Support the Arctic Sea Ice Forum and Blog

Author Topic: ENSO/ASI Correlation  (Read 3470 times)

Killian

  • Frazil ice
  • Posts: 354
    • View Profile
  • Liked: 80
  • Likes Given: 8
ENSO/ASI Correlation
« on: September 18, 2024, 10:05:57 AM »
Well, the theory with no name that has since been confirmed by two papers holds true. If you recall, the idea is simple: The summer of or the summer following a strong El Nino will usually lead to a new 1st, 2nd, or 3rd low level of ASI. Formulated in August 2015, it has held true through 2016, '20 and now '24 - even though '24 ended up being a strong, but not super, EN. So far, we have 3rd lowest ASI Area and... was it concentration?... from MSIDC. I am curious as to whether PIOMAS will join the fun. I guess we won't know until October.

The window is still open till next year as the ocean currents take time and energy takes time to propagate. Frankly, with all the ocean heat content of '23 and '24, I think we were darned lucky not to challenge 2020 or second place. That may change next summer, but with a La Nina brewing, maybe not.

Why this matters: 1. The better we understand how ASI melts, the better we can model it and be warned about future changes.

2. Extremes drive step changes in non-linear systems; they trigger tipping points. There;s a reason there is so much interest and trepidation about Blue Ocean Events.

3. Knowing El Nino is likely to trigger ASI lows can help us all with our ASI level polls. Big EN coming? Good bet it'll be a new Top 3 low or close to it!

Cheers

El Cid

  • Young ice
  • Posts: 2623
    • View Profile
  • Liked: 1003
  • Likes Given: 234
Re: ENSO/ASI Correlation
« Reply #1 on: September 18, 2024, 05:40:54 PM »
Big melt years this millenium (September piomas ice volume)

basically 2002-7  every year, 2010-2012, 2016, 2019 and probably this year

El nino years: 2009/10, 2015/16, 2023/24

None of the big melt years 2002-7 qualify. 2012 does not qualify, 2019 does not qualify. You have 2010, maybe 11 , 2016 and 2024

I am not convinced

Killian

  • Frazil ice
  • Posts: 354
    • View Profile
  • Liked: 80
  • Likes Given: 8
Re: ENSO/ASI Correlation
« Reply #2 on: October 03, 2024, 12:44:42 AM »
Big melt years this millenium (September piomas ice volume)

basically 2002-7  every year, 2010-2012, 2016, 2019 and probably this year

El nino years: 2009/10, 2015/16, 2023/24

None of the big melt years 2002-7 qualify. 2012 does not qualify, 2019 does not qualify. You have 2010, maybe 11 , 2016 and 2024

I am not convinced

That would worry me if you understood what you were posting about, but you're way off in a couple ways.

1. Why would you use 22 years of a record 3x longer than that? Also, I think you are off on a couple of your EN years.

2002–03 '02 new record, '03 new 2nd, '04 new 2nd
2004–05 New Record, '06 new 3rd
2006–07 New Record, '08 new2nd
2009–10 '09 new 3rd, '10 new 3rdm '11 new 3rd
2014–16 '15 new 3rd, '16 new 2nd
2018–19 '19 new 2nd, '20 new 2nd
2023–24 '24 Oops! So far... still gotta see how '25 goes.

HOWEVER...

That's only Extent. We have reached the point, as expected for years, where the degree of melt allows the ice to move more freely and extent is probably not as useful a metric as it has been.

I haven't ever correlated for volume and area because with extent I got @ a 65% correlation, IIRC, so it wasn't worth my time. (I'm not a stats guy or a math guy. This stuff takes a LOT of time for me because I have to do it all without any stats skills, etc.) That was before counting the ENs since 2015, so that % would be higher now.

This season, area and volume hit new 3rd lowest levels, meeting the criteria. The model covers the year of the EN and the following summer as it's assumed it takes time for heat to propagate from the equator to the Arctic. We have to wait till '25 to see if there are additional lows for any of the three measures.

2. You don't seem to understand the model.

'12 almost qualifies, but would not be included even if it did because that massive drop was caused by massively irregular conditions that have no repeated examples. It was a true outlier. Even if there had been an El Nino in 2011, I would asterisk '12 as far too noisy to claim it supported the model.

Thanks for chiming in, but you clearly don't understand the parameters of the model and used far too small a sample size. FYI, there is a paper from 2018 finding a link and a paper from 2021 identifying a mechanism that is almost certainly part of the process, the "heat bomb" paper. I identified the potential pattern in 2015 and predicted the 2016 low. And this year's. I don't recall even knowing there was an EN in 2019-20, but did comment about 2019 and 2020 indicating a regime change/signs of a tipping point because they didn't seem to have much reason for happening. Turns out, heating took a sudden upward swing starting in 2014-'15 that continues to today. A major tipping point is occurring in my opinion.

You said you weren't convinced, yet this list is highly correlated, so not sure why you are unconvinced. Please note, it has never been claimed every EN will result in a new 1st, 2nd or 3rd low. It's *an* effect on ASI, not *the* effect. We all know the largest variance occurs due to conditions each summer. 2012 is *the* exemplar of that with a big melt in early spring nobody has ever explained or even addressed, the GAC in August, and especially the dipole winds that blew huge amounts of ice through the Fram.

Cheers
« Last Edit: October 03, 2024, 12:57:10 AM by Killian »

El Cid

  • Young ice
  • Posts: 2623
    • View Profile
  • Liked: 1003
  • Likes Given: 234
Re: ENSO/ASI Correlation
« Reply #3 on: October 03, 2024, 09:47:49 AM »
Now I became curious so I did the volume numbers (since 1980).

And turns out you are right.

If we divide the years into neutral, nino and nina, the average (and median) annual and September volume loss are as follows:

Annual change in Piomas volume ths km3

Neutral: 0,03 (0,25)
Nino: -0,66 (-0,58)
Nina: -0,09 (-0,28)

Interestingly Nina years are also negative !

September volume change vs previous September:

Neutral: 0,09 (-0,13)
Nino: -0,89 (-0,89)
Nina: 0,13 (-0,05)


« Last Edit: October 03, 2024, 12:22:52 PM by El Cid »

johnm33

  • Grease ice
  • Posts: 919
    • View Profile
  • Liked: 155
  • Likes Given: 130
Re: ENSO/ASI Correlation
« Reply #4 on: October 03, 2024, 10:22:06 AM »
Quote
and a paper from 2021

Do you have a link?

Killian

  • Frazil ice
  • Posts: 354
    • View Profile
  • Liked: 80
  • Likes Given: 8
Re: ENSO/ASI Correlation
« Reply #5 on: October 06, 2024, 01:06:46 AM »
Quote
and a paper from 2021

Do you have a link?

No. I always refer back to the video they put out.



Here's an article they link from the video description. (I've never read it. I assume it'll have a little more detail.)

Killian

  • Frazil ice
  • Posts: 354
    • View Profile
  • Liked: 80
  • Likes Given: 8
Re: ENSO/ASI Correlation
« Reply #6 on: October 06, 2024, 01:12:08 AM »
Now I became curious so I did the volume numbers (since 1980).

And turns out you are right.

If we divide the years into neutral, nino and nina, the average (and median) annual and September volume loss are as follows:

Annual change in Piomas volume ths km3

Neutral: 0,03 (0,25)
Nino: -0,66 (-0,58)
Nina: -0,09 (-0,28)

Interestingly Nina years are also negative !

September volume change vs previous September:

Neutral: 0,09 (-0,13)
Nino: -0,89 (-0,89)
Nina: 0,13 (-0,05)

Cool. I can't really do that sort of numbers work, so appreciate it being done. I did the original analysis just eyeballing an ENSO chart and an ASIE chart. Since ASIE started falling in the early fifties, that's where I started. I have to assume the effect has risen from within the noise as time has gone on and the melts have grown in magnitude, so the effect may not be noticeable in the first decades - especially since the first two were during the cooling from the 40's to the 70's.

I would also think the effect would be strongest WRT volume if we could accurately measure that over the entire 70 years.

RE: La Nina's, I'm not surprised given your time frame because La Nina's are now warmer than El Ninos used to be. Maybe any large shift of energy will have some effect?

Cheers

Jim Hunt

  • First-year ice
  • Posts: 6489
  • Don't Vote NatC or PopCon, Save Lives!
    • View Profile
    • The Arctic sea ice Great White Con
  • Liked: 1006
  • Likes Given: 91
Re: ENSO/ASI Correlation
« Reply #7 on: October 06, 2024, 02:11:38 AM »
Do you have a link?

Video & paper mentioned at:

https://GreatWhiteCon.info/2021/05/month-in-review-arctic-science-edition/

Quote
I don’t know who came up with the “Heat Bomb” misnomer, since it isn’t mentioned in the paper itself.
"The most revolutionary thing one can do always is to proclaim loudly what is happening" - Rosa Luxemburg

binntho

  • Nilas ice
  • Posts: 2332
    • View Profile
  • Liked: 958
  • Likes Given: 250
Re: ENSO/ASI Correlation
« Reply #8 on: October 06, 2024, 07:19:19 AM »
The article on which the "Heat Bomb" video is based (?) is very extensive and informative - but I find nothing in there that might fit the visuals of the video.

Understanding Arctic Ocean Circulation: A Review of Ocean Dynamics in a Changing Climate

Everything to do with the layering, movement, mixing and turbulence of the Arctic Ocean waters is comprehensively discussed, unfortunately a bit too technical at times. But there are plenty of interesting takeaways (Note that I have removed all bracketed references from the original text but included the section numbers at the end of each quotation):

Quote
There is relatively weak tidal forcing in the Arctic, and most of the region is above the critical latitude north of which the semidiurnal lunar tide can propagate freely. Topographic waves generated over bathymetric slopes and rough topography, forced by the tides, are the main source of energy for higher tidal dissipation observed over topography, (4.1)

Tidal movement does happen but is small and the tidal wave does not propagate (which is to say, the tidal movement fizzles out). But tidal movement accross bottom features can produce internal waves.

So how fast do the waters of the Arctic actually move? Can we talk about "surging" waters or should we be thinking more along the lines of "skulking" or "crawling"?

Quote
These intrusions have a lateral component of motion, driven partly by double-diffusive vertical buoyancy flux divergences, and carry warm Atlantic Water from the boundaries to the interior basins. Walsh and Carmack (2003) estimated lateral diffusivities associated with these thermohaline intrusions to be around 50 m2 s−1. In this way, diapycnal mixing can redistribute Atlantic Water heat laterally, with Atlantic Water intrusions taking around a decade to propagate across the Canada Basin. (4.1)

Diapycnal is movement at right angles to the isopycnal, and the latter refers to layers of equal density. How diapycnal mixing can operate laterally in a laterally layered ocean is a mystery to me! But it does, and very very slowly ...

Quote
Mooring measurements indicate Atlantic Water boundary current speeds to be around 2 to 4 cm s−1 (Woodgate et al., 2001). This is consistent with transient tracer data which suggest that Atlantic Water propagation from the Eurasian Basin to the southern Canada Basin (a distance of around 6,000 km) takes around 7.5 years.(5)

Slower than a snail ... normal walking speed for your average rambler is 50 times faster than this.

Ocean heat transport into the Artic is has two sources, the Atlantic and the Pacific.

Quote
The West Spitsbergen Current (WSC) carries relatively warm and salty Atlantic Water north (around 7 Sv) into the Arctic Ocean on the eastern side of Fram Strait ...Atlantic Water also enters the Arctic Ocean from the Nordic Seas via the Barents Sea Opening (∼2 Sv). Observations indicate that Atlantic Water heat transport to the Arctic Ocean is higher through the Barents Sea Opening (∼70 TW) than through Fram Strait (∼40 TW).(5)

Atlantic inflow at around 9.5 Sv, with a total energy influx of 110 TW.  One Sverdrup (sv) is one million cubic metres per second. It is interesting to see that the heat input via the Barents gap is almost twice as big as the Fram Strait, while bringing in only about 1/3 of the volume.

Quote
Heat transport from the Pacific Ocean through Bering Strait increased by 60% during 2001–2014, from around 10 TW in 2001 to 16 TW in 2014; this was attributed to increases in both volume flux and temperature(8.1).(

Section 3 estimates Barents inflow at around 1 Sverdrup, based on which one can conclude that inflow from the Pacific is roughly 1/10 that coming from the Atlantic.

But it is of course interesting to see that the Bering inflow is increasing, both in temperature and volume. But even so, the speed at which the water is entering is very low, not much more than slow walking speed when measured based on propagation of salinity front from Mercator visuals.
because a thing is eloquently expressed it should not be taken to be as necessarily true
St. Augustine, Confessions V, 6

Jim Hunt

  • First-year ice
  • Posts: 6489
  • Don't Vote NatC or PopCon, Save Lives!
    • View Profile
    • The Arctic sea ice Great White Con
  • Liked: 1006
  • Likes Given: 91
Re: ENSO/ASI Correlation
« Reply #9 on: October 06, 2024, 04:02:28 PM »
The article on which the "Heat Bomb" video is based (?)

You're quoting from another article I mentioned. This is the one that relates to the alleged "heat bombs":

"A warm jet in a cold ocean"

"The most revolutionary thing one can do always is to proclaim loudly what is happening" - Rosa Luxemburg

binntho

  • Nilas ice
  • Posts: 2332
    • View Profile
  • Liked: 958
  • Likes Given: 250
Re: ENSO/ASI Correlation
« Reply #10 on: October 07, 2024, 07:15:31 AM »
The article on which the "Heat Bomb" video is based (?)

You're quoting from another article I mentioned. This is the one that relates to the alleged "heat bombs":

"A warm jet in a cold ocean"

Thanks, that makes much more sense.

According to this article, the energy entering the Arctic through the Strait, when averaged annually, is only around 3 TW, compared to the previous paper which talks about 10-16TW. One explanation could be that the latter figure is for the summer months only, but I think a more likely explanation is that the 3TW figure is only for that portion of Pacific waters which enters along the bottom of the strait, through the Barrow Canyon, and does not include potential surface currents.

The hot water intrusions discussed in the paper are moving at 1 m/s which is a lot slower than the apparent surface propagation of salinity vortexes as can be seen in Mercator visuals, the latter moving at a brisk 100 cm/s (very rough approximation) which is strolling speed on a smooth surface for an average adult.

The annual average heat transfer is 3TW which equals 3 Tera-Joules per second. Summer insolation at 90N is 300 W/m2, or 300 Joules/m2 so one square kilometer receives 3 Giga-Joule per second of insolation.

Since the Bering heat influx is averaged over 12 months, while summer insolation is calculated for 3 months, we can then calculate that the heat influx through the Bering strait is similar to summer insolation over 4000 square kilometers. The Chukchi sea is 620.000 km2, so the Bering Strait summer heat influx through the Barrow Canyon is comparable to less than 1% of summer insolation in the Chukchi.

The conclusion: Not a lot of heat, and moving very slowly. Interesting, but not something that poses an existential threat to Arctic Sea Ice.
« Last Edit: October 07, 2024, 08:46:13 AM by binntho »
because a thing is eloquently expressed it should not be taken to be as necessarily true
St. Augustine, Confessions V, 6

Phil.

  • Grease ice
  • Posts: 574
    • View Profile
  • Liked: 85
  • Likes Given: 13
Re: ENSO/ASI Correlation
« Reply #11 on: October 07, 2024, 03:10:56 PM »
"The hot water intrusions discussed in the paper are moving at 1 m/s which is a lot slower than the apparent surface propagation of salinity vortexes as can be seen in Mercator visuals, the latter moving at a brisk 100 cm/s"

1m/s is 100cm/s?

John Batteen

  • Frazil ice
  • Posts: 259
    • View Profile
  • Liked: 67
  • Likes Given: 3
Re: ENSO/ASI Correlation
« Reply #12 on: October 07, 2024, 06:06:36 PM »
Must be a typo in there, I was wondering about the same thing.

Very interesting stuff in here though.  Thanks for sharing, everyone.

uniquorn

  • First-year ice
  • Posts: 5379
    • View Profile
  • Liked: 2325
  • Likes Given: 393
Re: ENSO/ASI Correlation
« Reply #13 on: October 07, 2024, 07:04:14 PM »
Had a look at the heat flux estimate from mooring 3 in the Bering Strait which, like insolation, is also seasonal. Peaked at 57TW in summer October 2016. Data is to 2022.

http://psc.apl.washington.edu/HLD/Bstrait/Data/BeringStraitMooringDataArchive.html#Overview
« Last Edit: October 08, 2024, 12:28:17 PM by uniquorn »

binntho

  • Nilas ice
  • Posts: 2332
    • View Profile
  • Liked: 958
  • Likes Given: 250
Re: ENSO/ASI Correlation
« Reply #14 on: October 08, 2024, 09:59:40 AM »
"The hot water intrusions discussed in the paper are moving at 1 m/s which is a lot slower than the apparent surface propagation of salinity vortexes as can be seen in Mercator visuals, the latter moving at a brisk 100 cm/s"

1m/s is 100cm/s?

Yes indeed, which is a normal human walking speed. I measured this some time ago from a mercator salinity video, example shown below. This shows salinity at 34 m depth, and there is one frame per day.The highly turbulent inflows from the Bering are easy to see, with the rest of the Arctic hardly moving at all.

What I did (with a higher resolution video) was to very roughly measure turbulence front propagation from one day to another and found several instances of a maximum just under 100 km/day which is very close to 1m/s.
because a thing is eloquently expressed it should not be taken to be as necessarily true
St. Augustine, Confessions V, 6

uniquorn

  • First-year ice
  • Posts: 5379
    • View Profile
  • Liked: 2325
  • Likes Given: 393
Re: ENSO/ASI Correlation
« Reply #15 on: October 08, 2024, 12:04:22 PM »
<>
2002–03 '02 new record, '03 new 2nd, '04 new 2nd
2004–05 New Record, '06 new 3rd
2006–07 New Record, '08 new2nd
2009–10 '09 new 3rd, '10 new 3rdm '11 new 3rd
2014–16 '15 new 3rd, '16 new 2nd
2018–19 '19 new 2nd, '20 new 2nd
2023–24 '24 Oops! So far... still gotta see how '25 goes.
<>

Some interesting results from this modelled study, in particular I hadn't come across Bjerknes compensation before.

Bering Strait Ocean Heat Transport Drives Decadal Arctic Variability in a High-Resolution Climate Model
Yuchen Li, Wilbert Weijer, Prajvala Kurtakoti, Milena Veneziani, Ping Chang
First published: 20 June 2024
https://doi.org/10.1029/2024GL108828

abstract and some extracts
Quote
Abstract

We investigate the role of ocean heat transport (OHT) in driving the decadal variability of the Arctic climate by analyzing the pre-industrial control simulation of a high-resolution climate model. While the OHT variability at 65°N is greater in the Atlantic, we find that the decadal variability of Arctic-wide surface temperature and sea ice area is much better correlated with Bering Strait OHT than Atlantic OHT. In particular, decadal Bering Strait OHT variability causes significant changes in local sea ice cover and air-sea heat fluxes, which are amplified by shortwave feedbacks. These heat flux anomalies are regionally balanced by longwave radiation at the top of the atmosphere, without compensation by atmospheric heat transport (Bjerknes compensation). The sensitivity of the Arctic to changes in OHT may thus rely on an accurate representation of the heat transport through the Bering Strait, which is difficult to resolve in coarse-resolution ocean models.
Key Points

    Ocean heat transport variability through the Bering Strait has an outsized effect on Arctic sea ice cover and surface temperature

    Atlantic ocean heat transport anomalies into the Arctic are compensated by atmospheric heat transport anomalies on decadal timescales

Introduction
<>
Decadal to multidecadal timescale variability in the Arctic—manifested for example, by changes in sea ice extent and surface temperatures—has been previously shown to be closely related to ocean heat transport (Jungclaus & Koenigk, 2010; Zhang, 2015). Further, it has been suggested that variability in total heat transport into the Arctic is reduced by a phenomenon known as Bjerknes Compensation (BC), whereby anomalies in meridional ocean heat transport (OHT) tend to induce roughly equal and opposite anomalies in meridional atmospheric heat transport (AHT). Bjerknes (1964) proposed that this result follows from energy conservation on timescales where the top-of-atmosphere (TOA) fluxes and ocean heat content remain approximately constant. A recent study (Y. Liu et al., 2020) indeed found evidence of decadal-timescale BC in several reanalysis data sets and confirmed the importance of surface heat fluxes in transferring OHT variability to the atmosphere. However, as they note, the exact causal relationships of decadal-timescale heat transport variability are very difficult to parse out in such short observational time series.

Figure 3 shows the linear regressions of anomalous sea ice concentration, turbulent heat flux, and surface temperature onto the OHT anomaly through Bering Strait and the Atlantic separately. During periods of anomalously high OHT, ice is lost in the marginal ice zone and anomalous heat is transferred from the newly exposed ocean to the atmosphere. The greatest changes associated with Atlantic OHT variability occur along the East Greenland Current while the greatest changes associated with Bering Strait OHT variability occur in the Chukchi Sea. The greatest changes in surface heat flux are found at the marginal ice zone, similar to what was found in Outten and Esau (2017), Jungclaus and Koenigk (2010), and Kurtakoti et al. (2023). Regression of anomalous sea surface temperatures (SST) onto OHT reveals a spatial pattern similar to the heat flux regression (Figure S2 in Supporting Information S1). This positive correlation between SST and heat flux indicates that on decadal timescales, anomalous turbulent heat fluxes are primarily driven by changes in SST, rather than the other way around.

Even though Bering Strait OHT is much smaller compared to Atlantic OHT at 65°N, it has an outsized impact on the local sea ice and heat flux variability. Heat fluxes associated with anomalous OHT in the Pacific sector are comparable in magnitude to those in the Atlantic sector. Furthermore, loss of sea ice in the Pacific sector during periods of high Bering Strait OHT generates a substantial surface air temperature anomaly centered on the same area (Figure 3e). This temperature anomaly extends vertically into the atmosphere, though the strongest changes are confined to the boundary layer (Figure S3 in Supporting Information S1). Thus, we emphasize that the local influence of Bering Strait OHT on the atmosphere cannot be neglected.

The outsized impact of Bering Strait OHT on Arctic sea ice has an obvious geographical reason: the climatological sea ice edge is located much further south in the Pacific sector than in the Atlantic sector (Figure 1). In this model, the ice edge in the Pacific sector is located 5° to the south of the polar circle, while in the Atlantic sector it is located 10° to the north. OHT variability across 65°N has therefore a much more direct impact on the ice edge in the Pacific sector than in the Atlantic sector, an effect that is amplified by the ice albedo feedback. That said, Bering Strait waters are significantly colder than the Atlantic waters, which reach almost all the way to the ice edge (Figure S4 in Supporting Information S1).
my emphasis

John Batteen

  • Frazil ice
  • Posts: 259
    • View Profile
  • Liked: 67
  • Likes Given: 3
Re: ENSO/ASI Correlation
« Reply #16 on: October 08, 2024, 12:19:34 PM »
"The hot water intrusions discussed in the paper are moving at 1 m/s which is a lot slower than the apparent surface propagation of salinity vortexes as can be seen in Mercator visuals, the latter moving at a brisk 100 cm/s"

1m/s is 100cm/s?

Yes indeed, which is a normal human walking speed. I measured this some time ago from a mercator salinity video, example shown below. This shows salinity at 34 m depth, and there is one frame per day.The highly turbulent inflows from the Bering are easy to see, with the rest of the Arctic hardly moving at all.

What I did (with a higher resolution video) was to very roughly measure turbulence front propagation from one day to another and found several instances of a maximum just under 100 km/day which is very close to 1m/s.

So how then is 1 meter per second a lot slower than 100 centimeters per second, when they are equivalent?

uniquorn

  • First-year ice
  • Posts: 5379
    • View Profile
  • Liked: 2325
  • Likes Given: 393
Re: ENSO/ASI Correlation
« Reply #17 on: October 08, 2024, 03:36:15 PM »
<>
The conclusion: Not a lot of heat, and moving very slowly. Interesting, but not something that poses an existential threat to Arctic Sea Ice.

Some calculations and a different conclusion (imo) from Rebecca Woodgate in 2010

The 2007 Bering Strait oceanic heat flux and anomalous Arctic sea-ice retreat

Rebecca A. Woodgate, Tom Weingartner, Ron Lindsay
First published: 07 January 2010
https://doi.org/10.1029/2009GL041621  Citations: 340

Quote
3. Implications for Arctic Sea-Ice Retreat

[10] How relevant is this amount of heat (3–6 × 1020 J/yr, i.e., 10–20 TW) in the Arctic?

[11] This much heat could melt 1–2 × 106 km2/yr of 1 m thick ice, maybe 1/3rd of annual arctic sea-ice retreat and comparable to interannual variability in the September ice extent – winter extent is ∼10 × 106 km2,; the 2006 and 2007 September minima were 6 and 4 × 106 km2 respectively (National Snow and Ice Data Center data).

[12] Pacific Waters (PW) are found over roughly half the Arctic Ocean. Averaged over that area (∼5 × 106 km2), the Bering Strait heat flux is 2–4 W/m2, a significant fraction of Arctic annual mean net surface heat fluxes (−2 to 10 W/m2, ERA-40 atmospheric reanalysis [Serreze et al., 2007, Figure 5]).

[13] The Bering Strait heat flux is also comparable to the solar input to the Chukchi Sea, ∼4 × 1020 J/yr (∼1300 MJm−2 yr−1, 1998–2007 range [Perovich et al., 2007] (and subsequent extension), Chukchi Sea area ∼350 × 103 km2).

[14] In fact, the Bering Strait heat flux is surprisingly large for its net volume. Although the Fram Strait inflow is about 10 times greater in volume, its estimated net heat input to the Arctic is ∼30–50 TW [Schauer et al., 2008], only about 3 times our Bering Strait estimate.

[15] Thus, purely in terms of heat, the Bering Strait contribution is large enough to be a significant player in sea-ice retreat. But other factors are also important, viz., the timing of the delivery of this heat to the Arctic; the volume throughflow itself, which may carry heat and ice northward; and (since we seek explanations for interannual change) the magnitude of interannual variability of these properties.
<>

4. Conclusions

[27] Using year-round data from in situ moorings and satellite-sensed sea surface temperatures, we quantify oceanic fluxes of volume and heat from the Pacific to the Arctic via the Bering Strait between 1991 and 2007 with special focus on 1998 to 2007. We find heat flux increases almost monotonically from 2001 to 2007. Reflecting both high volume transports and high temperatures, the estimated 2007 heat flux was the greatest recorded to date, 5–6 × 1020 J/yr (range reflecting uncertainty in depth of the summer surface layer). This is almost a doubling of the total 2001 heat flux and somewhat greater than the incoming shortwave solar input into the Chukchi Sea. Moreover, the interannual variability in the Bering Strait heat flux is slightly larger than that of shortwave solar input to the Chukchi.

« Last Edit: October 08, 2024, 09:52:52 PM by uniquorn »

Glen Koehler

  • Grease ice
  • Posts: 967
    • View Profile
  • Liked: 775
  • Likes Given: 1477
Re: ENSO/ASI Correlation
« Reply #18 on: October 08, 2024, 08:48:43 PM »
RE
"How relevant is this amount of heat (3–6 × 1020 J/yr, i.e., 10–20 TW) in the Arctic?"
"This much heat could melt 1–2 × 106 km2/yr of 1 m thick ice, maybe 1/3rd of annual arctic sea-ice retreat" etc.

     The exponents are mistakenly shown as digits in these examples and elsewhere in the excerpted text.  1020 should be 1020, 106 should be 106 

    Makes a big difference!
“What is at stake.... Everything, I would say." ~ Julienne Stroeve

uniquorn

  • First-year ice
  • Posts: 5379
    • View Profile
  • Liked: 2325
  • Likes Given: 393
Re: ENSO/ASI Correlation
« Reply #19 on: October 08, 2024, 09:31:59 PM »
Tedious, but done. In case I missed any, here is a screen shot.
« Last Edit: October 08, 2024, 09:39:05 PM by uniquorn »

Glen Koehler

  • Grease ice
  • Posts: 967
    • View Profile
  • Liked: 775
  • Likes Given: 1477
Re: ENSO/ASI Correlation
« Reply #20 on: October 08, 2024, 11:02:52 PM »
     I wasn't trying to bust your chops for graphical purity!  I just thought it would help people to realize why the numbers looked odd so they could interpret the article more easily!   ::)  I didn't expect you would go back through and put in superscript every time one appeared.  Tedious indeed. But thanks for doing the edits.
“What is at stake.... Everything, I would say." ~ Julienne Stroeve

binntho

  • Nilas ice
  • Posts: 2332
    • View Profile
  • Liked: 958
  • Likes Given: 250
Re: ENSO/ASI Correlation
« Reply #21 on: October 09, 2024, 10:55:39 AM »
"The hot water intrusions discussed in the paper are moving at 1 m/s which is a lot slower than the apparent surface propagation of salinity vortexes as can be seen in Mercator visuals, the latter moving at a brisk 100 cm/s"

1m/s is 100cm/s?

Yes indeed, which is a normal human walking speed. I measured this some time ago from a mercator salinity video, example shown below. This shows salinity at 34 m depth, and there is one frame per day.The highly turbulent inflows from the Bering are easy to see, with the rest of the Arctic hardly moving at all.

What I did (with a higher resolution video) was to very roughly measure turbulence front propagation from one day to another and found several instances of a maximum just under 100 km/day which is very close to 1m/s.

So how then is 1 meter per second a lot slower than 100 centimeters per second, when they are equivalent?

Doh! I see the error now, sorry about that. Should of course have been 1 cm/s which is definitely slower than 100 cm/s !!!!
because a thing is eloquently expressed it should not be taken to be as necessarily true
St. Augustine, Confessions V, 6

binntho

  • Nilas ice
  • Posts: 2332
    • View Profile
  • Liked: 958
  • Likes Given: 250
Re: ENSO/ASI Correlation
« Reply #22 on: October 09, 2024, 12:18:39 PM »
<>
The conclusion: Not a lot of heat, and moving very slowly. Interesting, but not something that poses an existential threat to Arctic Sea Ice.

Some calculations and a different conclusion (imo) from Rebecca Woodgate in 2010

The 2007 Bering Strait oceanic heat flux and anomalous Arctic sea-ice retreat

Rebecca A. Woodgate, Tom Weingartner, Ron Lindsay
First published: 07 January 2010
https://doi.org/10.1029/2009GL041621  Citations: 340

Quote
3. Implications for Arctic Sea-Ice Retreat

[10] How relevant is this amount of heat (3–6 × 1020 J/yr, i.e., 10–20 TW) in the Arctic?

[11] This much heat could melt 1–2 × 106 km2/yr of 1 m thick ice, maybe 1/3rd of annual arctic sea-ice retreat and comparable to interannual variability in the September ice extent – winter extent is ∼10 × 106 km2,; the 2006 and 2007 September minima were 6 and 4 × 106 km2 respectively (National Snow and Ice Data Center data).

This paper talks about 10-20 TW as compared to the 3TW that I was talking about. The former figure is the total, the latter the bottom component only. And I think the math is slightly off  - I won't dispute that this amount of heat can melt 1-2 million km2 of 1 m thick ice, but this equals 1-2 thousand km3 of ice as compared to the total volume loss this year of around 18 km3. So saying that this amount of heat can melt 1/3 of the annual ice melt is somewhat overstated.

I think we have to separate the surface heat influx through the Bering from the bottom component. The surface influx has a large and outsized effect, as is stated in the other paper you posted:

Some interesting results from this modelled study, in particular I hadn't come across Bjerknes compensation before.

Bering Strait Ocean Heat Transport Drives Decadal Arctic Variability in a High-Resolution Climate Model
Yuchen Li, Wilbert Weijer, Prajvala Kurtakoti, Milena Veneziani, Ping Chang
First published: 20 June 2024
https://doi.org/10.1029/2024GL108828

abstract and some extracts
Quote
The outsized impact of Bering Strait OHT on Arctic sea ice has an obvious geographical reason: the climatological sea ice edge is located much further south in the Pacific sector than in the Atlantic sector (Figure 1). In this model, the ice edge in the Pacific sector is located 5° to the south of the polar circle, while in the Atlantic sector it is located 10° to the north. OHT variability across 65°N has therefore a much more direct impact on the ice edge in the Pacific sector than in the Atlantic sector, an effect that is amplified by the ice albedo feedback. That said, Bering Strait waters are significantly colder than the Atlantic waters, which reach almost all the way to the ice edge (Figure S4 in Supporting Information S1).
my emphasis

The effect that Bering Strait surface heat has is mostly secondary, i.e. the surface influx speeds up the melting of ice in the Chuckhi and neighbouring seas, which significantly increases insolation over open ocean during the summer months, heating the surface even more before it interacts with the ice. This is the major reason why variation in the heat influx from the Pacific has the same correlation with ice melt as that from the Atlantic, even if the latter carries around 10 times the amount of heat as the former - the ice melted by Pacific influx is much further south, and thus causes a much bigger arctic feedback, than any ice melted by the Atlantic influx.

The primary effect of the heat influx, i.e. where the surface heat brought in through the Bering Strait is directly melting sea ice, is limited to the start of the melting season, and is thus never able to bring the full force of annual heat transfer to the ice. A significantly large proportion of the heat influx through the Bering ends up in open waters and is presumably lost to the atmosphere.

As for the bottom component, most of that heat stays below the surface and the heat dissipates slowly upwards, probably over several years, and this has most likely been ongoing for a very long time.

But what might be changing regarding the sub-surface heat transfer is the heat-bomb effect as described in the video posted by Killian a bit higher up. If the ice is broken up or dispersed, and hit by a storm, this heat can potentially be brought up to the surface and cause significant melt. But this is a one-time effect. The heat that has accumulated over the years can only be used once - and some proportion of storm-induced turbulent upward mixing of sub-surface heat will happen over areas with little or no ice, with the heat being lost to the atmosphere rather than brought directly to bear on the ice.
because a thing is eloquently expressed it should not be taken to be as necessarily true
St. Augustine, Confessions V, 6

uniquorn

  • First-year ice
  • Posts: 5379
    • View Profile
  • Liked: 2325
  • Likes Given: 393
Re: ENSO/ASI Correlation
« Reply #23 on: October 09, 2024, 12:54:41 PM »
<>
The conclusion: Not a lot of heat, and moving very slowly. Interesting, but not something that poses an existential threat to Arctic Sea Ice.

Some calculations and a different conclusion (imo) from Rebecca Woodgate in 2010

The 2007 Bering Strait oceanic heat flux and anomalous Arctic sea-ice retreat

Rebecca A. Woodgate, Tom Weingartner, Ron Lindsay
First published: 07 January 2010
https://doi.org/10.1029/2009GL041621  Citations: 340

Quote
3. Implications for Arctic Sea-Ice Retreat

[10] How relevant is this amount of heat (3–6 × 1020 J/yr, i.e., 10–20 TW) in the Arctic?


[13] The Bering Strait heat flux is also comparable to the solar input to the Chukchi Sea, ∼4 × 1020 J/yr (∼1300 MJm−2 yr−1, 1998–2007 range [Perovich et al., 2007] (and subsequent extension), Chukchi Sea area ∼350 × 103 km2).


I found the comparison between Bering Strait heat flux and the solar input to the Chukchi Sea interesting.

ocean heat flux  = 3–6 × 1020 J/yr
solar input         = ∼4 × 1020 J/yr

uniquorn

  • First-year ice
  • Posts: 5379
    • View Profile
  • Liked: 2325
  • Likes Given: 393
Re: ENSO/ASI Correlation
« Reply #24 on: October 09, 2024, 01:35:14 PM »
"The hot water intrusions discussed in the paper are moving at 1 m/s which is a lot slower than the apparent surface propagation of salinity vortexes as can be seen in Mercator visuals, the latter moving at a brisk 100 cm/s"

1m/s is 100cm/s?

Yes indeed, which is a normal human walking speed. I measured this some time ago from a mercator salinity video, example shown below. This shows salinity at 34 m depth, and there is one frame per day.The highly turbulent inflows from the Bering are easy to see, with the rest of the Arctic hardly moving at all.

What I did (with a higher resolution video) was to very roughly measure turbulence front propagation from one day to another and found several instances of a maximum just under 100 km/day which is very close to 1m/s.

So how then is 1 meter per second a lot slower than 100 centimeters per second, when they are equivalent?

Doh! I see the error now, sorry about that. Should of course have been 1 cm/s which is definitely slower than 100 cm/s !!!!

Along jet velocity looks mostly above 50cm/s to me. Where did you find 1cm/s?

https://www.nature.com/articles/s41467-021-22505-5#Fig2

binntho

  • Nilas ice
  • Posts: 2332
    • View Profile
  • Liked: 958
  • Likes Given: 250
Re: ENSO/ASI Correlation
« Reply #25 on: October 10, 2024, 06:01:27 AM »
Doh! I see the error now, sorry about that. Should of course have been 1 cm/s which is definitely slower than 100 cm/s !!!!

Along jet velocity looks mostly above 50cm/s to me. Where did you find 1cm/s?


I may well have misread the numbers. I looked at the article again and found this, from A warm jet in a cold ocean


Quote
Observations of a meandering warm jet
Satellite observations from 15 Sept 2018 show a jet of warm water meandering offshore from Barrow Canyon (Fig. 1b). The warmest surface waters in this image are observed directly over Barrow Canyon. The jet initially turns eastward, but then makes a sharp turn northward/offshore near an indentation in the topographic slope (blue contour). Subsequent meanders in the jet of ~50–100 km scale are consistent with previous observations39,40. The Sea Surface Temperature (SST) anomaly associated with the jet is visible with reduced amplitude over 100 km offshore, impinging upon a remnant patch of multi-year sea ice. Convoluted stirring patterns can be seen between the warmest waters of the jet and cooler Arctic surface water. Upper ocean currents measured during the ship survey (Fig. 1c, arrows) are consistent with the meandering pattern visually apparent in SST, with peak amplitudes of over 1 m s−1 in the core of the jet.

My emphasis. This agrees with what I have said earlier, the surface currents entering Chukchi from the Bering strait can reach a leasurely walking speed of 1 m/s. Note that in the above quotations they are talking about surface features based on satellite observations.

For the sub-surface waters, where I postulated 1 cm/s (and I did make a mistake of writing 1 m/s, and am unable to correct this in the original post) was based on the discussion in the latter half of the article, starting with "Subduction processes" and also with reference to image 4. But I am unable to find the actual "1 cm/s" that I thought I had seen, so perhaps that was just me imagining things ... not for the first time!

Quote
The predicted vertical velocities are of order 10 m per day and predicted cross-jet velocities are a few km per day.

This is a description of movement within rotating eddies of warm water. The above numbers do not say anything directly about propagation speed, but 1 km/day is very close to 1 cm/s. Propagation speed would presumably be lower than rotational speed, although I am not sure if that can be postulated. So lateral propagation could be of the order of a few cm/s but it could also be less than 1 cm/s.

But another article does give direct speed numbers for subsurface currents within the Arctic, this time of Atlantic waters. I had already quoted this earlier in the discussion:

url=https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2018JC014378]Understanding Arctic Ocean Circulation: A Review of Ocean Dynamics in a Changing Climate[/url]

Quote
Mooring measurements indicate Atlantic Water boundary current speeds to be around 2 to 4 cm s−1 (Woodgate et al., 2001). This is consistent with transient tracer data which suggest that Atlantic Water propagation from the Eurasian Basin to the southern Canada Basin (a distance of around 6,000 km) takes around 7.5 years.(5)

Conclusion: I may have imagined seeing actual mention of the 1 cm/s that I stated (or rather: intended to state). But it was based on looking at several lines of reasoning, and should quite possibly have been 2-4 cm/s based on other lines of evidence. But could also be signifiantly lower based on the discussion in the original paper.
« Last Edit: October 10, 2024, 09:41:55 AM by binntho »
because a thing is eloquently expressed it should not be taken to be as necessarily true
St. Augustine, Confessions V, 6

johnm33

  • Grease ice
  • Posts: 919
    • View Profile
  • Liked: 155
  • Likes Given: 130
Re: ENSO/ASI Correlation
« Reply #26 on: October 10, 2024, 10:58:20 AM »
Does a 'jet' imply something faster than a surge?

binntho

  • Nilas ice
  • Posts: 2332
    • View Profile
  • Liked: 958
  • Likes Given: 250
Re: ENSO/ASI Correlation
« Reply #27 on: October 10, 2024, 11:25:22 AM »
Does a 'jet' imply something faster than a surge?
I'm not sure if these terms are defined on any official scales. I would say yes, it is misleading to talk about a jet that to describer something that moves slower than a snail. But not all statements need to be taken literally.
because a thing is eloquently expressed it should not be taken to be as necessarily true
St. Augustine, Confessions V, 6

uniquorn

  • First-year ice
  • Posts: 5379
    • View Profile
  • Liked: 2325
  • Likes Given: 393
Re: ENSO/ASI Correlation
« Reply #28 on: October 10, 2024, 11:38:32 AM »
<>
This agrees with what I have said earlier, the surface currents entering Chukchi from the Bering strait can reach a leasurely walking speed of 1 m/s.
<>

Good luck walking 86km/day   ;)

binntho

  • Nilas ice
  • Posts: 2332
    • View Profile
  • Liked: 958
  • Likes Given: 250
Re: ENSO/ASI Correlation
« Reply #29 on: October 10, 2024, 11:54:08 AM »
<>
This agrees with what I have said earlier, the surface currents entering Chukchi from the Bering strait can reach a leasurely walking speed of 1 m/s.
<>

Good luck walking 86km/day   ;)

Speed does not equal distance!
because a thing is eloquently expressed it should not be taken to be as necessarily true
St. Augustine, Confessions V, 6

uniquorn

  • First-year ice
  • Posts: 5379
    • View Profile
  • Liked: 2325
  • Likes Given: 393
Re: ENSO/ASI Correlation
« Reply #30 on: October 10, 2024, 12:22:52 PM »
<>
This agrees with what I have said earlier, the surface currents entering Chukchi from the Bering strait can reach a leasurely walking speed of 1 m/s.
<>

Good luck walking 86km/day   ;)

Speed does not equal distance!


But you already measured the distance and implied the speed from it.


"The hot water intrusions discussed in the paper are moving at 1 m/s which is a lot slower than the apparent surface propagation of salinity vortexes as can be seen in Mercator visuals, the latter moving at a brisk 100 cm/s"

1m/s is 100cm/s?

Yes indeed, which is a normal human walking speed. I measured this some time ago from a mercator salinity video, example shown below. This shows salinity at 34 m depth, and there is one frame per day.The highly turbulent inflows from the Bering are easy to see, with the rest of the Arctic hardly moving at all.

What I did (with a higher resolution video) was to very roughly measure turbulence front propagation from one day to another and found several instances of a maximum just under 100 km/day which is very close to 1m/s.

binntho

  • Nilas ice
  • Posts: 2332
    • View Profile
  • Liked: 958
  • Likes Given: 250
Re: ENSO/ASI Correlation
« Reply #31 on: October 10, 2024, 12:56:32 PM »
But you already measured the distance and implied the speed from it.

Not "implied", it is what we in Africa call "calculation" - you take the distance travelled and divide by the time used, and the result is something called speed.

Δ(distance)/Δ(time) = speed, a non scalar measure of movement.
because a thing is eloquently expressed it should not be taken to be as necessarily true
St. Augustine, Confessions V, 6

uniquorn

  • First-year ice
  • Posts: 5379
    • View Profile
  • Liked: 2325
  • Likes Given: 393
Re: ENSO/ASI Correlation
« Reply #32 on: October 10, 2024, 01:15:56 PM »
"The hot water intrusions discussed in the paper are moving at 1 m/s which is a lot slower than the apparent surface propagation of salinity vortexes as can be seen in Mercator visuals, the latter moving at a brisk 100 cm/s"

1m/s is 100cm/s?

Yes indeed, which is a normal human walking speed. I measured this some time ago from a mercator salinity video, example shown below. This shows salinity at 34 m depth, and there is one frame per day.The highly turbulent inflows from the Bering are easy to see, with the rest of the Arctic hardly moving at all.

What I did (with a higher resolution video) was to very roughly measure turbulence front propagation from one day to another and found several instances of a maximum just under 100 km/day which is very close to 1m/s.

Good luck walking 86km/day  :(
« Last Edit: October 10, 2024, 01:21:12 PM by uniquorn »

John_the_Younger

  • Grease ice
  • Posts: 590
    • View Profile
  • Liked: 91
  • Likes Given: 196
Re: ENSO/ASI Correlation
« Reply #33 on: October 10, 2024, 07:39:31 PM »
The greatest distance walked in 24 hours is 228.930 km (Guinness)  A third of that would be achievable by a fit enthusiastic normal human being.  I once walked about 50 km in 17 hours in not-ideal conditions.

uniquorn

  • First-year ice
  • Posts: 5379
    • View Profile
  • Liked: 2325
  • Likes Given: 393
Re: ENSO/ASI Correlation
« Reply #34 on: October 10, 2024, 08:26:01 PM »
The greatest distance walked in 24 hours is 228.930 km (Guinness)  A third of that would be achievable by a fit enthusiastic normal human being.  I once walked about 50 km in 17 hours in not-ideal conditions.

The distance from Bering Strait to Barrow Canyon is over 885km, more than 10 days at a constant 'walking pace'. I think the often repeated analogy is a mistaken attempt to minimise the effect of ocean heat transfer, along with descriptions like 'something that moves slower than a snail'.
I'm fed up with it and repeat this:

<>
The conclusion: Not a lot of heat, and moving very slowly. Interesting, but not something that poses an existential threat to Arctic Sea Ice.

Some calculations and a different conclusion (imo) from Rebecca Woodgate in 2010

The 2007 Bering Strait oceanic heat flux and anomalous Arctic sea-ice retreat

Rebecca A. Woodgate, Tom Weingartner, Ron Lindsay
First published: 07 January 2010
https://doi.org/10.1029/2009GL041621  Citations: 340

Quote
3. Implications for Arctic Sea-Ice Retreat

[10] How relevant is this amount of heat (3–6 × 1020 J/yr, i.e., 10–20 TW) in the Arctic?


[13] The Bering Strait heat flux is also comparable to the solar input to the Chukchi Sea, ∼4 × 1020 J/yr (∼1300 MJm−2 yr−1, 1998–2007 range [Perovich et al., 2007] (and subsequent extension), Chukchi Sea area ∼350 × 103 km2).


I found the comparison between Bering Strait heat flux and the solar input to the Chukchi Sea interesting.

ocean heat flux  = 3–6 × 1020 J/yr
solar input         = ∼4 × 1020 J/yr


« Last Edit: October 10, 2024, 08:38:46 PM by uniquorn »

binntho

  • Nilas ice
  • Posts: 2332
    • View Profile
  • Liked: 958
  • Likes Given: 250
Re: ENSO/ASI Correlation
« Reply #35 on: October 11, 2024, 08:17:39 AM »
The greatest distance walked in 24 hours is 228.930 km (Guinness)  A third of that would be achievable by a fit enthusiastic normal human being.  I once walked about 50 km in 17 hours in not-ideal conditions.

The distance from Bering Strait to Barrow Canyon is over 885km, more than 10 days at a constant 'walking pace'. I think the often repeated analogy is a mistaken attempt to minimise the effect of ocean heat transfer, along with descriptions like 'something that moves slower than a snail'.
I'm fed up with it and repeat this:

I am sorry to have fed you up, my intention was never to minimize the effect of ocean heat transfer which is indeed significant as we all know. And the heat transfer through the Bering Strait is a very important factor in annual melt, with a 10 times bigger effect than that of the Atlantic front. Which makes the rapid increase in Bering Strait heat transfer even more serious.

My main point is that, when it comes to things oceanic, change happens slowly in the Arctic, and there are no huge day-to-day changes, or even diurnal tidal surges as some seem to think. Similarly, if there is a correlation between ENSO and ASI (as seems to be the case, and what this particular chat is all about), then that is unlikely to be due to sudden changes in ocean heat influx.

Specifically when comparing the "heat bombs" which are sub-surface heat fluxes that move very slowly and have a limited effect, and do not pose an exisistential threat to the ASI, to the significant and rapidly growing surface influx of warm waters which do pose a threat to the ASI.

And I did not comment on your comparison of insolation heat influx and Bering Strait heat influx, it is no suprrise that they are similar since most of the ocean surface heat coming in through the Bering Strait is from insolation in the Bering sea. When you add insolation in the Beaufort and the ESS seas, the effect becomes even begger. And it is all linked, without the significant surface heat coming through the Bering Strait, most of the insolation iin the latter seas would be reflected back to space.

Putting numbers on the speed with which things happen is very important in order to understand what is going on, and what can happen. The surface influxes at "walking speed" are the fastest flowing waters in the Arctic, but they lose their speed very soon after entering the Chukchi. Generally, the waters of the Arctic move at a significantly lower speed than this, in the region of 1/50th of walking speed, or a few kms per day. Which means that fast and sudden changes are unlikely.

But localized sudden changes from storm-induced Ekman pumping, or topographical eddies, are always a possibility every year and the former is definitely happening more often now than before.
« Last Edit: October 11, 2024, 08:24:00 AM by binntho »
because a thing is eloquently expressed it should not be taken to be as necessarily true
St. Augustine, Confessions V, 6