The date usually given for the start of the Kali Yuga, 18 February 3102 BCE, is not a scriptural date but an astronomical one, and its origin is worth setting out. Indian astronomers of the classical period, Aryabhata in the Aryabhatiya of 499 CE among them, worked with a model in which the planets returned to a common starting point at a great conjunction, and calculating backwards from observed positions gave an epoch in the early fourth millennium BCE.
The Kali Yuga epoch is therefore a computed result, adopted as the zero point of the Kali Yuga era still used in Indian calendrical work, where the current year is given as a count of elapsed years since it. It is also the base from which the traditional 432,000-year duration is measured, which places the end of the present age far in the future.
That last point is frequently lost. Popular writing regularly presents the Kali Yuga as ending soon or as having ended, and neither claim follows from the traditional reckoning, on which barely five thousand years of 432,000 have run. Where a modern author has proposed a shorter cycle, as Sri Yukteswar did in 1894, that is a modern proposal and the atlas names it as one.
There is a further caution about the epoch. That Indian astronomers computed a date does not make it a historical event, and the traditional accounts that place the death of Krishna at the transition are religious narrative rather than chronology. The atlas records the computed epoch, the reasoning behind it and the tradition attached to it as three separate things, because they are.