This paper focuses on the typical inherent variabilities for the basic properties of natural geomaterials. The issue of measurement error, which is closely related, is noted when necessary. The results of an extensive literature review on inherent variability are presented. This variability is modeled as a random field, which can be described concisely by the coefficient of variation (COV) and the scale of fluctuation. The COV of inherent variability for various test measurements is given in histograms to illustrate its population distribution. The means and standard deviations of the COV are shown with the total number of data groups, and the shapes of the histograms are discussed. The scale of fluctuation is discussed elsewhere.
The presence of uncertainties and their significance in natural geomaterials (soils and rocks) has long been appreciated, at least in a qualitative manner. However, in the overwhelming majority of geotechnical texts and courses, after making the observation that uncertainty is an important factor, it is then relegated to a minor position, and the remainder of the text/course is evaluated in a traditional deterministic fashion. The reasons for this unfortunate state of affairs range from the point that it is "easier" to do deterministic evaluations to the fact that it is somewhat difficult to embrace uncertainty easily and directly because it is complex in geotechnical problems. After all, uncertainties arise in the loads, geologic site interpretations, geotechnical properties, computation models, etc. Quantifying all of these components is a monumental task. To help mitigate this problem, this paper focuses on the typical inherent variability in some test measurements of natural geomaterials. This variability commonly is modeled as a random field, which can be described concisely by the COV (= standard deviation / mean) and the scale of fluctuation (Vanmarcke, 1977).