A conceptual deterministic analysis of the kriging technique in hydrology

摘要: A stochastic approach to interpolating sparse observation records in hydrology referred as the ‘kriging technique’ has recently been developed by Matheron (1969, 1970) and systematically used field Ecole des Mines de Paris. Distinct primary features of this method are supposed be its predictive reliability ability provide an estimation error. In present paper a deterministic analysis for kriging technique is given without resorting statistical framework. It turns out that close relationship with most traditional interpolation schemes may easily found. The formulation shows relation other commonly methods, any claim produce more accurate estimates fails proven, since arbitrary component necessarily introduced decision process selecting operative model. recognized distinct unique advantage assessment An attempt also made give interpretation main trend. results show use concept involves additional arbitrariness whose impact upon final outcome can hardly predicted priori. conclusions suggest practical applications model not improved trend, unless expression reflects some information related general hydrologic context but included available observations.