So, to make sure I understand what you wrote, internal variability really is randomness, and it makes it very difficult to make highly accurate predictions beyond a few days.  Also, what is considered internal variability depends on the time scale (with ENSO qualifying as this over a century).  Is this correct?

Based on that, I have a few follow-up questions:

* What are some factors that are considered drivers of internal variability?

* Why are these factors considered to be random?

* Are there ways of reducing the randomness of internal variability?  If so, what are they?

Thank you for reading this, and I look forward to hearing from you.