Abstract
The statistical relationships between lake freeze-up/lake ice break-up dates and air temperature means over various time periods are analyzed for 63 lakes in Finland. Mean temperatures for the individual months before the lake event dates are strongly correlated with these dates; significant correlations hold for periods up to five months in length before freeze-up. Regression coefficients depend on location, but are consistent within regions. Latitude and distance from the coast are the most important sources of variation in the regression coefficients. The regression coefficients are used to translate changes in lake freeze-up/break-up dates into estimated changes in air temperature. In southern Finland a five day change in freeze-up date would represent a 1.1°C change in November temperature of the same sign. A time series of November temperatures estimated from lake freeze-up dates is derived and compared with observations at Helsinki. The spatial pattern of temperature change over time is al... Abstract The statistical relationships between lake freeze-up/lake ice break-up dates and air temperature means over various time periods are analyzed for 63 lakes in Finland. Mean temperatures for the individual months before the lake event dates are strongly correlated with these dates; significant correlations hold for periods up to five months in length before freeze-up. Regression coefficients depend on location, but are consistent within regions. Latitude and distance from the coast are the most important sources of variation in the regression coefficients. The regression coefficients are used to translate changes in lake freeze-up/break-up dates into estimated changes in air temperature. In southern Finland a five day change in freeze-up date would represent a 1.1°C change in November temperature of the same sign. A time series of November temperatures estimated from lake freeze-up dates is derived and compared with observations at Helsinki. The spatial pattern of temperature change over time is al...