Lake ice time series consist of the dates of freezing and ice break-up for the ice phenology, and ice coverage and ice thicknesses for the severity of ice seasons. This work analyses the physical interpretation of the statistics of lake ice time series illustrated by available data. The results provide tools to understand variations in the past lake ice seasons and to make projections into the future. The freezing date is related to fall air temperature with correlation time scale depending on the lake depth. Ice growth is primarily determined by the air temperature and snow accumulation. Ice break-up depends on the seasonal maximum ice and snow thicknesses, solar radiation and air temperature; the onset of melting is the primary question. Relationships are derived for ice season characteristics and climate changes. Warmer climate also brings qualitative changes to lake ice seasons by moving boundaries of ice climate zones.