describing methodology and instrumentation

[chporter]

Need to think more on how to describe methodology and instrumentation.
Very simple method would be to add variables to metadata, linked to variable names.
But we may need a more complex way to handle long-term experiments (for example) which may go thru a succession of instrumentation over the period of the experiment.
Even more complex would be describing weather data which might be obtained from multiple sources, including some computed values.