El Nino 2023-2024

[bluewave]

50 minutes ago, GaWx said:

The relatively cool OHC is the main thing keeping me from going higher than 70% chance for a +2.0+ ONI peak. But I don't think ENSO SST forecasts overall are showing "excessive momentum" because the latest major ones outside of the already very warm BoM and Meteo-France substantially warmed from the prior month despite a tepid June-July 3.4 warming/~neutral SOI/big OHC drop. Also, the JMA, CFS, CANSIPS, and UKMET all actually had cooled the prior month.

My main concern with the model forecasts from July was the only models showing a super El Niño had August above +1.5.  August is set to come in under +1.5 on OISST which I believe has been running a little warmer than the official  ERSST. I know other models have warmed in August. But we don’t have the September verification yet to judge them by.
 

July models showing super El Niño and their August forecast

AUS…………..1.93

ECMWF……..1.67

Metfrance….1.95

[GaWx]

17 minutes ago, bluewave said:

My main concern with the model forecasts from July was the only models showing a super El Niño had August above +1.5.  August is set to come in under +1.5 on OISST which I believe has been running a little warmer than the official  ERSST. I know other models have warmed in August. But we don’t have the September verification yet to judge them by.
 

July models showing super El Niño and their August forecast

AUS…………..1.93

ECMWF……..1.67

Metfrance….1.95

The Meteo-France has been similar to AUS on being much too warm. I expect they'll both verify significantly too warm for peak. That July Euro will clearly verify much too warm for Aug and it does have a warm bias. Their May forecasts for peak ONI (whether - or +) have not verified too cool since way back in 2004! 

 The model run that has recently had the biggest impact is the JMA's warming from +1.76 to +2.2. The JMA had recently been lauded ITT as one that has done well. I then analyzed how it had done earlier this year and found it to not be warm biased and instead slightly cool with averaging a -0.1 miss for the subsequent month and a -0.3 miss for several months out.

[bluewave]

10 minutes ago, GaWx said:

The Meteo-France has been similar to AUS on being much too warm. I expect they'll both verify significantly too warm for peak. That July Euro will clearly verify much too warm for Aug and it does have a warm bias. Their May forecasts for peak ONI (whether - or +) have not verified too cool since way back in 2004! 

 The model run that has recently had the biggest impact is the JMA's warming from +1.76 to +2.2. The JMA had recently been lauded ITT as one that has done well. I then analyzed how it had done earlier this year and found it to not be warm biased and instead slightly cool with averaging a -0.1 miss for the subsequent month and a -0.3 miss for several months out.

The 15-16 super El Nino was the only warm ENSO event in the last decade that the models didn’t have a warm bias for. So that’s why  I like to take a wait and see approach as to how strong this event gets. Looks like after the current WWB coming through we go back to some weaker trades again. This current WWB was also displaced closer to 1+2 and 3 and less impressive near the Dateline like we have been seeing.
 

 

[GaWx]

1 hour ago, bluewave said:

The 15-16 super El Nino was the only warm ENSO event in the last decade that the models didn’t have a warm bias for. So that’s why  I like to take a wait and see approach as to how strong this event gets. Looks like after the current WWB coming through we go back to some weaker trades again. This current WWB was also displaced closer to 1+2 and 3 and less impressive near the Dateline like we have been seeing.
 

 

-But those forecasts were made in April
-August JMA forecast miss for DJF ONI last 10 years: avg +0.1;

-average dynamic models Aug miss for DJF ONI last 10 years +0.3

-the same but for last 7 El Niños alone: JMA -0.1; avg dyn: 0

[bluewave]

46 minutes ago, GaWx said:

-But those forecasts were made in April
-August JMA forecast miss for DJF ONI last 10 years: avg +0.1;

-average dynamic models Aug miss for DJF ONI last 10 years +0.3

-the same but for last 7 El Niños alone: JMA -0.1; avg dyn: 0

We have seen some pretty big errors made by the JMA during recent Augusts. While the ENSO state was different, it was off by about +0.6 last winter and around +1.1 too warm for DJF  21-22. Many times the ENSO models in recent years didn’t get the correct Nino 3.4 temperature for the winter until already into SON. So these ENSO models are often repeater models that just carry forward and build off of what ever month they were initialized in.

[GaWx]

1 hour ago, bluewave said:

We have seen some pretty big errors made by the JMA during recent Augusts. While the ENSO state was different, it was off by about +0.6 last winter and around +1.1 too warm for DJF  21-22. Many times the ENSO models in recent years didn’t get the correct Nino 3.4 temperature for the winter until already into SON. So these ENSO models are often repeater models that just carry forward and build off of what ever month they were initialized in.

-I have Aug JMA for DJF 22-23 -0.43 vs actual -0.67 for a miss of +0.24

-I have Aug JMA for DJF 21-22 -0.16 vs actual -0.97 for a miss of +0.81 (but admittedly still a big error)

During last 10 years:
-Avg Aug JMA miss when DJF <-0.5: +0.5

-Avg Aug JMA miss when DJF >-0.5: -0.2

 If we were in/headed for another La Niña, I'd lean to this last JMA run being too warm. And it of course may be. But with it not La Niña and with all 7 major models now +2.0+, it is hard for me to lean that way considering the recent non-La Niña years as well as how earlier months this year have gone. So, I'm neutral on its latest of +2.18 for DJF, especially with it rising so much even with a much cooler OHC in July and only a tepid 3.4 SST rise in June/July.

[bluewave]

18 minutes ago, GaWx said:

-I have Aug JMA for DJF -0.43 vs actual -0.67 for a miss of +0.24

-I have Aug JMA for DJF -0.16 vs actual -0.97 for a miss of +0.81 (but admittedly still a big error)

During last 10 years:
-Avg Aug JMA miss when DJF <-0.5: +0.5

-Avg Aug JMA miss when DJF >-0.5: -0.2

 If we were in/headed for another La Niña, I'd lean to this last JMA run being too warm. And it of course may be. But with it not La Niña and with all 7 major models now +2.0+, it is hard for me to lean that way considering the recent non-La Niña years as well as how earlier months this year have gone. So, I'm neutral on its latest of +2.18 for DJF, especially with it rising so much even with a much cooler OHC in July and only a tepid 3.4 SST rise in June/July.

I should have mentioned 20-21 also. Was looking at the wrong column for 22-23. 

Just weary of fully believing these super El Niño forecasts when the subsurface is much cooler than 16 and 97 and the models have been verifying too warm in 3.4.

Plus the general warm bias of these model forecasts in recent years. 

[40/70 Benchmark]

10 hours ago, GaWx said:

 I became curious about 19-20 after reading this. So, I checked the 19-20 threads and found that nearly every forecast ended up verifying much too cold. I'm placing some of the blame on some seasonal model forecasts that were much too cold. For example:

-JAMSTEC (11/1/19) a pure disaster (I think it is cold biased in the E US because it has had cold forecasts there for a good # of winters that ended up mild)

 -But this C3S was actually not bad though it wasn't warm enough:

-The CFS and CANSIPS maps were ~normal.

 -The forecasters largely went near normal to slightly colder than normal. The consensus among them was a neutral to -NAO, possibly influenced by the strong -NAO (-1.6) (second strongest -NAO on record) for May-Oct 2019. NAO ended up sharply reversing to strong +NAO all 3 of DJF.

-Though even he verified too cold, Don Sutherland did the best of those that I saw with no BN anywhere in US. He had near normal Desert SW/Great Lakes/NE and slightly AN most other areas other than solidly AN SE. The key to him not missing by as much: he predicted a +NAO.

-If Don were to go -NAO this winter, that would be quite encouraging because historical stats show no correlation of a -NAO summer to -NAO following winter.

I feel like part of the issue that year was people taking into a consideration a weak el nino when forecasting, but the atmosphere wasn't in el Nino mode. 

[bluewave]

8 hours ago, 40/70 Benchmark said:

I feel like part of the issue that year was people taking into a consideration a weak el nino when forecasting, but the atmosphere wasn't in el Nino mode. 

Yeah, the impressive +IOD lead to the very strong SPV and +NAO which muted the weak El Niño signal.

https://rmets.onlinelibrary.wiley.com/doi/full/10.1002/asl.1005

Northern Europe and the UK experienced an exceptionally warm and wet winter in 2019/20, driven by an anomalously positive North Atlantic Oscillation (NAO). This positive NAO was well forecast by several seasonal forecast systems, suggesting that this winter the NAO was highly predictable at seasonal lead times. A very strong positive Indian Ocean dipole (IOD) event was also observed at the start of winter. Here we use composite analysis and model experiments, to show that the IOD was a key driver of the observed positive NAO. Using model experiments that perturb the Indian Ocean initial conditions, two teleconnection pathways of the IOD to the north Atlantic emerge: a tropospheric teleconnection pathway via a Rossby wave train travelling from the Indian Ocean over the Pacific and Atlantic, and a stratospheric teleconnection pathway via the Aleutian region and the stratospheric polar vortex. These pathways are similar to those for the El Niño Southern Oscillation link to the north Atlantic which are already well documented. The anomalies in the north Atlantic jet stream location and strength, and the associated precipitation anomalies over the UK and northern Europe, as simulated by the model IOD experiments, show remarkable agreement with those forecast and observed.

 

[Terpeast]

34 minutes ago, bluewave said:

Yeah, the impressive +IOD lead to the very strong SPV and +NAO which muted the weak El Niño signal.

https://rmets.onlinelibrary.wiley.com/doi/full/10.1002/asl.1005

Northern Europe and the UK experienced an exceptionally warm and wet winter in 2019/20, driven by an anomalously positive North Atlantic Oscillation (NAO). This positive NAO was well forecast by several seasonal forecast systems, suggesting that this winter the NAO was highly predictable at seasonal lead times. A very strong positive Indian Ocean dipole (IOD) event was also observed at the start of winter. Here we use composite analysis and model experiments, to show that the IOD was a key driver of the observed positive NAO. Using model experiments that perturb the Indian Ocean initial conditions, two teleconnection pathways of the IOD to the north Atlantic emerge: a tropospheric teleconnection pathway via a Rossby wave train travelling from the Indian Ocean over the Pacific and Atlantic, and a stratospheric teleconnection pathway via the Aleutian region and the stratospheric polar vortex. These pathways are similar to those for the El Niño Southern Oscillation link to the north Atlantic which are already well documented. The anomalies in the north Atlantic jet stream location and strength, and the associated precipitation anomalies over the UK and northern Europe, as simulated by the model IOD experiments, show remarkable agreement with those forecast and observed.

 

Interesting! I hadn’t come across that. When I do my winter outlook, I will factor the IOD into it. 

Right now it appears to be weakly positive, no? CPC forecast maps re-warms the WPAC/MC pool leading into winter, so that may bring the IOD to neutral-ish. 

[40/70 Benchmark]

1 hour ago, bluewave said:

Yeah, the impressive +IOD lead to the very strong SPV and +NAO which muted the weak El Niño signal.

https://rmets.onlinelibrary.wiley.com/doi/full/10.1002/asl.1005

Northern Europe and the UK experienced an exceptionally warm and wet winter in 2019/20, driven by an anomalously positive North Atlantic Oscillation (NAO). This positive NAO was well forecast by several seasonal forecast systems, suggesting that this winter the NAO was highly predictable at seasonal lead times. A very strong positive Indian Ocean dipole (IOD) event was also observed at the start of winter. Here we use composite analysis and model experiments, to show that the IOD was a key driver of the observed positive NAO. Using model experiments that perturb the Indian Ocean initial conditions, two teleconnection pathways of the IOD to the north Atlantic emerge: a tropospheric teleconnection pathway via a Rossby wave train travelling from the Indian Ocean over the Pacific and Atlantic, and a stratospheric teleconnection pathway via the Aleutian region and the stratospheric polar vortex. These pathways are similar to those for the El Niño Southern Oscillation link to the north Atlantic which are already well documented. The anomalies in the north Atlantic jet stream location and strength, and the associated precipitation anomalies over the UK and northern Europe, as simulated by the model IOD experiments, show remarkable agreement with those forecast and observed.

 

The MEI and RONI did not reflect el Nino at all that season. 

[bluewave]

1 hour ago, 40/70 Benchmark said:

The MEI and RONI did not reflect el Nino at all that season. 

It was one of those seasons when the RMM index MJO 4-6 was a more reliable indicator  than the weak forcing near the Dateline. Just a classic La Niña pattern with the ridge north of Hawaii and SE Ridge over the US. The RMM is usually more in line with where the VP anomalies show up. But for some reason that wasn’t the case in 19-20. 
 

 

[Terpeast]

Anyone know of a MEI v2 dataset that goes back to 1950? I know it only goes back to 1979, but I wonder if anyone has adjusted MEI v1 data from 1950-1978 to reflect MEI v2?

[snowman19]

2 hours ago, Terpeast said:

Interesting! I hadn’t come across that. When I do my winter outlook, I will factor the IOD into it. 
Right now it appears to be weakly positive, no? CPC forecast maps re-warms the WPAC/MC pool leading into winter, so that may bring the IOD to neutral-ish. 

It’s projected to stay positive through January: 
http://www.bom.gov.au/climate/ocean/outlooks/#tabs=Graphs&region=IOD

[snowman19]

Very good explanation of why this Nino is staying very east-based/EP

[bluewave]

3 hours ago, Terpeast said:

Interesting! I hadn’t come across that. When I do my winter outlook, I will factor the IOD into it. 

Right now it appears to be weakly positive, no? CPC forecast maps re-warms the WPAC/MC pool leading into winter, so that may bring the IOD to neutral-ish. 

NOAA maintains an IOD index which they call DMI. The fall of 2019 was the strongest positive on record. But it faded by the winter as record SSTs returned north of Australia with the very strong MJO 4-6 that winter.
 

https://stateoftheocean.osmc.noaa.gov/sur/ind/dmi.php

[Terpeast]

24 minutes ago, snowman19 said:

It’s projected to stay positive through January: 
http://www.bom.gov.au/climate/ocean/outlooks/#tabs=Graphs&region=IOD

I don’t pay attention to the BoM model due to its known warm bias

[GaWx]

14 minutes ago, Terpeast said:

I don’t pay attention to the BoM model due to its known warm bias

 Although the BoM has been much too warm this summer, it actually averaged a neutral bias the prior ten years with several significantly too warm but others significantly too cool to balance it out. So, despite the 10 year neutral bias, it might have a relatively high average error though I didn't calculate it and thus don't know how its error has been relative to other models.

[roardog]

OISST took a nose dive in 1+2 on the daily readings recently. That’s strange because CDAS did that last week but then came back up. Now CDAS is warmer than OISST. Also Enso 4 is getting above 1.0 now on OISST. This is definitely a basin wide Nino at this point.

[snowman19]

29 minutes ago, GaWx said:

 Although the BoM has been much too warm this summer, it actually averaged a neutral bias the prior ten years with several significantly too warm but others significantly too cool to balance it out. So, despite the 10 year neutral bias, it might have a relatively high average error though I didn't calculate it and thus don't know how its error has been relative to other models.

The BOM warm bias this year has been with ENSO, I haven’t seen the same bias with the IOD

[bluewave]

18 minutes ago, roardog said:

OISST took a nose dive in 1+2 on the daily readings recently. That’s strange because CDAS did that last week but then came back up. Now CDAS is warmer than OISST. Also Enso 4 is getting above 1.0 now on OISST. This is definitely a basin wide Nino at this point.

This El Niño continues to do its own thing. 

 

[mitchnick]

57 minutes ago, Terpeast said:

I don’t pay attention to the BoM model due to its known warm bias

Other model forecasts can be found at this link, middle of the page.

http://www.bom.gov.au/climate/enso/#tabs=Indian-Ocean

[GaWx]

18 minutes ago, snowman19 said:

The BOM warm bias this year has been with ENSO, I haven’t seen the same bias with the IOD

Also, I should have specifically referred to this summer's warm bias of the BoM as having been in Nino 3.4, 3, and 4. I've never seen a BoM forecast for Nino 1+2.

[Terpeast]

7 minutes ago, mitchnick said:

Other model forecasts can be found at this link, middle of the page.

http://www.bom.gov.au/climate/enso/#tabs=Indian-Ocean

Thanks for pointing that out. Going through the monthly forecasts from now till January, it appears that METEO and ECMWF are the most aggressive with the +IOD, but all models have them peaking in Oct-Nov, then backing off in December to a neutral/weak positive by January. 

This is pretty consistent with a mild December / start to the winter, then turning colder into Jan-Feb.

[GaWx]

22 minutes ago, bluewave said:

This El Niño continues to do its own thing. 

 

Wow, OISST 1+2 anomaly cooled a whopping 0.5C in just 24 hours!!

[bluewave]

43 minutes ago, GaWx said:

Wow, OISST 1+2 anomaly cooled a whopping 0.5C in just 24 hours!!

Yeah, I also think 29.7 is near a record in Nino 4 for this time of year. While it won’t average this high for the whole month due to the steep incline, getting close to +30C is impressive.

 

 

 

 

[Stormchaserchuck1]

More west-based El Nino forcing here (+PNA)

https://ibb.co/mTn9ndp 

coinciding with central-subsurface warming

[snowman19]

Wow, OISST 1+2 anomaly cooled a whopping 0.5C in just 24 hours!!

If that reading isn’t a “glitch” in the OISST given the warmer CDAS and CRW, it’s not going to last long with the strong WWB ongoing in regions 1+2, 3 and eastern 3.4….

[bluewave]

38 minutes ago, snowman19 said:

If that reading isn’t a “glitch” in the OISST given the warmer CDAS and CRW, it’s not going to last long with the strong WWB ongoing in regions 1+2, 3 and eastern 3.4….

The subsurface has cooled quite a bit in the eastern regions this summer. So there will need to be another kelvin wave in order to maintain those levels. Absent a new kelvin wave, we are set for an unusually early peak in Nino 1+2 and 3. Which would probably lead to a peak in 3.4 about 4-6 weeks after that. So not the usual El Niño timing if the subsurface counties to cool. None of the very strong El Niño events of the past had a subsurface cooling trend during the summer. 
 

https://www.cpc.ncep.noaa.gov/products/analysis_monitoring/enso_update/wkxzteq.shtml

[Typhoon Tip]

2 hours ago, Terpeast said:

Thanks for pointing that out. Going through the monthly forecasts from now till January, it appears that METEO and ECMWF are the most aggressive with the +IOD, but all models have them peaking in Oct-Nov, then backing off in December to a neutral/weak positive by January. 

This is pretty consistent with a mild December / start to the winter, then turning colder into Jan-Feb.

Most tend to hoot at the notion but the climate change signal is large swing variability with increased short duration cold latter October through November into early Dec ... Then, gradient intense hemisphere with higher tropospheric torque during mid terms lending to more rain/ice.  

It seems to be taking place despite the antecedent ENSO variance, too.  

We'll see.