My approach for initiating discussion of "large data"
visualizes "Data Assimilation": the numerical blending
(a) all Observations of a system with (b) "Ideas" (expressed
as model simulations based on laws/theories)..
This results in optimal analysis of a system state, which
can be used for
(a) diagnosing system mechanics,
(b) providing initial conditions for predictions into
(c) identifying errors in observations or models,
(d) predicting forecast chaotic uncertainty,
(e) understanding mechanisms of change (e.g.
natural climate variability vs. human-induced change)
(f) clarifying statistical & mechanistic
relationships (e.g., power law pdfs)
(a) Observations from a huge variety of sources,
with increased spatial resolution
(b) Models of higher resolution, with more
precisely & complete algorithms