A bootstrap method for testing equality of peak months

When ecological traits differ seasonally among biological groups, environmental adaptation is expected. However, current analytical methods for such seasonal data ignore the circular nature of data and therefore are likely to be flawed. We propose a simple bootstrap hypothesis test which statistically quantifies the presence/absence of differences in peak months for multiple groups taking the circular nature of the data into account. The test is based on a robust distribution-free method. Simulations showed that the test gives a satisfactory performance. The test is illustrated using data for anchovy and sardine egg abundances in the western North Pacific.