New paper attempts to link solar cycles and streamflows

Mike Wallace writes about his new paper, with some references to his previous posts here:

First, a post on streamflow forecasting using solar cycles for high altitude catchments.

Second, a post on the synoptic scale quasigeostrophic continuum. Third, a piece on atmospheric moisture waves.

The new paper covers other directions as well, but perhaps of greater interest is its identification of a different ENSO parameter which ties everything together, namely the trade wind velocities at 800 mb across the western equatorial pacific. Until this point, solar climate researchers have not been able to identify a ‘bottom up’ or ‘top down’ solar cycle signature. The intermittent evidence challenged a comprehensive picture. This paper identifies a lagged solar cycle
signature from the bottom through to the TOA across the footprint of note.

The paper is here:

https://www.tandfonline.com/doi/full/10.1080/02626667.2019.1567925

Trade winds localized within the Western Equatorial Pacific express lagged and statistically significant correlations to sunspot numbers as well as to streamflow in rivers of the Southern Rocky Mountains. Both correlation sets were integrated in a linear regression analysis to produce relatively accurate sub-decadal streamflow forecasts for an annual and a 5-year average. In comparison to the autocorrelation technique, the prototyped method yielded the highest correlations, the highest goodness-of-fit scores, and the lowest root mean squared errors, for both the 5-year average and the annual average assignments. Of all of the cases examined, the highest Kolmogorov-Smirnov test scores between observation and prediction were found for the single solar-based forecast 5 years in advance for the 60-month average streamflow of the Animas River in New Mexico.

From the paper:

Figure 2. Normalized 5-year trailing average time series observations for non-continental and continental locations: (a) solar and Pacific Ocean atmosphere parameters, and (b) streamflow observations in the Himalaya and Rocky Mountain basins of the Northern Hemisphere.

Summary and conclusions

Considerations of past research regarding solar cycles, Hadley and Walker circulation patterns, and streamflow characteristics of several mid-latitude, high-altitude watersheds have pointed to the potential for improved multi-annual to sub-decadal forecasting of streamflows in targeted locations. Such conditions appear to apply to
Northern Hemisphere watersheds of the Himalayas as well as the Southern Rocky Mountains of the Western USA.
Conditions also appear favorable in some Southern Hemisphere watersheds of the Andes, with the caveat that forecast spans are expected to be shorter in potential. The initial examples studied do not yet define any limit throughout the American Cordillera.

In the development of these conclusions, a set of correlations and linear regressions were explored for key sequential features based on previous published research and currently available solar (SSN), trade wind (TWWP), outgoing longwave radiation (OLR), geopotential height (Z), divergence of latent heat (LEDIV), and other indexes. Equivalent exercises were applied towards potential connections of some of those parameters to streamflow datasets for candidate streams of the RockyMountains. A two-stage regression-based forecasting approach was then applied to exploit some of the highest lagged correlations that were identified. The forecasts were compared to forecasts for the same streamflow
datasets via a conventional autocorrelation technique.

The training forecasts based on the new CRMA applications were generally the most accurate of all featured methods, for the series considered under a 5-year trailing average. Through a sequence of solar and trade wind regression exercises, forecasts for this set were advanced as far as 6 years into the future. The forecasts under the new method were also found to be more accurate than the conventional method under an annual average with a 2-year lead forecast approach, although the fidelity of all results was diminished in comparison to the 5-year average set of forecasts. An additional but limited investigation demonstrated high-fidelity 5-year trailing average forecasts for the Animas River based directly on solar cycles taking place 5 years in advance.

The success of the proposed methodology is expected to apply to other regions meeting the target criteria, including the Ganges River in India and the Rio Biobío in Chile. Subsequent exploration of monthly correlations between the TWWP and sunspot cycles suggests that, for appropriate locations, advances of hydroclimate forecasting accuracy with monthly resolution yet multi-annual lead times may also be possible through the new technique.

Advertisements

124 thoughts on “New paper attempts to link solar cycles and streamflows

  1. Heh, Joan Feyman et al and the correlation of Nile River levels with aurorae boreales.

    There are long data series and good data. Told ya there was something there.
    =========================

    • Beth the Serf, in an earlier thread found a piece by Will Alexander correlating sunspots with river levels in South Africa. So now there is correlation of something about the sun with river levels in North America and in North and South Africa.

      I don’t pretend to understand the mechanism, but I’m getting a feeling.

      Once upon a time, I could make it rhyme,
      With rhythm and reason, with lemon and with lime.
      ====================================

      • Apparently Will Alexander made predictions. Is there anybody out there capable of verifying that and checking out his predictions?
        ==========================

        • Hi
          In 2007 I was emailed by Mr. Will Alexander about some work I was doing on the solar magnetic (Hale) cycle at the time. Conclusion of our exchange was that the 22 year magnetic polarity cycle appears to be more relevant than the better known 11 year sunspot cycle. Since than the cycle has completed only about one half of its period, so even if he made a prediction I would think it is far too early to make a judgement.

      • “Once upon a time, I could make it rhyme,
        With rhythm and reason, with lemon and with lime.”

        Still can, Kim. Still do!

        pg

      • Kim,
        Yup he suggested a drought reversal and floods in 2016, which actually happened.
        The dumb ANC government there is so incompetent and useless that they never even noticed – just kept begging for global warming handouts.

      • That’s a very interesting article. It also puts forward a model for sunspots due to acceleration and deceleration of the Sun in the Solar System path around the Galaxy. I am not sure that is correct, but it is interesting nonetheless. Apparently the Sun advances and lags with respect to the SSB (barycenter) in a ~22 year cycle with solar minima at the points when it reaches zero acceleration at 180° directly ahead or behind the SSB.

        I guess the Solar System movements look very different when considered from the center of the Galaxy. The planets describe spirals along the SSB line of movement while the Sun moves to compensate their combined mass and position by being in the exact opposite. I would not be surprised if solar activity is related to that.

        • “…not be surprised if solar activity is related…”. Let’s not forget the 11 1/2 year orbital period of Jupiter.

        • Leif always claimed the movement around the barycenter of the solar system was trivial. Is that so around the barycenter of the galaxy too?
          =================================

          • Another way to ask this is if the effect on the gas giants of movement around the SS barycenter is not trivial, then the movement of the sun around the galactic barycenter may not be trivial in its effect on the sun. But is it so?
            ============

          • kim, Leif as usual was right. The key is to understand that objects freely floating in space cannot detect varying gravitational fields, because everything is equally affected. So the gravitational effect of say Jupiter on the earth is zero, because we’re freely floating in the varying gravitational field caused by all matter in the universe near or far.

            The trivial remaining effect is tidal. If we are standing and the sun is right overhead, our heads are closer to the sun than are our feet. This means the attraction due to gravity is stronger at our heads than our feet. This difference in force stretches out our bodies, making us stand slightly taller at noon than at sunset.

            And the tidal effect from say Jupiter is many, many orders of magnitude smaller than the tides of the sun and the moon. Gravity falls off as one over the square of the distance. Tidal forces fall off as one over the cube of the distance, a much quicker dieoff.

            You might enjoy my post, Canute Ponders The Tides.

            Best to you,

            w.

          • Kim,

            I am not a big fan of the barycenter explanation, but how could we rule out something as trivial when we don’t know if it has an effect or not?

            For what I have read there are several different possible explanations on how the movement of the planets could affect the Sun. There is another one based on the torque exerted by the planets, that does not rely on tidal effects. And there could be another one based on the positive and negative acceleration on the speed of the sun in the galactic path as it compensates for the movement of the planets.

            As it happened with Wegener’s continental drift, it could be a case where the evidence is produced before there is sufficient knowledge to explain how it happens.

          • Landscheit’s correlation have long struck me as in need of explanation. And now so is this one about the accel/decel of the sun correlated with the sunspot cycles.
            =========================

    • Interesting again kim. I’ve already benefited from your noting of the European solar cycle and precipitation study. I’d be interested to learn more about your insights and/or perspectives. Please feel free to email at mwa@abeqas.com. Or if you decide to author a post or have already done so I’d be interested to read it. There appears to be a persistent and extensive body of solar climate work for those who have the interest to explore it. I’ve been focused on a narrow objective but now that this paper is out, I’m planning to expand my reading beyond the fine but limited Hoyt and Schatten and Gray et al. publications.

    • Additionally his SSN record (yellow line) appears to be off (lagging by about 1 year).
      Compare the yellow line SSN minimas to the actual dates.
      SC 20/21 min: 03/1976
      SC 21/22 min: 09/1986
      SC 22/23 min: 08/1996
      SC 23/24 min: 12/2008

      • My suggestion would be for Mike Wallace to substitute the historical Oulu Neutron count record for SSN.

        (BTW, I’m not suggesting I agree with the Svensmark Hypothesis.)

    • Thanks for spotting that Joel. Somehow when I downloaded that longer trade wind time series which dates back to about 1948, I only captured up to 2012. I’ve also been relying upon the satellite resource for the same parameter. The records from that reach as long as the others in that image. I took a look again and I can say that the trade wind pattern swings back up after that current end point, just as the others do. Later in the full paper, I exclusively rely on the satellite trade wind coverage, from https://www.cpc.ncep.noaa.gov/data/indices/wpac850

  2. Here is an alternative explanation that has also been partly published in the peer-reviewed literature:

    Our lunar tidal model attempts to link the cycles seen in the lunar tidal alignments on inter-decadal to centennial time-scales with short-term lunar cycles that initiate El Nino events.

    https://astroclimateconnection.blogspot.com/2019/02/the-lunar-tidal-model-part-4.html

    Part 4 explains the lunar model. However, if you want to understand how this is linked to the cycles seen in the lunar tidal alignments on inter-decadal to centennial time-scales (i.e. thePerigean New/Full Moon Tidal cycle) you will have to slog your way through parts 1, 2, and 3.

    [Note that I am not claiming that there is only one mechanism that can explain everything. All I am saying is that the long-term quasi-cycles in the lunar tides play a significant role in addition to that of the Sun].

    • Ian it is clear that both the Sun and the Moon have an effect on climate, and I would dare to say that the effect of the Moon coupled to the Sun is mostly through atmospheric and oceanic lunisolar tides. I think we can agree on that.

      The problem is the Sun affects the climate through multiple pathways, including UV, TSI, solar wind, magnetism. And it affects the system at multiple levels, the thermosphere, the stratosphere, the polar vortex, tropospheric zonal wind patterns, and the surface and subsurface of the oceans.

      So the effect of solar variability on climate is multiple times that of the Moon. So if the sun is neglected (as Fritz Vahrenholt puts it), then the Moon is neglected at the n power. The bright side is that you have no competition in your research.

      I have been looking in detail to the lunar role on Dansgaard-Oeschger events and on the 1500-year cycle. I’ll let you know what I got when I am done.

  3. Sorry, I forgot to post this as well.

    Everyone seems to acknowledge that the 9.1-year periodicity that is observed in the world’s mean temperature is probably due to some form of lunar tidal forcing but no one seems to want to investigate how this forcing works.

    *******
    Like Michael Wallace, I have made predictions of climate conditions using another version of the Lunar tidal model that is more suited to the mid-to-high latitudes:

    Here is my submission to the Australian (Federal) Senate Committee on Recent Trends in and Preparedness for Extreme Weather Events in 2013.

    Wilson, I.R.G., 2013, Personal Submission to the Senate
    Committee on Recent Trends in and Preparedness for
    Extreme Weather Events, Submission No. 106

    https://www.aph.gov.au/Parliamentary_Business/Committees/Senate/Environment_and_Communications/Completed_inquiries/2010-13/extremeweather/submissions

    Go down to submission number 106 to download the pdf file.

    The submission in 2013 had the following words prominently displayed on the front cover:

    “South-Eastern Australia needs to prepare for hot dry conditions in the summer of 2019 and possible extensive flooding in 2029”

    Needless to say, my submission was totally ignored by the committee.

  4. I live in Central Ohio. The linked app is unavailable. But, I looked a the maps in the paper. As near as i can tell I am moving to Evansville IN. But, if it is RCP8.5, I am going to Cape Girardeau MO. I don’t see either of those fates as being worth any effort to avoid. I certainly would not support a tax increase for that purpose.

  5. .
    ❶①❶①❶①❶①❶①❶①❶①❶①❶①❶①❶
    ❶①❶①
    ❶①❶① . . . Temperature and Population by Country . . .
    ❶①❶①
    ❶①❶①❶①❶①❶①❶①❶①❶①❶①❶①❶
    .

    How many people will die, if we reach the +2.0 degrees Celsius temperature limit?

    Does anybody know?

    Even an approximate number?

    It is difficult to give an accurate number, because it is a totally new situation.

    But I have found a way to estimate the number of deaths.

    It took quite a bit of work. But in the end, the answer was obvious.

    The answer is so obvious, that I am not going to tell you the answer.

    I have done all of the work so far. It is about time that you pulled your weight.

    Don’t worry. I am only asking you to look at a graph. Do you think that you could manage that?

    ====================

    This is the only graph that you need to look at, to fully understand global warming.

    It even comes with a money back guarantee.

    So what are you waiting for, click the following link:

    https://agree-to-disagree.com/temperature-and-population-by-country

  6. Sadly all the real facts in the World will not stop the Greens and especially the Media support for CO2 does it all. Its the classic “Keep it simple stupid”.

    Complicated facts as with this article will just result in the ” Eye glazing” effect on the part of the general public.

    Perhaps the long awaited Red team versus Blue team, followed b y the EPA commencing a Court case to “Prove” that CO2 s a good and essential gas might help, but otherwise ii its a case of having to wait till the lights start to go out. .

    MJE

  7. In males a tickled scrotum is often correlated, after a lag, with an erection. Is this often-seen event worthy of a scientific investigation with associated societal costs, any more or less than these sunspot variations? Geoff

    • if sunspot variations are more controlling than CO2 then the premise of CO2 controlling climate is wrong; and using CO2 to control society is a failed concept.

      for your analogy to work, tickling would be the primary controlling factor … regardless of who the tickler is. This obviously not the case. So, no, the tickling premise is not worthy of scientific investigation (unless you just want trough work and you throw in a burka wearing tickler to enhance your grant opportunities).

  8. The jet stream from the north prevents the formation of El Nino.
    http://tropic.ssec.wisc.edu/real-time/mtpw2m/product.php?color_type=tpw_nrl_colors&layer=tpw&prod=midpac&timespan=24hrs&anim=html5
    https://www.tropicaltidbits.com/analysis/ocean/nino12.png
    The weak polar vortex in the lower stratosphere over North America will cause the inflow of Arctic air far to the south of the US.
    https://earth.nullschool.net/#2019/02/18/1200Z/wind/isobaric/70hPa/orthographic=-108.60,54.37,495

  9. Well, I went to see the paper … paywalled. Let me get this straight. The author of the paper wants to publicize the paper but it’s paywalled … grrrr.

    So I finally dug up a copy of the paper, gotta love Sci-Hub, it’s here. The paper claims a correlation between the wind at 800 mb over the Western Equatorial Pacific, which it identifies 5°N/5°S latitude and 135E to 170W longitude.

    And how do we know what the winds are doing up two kilometres (1.2 miles) up in the troposphere in the middle of the Pacific Ocean?

    Why … a climate model, called a “reanalysis” model. Here’s the first problem. Climate models output a lagged and resized version of what they are input. They are input with solar variations. This means that we are very likely to find a solar signal in their output … but the real world is not nearly as linear. Instead, it is chaotic and heavily damped, and inputs may just disappear.

    Next problem. Where there are lots and lots of observations, the reanalysis models cannot get too far away from reality. But where there are very few observations, the model is just guessing as to the values. So think about this … how many observations of trade winds two kilometers above the surface are being done in the middle of the Pacific Ocean? You’re right …

    Here’s Pat Frank on the subject of reanalysis computer models …

    I think all reanalysis is indeed unreliable. The reason is that no climate model deploys a valid theory of climate.
     
    Even where reanalysis is of the known climate, for which the model has been parameterized to reproduce certain observables, the uncertainty remains in the reanalysis because the parameters merely are tuned to have offsetting errors. Other sets of parameters, reflecting different physical relationships, will reproduce the same set of observables.
     
    That is, the underlying physical theory is incomplete or wrong or both, no matter whether the tuned parameters reproduce known observables, or not. Therefore large uncertainties remain in the calculational product. The uncertainties are merely hidden because of the parameter tuning.
     
    No one in the modeling community seems to pay attention to these absolutely critical details of scientific rigor. By excluding proper physical error analysis, climate modelers are claiming to know what they manifestly do not.

    Next, the author is working with trailing averages of highly autocorrelated series. When you start with a highly autocorrelated series and then apply a trailing average, the final output has a Hurst exponent of 0.98 to 1.0.

    This means, in turn, that his sunspot numbers and the like have an “effective N” of less than two … so we cannot even calculate a p-value for the proposal that two of them are correlated. You simply cannot do such analyses with smoothed data. See William Briggs, Statistician To The Stars, for a further discussion of the issues.

    Next, in his Figure 1, he has detrended the various indices. This has made the fit between them much better … but I don’t understand the justification for this procedure. He’s basically working with trendless indices (e.g. SOI, sunspots, AMO, PDO). So it would seem the only reason to detrend short sections of them is to increase the correlation … not that it matters, since the data is so smoothed that improved correlation is still meaningless.

    Finally, it appears that he’s found a couple of rivers with some small correlation to sunspots. He’s smoothed them as well, so it’s impossible to tell if there is a significant correlation. And one of them appears to be three rivers spliced together, which is an issue in itself.

    However, here’s the main problem. There are thousands of rivers all around the planet. And if you look at enough of them, one or more of them will have a good correlation with some given dataset … so what? Seriously, if you look long enough, you can find a river flow that correlates to the changing value of the Japanese yen or any other dataset you choose. To take care of this you use the “Bonferroni Correction”. It requires that the further you look, the better a fit you need to be significant. It seems he has not considered this problem.

    But like I said … it makes no difference because he’s smoothed the data to the point where no correlations are significant …

    So I’m sorry, but I don’t find anything that is at all supportive of the idea that solar sunspot-related variations affect the climate down here at the surface.

    Best regards,

    w.

    • Willis,

      You never answered to the evidence I posted. As usual when I talk about solar control of ENSO the anti-solar crowd chooses not to speak.

      I use the ONI (Oceanic Niño Index) from NOAA:
      http://origin.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/ONI_v5.php
      And the monthly smoothed sunspot dataset from SILSO:
      http://www.sidc.be/silso/DATA/SN_ms_tot_V2.0.txt

      Selecting the periods when solar activity (sunspots) increases from 35% of maximum (for each cycle) to 80% of maximum, it turns out that the average ONI is -0.649, which is full fledged La Niña conditions. For six solar cycles.

      Here is the data:
      Month Sunspots ONI
      SC19 35-80%
      1955.874 103.5 -1.7
      1955.958 114.8 -1.5
      1956.042 125.8 -1.1
      1956.124 139.4 -0.8
      1956.206 154.7 -0.6
      1956.29 168.1 -0.5
      1956.373 180.4 -0.5
      1956.456 193.9 -0.5
      1956.54 206.1 -0.6
      1956.624 211.8 -0.6
      1956.708 214.5 -0.5
      1956.791 220.6 -0.4
      1956.874 226 -0.4
      SC20 35-80%
      1966.538 71.4 0.2
      1966.623 80.3 0.1
      1966.707 89.5 -0.1
      1966.79 95.8 -0.1
      1966.874 99.4 -0.2
      1966.958 103 -0.3
      1967.042 106.2 -0.4
      1967.123 111.6 -0.5
      1967.204 116.5 -0.5
      1967.288 119.8 -0.4
      1967.371 123.9 -0.2
      SC21 35-80%
      1978.204 98.7 0.1
      1978.288 109 -0.2
      1978.371 117.8 -0.3
      1978.455 126.6 -0.3
      1978.538 138 -0.4
      1978.623 147.3 -0.4
      1978.707 153.6 -0.4
      1978.79 157.3 -0.3
      1978.874 160.4 -0.1
      1978.958 166.7 0
      1979.042 175.2 0
      1979.123 185.4 0.1
      SC22 35-80%
      1988.206 84.9 0.1
      1988.29 93 -0.3
      1988.373 101.4 -0.9
      1988.456 114.3 -1.3
      1988.54 128.5 -1.3
      1988.624 141.1 -1.1
      1988.708 151.2 -1.2
      1988.791 156.9 -1.5
      1988.874 164.4 -1.8
      SC23 35-80%
      1998.204 72 1.4
      1998.288 76.9 1
      1998.371 80.8 0.5
      1998.455 85.4 -0.1
      1998.538 89.8 -0.8
      1998.623 93.5 -1.1
      1998.707 96.4 -1.3
      1998.79 98.2 -1.4
      1998.874 102.3 -1.5
      1998.958 110.4 -1.6
      1999.042 118.4 -1.5
      1999.123 122.5 -1.3
      1999.204 122.3 -1.1
      1999.288 125 -1
      1999.371 132.6 -1
      1999.455 136.3 -1
      1999.538 138.1 -1.1
      1999.623 142.9 -1.1
      SC24 35-80%
      2010.958 42.5 -1.6
      2011.042 45.7 -1.4
      2011.123 48.8 -1.1
      2011.204 53.8 -0.8
      2011.288 61.1 -0.6
      2011.371 69.3 -0.5
      2011.455 77.2 -0.4
      2011.538 83.6 -0.5
      2011.623 86.3 -0.7
      2011.707 86.6 -0.9
      2011.79 87.4 -1.1
      2011.874 89.4 -1.1
      2011.958 92.5 -1
      Average ONI -0.648684211

      Definitely statistically significant.

      With the data above, my 100,000 Monte Carlo runs analysis with the average of six random 12-continuous-month periods in the ONI dataset only came below -0.64868 in 0.696%, significant at p<0.01.

      The occurrence of La Niña when there is a 35-80% increase in solar cycle activity is not due to random.

      Solar variability control of ENSO is a fact. Given the importance of ENSO to tropical climate we can say that solar variability is an important climatic factor.

      • too funny,

        smoothed some solar data and anything will pop out.

        try a test for granger causality.

        you wont.

        • smoothed some solar data and anything will pop out.

          No it won’t. My Monte Carlo analysis says this only happens 0.7% of the runs. And this is only for La Niña when solar activity is increasing. Then we have El Niño starting right around the solar minimum, and La Niña that precedes it. The probabilities of all that due to chance are astronomically low.

          try a test for granger causality.

          Why should I try any test you come up with? Because otherwise you won’t believe it? Like I care.

        • Says the king of smoothing, who just spent his professional life defending the spatial averaging of regional temperature data in order to form a singular global image!

      • Javier:
        Your response to Willis’s and Mosher’s comments are strawman distractions, not direct responses to their points.
        Why?

        “Javier February 14, 2019 at 1:51 am
        Willis,
        You never answered to the evidence I posted. As usual when I talk about solar control of ENSO the anti-solar crowd chooses not to speak.”

        That gripe sounds like a persecution complex, not a reasoned rational analysis.

      • Javier,

        Your theory says with 35-80% increase in sunspot, you get La Nina. You show us 76 samples. I counted 30 not La Nina (ONI > -0.5) The success rate of your theory: (76 – 30)/76 = 0.60
        In empirical science, null hypothesis can be rejected at P < 0.05
        That means success rate of prediction greater than 0.95 or 2-sigma significance. In physics, the gold standard is 5-sigma.

        Try to falsify your theory by counting how many La Nina when sunspot increase is LESS than 35-80%. Null hypothesis can be rejected if:
        (No. of La Nina) / (No. of samples) < 0.05

        • I guess you didn’t understand. From >35% to <80% in the ascending phase, on average La Niña conditions are prevalent (ONI < 0.0). You cannot count every month within the period as a separate instance. Each solar cycle is a separate instance and has to be analyzed as a whole. The average of the six solar cycles is below –0.5. Try to get that by chance.

          • If you count each 11-yr cycle as one, you only have six cycles (1950-2018). You can’t draw much statistical significance from 6 samples. Get more than 30 samples to analyze

          • That’s what you say. The average ONI value of six periods of 12 consecutive months randomly chosen should approach zero and follow a near gaussian distribution. Obtaining a value of –0.65 is not random at 99.3% statistical significance (p<0.007).

            It seems you don't understand because you don't want to understand.

          • I understand very well but don’t take my word. Try to publish your theory in solar physics journal (not the biased “climate science” journals). See if the reviewers will accept your claim that the evidence you present is statistically significant. Your paper should have no trouble getting published if what you claim is true.

          • Probability and Hypothesis Testing

            Let: L = La Nina (ONI -0.5)
            Observed outcome of six cycles (averaged 35-80%): L, N, N, L, L, L

            Your hypothesis did not predict the outcome L = 4, N = 2. Had your hypothesis predicted it, the probability P of it occurring by chance is P = 1/c
            c = number of possible combinations:
            L = 0, N = 6
            L = 1, N = 5
            L = 2, N = 4
            L = 3, N = 3
            L = 4, N = 2
            L = 5, N = 1
            L = 6, N = 0
            c = 7, P = 1/7 = 0.14

            Your hypothesis did not predict the outcome: L, N, N, L, L, L (La Nina on 1st, 4th, 5th, 6th cycles, No La Nina on 2nd, 3rd cycles). Had your hypothesis predicted it, the probability P of it occurring by chance is P = 1/p
            p = number of possible permutations = 2^6 = 64
            P = 1/64 = 0.016

            Therefore, your hypothesis did not predict a priori the observed outcome. The a priori probability of random events is not in favor or against your hypothesis. The probability P of an outcome occurring after the fact is P = 1. Your hypothesis sort of “predicted” the outcome after the fact. Good luck on the peer review!

          • The symbols did not come out. Here it is again.
            Let: L = La Nina (ONI less than -0.5), N = No La Nina (ONI greater than -0.5)

          • The hypothesis is solar activity causes negative ONI in the average of every solar cycle 35->80% activity phase.

            You don’t define the hypothesis, I do. And this is not a binary test Niña/Non Niña with probabilities assigned by combinatorial analysis. ONI values for a period follow a Gaussian distribution, so the probabilities are calculated from the value obtained.

            According to your simplified, silly analysis, getting a -0.4 ONI value in every solar cycle would mean a failure of the hypothesis that solar activity has no effect on ENSO, yet getting a -0,4 in every cycle would be a very unusual result. Getting a -0.65 is even more unusual. It cannot be explained by chance despite your pathetic attempts.

            And this is only considering the Niña that happens at the beginning of the solar cycle. When considering also the skewed probabilities for other outcomes, like the Niño during the solar minimum, the Niña that precedes it, and so on, it is impossible to reject statistically the conclusion that solar activity has a very strong effect on ENSO.

            You don’t need to worry about it being published. Other scientists will be able to decide if this is convincing to them or not.

    • Gosh

      please dont quote Pat frank on Reananlysis, he is dead wrong.

      “Even where reanalysis is of the known climate, for which the model has been parameterized to reproduce certain observables, the uncertainty remains in the reanalysis because the parameters merely are tuned to have offsetting errors. Other sets of parameters, reflecting different physical relationships, will reproduce the same set of observables.”

      This is NOT how reanalysis code works.

      Next

      “And how do we know what the winds are doing up two kilometres (1.2 miles) up in the troposphere in the middle of the Pacific Ocean?

      Why … a climate model, called a “reanalysis” model. ”

      Err NO. Reanalysis will assimilate data from all sources. FFS

      Example, on a daily basis commercial aircraft supply hundreds of thousands of measurements of wind, temperature, humidity, vertical wind gust, etc.

      Reanalysis takes all these measurements and “interpolates” using physics rather than statistics.

      Pat is wrong. We know Pat is wrong because you can test reanalysis and what he says is untrue,
      both untrue form an actual programming approach and untrue from a verification stand point.

      and yes re analysis has errors.

      With respect to wind at 800mb Here is what you would have to do.

      1. get the reanalysis.
      2. Look at their user guide for the variable in question .. they will tell you the confidence.
      3. Look at their data sources for assimilation

      THEN you might have a specific valid objection.

      But quoting Pat franks un informed opinion? no cookie.

      That said your other criticisms are spot on.

      • It is pretty hard to believe that these scientists didn’t take the time to appreciate what Steven Mosher might think about their methodology!

      • Thanks Stephen, I appreciate those straightforward descriptions of reanalyses (and I use many such products in this paper) but I respectfully disagree with your final sentence. I welcome any specific or general critiques, “spot on” or otherwise. Please now take the time to read the paper in depth if you must. It premieres new takes on old concepts, explains things that no other climate scientist has even bothered to explore, identifies for the first time many solar cycle climate connections of high correlation and high significance, including apparently for the first time significant surface climate connections, it includes some of the world’s most accurate forecasts (both training and testing states), and is responsible regarding its citations to prior art.
        It also raises many questions and it is by no means perfect. In any case, I’m happy to explore its warts and more intriguing features with serious readers for a time that this post is open to comments. I’ll return to this in the evening after I’m done working.

        • and is responsible regarding its citations to prior art.

          It is not. The Ruzmaikin/Feynman et al. studies and the Alexander et al. studies mentioned in comments have not been cited. Actually the number of citations for such a long work (61 pages manuscript) on such an ample subfield (solar variability effect on precipitations), is very low. Classical works like those of Rhodes Fairbridge in the 80s, and modern ones like Czymzik et al., 2016 are not mentioned.

    • Willis did the hard part, but I still want to read the paper and the link he found did not work for me at 16:46 EST 14 Feb 2019.

  10. Wallace findings about trade winds in the Pacific are totally consistent with solar control of ENSO that I have been reporting at WUWT for the past seven months:

    https://wattsupwiththat.com/2018/07/05/solar-minimum-and-enso-prediction/

    ENSO probabilities are skewed by solar activity to the point of making it highly predictable at certain times.

    https://i.imgur.com/mfCuhAt.png

    Now we can discuss how the sun controls the Pacific troposphere responsible for driving ENSO, the principal mode of climate variability in the tropics that affects the entire planet, because it is clear that it does.

  11. https://2.bp.blogspot.com/-RsGVz_DLr2w/T-nHkBIbjjI/AAAAAAAAAOw/E0jKXoKp88E/s400/Adelaide_Sp.JPG

    If you use SSA to investigate the de-trended maximum temperature time series (see figure 2), you find that there are spectral peaks at all sub-multiples of the 22.3 year Hale (H) cycle from H/2 to H/10. The most prominent sub-harmonics are those at H/3, H/6, H/9, and H/10.

    This result strongly suggests that the long-term median summertime maximum temperatures in Adelaide are primarily being driven by factors that are associated with the 22.3 year Solar Hale Cycle.

    The presence of sub-harmonics in the temperature record is indicative of the fact that the ~22-year forcing term must have a broadened temporal structure that is triangular like in appearance.

    This result is broad general agreement with the results of
    Thresher (2002) who finds that the variability in the strength of the zonal west winds (along their northern margins) broadly correlate with the 22-year sunspot cycle [see abstract below].

    Hence, the most likely causation sequence is:

    22 Hale cycle —> strength of zonal west winds —> the median summer maximum temperatures in Adelaide

    with the strength of the zonal west winds depending directly on the strength of the wind vorticity around low and high-pressure cells in the Southern Hemisphere.

    • If you want to prove Willis Eschenbach’s assertion that the 11/22-year solar cycle is not visible in any meteorological time-series records, simple do the following:

      The plot that I have posted is reproducible by anyone with even a modicum of scientific training.
      Go to the FTP site:

      ftp://ftp.bom.gov.au/anon/home/ncc/www/change/HQdailyT

      Download the tmax and info files and decompress.

      Name___________________________Size________Date_____ Modified

      HQ_daily_tdr_txt.tar___________11.9 MB____4/16/13,__10:00:00 AM
      HQ_daily_tmax_txt.tar_________12.0 MB____4/16/13,__10:00:00 AM
      HQ_daily_tmean_txt.tar________11.9 MB____4/16/13,__10:00:00 AM
      HQ_daily_tmin_txt.tar__________11.9 MB____4/16/13,__10:00:00 AM
      HQdailyT_info.pdf______________36.9 kB_____1/8/08,___11:00:00 AM

      Look for the data for this station near the centre of Adelaide, South Australia

      Number__Lat._____Long.___El.(m)____Name
      023090___-34.92__138.62__ 0048.0___ KENT TOWN

      Summers in Adelaide are characterized by frequent hot-spells with maximum temperatures exceeding 35 degrees Celsius. These hot-spells generally last for a day or two before temperatures are moderated by flows of colder air from the south.

      The highly variable nature of maximum summer temperatures in Adelaide means that the best way
      to characterize the average summer [December/January/February (DJF)] maximum temperature is to use the median rather than the mean.

      Create a time series of the median maximum temperature for the period between Dec 1st and the end of February. Do not smooth this time series consisting of annual data.

      Subject the time series to Singular-Spectra_Analysis (SSA).

      Use the technique to remove a slight long-term smooth parabolic variation in the data set i.e. subtract principal components 1 & 5 from the sum of the first 25 principal components.

      Re-analyse the moderately detrended data with the SSA technique.

      • Note- that I am using negative psychology here. If I asked you to disprove Willis assertion – no one would bother. By asking you to prove his assertion, I am hoping that some of you will use this as a cover to have a go!

      • Ian Wilson February 14, 2019 at 8:10 pm

        If you want to prove Willis Eschenbach’s assertion that the 11/22-year solar cycle is not visible in any meteorological time-series records, simple do the following:

        The plot that I have posted is reproducible by anyone with even a modicum of scientific training.

        So out of the thousands of temperature stations around the planet, you pick one of them.

        That’s not good enough, however, so you pick one season out of one of them.

        Still not good enough, so you pick just the max temperature, not the min.

        Then you say that means don’t work for your method, you need to used medians.

        Then you remove a “slight long-term smooth parabolic variation” by subtracting the PC1 and the PC5 from what you got from all of the previous cherry picking.

        And finally, you use Singular-Spectrum-Analysis on the results …

        Seriously? This is your evidence that the sunspot-related solar variations make a significant difference here at the surface?

        w.

        • Willis,

          Thank you for giving a reasoned set of points that I can discuss.

          Willis said: “So out of the thousands of temperature stations around the planet, you pick one of them” & “That’s not good enough, however, so you pick one season out of one of them.”

          No, I picked one site whose maximum summer temperatures are specifically influenced by a known meteorological phenomenon.

          During the summer months [DJF] over Australia, the southern hemisphere sub-tropical high-pressure ridge is located just off of the southern coast. The ridge often manifests itself as a semi-permanent high-pressure cell that is either located in the southern Tasman Sea (off the SE corner of Australia) or just to the south of Victoria, Tasmania and South Australia. The latitude and strength of these semi-permanent highs are a known contributing factor to the development of summertime [DJF} heat waves in cities like Adelaide, that are located along the eastern parts of Australia’s southern coast.

          Wilis said; “Then you say that means don’t work for your method, you need to used medians.”

          The day-to-day summertime maximum temperatures of the city of Adelaide vary by up to 10 C in a matter of a few days. This means that the median is a much better representation of the central tendency than the mean since the latter it is far more sensitive to outliers.

          Willis said: “Then you remove a “slight long-term smooth parabolic variation” by subtracting the PC1 and the PC5 from what you got from all of the previous cherry picking.”

          You can actually see the effect of the removal of the P1 and P5 principal modes upon the final spectrum and it only has a significant effect upon the spectral power for periods longer than about 11-15 years. So, no, this step does not affect the final conclusions.

          Willis said: “And finally, you use Singular-Spectrum-Analysis on the results …”

          The SSA is done on the raw time series without any smoothing. As far as I can tell, the SSA technique is a valid method for analyzing time series.

          All I am claiming is that if the Sun’s magnetic cycle can have an influence upon a known regional meteorological metric (i.e. the strength and vorticity of high-pressure cells in the Southern hemisphere) then it is possible that it could have an influence further afield.

          Cheers,

          Ian Wilson

  12. It has been conclusively proved that the sun and the Earth are linked by electro-magnetic plasma circuits, variously named ‘magnetic ropes’, ‘portals’ etc. with known effects in the equatorial ionosphere and polar regions in form of geomagnetic storms, etc.
    Similar, but even stronger circuits (as function of the planetary magnetic field strength) exist between the sun and the gas giants. Since all of the magnetic planets orbits are nearly coplanar, the Earth traverses such circuits annually for more distant, every 13 months 19 years for the two most noteworthy, with 11 year rise and fall in the intensity and 22 year magnetic polarity reversal cycle, etc.
    If effects on the terrestrial climate exist, some researchers think that there is no doubt about it, while others think they are negligible or there are none, with the above multiplicity of periodicities it is difficult if not impossible to establish convincing correlation with any particular terrestrial climate event such as el Nino, polar vortex effects, polar temperature amplification, global temperature rise or fall etc.

  13. My stream is pretty consistent no matter the solar cycle. However, it does depend on how much water I drink.

  14. “Normalized 5-year trailing average time series observations for non-continental and continental locations: (a) solar and Pacific Ocean atmosphere parameters, and (b) streamflow observations in the Himalaya and Rocky Mountain basins of the Northern Hemisphere.”

    What an amazing graph!?

    Why are the streamflows not overlain on the alleged atmosphere observations?
    Why? Because it is rather clear that they do not align, except at occasional points. Even then, some appear lagged and others appear anticipatory.

    Then, the author’s claims,

    “Both correlation sets were integrated in a linear regression analysis to produce relatively accurate sub-decadal streamflow forecasts for an annual and a 5-year average. In comparison to the autocorrelation technique, the prototyped method yielded the highest correlations”

    “relatively accurate sub-decadal streamflow forecasts”, relatively accurate? Subjective analysis and assumptions?

    “comparison to the autocorrelation technique, the prototyped method yielded the highest correlations”
    This smacks of data tortured until results the author desires are found. e.g. a comparison to one method of analysis technique is allegedly compared to another?
    Are the prototyped “highest correlations” simply numerically superior? To what extent?

    All of which ignores and distracts from the author’s direct claims; i.e. streamflow predictions.

    “Equivalent exercises were applied towards potential connections of some of those parameters to streamflow datasets for candidate streams of the RockyMountains{sic}. A two-stage regression-based forecasting approach was then applied to exploit some of the highest lagged correlations that were identified. The forecasts were compared to forecasts for the same streamflow datasets via a conventional autocorrelation technique.”

    “candidate streams”?
    Some of the streams? Why not the entire drainage area? i.e. every flow of water within the drainage area?
    “Candidate streams” implies subjective streamflow selections.
    Nor does allegedly finding some level of correlation for selected streamflows immediately make the analysis regionally or globally valid.

    “The training forecasts based on the new CRMA applications were generally the most accurate of all featured methods”

    “training forecasts”?
    In other words, input parameters to force the model’s results in one direction or another?
    “generally”? More subjective selection?

    “The forecasts under the new method were also found to be more accurate than the conventional method under an annual average with a 2-year lead forecast approach, although the fidelity of all results was diminished in comparison to the 5-year average set of forecasts”

    “forecasts under the new method” are allegedly more accurate? Provided multiple caveats are allowed.

    “An additional but limited investigation demonstrated high-fidelity 5-year trailing average forecasts for the Animas River based directly on solar cycles taking place 5 years in advance”

    Forecasts for one river are “high-fidelity”?
    Is that claim verified for the entire river and all feeder streams?
    Is that claim verified over many solar cycles and “Trade winds localized within the Western Equatorial Pacific express” cycles?

    “The success of the proposed methodology is expected to apply to other regions meeting the target criteria”

    Of course, the usual claim of potential greater value. Which apparently means the author wants to make a career of this analysis, send lots of money.

    All of the author’s manipulations to achieve “statistically significant correlations” appear to be searching for preferred statistical results.

    • Thanks for your first impressions Atheok. I recommend now that you actually read the full paper and you will find how out of context those impressions are. Then by all means return with any remaining concerns and I’ll try to address.
      Mike W.

      • What?
        Do you get a portion of the paywall fees?

        “Article Purchase 24 hours access for USD 50.00
        Issue Purchase 30 days access for USD 109.00”

        I wrote using the abstract, you posted.
        “recommend now that you actually read the full paper”, then post the whole paper, and I might consider wasting more of my time, provided there are obvious correlations and at least some proof of causation.

        I also find it interesting that you throw out an implied ad hominem strawman instead of responding directly to my comments.

        • Sorry ATheoK, no fee was given, and I didn’t intend to ad hominize. I posted here to share this good news, engage in serious discussion, and to learn from critical feedback. I’ve already found such from other comments which I appreciate. I’m glad again that you took some time to read the summary of the paper here and to write your first impressions.

          Now if you are serious, I ask you again to read the paper. For one example, you do not seem to understand the lagged part of the paper which addressed the first two time series of your interest.

  15. No need for “streamflow forecasting using solar cycles for high altitude catchments.”

    when we already no “no sunspots –> drought years”

    as your graphs show:

    https://4k4oijnpiu3l4c3h-zippykid.netdna-ssl.com/wp-content/uploads/2019/02/wallace-stream-flow.png

    https://www.google.com/search?client=ms-android-samsung&ei=wWplXPuON8eorgSZrJe4BA&q=lesser+coronal+ejections+sunspots+&oq=lesser+coronal+ejections+sunspots+&gs_l=mobile-gws-wiz-serp.

    drought years frequency ~ x times eleven years

  16. already no –> already know

    No need for “streamflow forecasting using solar cycles for high altitude catchments.”

    when we already know “no sunspots –> drought years”

    as your graphs show:

    https://4k4oijnpiu3l4c3h-zippykid.netdna-ssl.com/wp-content/uploads/2019/02/wallace-stream-flow.png

    https://www.google.com/search?client=ms-android-samsung&ei=wWplXPuON8eorgSZrJe4BA&q=lesser+coronal+ejections+sunspots+&oq=lesser+coronal+ejections+sunspots+&gs_l=mobile-gws-wiz-serp.

    drought years frequency ~ x times eleven years

  17. One (W.E.) wrote “So I’m sorry, but I don’t find anything that is at all supportive of the idea that solar sunspot-related variations affect the climate down here at the surface.”
    I hope you will continue to read the paper more closely. You also speculated on winds over the WEP at a higher elevation. There is a wind data set for the same location at 200mb if it interests you. In any case 800 mb is close enough to the surface for the purposes of my paper, which as all can see, found an equally strong correlation for two other parameters, OLR and divergence of latent heat across the same footprint. Accordingly the connection to sunspots is not only at the surface here, it can be found within the column all of the way to the top of the atmosphere TOA (roughly). That surface-to-top correlation of the atmosphere to solar cycles is a scientific first if I’m not mistaken. I think it was a reason for the lengthy review period, given that this paper went through 4 individual review cycles over two years. Readers will note the three appendices which were included purely to add supporting information.

    By the way, there is a key feature about the latent heat signatures in this paper that should be mind blowing relating to the sun, but so far no reader has noted. Also equally unsettling (amazing?) is the strong correlation of the OLR to the solar cycles. In other words clouds have now been directly correlated to solar cycles. Both are apparently scientific firsts.

    In any case, a key purpose of the paper was to more accurately predict streamflows, for a longer projection span, where that can be done. That was documented in the paper, with statistical features for any to explore and test if they wanted. Streamflows happen to lie on the surface of the planet and remain among the primary indicators of climate. Accordingly they are the second set of surface features (adding to the trade winds) I have identified which shift the current solar vs climate paradigm.

    For what it is worth, the paper did not limit to a few streams or cherry pick the ones identified. It sought streams that emerged from mid latitude high altitude catchments for a subtle solar signal. It also confirmed where the signal diminished, (which happens to be where the catchments are at a lower altitude and latitude) and it cited a number of papers which looked at many additional streams in adjacent catchments and saw intriguing relations. The paper also recognized the same potential within the Andes and the Himalayas and included examples.

    The paper also was mindful of smoothing issues in time series and the potential for misleading correlations. In fact the paper considers time spans ranging from decadal to monthly. Finally the only feature that was detrended was figure 1a and that was simply done for a common starting point (and because it was requested by a peer reviewer ) given that this practice has been a conventional approach by past solar climate researchers.

    I only cite a paywalled paper because of its context to past posts, because I wrote it, and because I did not pay the fee to make it free for all. I know that WUWT readers can be resourceful and if necessary can go to a library to find. I’m not sure why there was so much resistance by Charles the Moderator to even disclosing this contextual news with its scientific firsts. I think a month had gone by since I first submitted this news. I do understand that this work strongly repudiates claims by skeptics as well as alarmists. I will continue to learn and welcome further challenges to this paper.

    • I think CtM does great job, if anything he is too tolerant. I often post comments which can not be substantiated by good correlation or prevailing theories, so I think you must have come to a wrong conclusion about CtM.
      On subject of your paper, my knowledge is not at the level I could make a valid judgement, I would just point out that your graphs cover only 5 or 6 sunspot cycle and even more critically only one 60 year AMO or quasi-PDO cycles, which is good start for discussion, but may not be sufficiently long for formulating a hypothesis.
      Any contribution to the science, based on the observational data, proved correct or not, is a step in the right direction, so thanks for your efforts.

  18. I’m not sure why there was so much resistance by Charles the Moderator to even disclosing this contextual news with its scientific firsts.

    I doubt there was any problem with the content of your article. Solar-climate articles have no problem being published in WUWT. David Archibald publishes regularly on it. Then usually a fight ensues in the comments providing much entertainment to WUWT readers.

      • Thanks Krishna. Rough sledding to the data but found it. Tropical Western Pacific better documanted than anywhere on earth thanks to three stations; Darwin, Manus and Nauru.

  19. In the meantime, the Pacific Ocean skin temperatures have cooled dramatically over the past 3 months, especially around all of Australia, western and central northern Pacific, central southern Pacific, equatorial Pacific and especially off Argentina and west.

    The air temperature lag from ocean cooling is months, so let’s see how we go from here. If this continues, global temps are going down and even Stephen Mosher won’t be able to hide it (as much as he will try).

  20. There are lot of papers around finding relationship between sun activity and riverflow, a short look offered me Japan, China, Germany, Italy, South America…

  21. Why are some stream systems correlating, but not others? Do streams have some intrinsic mechanism allowing them to choose to be in step or out of step with a forcing?
    Geoff

  22. Javier February 14, 2019 at 1:51 am

    Willis,

    You never answered to the evidence I posted. As usual when I talk about solar control of ENSO the anti-solar crowd chooses not to speak.

    Pass. No idea.

    I use the ONI (Oceanic Niño Index) from NOAA:
    http://origin.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/ONI_v5.php
    And the monthly smoothed sunspot dataset from SILSO:
    http://www.sidc.be/silso/DATA/SN_ms_tot_V2.0.txt

    Thank you kindly. Having the full data info lets me see what you’ve done.

    Selecting the periods when solar activity (sunspots) increases from 35% of maximum (for each cycle) to 80% of maximum, it turns out that the average ONI is -0.649, which is full fledged La Niña conditions. For six solar cycles.

    Definitely statistically significant.

    Well … no. Here is an overview of what you’ve done. Curiously, your analysis suffers from the same problems as in the head post, Bonferroni and smoothing.

    The vertical blue lines show the 10% of the ONI data that you’ve selected for your analysis. Because you are analyzing subsets, you need to apply the Bonferroni correction. It will be on the order of 10 because you’ve selected a tenth of the data.

    That means to find statistical significance, you need to find a p-value of less than 0.05 / 10 = 0.005 … and your correlation is far from that.

    Curious fact: If you’d picked the period starting ~14 months earlier all the way through, with the effect of shifting the blue vertical lines in the graph above 14 months to the left, the ONI average would be +0.6 instead of -0.6 …

    Second, autocorrelation. Your ONI and sunspot data both have high Hurst coefficients. This leaves them with an effective N of only about 3 data points … huge problem.

    Finally, there is no correlation between your selected smoothed SSN and the selected ONI, viz:

    Call:
    lm(formula = oni ~ ssn)
    
    Residuals:
         Min       1Q   Median       3Q      Max 
    -1.15421 -0.45153  0.09752  0.34907  2.05181 
    
    Coefficients:
                  Estimate Std. Error t value Pr(>|t|)   
    (Intercept) -6.565e-01  2.127e-01  -3.087  0.00285 **
    ssn          6.509e-05  1.673e-03   0.039  0.96907   
    ---
    Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
    
    Residual standard error: 0.6143 on 74 degrees of freedom
    Multiple R-squared:  2.046e-05,	Adjusted R-squared:  -0.01349 
    F-statistic: 0.001514 on 1 and 74 DF,  p-value: 0.9691

    p-value of 0.96 …

    Regards,

    w.

    • Until recently, all the predictions foretold with certainty El Niño. However, they did not take into account the minimum of the solar cycle.

      • Sorry.
        Until recently, all forecasts predicted with great certainty El Niño. However, they did not take into account the minimum of the solar cycle.

      • “ren February 14, 2019 at 3:12 pm

        Until recently, all forecasts predicted with great certainty El Niño. However, they did not take into account the minimum of the solar cycle.”
        _______________________________________________

        Right, ren.

        But there’s 2 different “forcasts”:

        – La Niñas ‘forecast’ the WEATHER for the next ~7 years.

        – sunspot minima predict DROUGHT YEARS every ~ 11 years.
        _______________________________________________

        Regards. Hans

    • you need to apply the Bonferroni correction.

      No, I need not. The Bonferroni correction is to correct in multiple testing for the probability that one of the tests gives a positive. In this case there is a single hypothesis that would not be proven correct for a Niña in one of the intervals. It requires a Niña in all intervals, and therefore the Bonferroni correction is not adequate.

      Curious fact: If you’d picked the period starting ~14 months earlier all the way through, with the effect of shifting the blue vertical lines in the graph above 14 months to the left, the ONI average would be +0.6 instead of -0.6

      Not curious at all. By doing that you move to the solar minimum when Niño conditions predominate. It is further proof that ENSO is under solar control.

      Finally, there is no correlation between your selected smoothed SSN and the selected ONI

      Why there should be correlation between them? The solar effect detected on ENSO is that at certain times during the solar cycle the probability is strongly skewed towards a certain outcome. And the statistical analysis confirms it despite your arm waiving.

  23. steven mosher February 14, 2019 at 2:38 am

    Gosh

    please dont quote Pat frank on Reananlysis, he is dead wrong.

    “Even where reanalysis is of the known climate, for which the model has been parameterized to reproduce certain observables, the uncertainty remains in the reanalysis because the parameters merely are tuned to have offsetting errors. Other sets of parameters, reflecting different physical relationships, will reproduce the same set of observables.”

    This is NOT how reanalysis code works.

    Steve, always glad to hear from you.

    Please correct me if I’m wrong. My understanding is that a reanalysis model is a computer model that is constantly “nudged” back on to the right path by inputting whatever data we might have. The data comes from a host of sources.

    It takes all that various data as input, and from that, the computer makes its best guess as to what is happening where there is no data. However, the output is model output, NOT data.

    And as Pat Frank said, with different parameters we can get different answers using the exact same data as input. That’s why NCEP and NCAR and various versions of reanalysis model output exist.

    Next

    “And how do we know what the winds are doing up two kilometres (1.2 miles) up in the troposphere in the middle of the Pacific Ocean?
     
    Why … a climate model, called a “reanalysis” model. ”

    Err NO. Reanalysis will assimilate data from all sources. FFS

    Example, on a daily basis commercial aircraft supply hundreds of thousands of measurements of wind, temperature, humidity, vertical wind gust, etc.

    Reanalysis takes all these measurements and “interpolates” using physics rather than statistics.

    I understand all that. But the area in question, 135E to 170W and 10°N/S, gets very few flights. It’s not on a main path to anywhere. And the few it has will be way overhead. You don’t fly down low over the ocean.

    Pat is wrong. We know Pat is wrong because you can test reanalysis and what he says is untrue,
    both untrue form an actual programming approach and untrue from a verification stand point.

    What you quoted from Pat is true, as I pointed out. Not sure what else you mean.

    and yes re analysis has errors.

    With respect to wind at 800mb Here is what you would have to do.

    1. get the reanalysis.
    2. Look at their user guide for the variable in question .. they will tell you the confidence.
    3. Look at their data sources for assimilation

    THEN you might have a specific valid objection.

    But quoting Pat franks un informed opinion? no cookie.

    Both to me would have value. And you are right, if I did that I could see their input observations.

    However, that still says little about their output …

    That said your other criticisms are spot on.

    Thanks,

    w.

    • Steve, an interesting post here. Money quote:

      Simulated trajectories are computed starting from the positions of the balloon and advected using the ECMWF and the NN50 velocity fields. The simulated trajectories are compared to the real balloon trajectories. The spherical distance between the real and simulated positions exceeds 1000 ± 700 km on average in just 5 days using NN50 and after 10 days using ECMWF. The distances between the simulated and real balloons are found to increase faster in November and December, owing to the strong Rossby-wave activity in the stratosphere.

      Regards,

      w.

  24. Reanalysis is the best of both worlds. Real data keeps the models anchored on reality. The modelling gives a huge flexibility in the type of questions that can be asked. Unlike temperature datasets it is almost impossible to tamper with reanalysis as the data is almost real time, and as weather forecast is done through it, it is the type of data that lives depend on. Besides one of the main reanalysis tools is by ECMWF, a consortium of over 30 countries not susceptible to national politics.

    Reanalysis is rapidly becoming the tool of choice in science. How could anyone do research based on GISS when they can change the temperature of the past and render your research invalid? GISS is useless, redundant and unreliable, and should be terminated.

    • With extreme prejudice.

      And its perpetrators sent to polar regions to collect actual data instead of making them up in NYC.

    • I couldn’t agree more Javier, but I’ll try. I agree! The reanalysis work I explore already independently corroborates (at least to me) jet streams, polar vortexes, the Equatorial Trough, isotopic atmospheric hot spots, and of course the numerous solar connections to climate featured in my paper.
      Mike W.

  25. According to researchers such as Ahluwalia (2012), the contemporary plunge in sunspot numbers is suggestive of the advent of another grand minimum, such as the Dalton or Gleissberg minima.

    Lol. Researchers such as Ahluwalia 2012 in the Indian Journal of Radio & Space Physics. Was that the best citation you could find? Just by reading WUWT from time to time you might know about Valentina Zharkova who is a Media darling predicting an approaching Solar Grand Minimum in the style of Game of Thrones’ “Winter is coming.” Problem is she is wrong and a SGM is not coming, but anyway, Ahluwalia? Seriously?

    And Dalton or Gleissberg minima are not Solar Grand Minima. I don’t doubt your great expertise in hydrology, but you should have researched the solar part a little bit more, or have somebody that knows that part read your manuscript. Anybody with a little knowledge will know you have no clue. If you want to know what a solar grand minimum is you have to read Usoskin.

    • Thanks Javier, I welcome this opportunity to clarify.

      One reason I chose to cite Dr. Harjit Ahluwalia was because he was my Ph.D. advisor at the University of New Mexico. He remains the only scientist at the university that I know of, other than myself, who ever publicly opined that greenhouse gases may not hold the answers (at least I think he did, since he once sponsored Dr. Fred Singer to speak here). As a condition of my acceptance as his protege, he directed me to stop looking into interstellar dust and the heliosphere (and ice ages) and rather focus on terrestrial modern and contemporary hydroclimate signatures with an eye to any possible solar cycle correlations. That led directly and ultimately to my paper. Many many hours of mine over several years were spent learning about his views on solar cycles, and following up to read numerous solar related papers, ranging from the theories of the solar dynamo, planetary orbital – solar cycle theories, solar winds, and much more. So I have indeed read many scientific papers on solar cycles and their potential causes, and I plan to continue to read in that direction as possible.

      From that experience, I do know that many solar researchers engage in predicting the next significant solar minimum, and he is among them. I think citing him is accordingly well justified.

      I do regret that I didn’t define the terms as accurately as I could have and I thank you for bringing that to my attention. Perhaps my advisor would have caught that, but he declined to review the drafts, as a divide grew between us that I have never understood. Ultimately we parted ways while my paper crawled through a 2 year long, 4 cycle, anonymous peer review. In any case the jargon deficiency doesn’t undermine anything technical about the paper. To show how behind I may be on this minima topic, I have never heard of Valentina Zharkova either. I’ll check her out!

      I do happen to enjoy entertaining a notion regarding the Maunder Minimum at http://www.abeqas.com/the-maunder-minimum-and-the-inquisition/

      Thanks again for your time and comments. Are you a scientist? Either way is fine with me but I often find it is helpful in communicating my responses. Most scientists I know would be somewhat less agitated over a jargon correction, but in a feisty blog I understand that it has its benefits.

  26. Johann you are spot on at least for certain areas, and I also appreciate the reference.
    My paper also recognizes some regions which are more sluggish and muted in their responses than the high altitude catchments and across the limited WEP.

      • Leif has spent a lot of time and absorbed a lot of grief on blogs in his search for a solar/climate connection. For instance, see three mammoth threads on Climate Audit ~10-12 years ago. Perhaps his absence here on this thread is from a suspicion that a mechanism is finally being explicated. He’s certainly cautious enough not to jump in precipitately when he’s not sure.

        Nonetheless, evidence accumulates, and that European study is nicely timely.

        Good luck, Mike, and happy trails to you.
        ==============================

        • Thanks kim! I’m looking forward to learning more about Lief’s topics (what is his posting name?) and this European study. I’m now communicating with those authors as well.

Comments are closed.