Optimization of mobile radioactivity monitoring networks

  • Authors:

  • G. B. M. Heuvelink;Z. Jiang;S. De Bruin;C. J. W. Twenhofel

  • Affiliations:

  • Environmental Sciences Group, Wageningen University and Research Centre, Wageningen, the Netherlands;Environmental Sciences Group, Wageningen University and Research Centre, Wageningen, the Netherlands;Environmental Sciences Group, Wageningen University and Research Centre, Wageningen, the Netherlands;National Institute for Public Health and the Environment, Bilthoven, the Netherlands

  • Venue:
  • International Journal of Geographical Information Science
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

In case of a nuclear accident, decision makers rely on high-resolution and accurate information about the spatial distribution of radioactive contamination surrounding the accident site. However, the static nuclear monitoring networks of many European countries are generally too coarse to provide the desired level of spatial accuracy. In the Netherlands, authorities are considering a strategy in which measurement density is increased during an emergency using complementary mobile measuring devices. This raises the question, where should these mobile devices be placed? This article proposes a geostatistical methodology to optimize the allocation of mobile measurement devices, such that the expected weighted sum of false-positive and false-negative areas (i.e. false classification into safe and unsafe zones) is minimized. Radioactivity concentration is modelled as the sum of a deterministic trend and a zero-mean spatially correlated stochastic residual. The trend is defined as the outcome of a physical atmospheric dispersion model, NPK-PUFF. The residual is characterized by a semivariogram of differences between the outputs of various NPK-PUFF model runs, designed to reflect the effect of uncertainty in NPK-PUFF meteorological inputs (e.g. wind speed, wind direction). Spatial simulated annealing is used to obtain the optimal monitoring design, in which accessibility of sampling sites (e.g. distance to roads) is also considered. Although the methodology is computationally demanding, results are promising and the computational load may be considerably reduced to compute optimal mobile monitoring designs in nearly real time.