This study is concerned with the impact of an actual forecast of total water supply available (TWSA) made by the Bureau of Reclamation for irrigators in the Yakima Valley (Washington). Estimates of TWSA (legally calculated by combining natural flow estimates, water storage, and other sources, i.e., return flow) were issued in February 1977, suggesting that less than half of the long‐term average water supply would be available for the coming irrigation season. The Bureau allocated the estimated TWSA as follows: those with nonproratable (senior) water rights were to get 98% of their normal water allocation, while those with proratable (junior) rights were to receive about 6%. The TWSA estimate and subsequent water allocations prompted irrigation districts, as well as individual farmers and federal, state, and local government agencies, to take a variety of measures they might not otherwise have taken. By May 1977 the Bureau realized that there was more water in the system than they had expected and that some of their assumptions had in fact distorted their projections of TWSA. Many farmers took legal action against the Bureau to recover costs they claimed were incurred by their responses to the erroneous water supply projections. This study raises questions about societal responses to seasonal water supply estimates and about the benefits to society of good forecasts as well as the costs of erroneous ones. It also raises the issue of who, if anyone, should bear the responsibility for the societal implications of such forecast