Multivariate Modeling for Spatio-Temporal Radon Flux Predictions

Area of Science:

• Environmental Science
• Geostatistics
• Environmental Monitoring

Background:

• Radon flux data are crucial for monitoring greenhouse gas emissions and assessing indoor radon exposure.
• Radon flux is influenced by uranium/radon deposits and atmospheric variables like humidity, temperature, and precipitation.

Purpose of the Study:

• To model and predict the spatio-temporal distribution of radon flux densities in the Veneto Region, Italy.
• To estimate radon flux at unsampled locations and times, aiding environmental risk assessment.

Main Methods:

• Utilized multivariate geostatistics and the spatio-temporal linear coregionalization model.
• Employed joint diagonalization of empirical covariance matrices at various spatio-temporal lags.
• Generated predicted radon flux maps and probability maps for future risk assessment.

Main Results:

• Successfully modeled the spatio-temporal distribution of radon flux densities.
• Produced monthly predicted radon flux maps and probability maps indicating elevated summer risks.
• Provided a comparison with alternative univariate and multivariate models.

Conclusions:

• Multivariate geostatistics effectively models complex spatio-temporal radon flux data.
• The developed models and maps are valuable tools for environmental monitoring and public health risk assessment.
• Identified higher radon exhalation risks during summer months requiring targeted attention.