Yeah. There were also come claims with Cl-36, but multiple measurements have the effect in opposite directions and different magnitudes (http://arxiv.org/abs/1208.4357, so they seem more likely to be due to instrumentation effects than real differences
This is one of those "extraordinary evidence" things, and we aren't there yet. Annual variation is always suspect because experimental conditions can change subtly with the weather.
assuming a single image encompasses the entire sky
That assumption is quite false: it takes an exposure every 20 seconds and takes days to form an all-sky survey. The pixel size is 0.2", so Jupiter at its closest would be 250 pixels wide, not 4.
High resolution mass specs measure down to 0.0001 amu
My mass spectrometer (a Penning trap) routinely measures 1E-8 amu (10^-32 grams), and the best traps are pushing 1E-12 amu. They're getting to the point where they can see the chemical binding (mass-)energy between atoms in a molecule.
The method in the article is neat, but they've chosen a peculiar definition of "scale" in order to classify this as the most sensitive one.
Never test for an error condition you don't know how to handle. -- Steinbach