The number of tropical cyclones occurring per year in the western north Pacific basin is remarkably constant from year to year. Why is this so? The simplest explanation would be that some sort of negative feedback acts to regulate the number of tropical cyclones. If so, we should expect that when there is a period of strong tropical cyclone activity, this activity has some effect on the large-scale environment so that subsequent tropical cyclogenesis is inhibited. Such a signal should be observable. If it can be found, it would also provide some insight into the general problem of understanding environmental controls on genesis.

We attempt to address this issue using observations by looking at statistical relationships between different meteorological variables and an index of total tropical cyclone activity, the “accumulated cyclone energy” (ACE), computed for the western north Pacific and averaged to a weekly temporal grid. We examine the spatial and temporal structure of lag-correlation coefficients between ACE and weekly averaged sea surface temperature, outgoing longwave radiation, atmospheric temperature, and wind. The most immediately evident signal is a positive correlation between ENSO and ACE, presumably indicative of a causal influence of ENSO on tropical cyclone activity. When this signal is removed by partial correlation, other signals remain, which have space and time scales smaller than those associated with ENSO, and appear to be more indicative of the immediate effects of tropical cyclones on their environment. Possible mechanisms by which these effects might play a role in regulating subsequent genesis will be discussed.