GaWx

Look how much the forecasted mean GEFS NAO has risen from its Oct 13th forecast to today’s forecast! Oct 13th: peaking at only +1.0 on Oct 15th and then plunging to -1/remaining down at -0.9 on Oct 27th: Today’s prog (Oct 19th): peaked not til today way up at +1.73 and already after a dip to -0.8 back up to ~0 on Oct 27th: I’m assuming this drastic change is related to the actual strong storm offshore the E US that wasn’t even forecasted back on Oct 13th thus leading to lower actual SLP in Iceland today than progged 6 days ago. Daily NAO: https://ftp.cpc.ncep.noaa.gov/cwlinks/norm.daily.nao.gefs.z500.120days.csv

After a decent rise in last week’s release from +0.59 to +0.88, the latest weekly IOD update had no change. But that could easily just be a short-lived pause in the general upward trend since June 1st. Based on the history of IOD usually maxing out by Nov, I’m giving it another month to see if we get a very strong IOD peak this late autumn/early winter. One can see on this chart how much stronger was the +IOD in autumn 2023! At this time then, the IOD was already peaking way up at +1.91! Even in 2019 on this date it was way up at +1.94! https://www.bom.gov.au/clim_data/IDCK000072/iod_1.txt?1791781094025 @snowman19

@LibertyBellasked me this: “Are these similar to the naturally occurring doldrums at the horse latitudes that Columbus experienced?” ———— My answer: Good question. Quite possibly.

I closely followed the DMI site’s 80+N Arctic temps showing 2026 being THE coldest on record (back to 1958) by a good margin late May-early August and posted many times about that here and in the Arctic Ice thread. There’s little doubt it was notably colder there than recent summers, especially N of Alaska, and it helped the Arctic ice. However, as@bluewave@OSUmetstudemphasized in discussions with me, they convinced me to think that comparing to DMI charts prior to 2002 per Chris (see his quote below) and possibly even to all years through 2025 per @OSUmetstud (due possibly to a cooler ECMWF bias introduced in May of 2026) MAY make the appearance of the summer of 2026 being significantly colder than all years 1958-2025 per the DMI graphs deceptive because of the DMI ECMWF data. Per Chris in Arctic Ice thread: “Even the less reliable ECWMF only starts in 2002, so an apples to apples comparison to any year earlier than 2002 can’t be made.” —————— However, I’m still very open-minded about just how cold it really was in 80N+ late May-early Aug relative to 1958-2025, especially 2002+, pending continued research of other knowledgeable people’s opinions. Look how cold June 2026 was N of 80N per actual ERA5 reanalysis: 1. only ~1/7 is in warmest 1/2 of years vs ~6/7 in coldest 1/2 2. a little >50% is in 4 coldest colors meaning coldest 11 of 87 while only ~6% is in warmest 11, which is <1/8 the size of coldest 11 3. ~25% is in coldest 3 of 8 while only ~1% is in warmest 3 of 87, which is a mere 1/25 of the size of coldest 3 This 80N+ close-up came from this image below: only the relatively small area (~1/7 of 80N+) of light green, beige, yellow, orange, and reds is warmer than the median while >50% (4 coldest colors or coldest 11 of 87) is 87%+ percentile cold! So, does this cold of an ERA5 reanalysis (in image above) actually support this cold of a DMI 80N+ June ‘26 (in image below)? Food for thought at least. Keep in mind that it says this at the DMI site, itself: “The 'plus 80 North mean temperature' graphs can be used for comparing one year to an other.”

Thanks. So, since this pattern of this offshore low and associated cooler air in the E US is likely indirectly related to the very strong Nino, is this possibly a sign of things to come this winter?

Is the offshore E US strong low bringing down the cool air into the E US this entire week likely related to the very strong Nino? 0Z 10/12 GEFS forecast for 6Z 10/21: no offshore storm and thus no cool air pinned in E US: 0Z 10/18 GEFS forecast for 6Z 10/21: much cooler E US in reality associated with offshore strong storm working in tandem with 3 Canadian highs keeping cool dry Canadian air pinned in E US: Cool air this week E US thanks to the above pattern consistent with Euro Weeklies’ significantly cooler change: (had been warm on runs from Oct 12-13 when no offshore storm)

^Further to the above: illustrates the E US cooling power of the offshore storm this week!

1. Temps: Euro Weeklies 10/12 run for 10/19-25: warm E US because there was no offshore storm on model yet 10/18 run for 10/19-25 is much cooler E due to strong offshore storm: thank you storm! ——————- 2. Precip for 10/19-25: 10/12 run: no offshore storm 10/18 run: strong offshore storm keeping it cool in E US!

The non-tropical low now forming gas gone orange: Tropical Weather Outlook NWS National Hurricane Center Miami FL 800 PM EDT Sun Oct 18 2026 1. Off the East Coast of the United States: A storm-force non-tropical low pressure area is developing along a frontal boundary tonight off the southeast coast of the United States. This low is likely to move slowly and erratically over the next couple of days while it initially remains non-tropical. Later this week, if the system stays over the Gulf Stream, environmental conditions could become more conducive for the low to develop some subtropical characteristics. A subtropical cyclone could form over the second half of this week as the system begins moving generally northeastward, offshore of the eastern United States. Regardless of development, this system will cause high winds, coastal flooding, and dangerous surf along portions of the U.S. East Coast. Please consult statements and warnings from local National Weather Service offices and the Weather Prediction Center for more information, along with High Seas Forecasts and Storm Warnings issued by the National Weather Service. * Formation chance through 48 hours...low...near 0 percent. * Formation chance through 7 days...medium...40 percent.

Thanks. Do you know whether or not it’s normal for +AAM to be prevalent within 30-50N at any point during a strong El Niño? Or does it tend to stick closer to the equator? Have you seen any actual data? The only other El Niño for which I can see latitude based relative AAM from this site’s data is 23-4 because it only goes back to 2022. Reviewing 23-4, the 30S to 30N region had the main +AAM only during Dec 1st-Jan 15th and it never exceeded +3 SD for any latitude. That paled in comparison to Sept 14th-Oct 3rd of ‘26, when it exceeded +4 SD in portions of 15S to 15N!

POSSIBLE LOW LEVEL INVEST NEAR 36.0N 71.0W FOR 20/1800Z.

Ray et al, A similar theme to what Ray emphasized last month remains even nearly a month later where the really impressive +AAM emphasis is mainly 30S to 30N: Is it normal during El Nino for the +AAM (atmosphere rotating faster than the solid earth as I understand it), which we know is characterized as El Niñoish, to usually be strongest mainly in the 30S to 30N latitudes (with the highest near the equator) since that is nearer to the Nino regions than latitudes outside of that region? If so, do the regions outside of 30S to 30N normally get consistently positive during strong El Niños? Do they ever consistently average +AAM for long periods or is that typically relegated to 30S to 30N even during the subsequent N hem. winter? ———————— By the way, the relative AAM by lat. above is not surprisingly still supporting a very strong global relative mean +AAM with it back >+3 since Oct 12th:

The many upcoming chilly nights throughout the SE that have started will accelerate the autumn leaf changes everywhere, even where it’s too warm for frost. Is anyone seeing significant colors yet? I expect higher elevations almost certainly are, but what about those at 1,000 feet or lower? Here, the deciduous trees are almost all still green. By the end of the week, however, I wouldn’t be surprised to see a few starting to change even this far SE near sea level.

About as beautiful and comfortable a full late Oct week as is possible for outdoor activities is in the cards this week for a large portion of the SE away from coastal NC with lots of sunshine, high temps in the 60s-70s, mainly 15 mph or lower winds after today, and dewpoints down in the 30s-40s. We can thank persistent mainly N to NE winds resulting from three sprawling areas of Canadian high pressure centered to our NE, helped by persistent low pressure offshore the E coast especially the first half of this week. Enjoy it while you can, especially with the amount of rainfall expected Nov-March due to the very strong Nino.

Thanks, Adam. I’m certainly not worried about a persistent SE ridge this winter!

Tropical Weather Outlook NWS National Hurricane Center Miami FL 800 PM EDT Sat Oct 17 2026 For the North Atlantic...Caribbean Sea and the Gulf of America: 1. Off the East Coast of the United States: A strong non-tropical low pressure area is expected to form by Sunday night off the southeast coast of the United States. This low is likely to move slowly and erratically through the middle of next week, and some subtropical development is possible during this time. Regardless of development, this system will cause high winds, coastal flooding, and dangerous surf along portions of the U.S. East Coast. Please consult statements and warnings from local National Weather Service offices and the Weather Prediction Center for more information, along with High Seas Forecasts issued by the National Weather Service. * Formation chance through 48 hours...low...near 0 percent. * Formation chance through 7 days...low...30 percent.

To add to the above, the ‘23 Jul-Sep MJO was much more MC/IO oriented than the nonexistent MC/IO of ‘26: ’23: ‘26:

Doldrums lead to ‘23 FL Keys coral disaster as it wasn’t just due to GW…it was warming of several degrees from the doldrums superimposed on background ocean warming of several degrees due to GW: Sea-surface temperatures in the Coral Sea off northern Queensland often increase during periods known as the "doldrums," characterized by weak winds, reduced cloud cover and dry, descending air. These conditions are unusual for the humid tropics, where we would typically expect plenty of cloud cover, thunderstorms and rainfall.

12Z UK: H in N ATL: NEW TROPICAL CYCLONE FORECAST TO DEVELOP AFTER 36 HOURS FORECAST POSITION AT T+ 36 : 32.7N 70.7W LEAD CENTRAL MAXIMUM WIND VERIFYING TIME TIME POSITION PRESSURE (MB) SPEED (KNOTS) -------------- ---- -------- ------------- ------------- 0000UTC 19.10.2026 36 32.7N 70.7W 1007 44 1200UTC 19.10.2026 48 33.7N 71.3W 1004 43 0000UTC 20.10.2026 60 35.0N 71.2W 1002 41 1200UTC 20.10.2026 72 36.0N 70.5W 1001 39 0000UTC 21.10.2026 84 36.0N 70.0W 1001 45 1200UTC 21.10.2026 96 36.0N 70.2W 998 42 0000UTC 22.10.2026 108 36.2N 70.5W 996 45 1200UTC 22.10.2026 120 36.9N 69.2W 992 42 0000UTC 23.10.2026 132 38.1N 66.5W 987 48 1200UTC 23.10.2026 144 40.9N 61.1W 978 67 0000UTC 24.10.2026 156 45.8N 56.8W 977 55 1200UTC 24.10.2026 168 47.9N 58.0W 996 43

Nothing’s even close to a certainty. 2026 and 2023 are not even close in magnitude and 2023 never even got stronger than high end moderate as well as being way less E based in winter (close to basin-wide) vs where 26-7 is heading: these are the official relative anoms: 1+2 3 4 3.4 2023 9 2.27 1.51 0.45 0.91 2026 9 3.94 2.97 -0.02 2.06 2023 12 0.68 1.48 0.73 1.38 2024 1 -0.02 1.15 0.70 1.03 2024 2 0.16 0.78 0.50 0.84 Also, 2026’s PDO is much less negative than 2023’s: 2023 -1.24 -1.79 -2.50 -3.13 -2.46 -2.67 -2.44 -2.02 -2.52 -1.88 -1.16 -1.20 2024 -1.48 -1.33 -1.27 2026 -0.87 -0.79 -1.18 -1.28 -1.64 -1.32 -1.35 -1.12 -0.74

The 0Z UK is nowhere near the E coast. But it has a very high end TS, which based on the 976 mb SLP might be too low: NEW TROPICAL CYCLONE FORECAST TO DEVELOP AFTER 36 HOURS FORECAST POSITION AT T+ 36 : 32.8N 71.4W LEAD CENTRAL MAXIMUM WIND VERIFYING TIME TIME POSITION PRESSURE (MB) SPEED (KNOTS) -------------- ---- -------- ------------- ------------- 1200UTC 18.10.2026 36 32.8N 71.4W 1010 44 0000UTC 19.10.2026 48 32.0N 70.5W 1008 42 1200UTC 19.10.2026 60 32.4N 70.3W 1008 39 0000UTC 20.10.2026 72 34.2N 70.2W 1005 39 1200UTC 20.10.2026 84 34.7N 70.0W 1002 40 0000UTC 21.10.2026 96 34.7N 69.9W 1001 41 1200UTC 21.10.2026 108 34.5N 68.9W 1000 39 0000UTC 22.10.2026 120 35.3N 68.6W 997 42 1200UTC 22.10.2026 132 36.5N 67.0W 994 44 0000UTC 23.10.2026 144 38.5N 64.9W 986 56 1200UTC 23.10.2026 156 40.2N 60.6W 980 58 0000UTC 24.10.2026 168 43.3N 53.8W 976 56

As my analysis shows above, the Oct Euro hasn’t even been close with its winter prog the last 4 years: 24 25: too warm 22 23: too cold Out of the last 9 years, it didn’t do well 6 of the 9 years. So, it’s likely pretty clueless based on this.

On 10/5, I posted a detailed analysis of how the Oct Euro did for winter for the 9 years 2017-25: Tally: -did well in 19 and esp. 17 -did pretty well 21 -did pretty badly 23, 22, 20, 18 -did terribly 25, 24 -too cold both El Niños in SE and 1 of 2 in NE -NE: too warm on 4, close on 3, too cold on 2 -SE: too warm on 3, close on 2, too cold on 4

In terms of the degree of it being E based, the @griteater graph below suggests it will likely be nearest 1998 but with 1983 and 1878 not being far off. When also considering that 1998 wasn’t nearly as good a QBO match as 1983 and 1878 is unknown, perhaps 1982-3 is the best analog of these. Your thoughts?

What do you think about the possibility of unmodeled negative feedbacks eventually stopping the warming and thus causing at least the most extreme climate model solutions to end up being way overdone? Is this a realistic possibility?