Temporal evolution of the performance evaluation of the laboratories at Spanish nuclear power plants in water samples

TL;DR

This study analyzed the performance of Spanish nuclear power plant labs over 9 years, focusing on radionuclide testing accuracy and trends.

Contribution

The study provides a detailed temporal analysis of lab performance for radionuclide detection in nuclear power plant water samples.

Findings

Performance in determining tritium (3H) was mostly satisfactory, except in 2010 and 2014.

Gamma emitters showed consistently high satisfactory evaluation rates above 93%.

Linear regression analysis revealed higher dispersion for 3H and 14C compared to other radionuclides.

Abstract

This study examined the performance evaluation and analysis of the laboratories of Spanish Nuclear Power Plants over the last 9 years of the 34 editions organized by CIEMAT. The participating nuclear power plants were Almaraz, Ascó I, Ascó II, Cofrentes, Santa María de Garoña, Trillo, and Vandellós II. The radionuclides analyzed were those necessary to ensure operational radiological protection and the control of liquid effluents from a nuclear power plant. The radionuclides are classified into beta emitters: 3H, 14C, and 90Sr, and gamma emitters: 241Am, 57Co, 137Cs, and 60Co. The results of the z score showed that the performance in determining 3H was 100% satisfactory in most years, except in 2010 and 2014, where percentages of 83% and 87.5% were recorded, respectively. 90Sr showed greater variability with a satisfaction range between 60 and 100%. Furthermore, the gamma emitters achieved satisfactory evaluations above 93% within the studied time range. The analysis of the distributions of the relative differences between the laboratory results and the reference values confirmed that the distributions were normal according to the Shapiro–Wilk test, with a percentage of outlier values below 5%. Additionally, linear regression analysis was used to evaluate the dispersion of the results relative to the reference values through the RSE parameter, being highest for 3H and 14C. The PomPlot graphs allowed for the visualization of the relationship between the obtained results and the reference values, indicating that, although high percentages of satisfactory evaluations were achieved, certain trends in the laboratory results must be considered to detect potential future deviations in the analytical methodologies.

The online version contains supplementary material available at 10.1007/s10661-025-13851-8.

Figures