Contribution of Spatial Data Analysis in the Evaluation of the Reliability of Radiometric Measurements at the Luiswishi-Likuni Mining Site in the Province of Haut-Katanga (Democratic Republic of Congo)

This study concerns the quality control of radiometric data measured in the Luiswishi-Likuni mining site. Before undertaking an assessment of the radiometric contamination of a site with the aim of providing decision support tools for its rehabilitation, it is necessary to ensure the quality of the data acquired. Thus, based on solid and reliable foundations, uncertainties during the different stages of the decision-making process will be better controlled and the final quality of the project improved. To do this, an acquisition of radiometric data using a Geiger counter allowed us to carry out 13 radiometric profiles with the aim of mapping the intensity of radioactive radiation in this area. Data processing consisted of carrying out a geostatistical analysis (estimation error, histogram, data dispersion, etc.) and improving the signal/noise ratio by frequency analysis ‘Fast Fourier Transform (FFT)’ of the signal. Kriging made it possible to calculate and categorize the estimation error into 3 precision zones: High, Medium and low depending on the spatial distribution of the measurements. The histogram and the QQ-plot diagram of the data confirm that they obey the normal law because we observe values aligned along the line with minimal dispersion at both ends of the line. The FFT low-pass filter, with a window of 3 and a cutoff frequency of 0.166 Hz, produced a smoothing well suited to the radiometric data which appeared noisy due to the fact that they were acquired near the city of Lubumbashi. This significantly increased the signal-to-noise ratio, making the data usable for any future modeling.