nology, which in the context of climate-change mitigation is the price of a technology
that is capable of completely nullifying CO2 emissions. In DICE, the backstop is
deployed if the control rate on CO2 emissions reaches 100%, so it is conceptually the
marginal cost of the last unit of emissions abatement. Such a technology could most
plausibly be a zero-emissions energy technology such as solar or geothermal power.8
The backstop price starts very high (mean = US$1170/tC), but declines over time.
Hence it becomes an important determinant of the cost of abatement in the long
run.
The sixth and seventh parameters in Table 1 capture important uncertainties in
climate science. At a very high level of abstraction, one can distinguish between (i)
uncertainties in climate modelling that derive from the cycling of carbon between its
various ’sinks’ (the atmosphere, the hydrosphere, the biosphere and the lithosphere),
which therefore render forecasts of the atmospheric stock of CO2 for a given pulse of
emissions uncertain, and (ii) uncertainties in the relationship between a rising stock
of atmospheric CO2 and temperature. In DICE, the carbon cycle is represented by
a system of equations, each containing several parameters. Here, uncertainty about
the carbon cycle is captured in a tractable way by focussing on a parameter that
determines the proportion of CO2 in the atmosphere in a particular time period,
which dissolves into the upper ocean in the next period.
3.1 Climate sensitivity
Uncertainty about the relationship between atmospheric CO2 and temperature is
captured by a random climate-sensitivity parameter. The climate sensitivity is the
increase in global mean temperature, in equilibrium, that results from a doubling in
the atmospheric stock of CO2. In simple climate models, it is critical in determining
how fast and how far the planet is forecast to warm in response to emissions. The
IPCC’s Fourth Assessment Report compiled a number of recent estimates of the
climate sensitivity (IPCC, 2007). It concluded that the best estimate of the climate
sensitivity is 3oC, that there is a greater than 66% chance of it falling in the range 2-
4.5oC (the IPCC’s “likely” range), and a less than 10% chance of it being lower than
1.5oC (“very unlikely”). This leaves around a 17% chance that the climate sensitivity
exceeds 4.5oC, and indeed a critical feature of all 18 probability density functions
of the climate sensitivity compiled by IPCC is that they have a positive skew, with
8It could also be a geo-engineering technology such as artificial trees to sequester atmospheric
CO2 , except that DICE has exogenous emissions of CO2 from land use.
13