The price (or premium) of a weather derivative may be determined by applying statistical analysis to historical records from first-order stations of the U. S. National Weather Service. Such records, however, often suffer from inconsistencies introduced by changes in station location, station elevation, instrumentation, and observation protocol. Additional uncertainties arise due to human and environmental factors, such as missing observations, urban heat islands, El Niño/La Niña, and global change. Some weather derivatives may be mispriced as a result. Inaccurate pricing may undermine confidence, lead to problems in market liquidity, and thus may deter potential new market players.

Statistics are presented on over 60 put and call options tendered during fall and winter 1998-1999. Bid and offer prices from various models are compared, and the roles of the meteorologist in pricing and tracking are discussed. It is demonstrated that by adjusting for the apparent biases in historical measurements and by applying the results from a state-of-the-art climate forecasting system, one can improve pricing strategy, increase confidence, and thereby gain an advantage in the weather derivatives market.