The seasonal peak timing gap between Superior and Michigan-Huron showed no clear change between 1948-1986 and 1987-2025

We tested whether the seasonal peak timing gap between Lake Superior and Michigan-Huron changed between two equal 39-year periods. We used coordinated monthly levels, version v2025. Each year had an annual harmonic and a separate linear drift. The peak of the mean Superior harmonic followed Michigan-Huron by 1.330 months in 1948-1986 and 1.347 months in 1987-2025. The change was +0.017 months, with a conditional 95% five-year block interval of -0.293 to +0.338 months. All five planned checks also gave intervals that included zero. Thus, the test did not resolve a change. It does not prove equal timing gaps or measure daily peak dates. The phase of a mean harmonic differs from the mean date of annual maxima. Levels alone cannot identify runoff, evaporation, flow or regulation effects. Short periods, long climate cycles, model assumptions and source revisions limit inference. The package gives monthly inputs, source hashes, code, uncertainty draws, numerical checks and a figure.

Great Lakes · seasonal timing · annual harmonic · Lake Superior · Michigan-Huron · block bootstrap · null result