way more balanced and thus not nearly as E based as the extremely E based ‘97.

Since yesterday, the non-AI ensemble progged Gulf tropical activity has diminished other than for perhaps the Bay of Campeche, which would be favored to landfall in MX. (The AI never picked up on anything much.)

Do you have this map for other states in the SE? TIA

Thanks, Chris -Long range model winter forecasts have, indeed, tended to be too cold on average based on my recollection although the Euro has missed on both sides. So, I wouldn’t be shocked if 25-6 also comes in warmer than today’s forecast, but that remains to be seen. -But keep in mind that you essentially are showing a super Nino sample size of just one (2015-6 done in August) since 2023-4 wasn’t a true super (RONI peaked at only +1.49). ONI had ~0.5C excess due to very warm surrounding tropical waters from GW inflating the ONI. -Also, other super Ninos had a fairly similar temp. pattern to what the NMME shows overall.

-Unlike Cansips, these NMME maps look typical/more believable for a strong Nino. -The strong Nino very typical wetness (starting as early as Oct on this run but I usually bank on the start in Nov) gives a lot of hope that the bad but somewhat improved drought in the SE will be mainly history by mid to late winter. - These maps are consistent with the El Niño winters that had one major to possibly historic SE US snowstorm like has occurred in many El Niños.

Nino 3.4 warmed only 0.2 in today’s release for last week averaged out vs the prior week from 0.5 to 0.7 vs my guess based on daily OISST levels/changes of it warming 0.3 (from 0.5 to 0.8). But this difference could possibly be mainly due to rounding. For example, perhaps the prior week was, say, +0.46 vs last week being, say, +0.74 or whatever. It has gotten more E based the last 5 weeks but (as per what I recently posted) the Euro doesn’t have it getting more E based overall from this point forward and thus keeps it from getting anywhere near as E based as 1997-8: 1+2 3 3.4 4 29APR2026 0.6 0.4 0.4 0.5 06MAY2026 1.0 0.5 0.4 0.5 13MAY2026 1.3 0.6 0.4 0.6 20MAY2026 1.6 0.7 0.5 0.6 27MAY2026 1.7 0.8 0.5 0.7 03JUN2026 2.1 1.0 0.7 0.7

Thanks. I realize that the RDU sensor is often a hot spot under sunshine in the warm season per past discussions I’ve read here and elsewhere. Do you know if your house had temps that jumped around ~6 F within just a few minutes of 3 PM? That seems very high with little change in sky condition (mostly sunny as opposed to large amounts of cloud cover suddenly moving in or out) and no big wind direction shift!

Fake news from RDU??TODAY MAXIMUM 99 259 PM CLI from KRAH ——————-NORTH CAROLINA WEATHER ROUNDUP NATIONAL WEATHER SERVICE RALEIGH NC 300 PM EDT SUN JUN 07 2026 RALEIGH-DURHAM MOSUNNY 93 57 29 VRB3 29.93F RWR from KRAH ——————-So, the 99 high was recorded at 2:59PM whereas the 3PM hourly reading was only 93. Sure, yeah right. This is truly bizarre! And the 99 seemed too high when considering surrounding stations!

Are we headed for another 2023-4 where SSTs are warm around Australia? Per JB, the models that showed for winter 2023-4 at H5 the mean E US trough/Aleutian low (most of them) didn’t have it that warm there and thus he blamed the unforecasted warmth around Australia.

After a very strong warming the prior 5 days averaging a very impressive 0.1C/day, the latest OISST 3.4 finally slowed its warming although it still warmed slightly (.014). Now that the full week’s data is in, I’m guessing 3.4 will be warmed by 0.3C to +0.8C in tomorrow’s release:

The MJO ended up going strongly into phase 8 (see image below) after all despite models a couple of weeks ago going only modestly into it and extended ensembles a month ago turning left in 7 toward the circle/missing 8. Bastardi back in April predicted W Car/Gulf TCG in June based on his expectation the MJO would go into 8 despite the extended ensembles not showing that. He more often than not predicts early activity. So, that’s not unusual. But he was right about the MJO. Will he be right about the Atlantic basin June TCG? Models/ensembles are hinting at the chance for this late this week:

Thanks, snowman. A key will be to see whether or not 3.4/4 fall severely behind regions 3/1+2 to determine whether or not this will be a very strong form of E based. For those who prefer it be/not be overly E based like ‘97, you’ll want 3.4 to not warm too steadily/keep warming steadily through the summer/early fall overall.

The Memorial PGA event, hosted by Jack Nicklaus, in Dublin, OH, has a Hickory, NC native, JT Poston, with a big 4 shot lead over Raleigh, NC native Ryan Gerard headed into rd 4. The Memorial is easily considered a top 3 event in terms of field, prestige, and difficulty of course after the Majors/TPC. Yesterday it not only had 2 rain delays, it had from a supercell penny sized hail falling live on TV! That’s a first for me to see at any golf tourney. One thing that contributes to my enjoyment of watching golf is the strong wx influence.

Thanks. The 0Z EPS cut back significantly on the % of members with Gulf TCG vs yesterday’s 12Z run. It will be interesting to see how future runs compare. The 6Z GEFS has significantly increased activity vs earlier runs. Edit: The 12Z GEFS is significantly quieter than that more active 6Z GEFS.

Yesterday’s 12Z EPS had nearly 50% of its members with a TCG in S Gulf/NW Car late next week with most of those then heading to the U.S. Gulf coast the following week. However, the 0Z EPS cut back significantly on the % of members with Gulf TCG vs yesterday’s 12Z run. It will be interesting to see how future runs compare. The 6Z GEFS has significantly increased activity vs earlier runs.

It sometimes updates later but, yes, it often does update in the middle of the night our time. The 24 hour period that each daily is based on evidently ends at about the average middays of Darwin and Tahiti.

It turns out that the models were right on as today’s Darwin SLP is 1017.70, the earliest that high so early in the season and the 8th highest on record: 7 Jun 2026 1014.05 1017.70 -34.73 Darwin will be lower tomorrow. Based on history, that 1017.7 could easily end up the highest daily Darwin SLP of 2026. But there’s no telling for sure as an SLP that high remains possible through September (end of their winter).

Thanks. If a Gulf TC were to materialize in mid June, then JB will get a big win on a call he made way back in late April as I recall based on projecting MJO phase 8 in early or mid June though he had been leaning more toward a week earlier as he even admitted.

Thanks, Chuck. Consistent with that, 2025 has by a large margin the strongest +AO for the period I was analyzing, May 23rd-June 4th from 1958 through 2026. That period in 2025 was chilly (~1.5C BN) though not nearly as cold as the ~-3.5C of 2026. By the way, 2026 had the 4th strongest +AO of this period back to 1958 thus intuitively helping it be one of the coldest. 2023 had the 10th strongest +AO and was ~1C BN. 1994 had the 2nd strongest +AO but it was only ~0.5C BN. 1996 had the 3rd strongest +AO, but it was NN rather than cold. So, the correlation of +AO to cold N of 80B is far from perfect due to other factors coming into play as you very likely realize even though there’s still a decent partial correlation.