STOCHASTIC-ANALYSIS OF SPATIOTEMPORAL SOLUTE CONTENT MEASUREMENTS USING A REGRESSION-MODEL

A regression model is used to study spatiotemporal distributions of so lute content ion concentration data (calcium, chloride and nitrate), w hich provide important water contamination indicators. The model consi sts of three random and one deterministic components. The random space /time component is assumed to be homogeneous/stationary and to have a separable covariance. The purely spatial and the purely temporal rando m components are assumed to have homogenous and stationary increments, respectively. The deterministic component represents the space/time m ean function. Inferences of the random components involve maximum like lihood and semi-parametric methods under some restrictions on the data configuration. Computational advantages and modelling limitations of the assumptions underlying the regression model are discussed. The reg ression model leads to simplifications in the space/time kriging and c okriging systems used to obtain space/time estimates at unobservable l ocations/instants. The application of the regression model in the stud y of the solute content ions was done at a global scale that covers th e entire region of interest. The variability analysis focuses on the c alculation of the spatial direct and cross-variograms and the evaluati on of correlations between the three solute content ions. The space/ti me kriging system is developed in terms of the space direct and cross- variograms, and allows the separate estimation of the regression model components. Maps of these components are then obtained for each one o f the three ions. Using the estimates of the purely spatial component, spatial dependencies between the ions are studied. Physical causes an d consequences of the space/time variability are discussed, and compar isons are made with previous analyses of the solute content dataset.