This is a mis-framing of the climate problem that we're trying to solve. ENSO does lose predictability rapidly with time (less than years) just as weather loses predictability on timescales of a week or so, a tornado on timescales of less than several hours, etc. All of this results from chaos theory that we're all aware of. - - - Updated - - - It's a well-known phenomenon, call it a known known, but I'm glad some engineers recognize it! It may surprise people that it does take some training in atmospheric science to understand atmospheric science (note that I have no intention of telling you guys how to build a bridge, or whatever you do).
The original models which all of CAGW is based on did not couple ENSO and AMO since they did not believe it had much effect.
You confuse humbleness with arrogance. Engineers know that the world is chaotic and very unpredictable. We know that there are both known unknowns and unknown unknowns. We dont pretend that we have figured everything out. It is from this humbleness that we dislike academics who never work in the real world and are arrogant enough to think that they have figured everything out. How many times has Chris tried to claim that because academics hace written papers on a subject that tgd subject must be well understood? That is arrogance.
Bull(*)(*)(*)(*)!!! Dedicated oceanic models desinged to model ENSO are failing to make correct predictions based on present data. These are far more advanced models than the approximations used in GCMs. The dedicated ENSO have failed so often recently that the Australian Bureau of Meteorolgy has stated to the press that they have been burnrd too many times. The ENSO is a known unknown. You pretend that its well known because some guys wrote papers. If you pay people to write papers on a subject you will get papers on that subject. The complete and abject failure of dedicated oceanic models to predict near term ENSOs proves that at present science understands little about the ENSO.
You're being ridiculous. I never said that papers written on a subject mean it's well or completely understood, but reading what experts actually know about a subject is a pretty good start in advancing your knowledge on the topic. That's common sense, although requires a bit more effort than reading just this forum, Pretending that you can just ignore what we know because a topic is complex or incompletely known is just ignorant (and arrogant), but you insist on doing this because you're an engineer who took a class on systems. If that's your philosophy in life, then whatever... Right now, it is very evident that you know very little about climate or how we model it, and are getting far too much information from wingnut sources...which is why I politely recommended you do a little reading of the scholarly literature before making absurd claims. For instance, ENSO does not arise out of GCMs because of some random number generator telling it to do so (in fact, this doesn't even make sense). Until you do a bit of honest homework, you are not actually being a "skeptic" or a constructive critic. What you don't understand (and what I replied to Hoosier about) is that even if we knew "everything" about ENSO (whatever that would mean in some ideal world), there are fundamental predictability limits on it in a practical universe. Just like we could never model your hometown weather three weeks from now, I cannot tell you what the state of El Nino will be like 12 years from now. Any errors in the initial conditions in the model, even on the order of 10^-10 degrees, will eventually propagate onto the large-scale circulation and change the trajectory of both the real and model world weather (including ENSO). The reason "dedicated oceanic models" will have superior performance to those used in climate GCMs in this respect isn't necessarily because of their physics or degree of approximations, but because weather and short term decadal forecasts are being initialized to observations as best as possible in order for the model oceanic/atmosphere to follow a unique trajectory. For good weather models with good initializations, it turns out that a good weather forecast time horizon will typically stretch out to a week or so in Earth's atmosphere. For climate GCMs, the historical or future experiments may begin in 1850 AD or so, and thus all memory of initial conditions becomes lost and the problem becomes one of a satisfactory climatology. The statistics of this climate are well-behaved in the sense that model climates (or those we observe) do not diverge into alternative states based on initial condition uncertainty, i.e. that the statistics of the system are “ergodic”. Therefore, in climate projections the question is whether the signal of the forced component to greenhouse gases can emerge against 1) the noise of internal variability inherent in the climate system, 2) to natural external forcings or "unknown unknowns." To cover these bases: 1) Internal variability includes ENSO, NAO, AMO, etc that emerge as a consequence of the interacting ocean-atmosphere dynamics. The forced response in this sense includes some perturbation to the system- whether it be orbital (let's think day/night, seasons, etc), changes in energy input from the Sun, changes to greenhouse gases, volcanic eruptions. Whether or not the development of El Nino will kill the usefulness of our forecast depends on the size of the signal and the timescale we are interested in. For the seasonal cycle, summer is consistently warmer than winter (statistically), because the signal associated with higher solar irradiance in the hemisphere tilted toward the Sun is very strong. No plausible uncertainty in initial condition will disrupt this conclusion, even if it happens to kill the weather forecast three weeks from now. Similarly, if we moved the Earth closer to the Sun, its climate will warm in the long-term, and we can arrive at this conclusion independent of any detailed consideration of ocean dynamics (based on planetary energy balance)...it is the top of the atmosphere energy budget that shackles a planet's climate; all weather this climate experiences is the particular trajectory of the system through state space and its climate is the manifold on which that trajectory moves. From this, it becomes obvious that climate is a boundary value problem. Therefore, what you need to show is that the decadal to centennial statistics of the system response to external forcing (e.g., CO2 rise ) can be swamped by the uncertainty in the initial conditions or uncertainty due to ENSO, PDO, etc. This is also why climate modelers (and also the groups that forecast developing El Ninos) run an ensemble of simulations in order to sample uncertainty-space associated with initial conditions. For greenhouse gases, ensemble runs do fluctuate about the mean upward trend, but the trend is clearly beyond the difference amongst the ensemble members. Until you can pony up a credible atmosphere-ocean-glacier mechanism that can generate chaos or strong sensitivity to uncertainties of the sort to disrupt this conclusion, and also has explanatory power paleoclimatically, your concerns are bankrupt. 2) No one pretends all things are known or that unknown unknowns can't exist. It is of course conceivable that models are missing some process entirely, but the observational or paleoclimate record of Earth's history doesn't hint at all that such a major process is missing from our understanding...at least when it comes to the big picture question of planetary temperature, which is understood rather well. There is strong evidence that Earth's geologic history is a strong function of its CO2 history. Obviously if a volcanic eruption goes off, any near-term projection will no longer be valid (although the impacts are much shorter lived than the CO2 lifetime in the atmosphere, same with typical solar fluctuations, which tend to be very climatically small). If an asteroid hits the Earth, the long-term forecast will be useless as well. Unfortunately, we cannot include these factors anymore than an engineer can include the prospect for an asteroid hit in their bridge design plans. What we do hope to do is arrive at conclusions that ought to be robust for a plausible range of uncertainty-space.
Ha Thats a hoot you spend such a diatribe to close with yeah there are known unknowns and unknown unknowns but we knoe just about everything. What arrogance! News falsh thd PDO wasnt identified until the 80s, the AMO, and AO after that. The origin of most climate models actually precede their discovery. These were major unknown unknowns. They are still major known unknowns. You are saying nithing different than climate modlers were saying in tge 80s. Oh yes there are some thing we dont know but we kniw most everything. Academics like to think that they are so different tgan those that preceded them but they arent. They are just as fricking arrogant. That is the difference between an engineer and ab academic. The academic has to make bold claims of certainty for their own wellbeing an engineer has to make bold claims of uncertainty for the wellbeing of others.
Re-read what I wrote. I guess you've been reduced to making things up about what I've claimed. Not interested anymore.
I read what you wrote. Its crap that says very little. You are confusing complex oceanic models like the the ones the ABM and NOAA use to forecasts near term el ninos, and also dont work, with those of long term projection GCMs in which such events are ti quote NCAR "Its a more random and statistical representation of events rather than actual. Therefore, a specific event like an el nino is present in the model only in a statistical sense." As I said you are playing SimCity. Godzilla attack run!!!! "
Accurately modeling ENSO only improves the short term accuracy, it doesn't change the fact that the long term trend in global temperature continues to increase. http://arstechnica.com/science/2014...got-el-nino-right-also-show-warming-slowdown/
They may be getting close. Go to ContextEarth and read back about a year or so on his SOI model. You may be surprised; I was.
