Surf Forecasts and Marine Weather - No Hype - Just the Facts!
PDO Hits All Time Low - Inactive MJO To Pulse! - Video Forecast HERE (11/17/24)
Buoys | Buoy Forecast | Bulletins | Models: Wave - Weather - Surf - Altimetry - Snow | Pacific Forecast | QuikCAST | El Nino | Tutorials | Great Circles | Video

Google

Stormsurf Mobile App

Create Your Own Surf Forecast
Swell Calculator
Swell Decay Tables
Sea Height Tables
Swell Category Table
Convert from GMT:
 
 to timezone:

---
El Nino Southern Oscillation (ENSO) Forecast

Issued: 8/19/2009
Updated as Warranted

See ENSO page for Links and Current Data
ENSO Archives

Weak-Moderate El Nino Developing
Solid Warm Pattern in-Play Over the Equatorial East Pacific

Overview: Early 2008 saw a marginal El Nino try to take control of the tropical Pacific from March into July, fueled by several active phases of Madden Julian Oscillation (MJO).  The MJO faltered and La Nina returned, cutting the legs off the developing warm pattern. But the MJO was strong enough for long enough to cause anomalously warm warm to appear off Central America and change the atmosphere creating momentum that fueled the development of a decent Fall storm pattern, holding on into late November 2008.  But right when things were starting to get in a groove, the upper level (jetstream) winds that were driving the machine and fueled by the warm waters off Central America during the summer months, disintegrated.  A .cgiit jetstream flow took over totally shutting down the storm pattern.  The remainder of the Winter and Spring 2008/2009 season were completely.cgiacid. 

With that in mind, we moved into the Spring of 2009 seeing a similar pattern. That is, the MJO came-on strong again with a solid series of consecutive Active Phases and a marked increase in westerly wind anomalies in the West Pacific resulting in anomalously warm subsurface waters tracking east towards Central America (Kelvin Waves). As the Kelvin Waves erupted along the Central American coast, warm surface waters began to appear, much like the year previous, but with more vigor. It is with that background scenario that we look to the coming Fall and Winter of 2009/2010 with higher expectations and the hope for real surf for the first time in over a decade for Hawaii, Canada and the US West Coast.  The paragraphs below describe the state of various indicators used to assess long-term global weather conditions (be it either El Nino or La Nina), followed by a revised set of criteria for assessing it's impact on surf generation potential for the future.

SOI: First we look at the Southern Oscillation Index (SOI). This number compares surface pressure over Darwin Australia with pressure over Tahiti. If this value is negative for an extended period of time, that indicates average surface pressure is lower over Tahiti and higher over Darwin, symptomatic of El Nino (or over short durations, ~ 20 days, the Active Phase of the MJO). Positive values over longer times indicates La Nina (or the Inactive Phase of the MJO for shorter durations). The greater the negative or positive value, the stronger the ENSO (El Nino Southern Oscillation) configuration (be it El Nino or La Nina). During El Nino episodes wind flows from generally high pressure over Darwin towards generally lower pressure over Tahiti, which is a reverse of what a normal state is. When it's positive, the reverse it true, with higher pressure over Tahiti and lower pressure over Darwin, typical of La Nina, with wind flowing east to west, typical of the trade wind pattern for this region, only more so. The first 3.5 months of 2009 recorded nearly consistent strong daily SOI values in the positive range, with hardly any negative readings, typical of a moderate La Nina. In fact, the Winter of 2008/2009 and early Spring of 2009 were  dominated by a La Nina pattern. 

But by mid-April the pattern abruptly changed with negative SOI readings appearing and holding consistently through late June, consistent with the active phase of the MJO. And closer inspection revealed that a series of 3 consecutive and separate Active Phases of the MJO occurred, with no Inactive Phase in-between.  Most unusual. A weak Inactive Phase developed in late June holding into mid-July, followed by another Active Phase (negative readings in the SOI) from mid-July into early August. The net result is that all the effects of the previous years La Nina were erased with the 90 day average SOI not only returning to a neutral value of '0' but dipping consistently into the negative range (-2.0 or so). Other indicators suggest that El Nino is clearly in-effect, but the failure of the SOI to dip firmly into negative territory (-10 or greater) with any significant magnitude remains pe.cgiexing. NOAA suggests that during moderate to strong El Nino and La Nina events the MJO signal weakens with pulses one way or the other lessening in intensity. That would e.cgiain the general trend of late of weaker MJO activity with less effect on the SOI, but the overall value of the SOI still being near neutral is puzzling. As of today, a very weak inactive Phase of the MJO was trying to get a foothold and modeled to last for 2-3 weeks.  

Anomalous Sea Surface Temperature: Looking at current seasonally adjusted equatorial Pacific Sea Surface Temperatures (SST), the pattern clearly delineates El Nino. A well defined pool of warmer than normal water extends over the width of the equatorial Pacific starting above New Caledonia continuing east over the dateline about 0.5 deg C above normal then building while pushing to Ecuador with maximum temperatures up to 3 deg C above normal there. This is the classic signal of El Nino and is the exact opposite of what was occurring even in the early Spring of this year when cooler than normal water was over this exact same stretch of real-estate. Today's warmer waters migrated from the West Pacific during the three consecutive active phases of the MJO in the Spring and early Summer, with a steady flow of milder warm water moving east after those events. Also warmer than usual water was migrating north up the North American coasts of Mexico, Baja and into Southern CA, the result of suppressed trades in the area allowing the tropical warm waters of El Nino to head poleward.  Also a pool of much warmer than normal water was pushing northeast off Japan and quite pronounced in the Gulf of Alaska. 

Of interest is a clearly defined flow of cooler than normal water pulsing off equatorial Africa in the equatorial Atlantic, the exact opposite of what is occurring in the Pacific. Looking back in the historical record a similar pattern occurred during the record breaking El Nino of 1997/98.  And during lesser El Nino years a similar but weaker pattern is evidenced, but not during La Nina years.  This could possibly be the result of the Walker Circulation forming over the Pacific warm pool near Ecuador, drawing in surface air from Africa across the Atlantic causing upwelling. This flow of cooler than normal water ought to hinder tropical development in the Cape Verde Storm Corridor. A reverse teleconnection between the Pacific and the Atlantic is known to exist. That is, when the Pacific Ocean is active the Atlantic is inactive from a storm perspective, and visa versa.  It will be interesting to see how this.cgiays out late this summer, now that we're approaching the core of the Atlantic tropical season. Expect the Atlantic storm season will be much less active than normal. The South Pacific remains near neutral if not slightly on the cool side. Overall the signs of El Nino are quite evident, with the classic warm pool in.cgiace off Central America. 

Wind Anomalies: Wind anomaly analysis indicates that on average trades over the equatorial Pacific have been blowing weaker than normal, especially in the far Western Pacific to the dateline. In fact, there have been several instances of a total reversal of the trades there.  Rather than blowing east to west they have been blowing west to east, and at other times they have been blowing less strong than normal, which has the same effect as a reversal of trades in that it supports the generation of a Kelvin Wave (more about that below). With trades being suppressed, this reduces the amount of upwelling along the coast of South America, allowing the collection of warmer water there. Stronger than average winds there increases upwelling producing cooler than average waters. The 4 solid Active Phases of the MJO that have occurred so far this Spring and Summer caused trades to reverse themselves, each pushing a substantial amount of warmer then usual water to the east and depositing it off the Central America coast. These instances of reversed trades/westerly winds are called Westerly Wind Bursts (WWB).  These have been i.cgiicated in not only transporting warm water east, but also help to fuel tropical development the the Central and West Pacific. Of note: The MJO still has Inactive Phases even during El Nino years, just not as prolonged or strong. 

A Westerly Wind Burst (WWB) is an extended duration of wind that blows from west to east along the equator in the West and or Central Pacific contrary to normal trade winds, forcing warm surface waters to start moving in the same direction as the wind (details below), a hallmark of a strong Active Phase of the MJO and a precursor to El Nino. The MJO events of the Spring moderated some in intensity moving into the summer (as did subsequent Inactive Phases), but did not fade as they did the year before. In fact, the late July Active Phase produced a solid WWB (7/25-8/2) resulting in yet another decent pocket of warm water  feeding into the eastward moving conveyor, bound for Central America. Historically if either El Nino or a strong burst of the Madden-Julian Oscillation is in-effect, trade winds that normally blow from east to west reverse themselves and blow west to east. That is, when the MJO is in an Active Phase, the trades reverse themselves in the West Pacific, and when the MJO is not active, trades return. During La Nina, trades blow much stronger than normal.

Pacific Isotherm: Another key indicator in the evolution of either an El Nino or La Nina event is the depth and profile of the 20 degree isotherm (thermocline). During La Nina events, warm subsurface water remains pooled up in the far equatorial West Pacific while cold surface and subsurface waters dominate the East Pacific, resulting in a steep angle from east to west, or from a shallow pool of warm waters in the east to a deeper pool of warm water in the west. In El Nino events, warm subsurface water (i.e. Kelvin waves - more below) migrate from the West to East Pacific and the angle flattens and the depth of warm waters becomes more uniform across the entire equatorial Pacific. Latest data indicates a solid pocket of warmer than normal water (+2 deg C) tracking from just east of the dateline down about 125 meters, bound for Central America. And it is intensifying as it progresses east. This is the fourth such Kelvin wave to occur this year (at least) and was produced by a WWB that developed just west of the dateline in late July. This was a positive development, coming right on the heels of the mid-June Inactive Phase and what was thought to be a lessening in the intensity of El Nino, possibly signaling a fade like what occurred the year before. But that now looks not to be the case.  

Though the MJO signal has remained weak, it appears that the Active Phases are more productive than the Inactive ones, tipping the scales in favor of El Nino.  Warm water that successfully makes the multi-thousand mile subsurface journey east to Ecuador erupts to the surface just off the coast there, and weaker than usual trades then blow the warm surface water off to the east, also radiating north up the Mexican coast reaching into California during significant El Nino events. This results in a pool of warm water forming off Central America rather than it's usual.cgiace in the West Pacific, flattening the angle of the 20 degree C isotherm across the equatorial Pacific. At this point an El Nino pattern is clearly in control with no sign of cooler than usual waters off Central America and flatter than usual isotherm angle. 

Kelvin Waves: A Kelvin Wave is a pocket of warm water that travels under the oceans surface from west to east at a depth of about 150-200 meters. It is generated by a burst of strong westerly winds blowing over the equator (a.k.a. Westerly Wind Burst (WWB) in the West Pacific and is typically associated with the Active Phase of the MJO. As the warm surface water gains eastward momentum by forcing of the WWB, it sinks near the dateline and travels well under the oceans surface, only to reappear at the surface when it impacts the South America Coast. This results in the sudden appearance of warm waters along the coast of Peru and Ecuador. Occasional eruptions are normal. Large and consistent eruptions are the hallmark of solid El Nino events. The source of Kelvin Waves, a negative SOI and reversed trades, is directly related to the strength and frequency of the Active Phase of the Madden Julian Oscillation (MJO). This weather pattern is responsible for the periodic strengthening of the anomalous westerly winds in the West Pacific which drive production of subsurface Kelvin waves, and also drive the SOI negative. 

The MJO runs in rough 40-60 day cycle, roughly 20-30 days of the Active Phase followed by 20-30 days of the Inactive Phase.  The cycles alternate continuously year after year and are stronger during the northern hemisphere Fall, Winter and Spring season and lessen during the summer. When the strength of Active Phases of MJO dominate, El Nino indicators strengthen. When the Inactive Phase dominate, La Nina moves to the forefront. That is not to say that the MJO causes either El Nino or La Nina, only that it appear to support the development of either extreme of the ENSO cycle. In other words, when the balance of energy favors the Active Phase, the odds of El Nino developing increases.  When the balance shifts towards the Inactive Phase, La Nina becomes more likely.  Currently there is clear evidence that the most recent Active Phase of the MJO (in late July) produced a Kelvin Wave that is currently working it's way across the equatorial Pacific with no cooler water pockets in it's way to block it's progress.

Pacific Equatorial Counter Current: There are three currents that run along the Pacific equator. Two run just abreast (a few degrees north or south) of the equator flowing east to west driven by the semi-permanent high pressure centers located in the center of the North and South Pacific.  These high pressure system are what drives equatorial trade winds too. But sandwiched between them is the Pacific Equatorial Counter Current, which flows against the two, running west to east.  Satellite based sensors are used to measure the strength/speed of the counter current. That data can be analyzed to determine if the current is flowing stronger or weaker than in years previous.  In essence,  anomalies in the current can be detected.  A curious fact becomes apparent when looking at long-term trends:  During El Nino years the counter current runs harder than usual to the east, and during La Nina years it runs harder to the west. This is what one would expect, especially since the exact same pattern appears when analyzing tradewinds. 

Looking at data for this year it is apparent the current is running stronger than normal towards the east, a signal of El Nino. Looking at this years data compared to previous years (the record starts in '93) it is in the range of the '94, '02 and '06 events in terms of the size of the anomaly. But this years event bested all those but 2002 when you looks at the time of the year it started (Jan) where the others started at least 3-4 months into the year. In these previous events (including the record year of '97) it was not really till the Fall months that the current bloomed to it's full potential.  This suggests that more strengthening is yet to occur. And interestingly most had some sort of a pause or significant decline in the flow in the July/August timeframe, before the final push into Fall (except 2006). 2009 appears to be no exception, with that pause occurring now. In many cases, the current changed before the SOI dropped for the first time that year into it's first run of  sustained negative numbers (i.e. the first Active Phase of the year): 2002 current change was Feb 1/Negative SOI starting mid-March, 2006 current change mid-Feb/SOI early May, 09 current mid-Jan/SOI negative late April/though dropping since Feb. 1994 was excluded from the review because it appeared that a multi-year ENSO event was occurring, carried over from '92/93.   This is all new data and the record is very short, so no solid conclusions can be drawn, but it provides some foresight for future research. Regardless, this El Nino sits in the middle of the pack with other like-sized weak-to-moderate events.            

OLR: When El Nino events unfold at the oceans surface, increased cloudiness/precipitation will develop in the atmosphere above the warmer surface waters, since warmer water supports higher condensation rates above it. The presence of consistent cloudiness or precipitation where it historically shouldn't be is a hallmark of El Nino. Satellite based Outgoing Longwave Radiation (OLR) measurements track reflectivity from clouds over time.  The greater the reflectivity, the less sunlight is being absorbed by clouds, and so the lower the amount of reflection off the oceans surface. Low reflectivity values represent greater cloud absorption. Current satellite data indicates a slight increases in cloud cover over the eastern equatorial Pacific associated with the developing warm pool there. An area of persistent cloudiness over the Western Pacific a few weeks before was likely associated with the Active Phase of the MJO, and has since dissipated consistent with the recent turn towards the Inactive Phase. Drier than normal weather is suggested over the West Pacific, specifically Australia northward to the equator. But again suspect that is more associated with the MJO's current Inactive Phase. This is marginally consistent with weak a El Nino.

Analysis: Reviewing all the data over the past several months, the evidence clearly indicates some form of El Nino event has developed and appears to be continuing to evolve. It started in the Spring with consecutive Active MJO pulses generating WWB's that in turn created Kelvin Waves transporting warmer than normal water to the east.  There was a short break in June followed by another WWB in July and another Kelvin Wave that at this time is still working it's way to the east. Warm water 1.0-1.5 degrees C above normal is in.cgiace off Central America with no sign of dissipating anytime soon. We now are entering another Inactive Phase, but would hope to see another WWB in the next scheduled Active Phase in early to mid September. Looking at satellite based historical ocean surface temperature anomaly measurements, waters have not been this warm at this time of the year since the record setting El Nino of 1997/98. And even further back only the the strong El Ninos of 1982 and 1987 exceeded the magnitude of the current event in the months of August. In other words, this is the strongest warm pool that's occurred in the last 12 years. Make no mistake though, it is not on-par with the super events of 1982, '87 or '97. 

But what is does have in common with them is that it is a 'classic' El Nino as compared to a Modoki El Nino. A Modoki El Nino (Modoki is Japanese for 'the same but different') forms more in the center of the equatorial Pacific (south of Hawaii), in the Nino 4 region, and evolves there, rarely making much eastward headway through it's life and not typically reaching the Ecuador coast. These El Nino events tend to be more weak-to-moderate in strength too, with lesser impact on the northern hemisphere Fall, Winter and Spring storm pattern. They still have an enhancing impact, just not as strong. The Classic flavor of El Nino starts forming it's warm pool directly adjacent to Ecuador and expands westward as it matures. The environmental impact tend to be more severe.  This current event is of the classic variety, only one of 5 that have occurred since 1980 (82/83, 87/88, 91/92 [debatable], 97/98 and now 09).  

This current event is the weakest of the bunch, but then all but the 91 event were exceedingly strong.  In all instances the classic form of El Nino evolves and builds from Aug into Dec.  This suggests more strengthening is historically not only possible, but likely. Conversely, the Modoki flavor of El Nino tends to start either weak or weak-to-medium in strength (relative scaling based on July water temp anomalies) and have 50/50 odds of strengthening into December. Recent Modoki ENSO events occurred in 86, 90, 93, 94, 02, 04 and 06. That strengthening was generally moderate, with none reaching the strong category by Dec. But in all cases, they did not fade, just never developed particularly stronger than what they were in July.  Given that historical background, this event appears to best any but the strongest ENSO events over the past 27 years ('82, '86, perhaps '91 and '97).  That is the good news.  

The more disconcerting news is the lack of a clear response by the SOI.   One would expect that if a solid El Nino was developing the SOI would be negative, and not just a little but more like in the -10 to -15 range rather than the current 90 day average of ~ -1.5. Relatively this is much better than anything that has occurred in the past 3 years, with the start-off spike for this event well into negative values in late April and May most impressive.  But it could not be sustained and even now struggles to stay negative.  OLR data is inconclusive too. All this leads to what could be considered a muddled picture. But assuming the historical trend presented above is accurate for classic El Ninos, one would expect to see strengthening over the coming Fall months and a deepening (into negative territory) of the SOI over time.  That remains to be seen.

Regardless of the statistical indicators, the fact remains that there is a moderate and building pool of warm water off Central America that has been in.cgiace since June and is showing no sign of dissipating. And since it is the presence of anomalously warm water off Central America that drives changes in the atmosphere, and the classic form of El Nino has the most impact, then the fact that this current event exists at all and is as warm as it currently is trumps all other indicators and analysis. In short, the argument is "It exists so therefore it cannot be denied". The only remaining question is: "How long will it last and how strong will it become?" 

Always of interest is the relative activity level of the Atlantic hurricane season. Hurricane activity in the Atlantic is inversely tied to the strength and duration of El Nino and La Nina in the Pacific. A classic El Nino produces strong shearing winds over the Atlantic that tear the tops off developing tropical storms rendering them weak and ineffective at evacuating warm moist surface air up high into the atmosphere through the storms eye. In effect, a hurricane is the atmosphere's attempt to create equilibrium, or to restore balance to a system that is too warm, by creating a chimney to vent off the hot air to the upper atmosphere. So if an an inordinate number of hurricanes occur in the Atlantic, or if they are unusually strong, one could conclude that there is a build up of latent heat energy in the ocean and the shearing effects of El Nino are not in.cgiay, which suggests at least a moderate La Nina might be in effect. Looking at the Atlantic hurricane activity to date, there had been effectively no activity until 8/14.  Quiet as a mouse. since then three systems formed quickly, all dying fast other than Hurricane Bill. Dr Grey and associates at the University of Colorado forecast 10 tropical storms and 4 hurricanes, 2 of which would be intense this season resulting in a total of 85% the normal tropical activity (8/4 forecast). A normal year produces 9.6 named storms and 5.9 hurricanes.  At this point odds are likely that even that estimate might be high.  Conversely, if this were a Modoki El Nino, the impact on tropical development in the Atlantic is theorized to be much reduced, if not actually enhancing the odds for development (we have our doubts about that, especially after reviewing historical data on number of storm days during those years). The presumed theory is that since the Modoki  El Nino forms in the Central Pacific, in-flow to the associated Walker circulation east of the core of the warm pool which normally would be over the Tropical Atlantic is shifted west, over South America. This results in in less shearing if not actually supporting a consistent east to west flow both at the surface and at upper levels. But again we have our doubts.  Also of note, the presence of a classic El Nino in the tropical Pacific supports the development of tropical storms both in the East Pacific (off Mexico) and during WWBs and the Active Phase of the MJO, in the far West Pacific.  That trend has been in evidence, with a steady pack of stronger systems already on the books in the West Pacific, but also a steady stream of lesser storms in the East with no end in sight. This is a significant change from the past three years, which have had below normal activity levels.        

The latest El Nino discussion from the Climate Prediction Center/NCEP (August 6, 2009) states that a weak El Nino was present during July 2009 and expected to strengthen and last through the northern hemisphere winter of 2009-2010.

Models: Looking at the MJO models, a rather vigorous pulse of the Inactive Phase of the MJO was reaching over the dateline while a weak Active Phase was exiting over Central America.The Inactive Phase is to become dominant and peaking out by 8/18, then slowly fading through 8/28 and all but gone by 9/2 while a new Active Phase starts building in the Indian Ocean. The projected strength of this Inactive Phase is somewhat surprising (at least on the models) given that El Nino is supposedly building, and will likely take a pretty good bite out of the developing warm pool off Ecuador. One last solid Kelvin Wave is en-route to Central America, likely to help provide reinforcements, but then it will be a long haul till another follows, unless some unforeseen event generates another. With 20-25 day cycles between each phase change, it will be mid-Sept before another could expected to be generated.

Even further out, of 18 ENSO models run in July 16 2009, all but 3 indicated some form of El Nino developing with waters temps greater the 0.5 deg C above normal. the remaining 3 indicated a rise less than 0.5 degs. none indicated La Nina. The average of the peak was 1.25 degrees above normal occurring in mid Nov. with anomalies 1 deg or grater into April of 2010. 

LONG-RANGE NORTH PACIFIC STORM AND SWELL GENERATION POTENTIAL FORECAST

Fall/Winter 2009-2010 Swell Generation Potential (for California & Hawaii) = 7.5
Rating based on a 1-10 scale: 1 being the lowest (small and infrequent winter surf conditions), 5 being normal/average, and 10 being extraordinary (frequent events of large long period swells)

Updated Methodology (from 2008) : We have upgraded our methodology for making long term predictions. In the past we looked solely upon the presence El Nino using the approach that El Nino typically enhances the size, strength, frequency and duration of winter North Pacific storms in and around the Gulf of Alaska, thereby improving the likelihood for large winter surf in California and Hawaii. And that La Nina typically decrease the size strength, frequency and duration of such systems. But after reviewing data from many such years, we are turning more to the opinion that either strong El Nino or strong La Nina events have enhancing effects on net North Pacific storm formation. That is, the more extreme the divergence away from a neutral state, the greater the propensity for weather systems to try and return the system to a state of equilibrium in the form of storms (which create winds and therefore waves). In the El Nino state, the focus of the storm is centered more upon the dateline and the Western Gulf of Alaska, where in a La Nina scenario the focus is more on the Eastern Gulf. Therefore, the swell source moves east, making Hawaii a less likely target.

Also the relative strength of the MJO has an impact on the transport of tropical moisture from equatorial regions of the West Pacific northward to ultimately fall under the influence of the jetstream, increasing the probability for storm formation moving over the dateline and into the Gulf of Alaska.

In addition we are become more convinced there is a teleconnection between activity in the southern hemisphere and that in the north 6 months later. That is, in years where the net storm activity is up in the southern hemisphere winter (summer in the N Hemi), a corresponding increase in activity could also be noticed in winter in the northern hemisphere (6 months later). The only exception is when there is a strong El Nino or La Nina up north in the winter, then that translates into a net increase in winter activity in the southern hemi 6 months later. Most of this focuses on the strength of the MJO, and seeing how the area it directly impacts is the equatorial Pacific which straddles both the north and south hemisphere's, it would seem reasonable to have an impacts at both poles.

And yet one more possible early indicator is the configuration of the jetstream over the North Pacific starting late July into August over the North Pacific. There is some evidence to suggests a healthy consolidated flow over the NPac early in the Fall season might lead to a continuation of that pattern through the Winter season, and that if an early season .cgiit pattern develops, it will continue in that mode through Winter and Spring. A .cgiit Northern hemi jetstream does nothing to support surface level gale development.

Forecast Conclusion: There is.cgienty of data to support a thesis that weak to moderate El Nino conditions will persist over the next 6+ months. The presence of a warm pool positioned in the classic El Nino corridor and a look at what happens historically when such a pattern is preset at this time of year is all the evidence required. So the issue then turns to a discussion on whether it will develop more, or just hold. At this time it does not seem likely that it will evaporate, though that is always an outside possibility. Historically Classic El Ninos continue to develop through the Fall months, peaking out in December with effects on the environment holding well through Spring, and if strong enough, affecting Summer and Fall of the following year (or longer). But that is only in the case in an extreme event (i.e of magnitude similar to '82/83, '88/88 and the '97/98 events). There is no evidence whatsoever to suggest this years event will come anywhere close. Rather a moderate event is the best likely outcome, with reasonable odds it could actually hold and become no stronger than it is today (weak to moderate). 

The biggest concern is the strength of the current Inactive Phase of the MJO (8/14/2008) and it's possible detrimental effects on the developing warm pool. There is always the possibility that this is the start of a series of stronger pulses, with a possible solid Active Phase developing in September to counteract whatever is lost in August.  But such a discussion would be pure speculation and is unsupportable by any objective evidence. So we have to go with what is real now. Over the short run the Inactive Phase will likely have a limiting effect on the transport of moisture to northern latitudes limiting fuel for storm development, and putting a damper on what was starting to look like a decent though not spectacular North Pacific tropical season. The good news here is that El Nino appears to already be having some positive impact in the North. And it will only serve to extend the tropical season, likely into November or later if it evolves as expected.   

Down south, southern hemisphere net winter storm activity was been above previous years, more reflective of a historically 'normal' pattern, with 4 significant class storms/swells occurring and a series of solid utility class storms resulting in swell having hit Hawaii, though smaller along the US mainland. Reviewing rainfall records for Eastern Australia over the past 6 months, the area most often affected by El Nino during their winter months, a drought has been declared in Central Queensland covering 35% of the total land area with rainfall reported at 10% or less of normal and the same condition over a large swath of North-Central Australia. Anecdotal reports indicate a hard drought and wildfires are persistent. So there is sufficient evidence to support that El Nino is already affecting the atmosphere on at least a large regional level (Pacific Basin). And that affect is likely to expand, deepen and gain in momentum the longer the warm pool stays in.cgiace, branching out and starting to turn events in the Atlantic and Indian Ocean too.  And we suspect that is already occurring, as evidenced by the strong cool flow present off tropical West Africa and the apparent damper on tropical storm activity there. 

The North Pacific jetstream pattern looks good as of early August supporting transport of a steady stream of small but cohesive low pressure centers off Japan tracking over the dateline and lifting north into the Bering Sea near the Western Gulf of Alaska.  This is what one would expect to see if El Nino was in.cgiace. And this has been happening since late July. A pocket of cooler than normal waters is positioned in the northern dateline region, evidence of the mixing effect winds from these system have had in the region.The next step in this process would be to see tropical systems in the west turned hard east, transitioning to extratropical status and building while moving over the dateline towards the Gulf.  In the super El Nino of '97, the first Significant class storm of the year occurred in mid-Sept, an extratropical cyclone resulting in 65 kt winds over a large area aimed east producing swell of 10 ft @ 25 sec solid hitting the California coast on Sept 27. And lesser northwest swell producing systems had occurred as early as mid-August. Clearly that is not the expectation this year, but it provides some historical context and a model of what can occur in the perfect scenario. We are in that window now, so it is advised to keep a watchful eye to the west. 

We have assigned a swell potential rating of 7.0 for the coming Fall and Winter season, suggesting that there is better than average odds of historically longer runs of Significant Class Winter swells.  That means not only more storm frequency, but stronger and longer lasting ones producing larger and longer lasting swells. A temporary return to the glory days of the late 90's and early 2000's.  Compared to the absolutely abysmal surf pattern of the past 3 winter seasons where we would get one day, or even 12 hour swell events, this years pattern should stand in stark contrast and be a much welcome change. Plan your surf strategy for the long run, where endurance and stamina month after month outweght short feasts and 'go-for-broke' assaults.   Likewise weather will likely.cgian a factor in CA. More moisture than normal could be expected with increased snow pack levels in higher elevations of the Sierra on into Nevada and Colorado.  El Nino tends to shift the jetstream southward and flat over the continental US, favoring precipitation for the more southerly positioned ski resorts in the Southwest. But it too causes surface temperatures to rise, meaning the freeze line will increase in elevation with increased odds for rain at lower lying resorts.   

This is a preliminary assessment, based on what is known at this time. It could be ratcheted up or down as conditions in the MJO and the warm pool change through the critical Sept-Oct timeframe.   

And there remains the nagging question of whether we have moved into the Inactive Phase of the Pacific Decadal Oscillation (PDO). The PDO is in-effect is a 20-30 year cycle of weather than slightly mirrors El Nino in the active phase and La Nina symptoms in it's inactive phase. Our thinking is that if we are in-fact in the inactive phase of the PDO (probably since 2000, and will be for the next 20 years), then the odds for favorable winter storm generation conditions are stacked more in favor of La Nina than El Nina, since the inactive phase of the PDO mimics a weak La Nina. But this is mostly just pure speculation.  And there is other data that suggests that we have only been in a 'corrective pattern' since the big 97/98 El Nino. Since that ENSO event was so large and strong, the atmosphere has been trying to re-establish some form of equilibrium for nearly a decade since, and this is the first chance since then for a normal pattern to manifest itself. Our thoughts are the historical record is too short and it too soon to know with any certainty whether we are in a down phase of the PDO.  

So assuming a weak to moderate Classic flavor of El Nino is in.cgiay and will continue to hold through the Fall and Winter of 09/10, we calculate net storm activity will be greater than normal, with storms lasting longer in duration and moderately more intense than normal with the potential to cover more surface area, resulting in generally a larger fetch of increased duration and more intensity. This should results in more consistent, larger and longer period surf than has been seen in the past 3+ years, and perhaps better than anything in the last 10+ years.  

(This forecast is highly speculative and based on historical analysis of past La Nina/El Nino events and the latest long-range forecast models)

Sea Surface Temperature Anomalies
Courtesy: NOAA OSDPD
Notice a broad area of warmer than normal water temperatures (yellow and orange) extending along the equator from the West Pacific all the way east to Ecuador. Also notice the concentration of the warmest waters off Central America 1.5+ degs C above normal.  This is the location of what is called a Classic El Nino.  This is reasonably similar to what water temps have been since July, with not much increase or decrease since then.  Under normal conditions this water should be pooled up in the far West Pacific.  The reversal of the warm pool from the west to the east is caused by strong westerly winds blowing (west to east) over the Pacific equator pushing warmer water from the west to the east. These winds were caused by enhanced activity during the active phase of the MJO. Note cooler than normal waters flowing from Africa towards South America on the equator.  This is likely the result of inflow into the Walker circulation (east to west flow) with the updraft center over the warm pool off Ecuador, and the exact opposite of the flow that is occurring in the equatorial Pacific (west to east). Warmer than normal waters are also pooled up off the Pacific Northwest in the Gulf of Alaska, while a slightly cooler pattern reigns over the southern hemisphere. 

.

Sea Surface Temperature and Surface Wind Anomalies on the Equatorial Pacific
Courtesy: NOAA PMEL
In the top image notice winds blowing from east to west over the entire Tropical Pacific Ocean, the standard trade wind pattern for this time of year. There is no evidence of reversed trades in this image which would be associated with the active phase of the MJO or a Westerly Wind Burst (WWB). But notice that winds are near slack in the west. In the lower panel notice that surface temps are above normal over the equator south of Hawaii (155W) and slightly warmer over the entire equatorial region. The arrows indicate the strength and direction of wind anomalies, which are trending towards more of a west to east flow than the normal trade pattern.  several time this year fully blowing west winds had occurred lasting for a week or longer (a WWB).

 

20 Degree Thermocline Depth and Position Time Series
Courtesy: CPC NCEP NOAA
(Top Image) The core of warm subsurface water is centered just west of the dateline and making slow but steady headway to the east, symptomatic of El Nino. The thickness of the depth of warm waters in the east has increased, another indicator of El Nino.
(Lower Image) Notice the pronounced pocket of warmer than normal water 
(2+ deg C below normal) at 165W and 125 meters deep. This is a Kelvin Wave  that is traveling from west to east generated by a previous Westerly Wind Burst in late July. It typically takes 2 months for a Kelvin Wave to travel from the West Pacific to Ecuador, where it eventually breaks to surface there generating the pool of warm water presently holding off Central America.

.

Equatorial Pacific Sea Surface Temperature Forecast
Courtesy: NOAA/NCEP
Notice that the average of many separate runs of the NCEP model suggest generally another 1 deg C increase in water temperature are forecast off Ecuador by Jan 2010, consistent with development of a moderate El Nino.

.

Southern Oscillation Index (SOI)
Courtesy: BOM
The SOI depicts the difference in pressure between Tahiti and Darwin Australia. When it is consistently negative (that is surface pressure is lower in Tahiti than Darwin Aust), the trend is towards El Nino. And when it is positive the trend is towards La Nina. Notice that since February 2009 there was a steady downward trend, symptomatic of El Nino. A strong upward flux occurred in July associated with the Inactive Phase of the MJO, but has since dropped as the Active Phase took over later in July. Also notice that even within these broad trends there are distinct smaller up and down cycles at 30-45 days each. These are pulses of the MJO. Dips are the Active Phase of the MJO and rises are the Inactive Phase. The current Inactive Phase of the MJO is evidenced by a slight uptick at the tail end of the graph.



Pacific Countercurrent Anomalies
Courtesy: OSCAR/NOAA
This image depicts the zonal flow of the Pacific Counter current, which runs roughly on the equator between the West and East Pacific. When the flow is east to west (blue) as it normally is, this means nothing, unless is is tong, then that suggest La Nina.  When the flow reverses and moves west to east (red) that suggest El Nino.  
The top panel depicts the absolute flow and speed of the current. Notice the '97 El Nino event is obvious, but all others are more just hints or fragments of something occurring.
The bottom panel depicts anomalies in the current as compared to historical and seasonal averages. In this image departures from normal are clearly obvious, with strong red instances reflective of El Nino and strong blue of La Nina.  Again notice the clear signal of the '97 El Nino event. Not all red signals necessarily resulted in an El Nino (i.e 2008).  It appears the current change must be sustained over some minimal duration threshold. 


9189

.
Contact | About | Disclaimer | Privacy
Advertise/Content | Links
Visit Mark Sponsler on Facebook Visit Stormsurf on Instagram Visit Stormsurf on YouTube
Copyright © 2024 STORMSURF - All Rights Reserved
This page cannot be duplicated, reused or framed in another window without express written permission.
But links are always welcome.
Buoys | Buoy Forecast | Bulletins | Models: Wave - Weather - Surf - Altimetry - Snow | Pacific Forecast | QuikCAST | El Nino | Tutorials | Great Circles | Calculator