He's talking here about the IPCC's underestimation of the speed of events.
Speaking at the launch of the full 2007 IPCC report on the impacts of global warming, the co-chair of
Working Group 2, Professor Martin Parry, told his audience that: "We are all used to talking about
these impacts coming in the lifetimes of our children and grandchildren. Now we know that it's us."
He said destructive changes in temperature, rainfall and agriculture were now forecast to occur
several decades earlier than thought (Adam, 2007b).
The speed of change can in itself worsen impacts. Leemans and Eickhout (2004) found that species'
adaptive capacity decreases rapidly with an increasing rate of climate change: five percent of all
ecosystems cannot adapt more quickly than 0.1°C per decade over time. Forests will be among the
ecosystems to experience problems first because their ability to migrate to stay within the climate
zone they are adapted to is limited. If the rate is 0.3°C per decade, 15 percent of ecosystems will not be able to adapt. If the rate should exceed 0.4°C per decade, all ecosystems will be quickly destroyed, opportunistic species will dominate, and the breakdown of biological material will lead to even greater emissions of CO2. This will in turn increase the rate of warming (Kallbekken and Fuglestvedt, 2007). Temperatures are now increasing at a rate of more 0.2°C per decade with some IPCC scenarios showing the speed rising to 0.4°C per decade by mid-century, to which few species will be able to adapt. Another study of the IPPC report's low- and high-emission scenarios found 12-39% and 10-48% of the Earth's terrestrial surface may respectively experience novel and disappearing climates by 2100 AD (Williams, Jackson et al, 2007).
Speed of change and uncertainty impel us to consider the worse-case outcomes, not just the scenarios
considered to be the most likely currently. Pittock (2006) argues persuasively that "Uncertainties in
climate change science are inevitably large, due both to inadequate scientific understanding and to
uncertainties in human agency or behavior. Policies therefore must be based on risk management,
that is, on consideration of the probability times the magnitude of any deleterious outcomes for
different scenarios of human behavior. A responsible risk management approach demands that
scientists describe and warn about seemingly extreme or alarming possibilities, for any given scenario
of human behavior (such as greenhouse gas emissions), even if they appear to have a small
probability of occurring. This is recognized in military planning and is commonplace in insurance.
The object of policy-relevant advice must be to avoid unacceptable outcomes, not to determine (just)
the (apparently) most likely outcome."
It is something that has not always been done, leaving the science in crucial areas looking flat-footed and behind-the-times. Hansen sets the stage: "For the last decade or longer, as it appeared that climate change may be underway in the Arctic, the question was repeatedly asked: 'is the change in the Arctic a result of human-made climate forcings?' The scientific response was, if we might paraphrase, 'we are not sure, we are not sure, we are not sure…yup, there is climate change due to humans, and it is too late to prevent loss of all sea ice.' If this is the best that we can do as a scientific community, perhaps we should be farming or doing something else" (Hansen and Sato, 2007b).
EDIT
http://www.carbonequity.info/PDFs/Arctic.pdf