Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Estimating the spatial pattern of radionuclides in the environment using kriging

Conference ·
OSTI ID:5749666
The purpose of this paper is two-fold: (1) to increase awareness within the statistical community of kriging methods for estimating spatial patterns and amounts of environmental pollutants and other phenomenon, and (2) to illustrate using a large radionuclide data set some of the estimation biases and other difficulties that can arise when using kriging techniques. Kriging techniques are relatively unknown to statisticians in the United States except for those involved in geology, mining or environmental studies. The method has potential for application in some environmental pollution studies, but improved statistical methods for estimating the underlying spatial correlation structure and in quantifying the uncertainty in estimates of spatial pollution concentration contours are needed. Design questions should also be addressed. Kriging is applied to the radionuclide data set to obtain estimates of concentration isopleths (contours of constant concentration), confidence bands about those contours, average concentrations over 50' by 50' block areas, and the inventory (total amount) of /sup 241/Am in surface soil. The effect on these estimates of a nonstationary semi-variogram (a function giving the correlation structure) over the study site is studied by dividing the site into more homogeneous zones and using a separate semi-variogram model for each zone. 37 references.
Research Organization:
Pacific Northwest Lab., Richland, WA (USA)
DOE Contract Number:
AC06-76RL01830
OSTI ID:
5749666
Report Number(s):
PNL-SA-11221; CONF-830899-1; ON: DE83016741
Country of Publication:
United States
Language:
English