bluewave

The recent dry pattern stretching back to last summer has mostly affected the soil moisture and vegetation. It was really only parts of NJ that had issues with the very low reservoirs last fall. As the NYC watershed upstate didn’t run into any low water supply issues. Long range precipitation forecasting is very low skill. We had a 500 year drought back in the 1960s followed by a steady increase in rainfall. While the drought from 1999-2002 did result in NYC watershed restrictions and low reservoir levels, it wasn’t as bad as the 1960s. We entered a much wetter pattern back in 2003. All our dry periods since then have been short and the rebound to wetter following them was more impressive. So we will need more time to see if the major drought covering Canada and parts of the CONUS is some type of new pattern the last few years. In the mean time the dry pattern has been influencing or daily temperature range. The low temperatures have been running significantly cooler than the highs. So this has lead to the great radiational cooling in places like Walpack, NJ. But the high temperatures have been beating guidance on the days with close to full sun and offshore flow.

Both the GFS and Euro have 850 mb temps which would support 90° highs by later next week at the usual warm spots if the flow becomes SW with full sun. But they have been wavering on whether we get SW or more onshore flow and amounts of clouds. So if the flow stays more onshore, then the warm spots will probably only top out in the mid 80s.

The EPS longer range was too cool for yesterday. It originally had mid 70s from its older forecasts. But the warm spots it to the low 80s yesterday. Data for September 11, 2025 through September 11, 2025 Click column heading to sort ascending, click again to sort descending. SOMERSET AIRPORT WBAN 83 NEWARK LIBERTY INTL AP WBAN 82 Newark Area ThreadEx 82 TRENTON-MERCER AIRPORT WBAN 81 Trenton Area ThreadEx 81 MILLVILLE MUNICIPAL AIRPORT WBAN 80 ATLANTIC CITY INTL AP WBAN 80 TETERBORO AIRPORT WBAN 80 CALDWELL ESSEX COUNTY AP WBAN 80 SOUTH JERSEY REGIONAL AIRPORT WBAN 80 Atlantic City Area ThreadEx 80 PENNSAUKEN 1N COOP 80

Models bring back the mid to perhaps upper 80s again going forward for the usual warm spots. Much drier pattern returns as the recent rains were just a blip in the expanding drought pattern. The cool departures will begin shrinking. If we get a warmer Sep 21 to 30, then the month could finish with a + departure.

It didn’t work out this past winter and we saw one of the strongest +PNA patterns from December into January since 1948. But you are correct that prior to this winter it was a good indicator. Even though the CPC registered ad a +PNA in February, the 500mb PNA had a more negative look. In any event, the winter PNA has been trending more positive since 1948.

That’s why during the fall of 2023 I was pointing out the WPAC to IO would have more influence than we typically get with such strong El Niños. It lead to me doubting the Euro seasonal forecast of the typical Nino trough with colder conditions in the Mid-Atlantic and Southeast. The ridging in Canada was also stronger and more extensive than we typically see with El Niño events building down into the East which we usually see with La Ninas. So effectively a hybrid pattern with the warmest features of an El Niño and La Ninas merged together. We got early hints this would be the case back in May 2023 when the historic ridge developed over Canada leading to the record wildfire season. So there are clearly multiple areas of record SST warmth contributing to the overall pattern. In the old days we could just look at Nino 3.4 and have a reasonable assumption of what the pattern would look like. Now as I have been saying we have to deal with competing influences from these numerous areas of record SST warmth. This has been one of the reasons these seasonal models have been having so much of a challenge with the recent winter forecasts. The stock El Niño and La Ninas composites from the colder climate era are not manifesting in the same ways anymore. We first got a sample of this when the El Niño forecasts in 2018-2019 and 2019-2020 failed to couple due to the record SST warmth from the IO into the WPAC.

Yeah, the last cold and snowy month around our area in January 2022 actually had a strong +NAO pattern. We were able to get that great MJO 8 and have one cold snowy winter month. But the ridge axis over the Rockies was a little too far east so this favored coastal areas for the heavy snows instead of areas south and west of NYC.

The JMA seasonal just updated for the winter. It was the only model that got the 2013-2014 winter correct with its fall forecasts. This time it has a much different look. It has a more +PNA pattern than we typically get with a La Niña. So it has a Baja ridge instead of the deep trough we had there in 2022-2023. My guess is that this is a continuation of the ridging and warm water we have been seeing near the West Coast this summer. In the East we have the familiar -AO Greenland block linking up with the Southeast ridge. So a milder winter in the East than last winter. As always these extended forecasts much of the time are low skill. So I just include here for informational purposes. https://ds.data.jma.go.jp/wmc/products/model/map/7mE/map1/zpcmap.php Famous forecast from September 2013 showing the record TNH winter

The Euro lost the cutoff low for early next week. It’s replaced by a 591 dm ridge to our west. So it has mid 80s from Sunday right into next week. New run Old run

This is where we get into non-linear climate shifts as opposed to the more generic and bland sounding CC term. From around 1880 to 1982 we were more or less in a common climate era as the temperature didn’t change that much. Then once emissions began to accelerate after 1980 temperatures began to rise at faster rate. The 1970s were the last time the CONUS had top 10 coldest winters. The next era was from 1982 to 1997 when be began to notice a faster rate of warming than before 1982. Regionally we were still getting very cold winters during this period. But the cold wasn’t as widespread as back in the 1970s. The first real jump in global and national temperatures occurred in 1998. So the next era with similar temperatures occurred from 1998 to early 2015. Some regions like the Midwest and Great Lakes were able to have a very impressive cold and snow in 2013-2014. But this cold couldn’t reach the magnitude across the CONUS that it did during the 1970s. So it was more focused into your area. 2015-2016 was the greatest shift or jump in temperatures that we have ever seen. This marks the beginning of a much warmer era where none of the winters since then have been able to rival any of the snow and cold records experienced as recently as 2009-2010 to 2014-2015. This last decade was the warmest on record during the winters for the CONUS by a significant margin. So we went from CONUS widespread cold in the 1970s to more narrowly focused regional cold prior to 2015-2016. The cold outbreaks since then have been regionally focused more around 30-45 days rather than the 90 day + regional cold of 13-14 and 14-15. The geographic footprint and duration of the cold has been getting shorter. Following the 2015-2016 baseline temperature shift or jump, the cold outbreaks between have been more narrowly focused and of shorter duration. The area around Rockford, IL set a new all-time low in January 2019 but the duration of the extreme cold wasn’t that long. My area in the Northeast had a great 2 week cold period from late December 2017 into early January 2018. But this didn’t last and was followed up by our first 80° February warmth ever. Fast forward to February 2021 for the record cold focused in the Plains for around a month. But the magnitude of the warmth in December 2021 around DFW was greater than that of the cold in February 2021. Next we experienced another global and national baseline temperature shift or jump to higher during 2023-2024. So the core cold this winter was again more narrowly focused and didn’t rank very high as it followed the 4th warmest December on record for the CONUS. The common theme since 2015-2016 has been shorter and more regionally focused cold events surrounded by periods of higher ranking record warmth.

Yeah, hoping the Euro is correct about the heavy rain potential with the omega block and cutoff low next week.

The seasonal models at this very early stage seem to like the idea of a very amplified Aleutians ridge. I am not looking at these long range forecasts for the exact locations of the ridge-trough axis yet. But several solutions both keep the trough axis to our west. Would match recent years where we get a Southeast Ridge or WAR influence for the Northeast. I want to see what the MJO activity and the rest of the October pattern looks like before refining the outlook more than these generalities.

Yeah, the impressive cold outbreak over the last few weeks has been focused to our west again. The Columbus, OH area saw their 5th coldest 2 week Aug 25 to Sep 7 period. This has been a repeating theme since the 2015-2016 super El Niño global and national temperature jump. We saw this during this past winter when the coldest anomalies went to our SW and W. Then especially in February 2021 when the record cold went down the Plains and caused all the problems for Texas. Prior to this we had the record cold stay to our west in January 2019 when new all-time records for cold were set around Rockford, IL. The last impressive cold outbreak for a few weeks centered in the Northeast was back in late December 2017 into early January 2018. The last extended cold to be centered in the Northeast was JFM 2015. Time Series Summary for Columbus Area, OH (ThreadEx) top 5 coldest Aug 25 to Sep 7 Click column heading to sort ascending, click again to sort descending. 1 1885-09-07 61.9 0 2 1915-09-07 64.3 0 3 1891-09-07 64.6 0 4 1896-09-07 64.7 0 5 2025-09-07 64.8 0 - 1988-09-07 64.8 0 - 1967-09-07 64.8 0

The Delaware Water Gap area is a beautiful part of NJ. It’s too bad we don’t have a continuous record there going back as long as some of the other Sussex County observing sites. The Sussex airport has continuous records in August online going back to 2001. This August featured the lowest monthly reading of 43°. The old COOP which was nearby stoped posting data a few years ago. So it looks like it may have closed. Both the COOP and airport were operational back in August 2006. The airport had a low of 45° and the COOP 46°. Prior to the airport readings going online back around 2001, the coldest August reading at the Sussex COOP was 34° back in 1965. So if Walpack was 7° colder than the Sussex COOP back in 1965, then it could have fallen to around 27° in that much colder valley area than around the Sussex COOP and airport. Parts of NY and PA were also in the upper 20s during August 1965. That was the worst drought around the region in hundreds of years. Monthly Data for August 1965 for New Jersey Click column heading to sort ascending, click again to sort descending. NEWTON COOP 34 SUSSEX 1 NW COOP 34 CHARLOTTEBURG RESERVOIR COOP 35 HIGH POINT PARK COOP 35 BOONTON 1 SE COOP 36 LAYTON 2 COOP 36 INDIAN MILLS 2 W COOP 37 PEMBERTON COOP 37 FLEMINGTON 5 NNW COOP 37 LONG VALLEY COOP 37 MORRIS PLAINS 1 W COOP 37 TOMS RIVER COOP 38 CHATSWORTH COOP 39 ESSEX FELLS SERVICE BLDG COOP 39 WANAQUE RAYMOND DAM COOP 39 BELLEPLAIN STA FOREST COOP 40 ATLANTIC CITY INTL AP WBAN 40 HIGHTSTOWN 2 W COOP 40 LAMBERTVILLE COOP 40 NEW MONMOUTH COOP 40 CANOE BROOK COOP 40 BELVIDERE COOP 40 Atlantic City Area ThreadEx 40 Belvidere Area ThreadEx 40 SOMERVILLE 4 NW COOP 41 Monthly Data for August 1965 for New York Click column heading to sort ascending, click again to sort descending. ELIZABETHTOWN COOP 26 SPECULATOR COOP 27 ANGELICA COOP 27 SLIDE MOUNTAIN COOP 28 FRANKLINVILLE COOP 28 ONEONTA 3 SE COOP 29 ROXBURY COOP 29 Monthly Data for August 1965 for Pennsylvania Click column heading to sort ascending, click again to sort descending. COUDERSPORT 4 NW COOP 27 AUSTINBURG 2 W COOP 28 KANE 1NNE COOP 30 LAWRENCEVILLE 2 S COOP 30 FRANCIS E WALTER DAM COOP 31 ENGLISH CENTER COOP 31 MILANVILLE COOP 31 EAST STROUDSBURG COOP 32 KARTHAUS (RIVER) COOP 32 EMPORIUM 1 N COOP 32 CLERMONT 8 SW COOP 32 BRADFORD 4SW RES 5 COOP 32 TOBYHANNA POCONO MOUNTAIN ARPT WBAN 32

I haven’t heard any information suggesting that the new sensors being installed at the airports have an inherent cold bias during radiational cooling events. The coldest temperature in NJ since late August wasn’t from an airport but at the Walpack ,NJ climate station. Plus KFOK which is at an airport has always been a great radiational cooling spot out in the Hamptons. So very cold readings at airports which happen to located in great radiational cooling locations like the Pine Barrens with sandy soil have been a common occurrence long before the new sensors began to get rolled out.

Going from the 2nd highest number of hours above a 70° dewpoint in July to the most hours below 60° in August made all the difference.

The record warm SST pattern from Japan to California and across the entire Atlantic mid-latitudes is a function of the continuous 500mb ridge which has been in place this summer. So it’s just a reflection of the higher pressures and lighter winds creating record SSTs below. My guess is that this mid-latitude warmth has created the current lull in tropical activity in the Atlantic with cooler tropical departures relative to further north. The thing to watch for later in the season is the potential for close in development with the record Gulf SSTs. While we are on track to finish with less ACE than last season, it only takes one with hurricanes. Hoping the Gulf can catch a break from the record run recent years. But we always have wait and see what happens since late September and October have been so active in recent years.

Dry conditions snd high pressure really allowing the radiational cooling to overperform during recent weeks. So the low temperature departures have been more impressive. Also helps the high temperatures beat guidance at times.

This is also a general winter and seasonal forecast thread. My post about DC snowfall was only introduced when a poster asked me about it. You even liked the post. The way any ENSO state is being expressed in our sensible weather has been heavily influenced by the sudden jump in SSTs and global temperatures. So the WPAC warm pool along with other basins has been becoming a greater factor at times than ENSO. We saw this many times in recent years when El Niños in 18-19, 19-20, and 23-24 were heavily influenced by events in the WPAC and IO. So just looking at a one dimensional slice of ENSO east of the Dateline has become less relevant than it used to be. The first two winters couldn’t couple and we got more of a Niña background state. 23-24 had unusually strong WPAC and IO forcing with the record SSTs there and the typical El Niño cold trough in the East wasn’t able to develop. This past winter there was talk of a 13-14 analog. But the Pacific Jet was much stronger so places like Chicago had one of their least snowy winters instead of a 13-14 outcome. Plus the 850 mb cold pool was the smallest on record for the Northern Hemisphere preventing the cold trough formation in the Great Lakes like 13-14.

I was happy to get the PNA mismatch potential last October from the early MJO indicator back in October which was a key part of the seasonal 500 mb pattern. Then the under 4” snow last December around NYC signaled another below average snowfall season. I tend to score long range seasonal forecasts like how the batting average is regarded in baseball. All you need to have a shot at the batting title in any given year is to get a hit only around 33% of the time. So if you get at least 1 aspect of the 3 key elements including 500mb pattern, P-types and amounts with the storm track, and the temperatures correct then I will consider it a good showing from seasonal outlook.

The record ridge and drought in Canada generated strong high pressure and low dewpoints. This was followed up by the record trough over the Western Great Lakes ago. These very cold early season temperatures are actually being driven by the minimums allowing ideal early fall radiational cooling. We should continue with the comfortable early fall temperatures here right into mid-September as Canadian high pressure dominates. But much the CONUS is experiencing expanding drought conditions. So we could see a drought feedback warm up beginning to west while we enjoy the great early fall weather here. Unfortunately, the reliable models only go out 15 days. So the pattern from late September into October will depend on the pattern evolution. If we start getting more high pressure to our SW, then that drought feedback warmth could arrive here later September into October. Statistically we don’t see much 90° heat here at the warm spots like Newark that late in the season. We haven’t seen any 90° heat at these locations after September 20th since 2017 and 2019. But you never know if this will be the first 2020s year to pull it off. Especially if the winds can turn more SW. The big theme here in recent years has been record 80s warmth right into late October and early November like last year.

Top 5 coolest 8-25 to 9-7 period for the Great Lakes area. Similar theme to recent years with the coldest temperatures going to our west. Will be interesting to see how big the temperature rebound is in the coming weeks as these cooldowns over the years have been tough to sustain for very long. This was the coldest 8-25 to 9-7 period for Dayton, Ohio since record keeping began in 1935. Time Series Summary for DAYTON INTERNATIONAL AIRPORT, OH top 5 coldest 8-25 to 9-7 periods Click column heading to sort ascending, click again to sort descending. 1 2025-09-07 63.3 0 2 2017-09-07 63.9 0 3 1988-09-07 64.7 0 - 1935-09-07 64.7 0 4 1986-09-07 64.9 0 - 1946-09-07 64.9 0 5 1968-09-07 65.3 0

When I was a kid growing up, we were lucky if even the next few days forecast was correct. The computer models were really primitive during the 70s and 80s. But the Euro doing so well with March 93 and January 96 events began to give us some confidence in the 3-5 day forecast. Occasionally the model does really well on storm details day 6-10 like with Sandy. But the range beyond week 1 and week 2 is still very low skill. Maybe they can find a way for AI to bias correct the longer range NWP forecasts.

If we are looking for as close to perfection as we can get from an October seasonal forecast, then the JMA October 2013 record TNH forecast was pretty close. https://ds.data.jma.go.jp/wmc/products/model/map/7mE/map1/pztmap.php

I know the NYC micronet in Fresh Kills made it to 92° yesterday and Newark was 90° along with Harrison.The next area of 90° highs was spread out from SNJ to CNJ. Fresh Kills 92 67 Climatological Data for HARRISON, NJ - September 2025 Click column heading to sort ascending, click again to sort descending. 2025-09-01 80 61 70.5 -2.9 0 6 0.00 0.0 0 2025-09-02 80 60 70.0 -3.1 0 5 0.00 0.0 0 2025-09-03 79 58 68.5 -4.4 0 4 0.00 0.0 0 2025-09-04 81 62 71.5 -1.1 0 7 0.00 0.0 0 2025-09-05 80 65 72.5 0.1 0 8 0.89 0.0 0 2025-09-06 86 67 76.5 4.4 0 12 0.00 0.0 0 2025-09-07 90 61 75.5 3.7 0 11 1.29 0.0