• Title/Summary/Keyword: groundwater comtaminant

Search Result 2, Processing Time 0.015 seconds

SOME ASPECTS IN PRACTICAL ESTIMATION OF GROUNDWATER CONTAMINANT CONCENTRATIONS

  • Cho, Choon-Kyung;Ha, Bang-Soo;Kang, Sung-Kwon
    • Journal of applied mathematics & informatics
    • /
    • v.8 no.2
    • /
    • pp.417-428
    • /
    • 2001
  • Spatial distribution of groundwater contaminant concentration has special characteristics such as approximate symmetric profile, for example, in the transversal direction to groundwater flow direction, a certain ratio in directional propagation distances, etc. To obtain a geophysically appropriate semivariogram which is a key factor in estimation of groundwater contaminant concentration at desired locations, these special characteristics should be considered. Specifically, the concepts of symmetry and ratio are considered in this paper. By applying these two concepts, significant improvement of semivariograms, estimation variances, and final estimation results compared with the ones by conventional approaches which usually do not account for symmetry and ratio are shown using field experimental data.

STATISTICAL VALIDATION OF SYMMETRY IN ESTIMATION OF GROUNDWATER CONTAMINANT CONCENTRATIONS

  • Cho, Choon-Kyung;Sungkwon Kang
    • Journal of applied mathematics & informatics
    • /
    • v.13 no.1_2
    • /
    • pp.335-351
    • /
    • 2003
  • Spatial distribution of groundwater contaminant concentration has special characteristics such as approximate symmetric profile, for example, in the transversal direction to groundwater flow direction, a certain ratio in directional propagation distances, etc. To obtain a geophysically appropriate semivariogram which is a key factor in estimation of groundwater contaminant concentration at desired locations, these special characteristics should be considered. In this paper, a method for finding appropriate symmetric axes is introduced. Statistical analyses for the choices of symmetric axes and mathematical models for semivariograrns are performed. After implementing symmetry, the corresponding semivariograrns, kriging variances, and final estimated results show significant improvement compared with those obtained by conventional approaches which usually do not account for symmetry.