As an employee of the NRAO, I can tell you almost everything collected is publicly available. We do respect a proprietary period of one year, during which only

As an employee of the NRAO, I can tell you almost everything collected is publicly available. We do respect a proprietary period of one year, during which only the principal investigator's team can access the data, but after that those observations become public. Our internal storage cluster holds around half a petabyte. If anybody wanted to, they could certainly mirror it, but since we're talking about radio astronomy data here it's not likely to be interesting to anybody but a radio astronomer. You can of course take a CASA tutorial and learn how to reduce that data yourself! To the best of my knowledge, this is general policy for the NSF and other NSF-funded institutions have the same charge. Archive development is a big topic for all scientific institutions. It's actually a fairly significant part of my job.