118 Current Trends in Bayesian Methodology with Applications
deterministic, but extremely complex, function of other physical variables.
This kind of modeling retains a high degree of fidelity to the true process by
which the data is generated. However, it is inevitable that not all featur e s
of a sy stem as complex as climate will be measured or retained in vario us
forms of data r e c ords. Also, our pr e sent state of knowledge about how differ-
ent physical, atmospheric, geophysical and other variables interplay is limited,
as is natural in any scientific discipline. A very partial and incomplete review
of natural scientific modeling of climate may be o btained from [14], [21], [9],
[11] and several references therein. A Bayesian framewo rk wher