The Nucleon-Nucleon database
Our database consists of elastic nucleon-nucleon scattering data in the low and intermediate energy region, up to Tlab around 3 GeV. The database is created in order to use them in partial wave analyses (PWA's) of nucleon-nucleon scattering data.
It is our goal that the database, in the course of time, will be fully documented with hyperlinks to original online papers of references. This way it will become an electronically accessible mixture of the NNDC database, the Nucleon-Nucleon Scattering Data Tables of Bystrický and Lehar, and SAID.
The database is built with in mind its applicability in PWA's. PWA's are statistical data analyses in which we try to translate all information that is contained in the experimental data into phaseshifts; modelindependent quantities that provide us useful information about the nucleon-nucleon interaction.
The database is input as well as output of a PWA. One of the purposes of a PWA is actually to contruct a statistically sound database, free of ``wrong'' data and outliers.
What is in the database before an analysis
Input to the analysis is a raw database, which in our case is built up according to the following criteria:
In principle all elastic scattering data were used that are published in regular journals in or after 1955. That means that we do not use:
- Data that are published before 1955.
- Preliminary data ( = data to be published ). Unpublished data may be included in the database, however they will be flagged as such and will not be included in a PWA until publication.
- Data presented in conference proceedings, theses, or internal reports. These data may be included in the database, however they will be flagged as such and will not be included in a PWA until publication.
- Proton-proton total cross-sections. These are not unambiguously defined and there are differences between the way experimentalists treat the Coulomb-nuclear interference terms.
- Data obtained from quasi-elastic scattering (e.g. deuteron targets).
Of course there are occasional exceptions.
We think the database is fairly complete and up-to-date. We are aware of some data not yet included, but they eventually will. We are also aware of some other shortcomings and inaccuracies in the database. If you know of data not in our database, or if you have corrections or other remarks, please let us know.
What is in the database after an analysis
One purpose of a partial wave analysis is to decide on a statistically sound and consistent database. That means that some groups of data or some individual data are tagged and excluded from the restricted database that is used for the final analyses. When browsing these excluded groups or individual data are marked with (*) As we currently have analyses up to 500 MeV, any such remarks for data up to 500 MeV can be trusted. For data at higher data you should ignore them as yet.
Notice that new developments and new analyses may cause rejected data
to be included again and vice versa! Exclusion of data reflects our
most recent decision on whether or not to include data! Usually the
reason for exclusion of the data is given.
We used the following criteria for rejection of data. More detailed
information can be found in Bergervoet et al.,
Phys. Rev. C 38 (1988), 15.
- Single points are excluded when they have a χ2 > 9 in our final energydependent PWA.
- Groups of data are excluded when their χ2 is lower or higher than reasonably can be expected. (A 3 standard deviation criterion!) A table with the minimum and maximum allowed χ2 values is available.
- Data are excluded if their behavior compared to the solution of the final PWA is weird.
- Data of which it is `known' that they are not good or unreliable. For example if (after publication) the experimental setup turned out to behave different than expected, or when authors specifically mention that data are not suitable for inclusion in PWA's.
Notice that new developments and new analyses may cause rejected data to be included again and vice versa! In the course of time these criteria may change, data may change and data may or may not fulfill the rejection criteria in updates and extensions of analyses. Exclusion of data reflects our most recent decision on whether or not to include data! Usually the reason for exclusion of the data is given.