Quantifying the chemical variability of a Precambrian diabase from south Jordan using stochastic techniques: a proposal

A late Precambrian diabase from south Jordan is used as an introductory case study to examine the possibility of stochastically quantifying the chemical variability of given volcanic rock constituents based on their total alkali (TA) and silica (S) contents. The primary aim has been to quantify the chemical behavior of the rock constituents studied in order to evaluate whether their sample constituents are chemically interdependent, i.e., if knowing one tells us anything about the other, and if so over what range this occurred. The application of the autocorrelogram analysis revealed that most of the rock constituents exhibit specific chemical dependence between 0.40 and 0.80 wt.% along the silica direction and between 0.10 and 0.30 wt.% along the total alkali direction. The deduced autocorrelation functions were found to be correlated over certain chemical ranges with patterns of three basic types: typical; having a large zone of influence; and cyclic with nested structures. The application of semivariogram analyses, on the other hand, indicates that the rock constituents are chemically interdependent over a larger scale, that their interdependence is greater than that encountered when applying autocorrelation techniques (2–5% for major oxides and from 0.87 to 5.30% for Sr and Ni), and that many variables exhibit inherited random variability. The determined ranges of chemical dependence could be used to characterize differentiation trends which prevailed during rock formation, and to develop more precise predictive models regarding petrogenesis.