Tradeoffs between climate change mitigation and adaptation policies are explored under both certainty and uncertainty with learning using a numerical two-period decision model. We first replicate a version of the Adaptation in DICE climate model (AD-DICE) (de Bruin et al., 2009), which modifies the Dynamic Integrated model of Climate and the Economy (DICE) (Nordhaus and Boyer, 2000) to incorporate climate change adaptation explicitly into the traditional optimisation framework. Our model is then extended to include uncertainty over a probability distribution of expected Climate Sensitivity (CS) values and the implications for optimal mitigation and adaptation levels are then explored. In the certainty model runs, the results of previous studies that incorporated adaptation into the portfolio of climate change responses are largely confirmed. Modelling an uncertain CS with the same expected value as under certainty leads to several insights: before learning occurs, optimal levels of both mitigation and adaptation are lower under uncertainty than under certainty; in this same early period, optimal mitigation and adaptation levels are most sensitive to the respective cost of each strategy, with the mitigation level more dependent on adaptation costs than vice versa; variance in CS – a parameter with long-term effects – affects mitigation levels more than adaptation levels.