The conventional definition of temperature seasonality is the difference between the annual maximum and minimum temperatures (figure to the left).

Plant growth in the North does not depend on these two extreme values - *it depends on the seasonal cycle of temperature above some threshold* (figure to the right).

Therefore, we define temperature seasonality as the inverse of the cumulative value (shaded region) in the figure to the right.

Both definitions when based on temperature values averaged over a latitudinal belt around the Earth (50N to 65N in the figures above) are a function of Sun-Earth geometry only, i.e. Earth's axial tilt and Orbit around the Sun. Thus, we expect *no time trends in seasonality during our study period (100+ years) due to changes in sunlight and day length*.