Depending on how the Arctic acts this year, the dip in agriculture could be quite steep if the Arctic melts out as a result of its effects on the jet stream ridding the world of one of the breadbaskets.
I think its possible to say that the problems posed by food production and the consequences they entail may be able to sneak up without warning.
Sorry about the mismatched quote pairs - edited my original comment to fix.
I think the earlier stages of this will manifest more as a progressively more challenging agricultural situation with more extreme weather causing more diminished (or occasionally failed) harvests in specific regions. It ought to be noted even with the drought last year - the US still produced a fair amount of food (although the prevalence of drought globally was concerning).
In that sense, I think the effects are
already with us today - that they started to sneak in from 2010 onwards (that one can tell), with the Russian drought (and floods in Pakistan). The point here is really that we can look to what is already happening for clues as to the nearest future (extrapolation obviously increasingly invalid further into the future).
Interestingly a recent study concluded that the ongoing US drought was not caused by climate change, but by an unusual shift in the jet stream. I disagree vehemently with the first part of that assessment (not due to climate change part) as I think the second points to arctic amplification.
http://www.guardian.co.uk/environment/2013/apr/12/climate-change-not-2012-droughtTo that extent the consequences are already here and already sneaking up on us - the immediate questions are really:
1. How fast things get worse?
2. How big the rapid impacts on agriculture are? (noting the presence of a whole host of other longer term issues facing it)
3. How resilient is global civilisation to deal with the shock?
I think most people can agree one 1. that the Arctic sea ice is changing over a timescale of only years. 2 and especially 3 are far more contentious issues, and without solid scientific research supporting different views - really - the only way I can see that this discussion can move forwards much more is:
1. Proof by demonstration from the earth system
2. More scientific research
I prefer not to wait for 1 and not sure I have the time (and most likely the expertise and connections to academia) to attempt 2.
On that note, I'm not sure how the general discussion in this thread can really move forwards without just restating previously stated points of view? I think we can reach a reasonable consensus that:
1. Civilisation is potentially in serious trouble at some point in the future, whether years or decades away (and this surely demands action of some sort!)
2. We can constrain to some extent the range of probable scenarios, both near future and longer term (while leaving large amounts of uncertainty)
Is anyone aware offhand of good research showing the effects upon food prices of various yield loss or supply shortfall scenarios? If so, we could attempt to extrapolate a range of scenarios a bit more scientifically - taking that and trying to determine the instability of various nations and regions taking the research linking food prices and conflicts into account (ie macroscopically conflict is a predictable consequence of high food prices).
Theoretically I suppose one could construct a model of the world - and by model - I mean more a
human model than a climate model, to try to predict the general impacts. For example, if one identified at the macroscopic level particular resource dependencies (oil for instance) one could try to work out the amount of risk posed to production regions, in a model linking things together appropriately (to try to capture knock on effects).
Certainly not an exact science as the human factor is anything but predictable, but one might arrive at something useful as a statistically averaged view of the whole bag of wool is my thinking? (given we are discussing the big picture, not just for specific regions)
In summary it's hard to see how this discussion can advance beyond reiterated opinions at this stage without more scientific rigour being brought to bear? (which actually I'd like try to do, but I'm going to let the idea sit for some time to mull over)