Thursday, November 10, 2016

TempLS down 0.04°C in October; sea ice is low.

The October TempLS mesh index dropped a little from 0.721°C in September to 0.679°C in October. This establishes a new record in recent times - it is the first month since April 2015 that has not been the hottest of that month in the whole record. October 2015 was the first really warm month of the El Nino. Canada and Mexico were late reporting this month, so I waited for that.

The TempLS grid index showed a greater drop, from 0.734°C to 0.617°C. As usual, this reflects disparity in polar conditions, which remained relatively warm with more cooling elsewhere. Indices like HADCRUT and NOAA may follow this trend. The modest fall in TempLS mesh was very similar to the NCEP/NCAR index, and also UAH lower troposphere.

The main map features are a cool band across Siberia, and warmth in the USA.

Other interesting things are happening. November in the NCEP/NCAR index has been exceptionally warm, up about 0.22°C on October at this early stage. Arctic warmth is reflected in the Arctic sea ice, which I mentioned in the NCEP/NCAR post. That remains below other years, but will eventually freeze. The Antarctic ice has been exceptionally low since mid October, which may be more significant, since the melting season is starting from that low base. Here is a section of the SH radial plot:


  1. Interesting that both the Arctic and Antarctic ice extents are anomalously low at the same time. I have not been following ice extents all that long, but this seems unusual for the satellite record period. I also find it interesting that the low ice extent in the Antarctic is accompanied by anomalously cool sea surface temperatures, whereas they are anomalously warm in the Arctic. Maybe this pattern is related to melting with the onset of summer in the Antarctic but delayed freezing in the Arctic with the onset of winter? I assume the melting Antarctic ice would leave a less dense pool of fresh cold water at the surface near the melting ice. Extra melting for whatever reason could produce more cool water than normal and hence cool anomalies near the melting ice?

    Another oddity is that the daily global sea surface temperature anomaly estimate from the UM CCI portrayal of the NOAA OISST V2 has been plunging the last few weeks, as seen here:

    As of November 9, the estimated global sea surface temperature anomaly is now down to +0.14C relative to 1971-2000. It was as high as +0.38C on October 16. Recently a very large and intense cold anomaly has developed in the North Pacific extending from northern Japan to south of the Aluetians and into the Northeast Pacific. The North Pacific anomaly has dropped to +0.12C and the Tropical Pacific is down to +0.06C with the apparent onset of a weak La Nina beginning. These developments seem to contradict the low polar ice extents. Very puzzling.

    1. Every month a new data point comes in. Taken in isolation it means nothing, as any one point is subject to statistical fluctuations. If you have monthly data back to 1880, a single new month is just 1 part in over 1600 in terms of significance. IOW, one in a thousand is inconsequential.

      I don't get why everyone gets excited about doing this exercise other than they enjoy treating everything like a horse race.

    2. @Bryan: It's called weather. Currently, the Arctic is experiencing a very powerful positive cryosphere feedback (see Cohen et al. 2014 "Recent Arctic amplification and extreme mid-latitude weather" for context and explanations), whereas the Antarctic saw a shift in the prevailing circumpolar Rossby wave pattern earlier this year. The shift is acting such that more floating sea ice is pushed towards the continent as opposed to when sea ice is pushed away from the continent which causes strong positive extent anomalies as seen in the last view years. The circumpolar SSTs are cold because of the usual (perhaps even enhanced) upwelling which will continue for many decades to come. As an aside, this is one of the physical mechanisms that are poorly represented in many GCMs.

      Regarding Pacific SSTs: Two strong Trade wind bursts in the West Pacific have given La Nina a slight boost, but nothing extraordinary. The North Pacific cooling is very substantial and purely wind driven. Extremely strong cold air outflow/advection from Siberia on the one hand and a couple of very powerful extratropical cyclones in the Aleutian region caused the SSTs to nosedive. That's counterbalanced (wrt NH surface temperature) by the associated strong ridge over North America with temperature anomalies way above normal. Since such pattern tend to re-emerge, chances are that the North Pacific stays colder for a while. It's noteworthy that the warm Canadian outflow is increasingly warming up the North Atlantic. That's in total contrast to what we've been seeing over many of the most recent winter seasons.

    3. Web - watching the painters is more interesting than watching the paint dry.

      Oyashio current?

    4. That's a good one JCH. It's like playing sudoku instead of twiddling your thumbs .. marginally more entertaining.

      When US climate science research starts getting defunded thanks to the orange Hitler, I will continue to use zero funds to mathematically solve the sudoku-like puzzles which explain ENSO and QBO.

      There's really no fundamental reason that every major temperature fluctuation can't be predicted ahead of time, save for volcanic-related events. Its all deterministically locked in place due to the massive amount of inertia in the ocean and the wobbling earth.

    5. KarSteN, I agree the rapid changes we are seeing are primarily weather driven. I also agree with your assessment of the large cold anomaly patch in the North Pacific. However some features may be influenced by longer time-scale events like ocean cycles and certainly ENSO. I suspect the warm Arctic sea surface temperature anomalies in recent years may be related to longer term ocean cycles. But it might take hundreds of years of modern measurements to be able to identify these longer term patterns with much confidence. Wish I could live long enough to see that happen. Oh well...

    6. Bryan, PDV has a weak teleconnection with the North Atlantic (via PNA-NAO-link), but everything multidecadal is just forced pseudo-oscillation. In an ideal world, i.e. volcano-free and zilch anthropogenic emissions, where the climate system is in quasi-equilibrium, ocean cycles may introduce some random multi-decadal variability. But in a strongly radiatively forced world, internal variability is dwarfed by external forcing. The last 160 years (the instrumental period) show clearly attributable pseudo-oscillations only. PDV is limited to random phase changes that last no longer than 10-15 years (see Kosaka and Xie 2016 for a start). That's it. The Atlantic follows suit to some degree, but what is known as "AMV" (or AMO) is mostly externally forced. I know that you can list you a huge array of papers, all claiming the existence of AMO. However, not a single paper has put forth a convincing mechanism for why there should be such oscillation. Except of course those papers that argue that AMO is mostly externally forced.

      I guess you see where this is going. Of course I strongly believe that the Arctic SSTs are forced. In fact it's easy to prove. All you have to do is show that the atmospheric circulation wasn't particularly odd recently (last 10 years). Rest assured, it wasn't. We've seen weather conducive for accelerated melt such as we've seen patterns that favour growth or slowed melt. Same is true for mass transport out of the Arctic through Fram strait. Nothing out of the extraordinary. Last winter slowed the growth. This summer slowed the melt. It's only now that growth is hampered again. The resulting warm anomalies are simply due to open water. What makes it last (now 50 days in) is the positive feedback that kicks in. For how it works, please read Cohen et al 2014 (Box 2). This emerging NH cold season is a textbook example of the so-called cryosphere feedback. It's not gonna be the last time we've seen it. And it's not random natural variability that bestows such patterns upon us. It's a powerful Global Warming feedback. Perhaps the first which comes with a real potential for a tipping point scenario.

      On the plus side for you, you don't need to wait long to realise that it is indeed a forcing feedback and not some weird ocean cycle ;-)

    7. "But it might take hundreds of years of modern measurements to be able to identify these longer term patterns with much confidence. "

      That's like punting on first down! Don't give up so easily.

      It's becoming rather obvious that ENSO is predictable with an interval of just over 40 years required to predict the oscillations for many decades in advance. See the linked chart in my last comment.

      The intriguing aspect to this is that if the Chandler wobble periods were known as of 1950, the El Nino of 1998 could have been predicted 48 years in advance. The Southern Oscillation was discovered in the 1920's.

      PDO and ENSO are also very similar, with PDO featuring additional long term variability.

  2. KarSteN and @whut, thanks for the perspective and links. I studied meteorology and engineering in college up to the master's level, and before I recently retired, I had about 40 years of experience in weather and air quality forecasting, monitoring, quality assurance, and monitoring data validation. Since retiring, I have been trying to learn more about ocean and cryosphere influences on weather and climate. Living in Central Texas, I don't see much in the way of cryosphere, but when it rarely snows or we get freezing rain here in the Austin area, the city quickly grinds to a halt. :)

    So, thanks for the links. I will take a look. The ENSO model graph is interesting, although I see some glaring misses, especially before the training period, although I have serious reservations about ocean temperature accuracy prior to the satellite era and increasingly less confidence the farther back in time. And of course that model cannot account for strong explosive volcanic eruption influences like Chichon and Pinatubo or extreme explosive events like Yellowstone or La Garita (not to mention meteor strikes like the one that hit near what is now Chesapeake Bay about 35 million years ago).

    I would prefer to see a comprehensive land/ocean/atmosphere/solar/earth mechanical and orbital climate model for periods of decades to centuries, but I realize this is much easier said than done. Ideally, in the long run, climate models should be able to accurately predict glacial periods and interglacial periods, but that will be an extreme challenge and could take hundreds of thousands of years to achieve any kind of appreciable accuracy - although if Javier is right, maybe sooner:

    Of course, in the very long run, the sun will eventually bake the earth to a crisp and humanity will have to leave the earth to survive, but there should be many millions of years to deal with that problem if humanity survives long enough.

    I routinely review GFS weather forecast output and I am always amazed at how well it performs for a few days. NCEP runs the model for 384 hours, but I rarely see much accuracy beyond about 5 or 6 days. I expect that global climate models as they improve will have similar problems but over longer time scales. Regional climate forecasts that I have seen even for a year or less are currently abysmal. So I have little confidence beyond that time scale for now.

    1. That's pretty wishful thinking!

      I've followed Doug Smith's multidecadal forecast model at the Met Office since it started. Given the extreme difficulty and unlikelihood of success, I think it has actually done fairly well, but I have no idea how people are required to evaluate it. It forecast that 2014 would be a smashing record year, which it was, but barely. 2015 was perhaps more what their model forecast. So is it a complete fail for a multidecadal model to miss a by a year? I don't know, but I don't think so. They also predicted half the years after 2009 would be record warmest years.

      If people want to be harsh, then it's a flop, but to me, the basic picture of the years after 2009 - so far they have the basic picture right. In the early years of their prediction they had natural variation swamping AGW. That became the so-called pause in global warming. On Gistemp, half the years after 2009 are record warmest years.

    2. @Bryan: Just in case you weren't aware. GCMs do pretty well with glacial and interglacial periods (see e.g. Ganopolski et al. 2010). But that's more ice-sheets dynamics than anything. What's more important now is the fast response. With a few exceptions (ENSO response, Southern ocean upwelling, Cryosphere feedback), the models do a very good job reproducing the modern warming period wrt spatio-temporal changes over the last 160 years. That's even more true if we account for their known limitations and biases of which there are many. But they are explicable and traceable. And they don't change the big picture.

      Generally, it is well-known that there is a temporal gap in skill. Everything between 3-15 years is literally unpredictable as internal noise in the system overwhelms the climate change or external forcing signal. The first 3 years come with some "auto-correlative" skill due to ocean-atmosphere coupling. Beyond that the atmosphere-ocean-coupling wipes out the internal ocean signal. There might be skill in predicting the ENSO phase (if WHUTs hypothesis holds true), but the PDV phase can't be predicted (except under strong radiative forcing influence). After 15 years, the forcing signal will start to emerge as true (unforced) internal variability has a limited retention period.

    3. Bryan said:

      "... although if Javier is right, maybe sooner: "

      That guy Javier is both clueless and devious. Part of the skill in navigating climate science is in dismissing people that are trying to obfuscate and have an agenda. Javier is a slick huckster that I can easily ignore.

      My model of ENSO is of course a hypothesis, but it is based on an effect that must occur. The ocean's volume is a sensitive measure of any angular momentum variations in the earth's rotation rate. Those small variations will induce a sloshing in the ocean's volume -- if this does not happen, Newton's laws of conservation of momentum will be violated. It's essentially a matter of determining how big the effect is. My premise is that the effect is bigger than is imagined, and that the prevailing wind is not what causes the sloshing.

      Given that premise, I apply only known variations in the momentum such as from the Chandler wobble and the lunisolar rates. If one uses any other periods than these in the model for ENSO, the fit is horrible. This of course still does not prove that the effect occurs (because correlation does not equal causation) but it is substantiation for a potential mechanism. If the fit using the wobble and lunisolar periods was also horrible, it would easily be enough to dismiss the mechanism. Same holds true for QBO.

    4. @whut: "Part of the skill in navigating climate science is in dismissing people that are trying to obfuscate and have an agenda."

      I definitely agree. There is plenty of that going on in the climate science world. However, anyone can be right about one thing and wrong about another. So I try to evaluate hypotheses on merit and not on who made the hypothesis. But I admit that I run into some people that endlessly try to push hypotheses that I find have little merit.

      I also like to hear both sides of an argument and I try to avoid listening only to those I tend to favor, although sometimes this is easier said than done. It's very easy to fall into the pit of ignoring what does not fit your pre-conceived notions. Too many people put on blinders to see only what they want to see and I fall into that trap sometimes. Confirmation bias can sometimes be difficult to avoid.

    5. Bryan, I'd suggest that there are a bunch of Trump-like scientists doing climate modeling. The ones that spring to mind are Lindzen, Tsonis, Salby, and Curry. It's about as easy to tell that they are offering up scoops of dung as it is to determine that Trump is a blithering idiot. And these people write textbooks meant to teach college students about atmospheric science. Its really not about putting blinders on as just ignoring their work, and looking elsewhere.

      Here's a good quote by Pierrehumbert on Lindzen:

      ""... Lindzen’s theories that were wrong. Unfortunately, bad scientists don’t acknowledge their mistakes; Lindzen keeps inventing ever more arcane theories to avoid admitting he was wrong.""

      That's what we are dealing with and it makes no sense to even consider an amateur manipulator like Javier, and I'd rather take on the professional types since they have the clout.

      One guy that is extremely hard to corner is Roger Pielke Sr. I have looked at what he has written and can't find much fault, yet he is pretty harsh toward the AGW consensus. I have since looked at his writing on aliasing of climate measures and the math matches what I have been doing with QBO.

  3. Here is a good discussion of decadal forecast modelling.

    I would say the AMO/North Atlantic-based efforts failed miserably and will continue to do so.

    1. JCH, thanks for the link. Climate models definitely need to be able to replicate the real world and to do so on a regional basis as well as a global basis. There is obviously still a long ways to go in that regard.