[LHAPDF-announce] Updates for NNPDF23_(n)nlo_as_0119_qed central PDFs

[Andy Buckley]

Dear LHAPDF users,

Today it was discovered that the central/mean (index=0) member of the 
NNPDF23_nlo_as_0119_qed and NNPDF23_nnlo_as_0119_qed PDF sets had been 
fitted with the "forced positive" prescription of the NNPDF "MC" sets by 
accident. This meant that the PDF was in places equal to zero where it 
should have been negative (as is allowed for > LO PDFs).

The NNPDF team have produced an immediate fix, which I have just 
deployed to the collection on the LHAPDF web pages and in our AFS and 
(shortly) CVMFS data libraries. The previous (buggy) version is 
available in the "deprecated" subdirectory on the website. For those 
managing their own local PDF library, this update can be "automatically" 
downloaded and installed via the lhapdf script.

The expectation is that the physics impact of this bug and update should 
be very small, but since it is a change to the values of the central 
member of quite a mainstream PDF, we feel duty bound to let you all 
know. All the other members of each PDF set are unaffected, and any mean 
observable values computed via the replicas will not have been affected 
by this truncation.

Best wishes,
The LHAPDF team

-- 
Dr Andy Buckley, Lecturer / Royal Society University Research Fellow
Particle Physics Expt Group, University of Glasgow