Inter-comparison of Radiation Measurements
Inter-comparison of Radiation Measurements
HO Lok-ping, LEUNG Ho-ming
July 2023
The results of radiation measurements are often affected by different factors and there exists uncertainties in the results. How could organisations conducting radiation measurements ensure the reliability of their measurement results?
To ensure the reliability of the environmental radiation monitoring results, the Hong Kong Observatory not only conducts regular instrument calibration and goes through internal quality assurance procedures, but also participates in radiation measurement inter-comparison exercises and proficiency tests.
Since 1989, the Observatory has been participating in inter-laboratory comparison exercises and proficiency tests organised by international and national organisations. So far, the Observatory has participated in those organised respectively by the International Atomic Energy Agency (IAEA), the World Health Organization (WHO), the National Physical Laboratory of the United Kingdom (NPL), the China Institution for Radiation Protection (CIRP), the China Institute of Atomic Energy (CIAE), the Shanghai Radiation Environmental Supervision Station (ShRESS) and the Government Laboratory (GL) of Hong Kong.
Laboratory Proficiency Test
In July 2022, the Observatory participated in the laboratory proficiency test organised by the IAEA, measuring the activities of radionuclides including cobalt-60, cesium-134, cesium-137, tritium and strontium-90 in water samples. Since the characteristics of each radionuclide are different, the Observatory needed to use a variety of instruments for conducting the measurements.
As the first step, the Observatory measured the gamma ray energy spectrum of the samples using the High Purity Germanium Detector of the Gamma Spectrometry System (Figure 1). Through the analysis of the spectral characteristics, the gamma emitting radionuclides in the sample, including cobalt-60, cesium-134 and cesium-137, and their corresponding activities were identified. After finishing the gamma spectrometry analysis, chemical treatments of the samples were carried out to extract the radionuclides tritium and strontium-90 respectively. Tritium is a low-energy beta emitter and the Liquid Scintillation Counting System (Figure 2) was required for the measurement. Meanwhile, the Low Level Alpha-beta Counting System (Figure 3) was used to measure the activity of the pure beta-emitting radionuclide, strontium-90.
The decay of radionuclides is a random process, and a certain degree of uncertainty is bound to exist in the measurement of the radiation level of a sample. According to the requirements of the IAEA, the accuracy and precision of the measurement results would be evaluated for each participating organisation. The measurement of cesium-137 in one of the water samples of the laboratory proficiency test is taken as an illustration.
Radionuclide | Cs-137 |
Measurement Result of HKO (Bq/kg) | 24.2 |
Measurement Uncertainty of HKO (Bq/kg) | 0.9 |
IAEA Target Value (Bq/kg) | 24.2 |
Uncertainty of IAEA Target Value (Bq/kg) | 1.5 |
Relative Bias | 0% |
Combined Standard Measurement Uncertainty (P) | 7.23% |
Maximum Acceptable Relative Bias provided by IAEA | 20% |
If the Relative Bias is not greater than the Maximum Acceptable Relative Bias provided by IAEA, the accuracy of the measurement results is considered "acceptable".
In this measurement item, Relative Bias = (Measurement Result of HKO – IAEA Target Value) / IAEA Target Value x 100% = 0%, and the Maximum Acceptable Relative Bias provided by IAEA is 20%.
If the combined standard uncertainty of the measurement P ≤Maximum Acceptable Relative Bias provided by IAEA and Relative Bias ≤ k*P , the precision of the measurement result is considered "acceptable". k is the coverage factor, and k is 2.58 at the 99% confidence level. The combined standard uncertainty of the measurement can be calculated from the measurement uncertainties of the IAEA and the Observatory as follows:
In this measurement item, the combined standard uncertainty P is 7.23%, which meets the requirement of the IAEA for measurement precision.
Therefore, the measurement result of the Observatory successfully passed the accuracy and precision assessments. There is no difference between the measurement result of the Observatory and the IAEA target value, so the Observatory ranked first among more than 200 participating organisations in this measurement item. According to the evaluation report released by the IAEA, all the measurement results of the Observatory are within the acceptable range, which shows that the measurement capability of the Observatory's radiation laboratory complies with international standard.
Outdoor Measurement of Environmental Gamma Radiation Dose Rate Inter-comparison Exercise
In addition to laboratory proficiency tests, the Observatory also participated in the outdoor measurement of environmental gamma radiation dose rate inter-comparison exercise organised by the CIAE at the King’s Park Meteorological Station in Kowloon in April 2023. Other participating organisations included scientific research and government institutes from Mainland China. Because different natural radionuclides are present in rocks and soils, there will be natural variation in their radiation levels. To ensure the comparability of different radiation monitoring instruments under the same measurement method and environment, several measurement points including meadow, concrete floor and asphalt pavement were selected in advance. Each participating organisation would take turns to measure the ambient gamma radiation dose rate using their radiation monitoring instruments at each measurement point in order to compare the performance of instruments under different environments (Figure 4 and 5). In this inter-comparison exercise, the median (M) and the normalized inter-quartile range (NIQR) of all participants’ measurement results were calculated for each measurement point, then the Z-score (Z) of each participating organisation would be calculated according to the following equation:
where X is the measurement results of a particular participating organisation. Absolute value of Z-score less than 2 means that the measurement result matches with those measured by other organisations, absolute value of Z-score between 2 to 3 means that the measurement results deviate from those measured by other organisations, and absolute value of Z-score larger than 3 means that the measurement results are outlier compared to those measured by other organisations. In this inter-comparison exercise, the absolute value of Z-score of all participating organisations were less than 2, meaning that the measurement results of all organisations agreed with each other. Besides, experts from the participating organisations also conducted technical exchanges and shared their experience in radiation measurement during the inter-comparison exercise.
The results of inter-comparison exercises and proficiency tests participated by the Observatory are published in the annual "Summary of Environmental Radiation Monitoring in Hong Kong".