>I mean if this is not a measurement error then what would be your first idea bout possible causes?
I now have found the ultraviolett measurements at the extreme ultraviolett variability experiment
which is also at UColorado. The plots take at least as long as the Lisird ones plus the connection to UColorado from Europe may not the best, i.e. I could only look at very few samples. However I noticed that there seem to be a rather sudden increase in around mid 2013-2014 in irradiance in the line O III 526. Unfortunately the measurements seem to have had a major outage from beginning of Jun 2015 to around Dec. 20. 2016.
An increase in irradiance of an element might be due to more ionizing and recombination processes with that element in question, which in the turn appears mostly (?) due to more ionizing radiation or it could be due to an increase of the quantity of the element itself in the (more or less (depending on the nature of radiation)) outer layers of the sun or both.
So more ionizing radiation could mean that fusion processes take place in more "outer layers" but so
I am also asking now myself wether it could be that there are more really hot regions in the sun so that eventually more Bethe-Weizäcker Cycle
fusion instead of proton-proton fusion
than before may take place. Apparently the Cycle produces O15 instead of O16 (see here
(page 17)), i.e. there is one neutron missing to the normal oxygen atom. The O III line comes apparently from a doubly ionized oxygen
where alone for line broadening reasons it seems still assumed that isotopes have the "same" spectral line. Here is the NIST data
for atomic number 8, where the orbital transition for the 526 line is described.
Any comments on that?