Comparing weather and climate, it may not be more difficult to predict a long term climate pattern (months, years, or decades later) than predicting the weather in next week, if you are very confirmed about the critical factors determining the long term trend.
To simplify, there are only two important factors in determining long term global temperature: the sun activity, concentrations of greenhouse gas (GHG) and little aerosols. Nothing to do with El-Nino or La-Nina which are only influencing in a short term period.
In the past (thousands and million years), the atmospheric CO2 concentration and our Earth temperature correlates well as shown in Figure 1. Each ice age (the period between each CO2 and T peak) is about 100,000-120,000 years. Looking into the pattern, each ice age is with sharp sharp (over less than 10k years) increase to its peak, followed by gradual decrease (~ over 100k-120k years). This ice-age cycle is likely due to the long term periodic sun activity, though i did not find the evidence, since no one can estimate the sun activity that long time ago. The reason why the heat-up time frame is always much shorter than the cool-down time frame, could be strongly relates to the greenhouse gas feedback system. While the sun starts becomes more active to cause increase in T. The increased T evaporate more water vapor and CO2 (both are important GHG) from ocean into the atmosphere, resulting +ve feedback to accelerate the warming and result a sharp increase in T. However, the peaks tell that the accelerated heating is not keep going. The only possible reason I can think of is the lower sensitivity of T to GHG concentrations upon this T&CO2 peak (i.e. the GHG +ve feedback system is more influential when the temperature is low), since it is unlikely for the sun to immediately inactivate itself to brake the +ve feed back. The lower sensitivity, plus gradually less activated sun may be the cause of gradual decline in T, and cause more water vapor condensed and CO2 dissolved back into the ocean.
Actually, if the feed back system is as sensitive as with low T, we may already suffering 10F-15F (~5C-8C) higher in T than now. Look at the right most point in Figure 1, the CO2 concentration (blue) is approaching 400ppm, though the T (red) is only ~ 0.8C (~1.4F) (see Figure 2) higher than the baseline. Although I consider the sensitivity of T to GHG concentration could be already weaker, it is NOT so weak that we can neglect. In fact, although in the past 1800 years the T is considered quite stable, the recent 200 years (especially 70 years), with the anthropogenic fast deforestation and industrialization, the rapid increase in T alarms the climate scientists.
Figure 3 shows the recent warming due to net human influence and 11-year periodic solar activity (this is the confirmed short term solar activity cycle, different from the long term ice-age cycle I suggest in part 1). Besides, in the past 2 years (2008-2009) are in a centrally low solar minimum, which may be the cause why Arctic ice extent recovers a bit as compared to 2007 record.
Now the time frame is getting much shorter. I shall bring out my understanding on El-Nino and La-Nina. Not all the heat trapped by the GHG immediately reveals in immediate T rise, our ocean with its huge heat capacity, absorb a majority of heat . This absorbed heat will somehow in somewhere (at least partially) release, in a form that we call it El-Nino. In other words, El-Nino and La-Nina do not affect the total heat from the sunlight trapped on the Earth, but they are the results of self regulation of heat that is already trapped in the ocean: exchanging the heat between deep and shallow water and between different places of the ocean due to the flow.
Even though the arctic ice extent slightly recover in 2008-2009, the arctic ice volume actually keeps decreasing with thinner ice. In other words, the ice under the water keeps decreasing rapidly even during the solar minimum. The main source of heat provided to melt this underwater ice could come from the already trapped heat in the ocean.
With the start-recovering solar activity this year , both the Arctic ice volume and extent are breaking the records (Figure 4 & 5), though the time to break the yearly minimum ice-extent record may likely occur in Sep.
Now it comes to my prediction. With the new record breaking low ice-extent, the heat being absorbed will be even greater, since ice-free ocean is much darker and absorb most light into heat, compared to the ice which reflects the light back to space. This absorbed extra heat by the Arctic ocean will likely delay the ice from refrozen back in fall and winter. This heat will also bring over to other places, and possibly there will be more places without winter in North Hemisphere this year end.
The Arctic sea ice may be totally gone in the coming few years, especially during the coming solar maximum ~ 2012.
To be continued...
I'll find sometime (chance) to share my thought with people more professional in climate science.