Is the 2015/16 El Niño an El Niño Modoki?

AND Is that the Reason Why This El Niño is Not Suppressing the California Drought as Expected?

Guest Post by Bob Tisdale

El Niño events come in different flavors, and those different flavors can have differing impacts on regional weather around the globe. There are East Pacific El Niño events like the 1997/98 El Niño, and there are Central Pacific El Niños (a.k.a. El Niño Modoki) like the one in 2009/10. See the maps shown in Figure 1. Where does the 2015/16 El Niño fall and why does it matter?

Figure 1

Figure 1

If the 2015/16 El Niño has the characteristics of an El Niño Modoki (Central Pacific El Niño), it may provide a possible answer for why some parts of California have not been flooded like they were during the 1997/98 El Niño…have not had the drought conditions lessened as much as had been hoped when the 2015/16 El Niño developed last year. See the ABC News article El Nino Leaves Much of Drought-Stricken California in Dust. (Yes, I understand there are forecasts for significant rainfall and snow in California this weekend. See California Drought: Big weekend storms coming with rain, snow, strong winds in the San Jose Mercury News.) Also see Is El Niño AWOL? at the NOAA ENSO Blog for other discussions.


One of the methods used to define a Central Pacific El Niño is the El Niño Modoki Index created in Ashok et al. (2007) El Niño Modoki and its possible teleconnection. The El Niño Modoki Index was prepared in that paper using HADISST sea surface temperature data (from the UK Met Office) for three regions in the tropical Pacific. See Figure 2. Ashok et al. write (their Figure numbers):

Based on the EOF2 pattern presented in Figure 2b and the PC2 time series shown in Figure 3, we derive an El Niño Modoki index (EMI). Because of the unique tripolar nature of the SSTA, the index is defined as follows:

EMI= [SSTA]A-0.5*[SSTA]B-0.5*[SSTA]C (1)

The square bracket in Equation (1) represents the area-averaged SSTA over each of the regions A (165E-140W, 10S-10N), B (110W-70W, 15S-5N), and C (125E-145E, 10S-20N), respectively.

Figure 2 Corrected

Figure 2 – Corrected

UPDATE:  Blogger tadchem on the cross post at WUWT noted that my original Figure 2 (Cell b of Figure 2 from Ashok et al. (2007)) showed the wrong coordinates on the map for Region B.  I missed that error in the original Figure 2. I’ve replaced that map with one from the JAMSTEC El Niño Modoki webpage. Thanks tadchem.  [End update.]

The Japan Agency for Marine-Earth Science and Technology (JAMSTEC) has an El Niño Modoki webpage, with links to El Niño Modoki Index data in monthly and weekly formats. Notes: The sea surface temperature data and El Niño Modoki Index data on those webpages have not been standardized, while Ashok et al. (2007) standardized the El Niño Modoki Index when they determined the threshold for an El Niño Modoki. The monthly El Niño Modoki Index data at JAMSTEC appears to be based on HADISST data and the weekly data should be based on the standard weekly Reynolds OI.v2 sea surface temperature data.

Back to precipitation…

Figure 3 is Figure 11 from Ashok et al. (2007). It includes correlation maps of December-January-February (DJF) precipitation with two ENSO indices. Cell b includes the correlation of precipitation with one of the commonly used ENSO indices, NINO3 region sea surface temperature anomalies. Of course, NINO3 sea surface temperature anomalies were quite high this El Niño. See the graph here, comparing the NINO3 sea surface temperature anomalies in 2015/16 to those in 1997/98. So, based on the positive precipitation correlations in California with NINO3 data, we might have expected a higher-than-normal chance of precipitation in California this boreal winter.

Figure 3

Figure 3

However, Cell a shows that there is no correlation of DJF California precipitation with the El Niño Modoki Index.


Unfortunately, being based on HADISST data, the monthly El Niño Modoki Index lags a few months. So I’ve included the weekly El Niño Modoki Index data for the 2-year periods of 1997/98, 2009/10 and 2015/2016 (year-to-date) in Figure 4. I’ve included a dashed line at 0.7 (Standardized EMI) as the threshold of an El Niño Modoki. It must be kept in mind, however, that Ashok et al. used that 0.7 value for the 4-month period of June-July-August-September (JJAS) and the 3-month period of December-January-February (DJF).

Figure 4

Figure 4

The 1997/98 El Niño was a strong East Pacific El Niño, so it registered a negative El Niño Modoki Index value. The 2009/10 El Niño rises well above the 0.7 threshold for an El Niño Modoki, and referring back to Figure 1, we can see that the 2009/10 El Niño was a Central Pacific El Nino.

The 2015/16 El Niño has cycled back and forth across the threshold of an El Niño Modoki for about the last 18 weeks.

Using a 12-week running-mean filter, Figure 5, we can see that the most recent 12-week average is above the 0.7 threshold of an El Niño Modoki. In other words, for the months of December-January-February (DJF), the 2015/16 El Niño should qualify as an El Niño Modoki.

Figure 5

Figure 5


There had been hope, and there continues to be hope, that the 2015/16 El Niño will help to lessen the drought conditions throughout California. If the precipitation in California does not meet expectations, I suspect some researchers will point to the El Niño Modoki Index as the basis for those shortcomings.

39 thoughts on “Is the 2015/16 El Niño an El Niño Modoki?

  1. “If the precipitation in California does not meet expectations, I suspect some researchers will point to the El Niño Modoki Index as the basis for those shortcomings.” I respectfully disagree, they will point to anthropogenic CO2

    • Bob’s statement says ‘some researchers’. I think his analysis will stand as no researchers currently point to CO2 levels as an El Nino index driver. That privilege is reserved for politicians.

      • So why all the excitement over “the warmest February on record”? (Nearly typed warmist but the spell checker caught me!)

    • Current rainfall averages in “Northern” California are running 105% to 122% above average. It is easy to forget that there are at least 2 distinct regions in California. Next winter will end the drought.

      • Next winter we will most likely be La Nina. The sliver of hope would be a “Wet” La Nina where the typical above normal PAC NW precip intrudes a bit into CA. That may give places like The Trinities, Southern Cascades and Far Northern Sierra above normal. Unfortunately that is still not enough to dig us out of the whole state wide.

  2. A minor quibble(?): The area B definition B (110W-70W, 15S-5N) does not correspond with the rectangle on the graph B (110W-70W, 10S-5N) – their southern boundaries differ by 5°.

  3. Everything in nature (the universe) happens only once; then something else happens. None of the things that happen are exactly like our models. Maybe close, but bet never exactly like our models which are pure fiction.

    So it is not surprising that something that is not a match to one of our models, would be exactly like something else that only happened once.

    So Bob showed us three pictures of El Nino maps.

    I wonder if there are other El Nino event maps at other times that match any one of those three Bob posted here.


  4. Bob, your ENSO knowledge never ceases to amaze. I just learned something useful-not all El Ninos are similar in location and teleconnection.

    • Can we go one further and say that every El Nino is unique? Well, OK, I guess that’s trite.

      Taking a step back, what can we say for sure are the effects of all El Ninos?

  5. “… and those different flavors can have differing impacts on regional weather around the globe.”

    Maybe, if it propagates down and says hello with both hands …

    Met Office hedging its bets anyway.

  6. Analysis of recent California precipitation shows that it is dry but not extraordinarily so. But the USA destroyed 500+ damns in the 200 hundred years plus since the Revolution. And it tore down 500+ dams in the past 10 years. So water stored behind dams, for the “dry times” is simply not there to be used.

    The instigators for this state of affairs are the phony hopelessly anti-Science Enviros. Thank your local leftist Greenies for that wonderful state of affairs.

  7. Bob Tisdale,

    I liked the essay. Your graphs and maps suggest that the closer the warmest region of the El Nino is to the West Coast the more rain falls on the West Coast. Whether the 2015-2016 falls into a Modoki class or not, the continuous spatial distribution of the warmest surface area looks to be important.

    • Matthew, the proximity of warm ocean waters to the coast is most important for the atmosphere to pick up moisture and carry it inland without too much interfering with the process.

      • The Pineapple Express, by way of contrast, brings rain from quite long distances. My comment was really that the information provided by Bob Tisdale was on the whole more interesting than the question of whether this el Nino is or isn’t a Modoki. People are disposed toward classifications, maybe because the full continuity of all the variables is harder to think about.

    • Both Washington St and Oregon St had well above average rainfall this fall/winter. There was a good bit of flooding in areas. Northern California is also well above average in rainfall. It is only Southern California, which is in closest proximity to the ENSO regions, that is below average in rainfall so far.

  8. The Japanese researchers are years ahead of others, identifying this variation of El-Nino and importantly for us in Australia, the Indian Ocean Dipole. Once the IOD was bought into the equation forecasts for south eastern Australia became more reliable but it took our own weather services many years before they even accepted it even existed and longer before it began to be reflected in their forecasts.
    The work of the Japanese researchers is well worth paying attention to.

  9. Bob – Is there such a thing as a La Nina Modoki? I seem to recall a La Nina maybe 8-10 years ago that failed to deliver rain to E Australia.

    • Mike, the IOD needs to be taken into account when considering rainfall in Australia. The Pacific Ocean doesn’t work in isolation, and anyway weather generally moves west to east.

    • Mike, Ashok et al. do mention La Nina Modoki, though I don’t know the characteristics of them.

      Also keep in mind that precipitation patterns for ENSO events do vary. That’s why researchers will often note that precipitation “tends” to follow a pattern.


  10. Rough quote from retired hydrologist re Southern California rainfall: “We have had about as many dry El Ninos as wet La Ninas.” A Fisher’s Exact Test for southern oscillation index vs. rainfall above median for the water years comes out at about flip-of-the-coin. The rough quote is not rejected.

  11. The only conclusion I can draw is that weather is rather more complex than any AGW model, and AGW models do not have any relation to the real world weather.

  12. Bob,
    This probably explains why in the Eastern Australia we have been receiving continual rainfall where as many were saying we were in for a drought period. Parts of Oz are in drought but that is not unusual.

    • Frank,
      The far eastern coastline can often receive high rainfall in El Niño years as the blocking ridges that form in the Tasman sea directs moisture onto the coast in an easterly flow and the formation of east coast lows enhance the effect.
      If you lived in Sydney you could be forgiven for thinking this past summer was cool and wet. Not so for the bulk of the inland east.
      We have a prolonged rainfall defecit of over a decade in large areas of inland eastern Australia and that is not typical. Our area in the Victorian eastern Wimmera cropping belt has a sixteen year deficiency.
      California doesn’t interest me much but they are not alone in extended drought. It is probable that Modoki events leave Australia and California dry at the same time and that is Bob’s point I think.

      • uh huh been here 8+yrs and only for two did we manage to see some water in our lake..its all dry n crusted over again now;-(
        and they reckon SA is the driest state in the driest continent..hmm maybe not so much the last few yrs anyway.

  13. The drought in California is way overhyped.

    Blame the Palmer Drought Index that adjusts physical precipitation by applying the bogus warm temperatures.

    You see, a normal 7 inches of rain now is reimagined as drought since the higher reimagined temperatures mean plants would need more water…

    In the cold 70s drought, with about 2/3 the present population and more dam storage, we had mandatory no-flush on pee-only rules and no water served in restaurants and more. This time one hardly notices.

    California gets real drought when it is colder. We are actually cool now, so only a minor real drought.

  14. Bob
    I often read your articles. I can’t offer much on the subject of your article but I can offer some of the analysis work I have done on the Nino regions. Signal analysis is something I am familiar with in my career on rotating equipment.

    I have made use of Dr. Evans Optimal Fourier Transform (OFT) and my own least squares fit program for fitting curves to data. Generally, I use the output of the OFT as inputs to Marquardt analysis of the data that tries to fit the data.

    What is presented if for region 3.4 but the results are similar to the other regions that I have also looked at. Just the other day there was an update to the daily data. My analysis uses monthly data up to d014 and then uses daily measurements thereafter. You will see that in the figures.

    For region 3.4 I used 90 sinusoids. That means I am using 270 initial guesses from the OFT plus one addition value for an offset. I calculate two different correlation factors. The first is an overall correlation coefficient and the second is a correlation factor for more recent times. In this instance the overall correlation coefficient is 0.934. I consider this very good and you will see this in the figures presented below.

    Since I used sinusoids I feel it is possible to make projections out in time for a brief period. You will see in the figures it looks like we are headed toward a camel hump.

    I am sorry that I don’t know how to add figures to a comment so I provided a link to the figures.!12743&authkey=!AJtj3rtSud4H_Kg&ithint=folder%2c

    Perhaps, I am out there on a limb but as you will see in the following link things are heading in my direction.!12744&authkey=!AHnK6jqAGrC1X24&ithint=folder%2c

    I hope you can make use of this. I know there have been predictions of a forthcoming La Nina. I beg to differ.

  15. Just a passing observation (so likely wrong), but it seems that when the NINO 3.4 region cools, California receives moisture shortly after. I haven’t noticed much cooling of the 3.4 region for the past month.

    It would be interesting to see the derivative of the 3.4 region temperature anomaly correlated to central and southern California precipitation to see how much truth there is to this passing observation.

  16. California precipitation is by no means a direct indicator of the El Nino/La Nina cycle, which itself is quite irregular (broad-band). Thus it should come as little surprise that the current event is not convincingly explained by a simple bivalent conceptualization of its global effects. The classical viewpoint is the state of the waters off Peru. From that vantage point, it’s hard to imagine currently a “Modoki” event.

  17. Did the “bloob” sitting where El Nino normally produces a low pressure system have something to do with the lower amount of precipitation in California produced by the El Nino?

    • No. But the Polar Jet has repeatedly flipped from zonal to meridional. In other words, we’ve had what were essentially baby La Nina jets interspersed with the typically zonal El Nino Jet. Modoki may have something to do with that. The other thing I wonder about is modulation by PDO and possibly other even lower frequency oscillations that we have yet to identify. Bear in mind awareness of PDO is relatively recent. Lower frequencies might not have revealed themselves due to the overall short period of relatively reliable observations of SSTs at the ocean basin and global levels.

  18. 2015/16 seems to have an identity crisis in deciding whether it’s a Modoki or Canonical El Nino. Perhaps we should label it a Swinger!
    More seriously, it strays further from the path of the canonical (East Pacific) El Nino than it does from that of the Modoki and this is most evident from week 36 onward. Perhaps this is the reason why Roy Spencer’s UAH dataset has failed to find the kind of exaggerated response in the tropics which was so characteristic of 1997/98, finding instead a huge temperature response in the northern hemisphere extra tropics. I think the relative lack of response in the tropics in 2016 is symptomatic of the current El Nino’s Modoki meanderings; I’m guessing that the huge NH extra-tropical response is due to a conflation of other non El Nino factors, perhaps related to goings-on in the Atlantic circulation, but that really is only just a guess.

  19. What “Modoki” might mean is an el Nino that doesn’t really engage the Bjerknes feedback.
    The 1997-99 el Nino did see a huge Bjerknes feedback excursion with Peruvian upwelling cut off completely at the height of el Nino, and the trades consequently fully reversed. Thus the angry red SST anomaly plume being maximal right up against the Peruvian coast.

    This is a true el Nino, since el Nino does not just mean a warm equatorial Pacific anomaly, but it actually means an engagement of the Bjerkes feedback. The flip side of this is the following La Nina where the Peruvian upwelling comes back with a vengeance. As Bob has shown previously, in terms of effect on global climate it is the following La Nina that causes the warming step-up since the considerable warm water accumulated by the el Nino is pumped poleward by the resurgent Peruvian upwelling of the La Nina counter stroke (c.f. Willis E’s analogy of ENSO as a 2 stroke heat pump).

    By contrast, in the Modoki el Ninos, the event abd warm anomaly plume does not reach the Peruvian coast but is focused in the central Pacific. This is because such events fail to engage the Bjerknes feedback (which links Peruvian upwelling to the trades). The current 2015-2016 event is a good example. The Peruvian upwelling was never interrupted as evidenced by continued strength of the Peruvian anchovy fishery – a bellwether of Peruvian upwelling. Especially juvenile numbers remained strong indicating good forst feeding food item abundance and thus nutrient rich upwelling.

    Since the Bjerknes feedback gas not been engaged, not only was the upwelling never interrupted, but also the strong upwelling counter-stroke of La Nina will also be absent. So a long term step up in global temperatures will also not occur.

    One can go even further and suggest that, far from pumping and increasing poleward ocean heat transport, the 2015-2016 non-event actually takes place in the context of decreasing poleward heat transport globally. The North Pacific gyre has slowed slightly, as has the North Atlantic drift. In general such a slowdown, being part and parcel of an AMO downswing, is likely to cause some cooling globally.

    Not that we will necessarily know this – the climate datasets which are all in the hands of AGW activists will show endless warming regardless of what is really happening.

Comments are closed.