Depletion rate is a key factor and it really impacts timing of peak and consequently what each of us can do to prepare for peak.
Here's what my own model looks like while using a 2% depletion rate and 1.7% annual demand growth rate (as expected by both ExxonMobil and BP).
Here's what 5% depletion rate looks like.
And here's what the 6.9% depletion rate that the IEA looks like.
In the near term years, I've plotted the initial yields from new fields coming onstream and while they help to offset some of current depletion, the new fields for 2005-2008 don't really kick in til the second or third year of production.
Note that in my model I've also never accepted input from new fields as less than 3 million barrels per day clear out to 2030, even though this may be excessively optimistic. But it really doesn't matter since the depletion curve starts to eat at so many existing fields that we basically need to discover several new Saudi Arabias to continue growing. It's just not going to occur.
This drives us into doing what we can with alternatives as soon as we can. If the 2% depletion expectation is correct, we've still got a few years where the industrialized nations can "save themselves" even if they can't save the poorer nations of the world. But they better get busy right now since turn around time on capital investments of this magnitude is measured in decades
Here's another point to consider, which has been harped on by Saudi Arabia. The world is at refinery capacity peak right now too. Even if those fields come online we need the equivalent of at least 4-5 new "global" sized refineries every year for the next 10 years at least. Since these things take 5 years or so to build even if you cut out the 10 years of red tape, we're just not going to have that refining capacity until 2010 at the best.
Because of that, if we get the 2% depletion scenario, we get to watch the entire world stagnate in a low-growth (constrained by refinery capacity which constrains energy inputs to growth) economic scenario for the next 5 years. This is actually a good thing since it shoves peak out further in the future and gives the world more time to do something constructive about it. But it depends on actual depletion being as low as 2%.