The design of optimal environmental policy inherits model uncertainty. We investigate the consequences in a simple linear model, where the aim of the policymaker is to stabilise the atmospheric content of carbon. We study how decision-makers' concerns about robustness alters policy using the Hansen and Sargent (2003, 2008) approach. The analysis shows that a policymaker, who fears about model misspecification should react more aggressively to changes in the stock of atmospheric carbon and reduce emissions stronger.