Several long-running and geographically extensive survey programmes provide information used to make annual management decisions for North American waterfowl populations. Data from these programmes can also be viewed as resulting from long-term population studies and have formed the basis for a number of retrospective analyses. Short-term studies have been used to investigate mechanisms underlying results of the retrospective analyses. Results of the long-term and short-term studies complement each other nicely and have led to many useful inferences about North American waterfowl population dynamics. However, important questions remain unanswered, and it is suggested that many of these could best be addressed using an experimental approach.