[en] We review neutron cross section covariances in both the resonance and fast neutron regions with the goal to identify existing issues in evaluation methods and their impact on covariances. We also outline ideas for suitable covariance quality assurance procedures.We show that the topic of covariance data remains controversial, the evaluation methodologies are not fully established and covariances produced by different approaches have unacceptable spread. The main controversy is in very low uncertainties generated by rigorous evaluation methods and much larger uncertainties based on simple estimates from experimental data. Since the evaluators tend to trust the former, while the users tend to trust the latter, this controversy has considerable practical implications. Dedicated effort is needed to arrive at covariance evaluation methods that would resolve this issue and produce results accepted internationally both by evaluators and users.

[en] The Generalized Nuclear Data structure (GND) is being designed as a new standard for storing evaluated nuclear data. It is intended as an eventual replacement for the ENDF-6 format, and as such must be capable of handling at a minimum all the types of data supported by ENDF-6, including central values and covariances. Containers for storing covariances in GND are being developed, with the goal of representing covariance data in a simple and concise manner, that is easy to read, understand and use. This paper presents an overview of the status of covariances in GND, and on the tools for generating and using covariance data that are being implemented in LLNL's nuclear data toolkit ‘Fudge’

[en] Recent evaluations of neutron cross section covariances in the resolved resonance region reveal the need for further research in this area. Major issues include declining uncertainties in multigroup representations and proper treatment of scattering radius uncertainty. To address these issues, the present work introduces a practical method based on kernel approximation using resonance parameter uncertainties from the Atlas of Neutron Resonances. Analytical expressions derived for average cross sections in broader energy bins along with their sensitivities provide transparent tool for determining cross section uncertainties. The role of resonance-resonance and bin-bin correlations is specifically studied. As an example we apply this approach to estimate (n,γ) and (n,el) covariances for the structural material ⁵⁵Mn.

[en] We evaluated covariances in the neutron resonance region for capture and elastic scattering cross sections on minor structural materials, ^50,53Cr, ^54,57Fe and ⁶⁰Ni. Use was made of the recently developed covariance formalism based on kernel approximation along with data in the Atlas of Neutron Resonances. Our results of most interest for advanced fuel cycle applications, elastic scattering cross section uncertainties at energies around 100 keV, are on the level of about 7-10%.

[en] We evaluated covariances for neutron capture and elastic scattering cross sections on major structural materials, ⁵²Cr, ⁵⁶Fe and ⁵⁸Ni, in the resonance region which extends beyond 800 keV for each of them. Use was made of the recently developed covariance formalism based on kernel approximation along with data in the Atlas of Neutron Resonances. The data of most interest for AFCI applications, elastic scattering cross section uncertainties at energies above about few hundred keV, are on the level of about 12% for ⁵²Cr, 7-8% for ⁵⁶Fe and 5-6% for ⁵⁸Ni.

[en] An international effort is underway to design a new structure for storing and using nuclear reaction data, with the goal of eventually replacing the current standard, ENDF-6 (see the formats manual at http://www.nndc.bnl.gov/csewg/docs/endf-manual.pdf). This effort, organized by the Working Party for Evaluation Cooperation, was initiated in 2012 and has resulted in a list of requirements and specifications for how the proposed new structure shall perform. The new structure will take advantage of new developments in computational tools, using a nested hierarchy to store data. The structure can be stored in text form (such as an XML file) for human readability and data sharing, or it can be stored in binary to optimize data access. In this paper, we present the progress towards completing the requirements, specifications and implementation of the new structure. (orig.)

[en] We present the NNDC-BNL methodology for estimating neutron cross section covariances in thermal, resolved resonance, unresolved resonance and fast neutron regions. The three key elements of the methodology are Atlas of Neutron Resonances, nuclear reaction code EMPIRE, and the Bayesian code implementing Kalman filter concept. The covariance data processing, visualization and distribution capabilities are integral components of the NNDC methodology. We illustrate its application on examples including relatively detailed evaluation of covariances for two individual nuclei and massive production of simple covariance estimates for 307 materials. Certain peculiarities regarding evaluation of covariances for resolved resonances and the consistency between resonance parameter uncertainties and thermal cross section uncertainties are also discussed

[en] The nuclear reaction code EMPIRE has been extended to provide evaluation capabilities for neutron cross section covariances in the thermal, resolved resonance, unresolved resonance and fast neutron regions. The Atlas of Neutron Resonances by Mughabghab is used as a primary source of information on uncertainties at low energies. Care is taken to ensure consistency among the resonance parameter uncertainties and those for thermal cross sections. The resulting resonance parameter covariances are formatted in the ENDF-6 File 32. In the fast neutron range our methodology is based on model calculations with the code EMPIRE combined with experimental data through several available approaches. The model-based covariances can be obtained using deterministic (Kalman) or stochastic (Monte Carlo) propagation of model parameter uncertainties. We show that these two procedures yield comparable results. The Kalman filter and/or the generalized least square fitting procedures are employed to incorporate experimental information. We compare the two approaches analyzing results for the major reaction channels on ⁸⁹Y. We also discuss a long-standing issue of unreasonably low uncertainties and link it to the rigidity of the model

[en] The next generation of nuclear data infrastructure tools at the Livermore National Laboratory (LLNL) consists of pipeline of codes that read and process nuclear data from evaluated files saved in the new GNDS (Generalised Nuclear Data Structure) nuclear data format. The processing code FUDGE (For Updating Data and Generating Evaluations) is at the front-end of this pipeline as it reads and process the evaluated data for use in downstream transport codes. FUDGE is Python based with C and C++ extensions for computationally intensive tasks. As is the case for the evaluated data, the processed output is also saved in the GNDS format and the GIDI+ API is provided as the interface between the processed data and the transport codes. GIDI+ is a C++ based suite of codes and it includes GIDI (General Interaction Data Interface), a library for reading and writing GNDS data, and MCGIDI which is the cross section lookup, and reaction and product distribution sampling interface between Monte Carlo transport codes and the GNDS data. GIDI provides methods for easy access to the multi-group processed GNDS data and this is demonstrated through its implementation in ARDRA, the LLNL deterministic transport code. The evaluation and sampling methods in MCGIDI are available as both CPU and GPU methods which facilitates the use of MCGIDI in both traditional CPU-based as well as the next generation mixed model computational architectures. This is demonstrated through the GIDI+ implementation in MERCURY, the LLNL Monte Carlo transport code. (authors)

[en] We calculated ²³Na neutron cross sections for nuclear data assimilation using the nuclear reaction model code EMPIRE and its latest built-in capabilities. Cross sections were based on nuclear model parameters: in the resolved resonance region the cross section was determined using the multi-level Breit-Wigner model, and in the fast neutron region using the optical potential, level densities, and pre-equilibrium emission. An estimate of the variances in the nuclear model parameters was obtained in the low energy range (up to 1 MeV) using the ATLAS of neutron resonances, and in the fast neutron region up to 20 MeV the parameter covariances were estimated by the Bayesian code KALMAN. Overall, more than 100 model parameters were perturbed in order to generate sensitivity coefficients in a multi-group representation. Uncertainties of multi-group cross sections were calculated by combining the covariances of the mentioned model parameters with the sensitivity coefficients. (authors)