Geostatistical methods were used to find efficient and accurate means for salinity assessment using regionalized random variables and limited sampling. The random variables selected, sodium absorption ratio (SAR), electrical conductivity (EC), and clay content were measured on samples taken over an area of fifteen square miles. Ordinary kriging and co-kriging were used as linear estimators. They were compared on the basis of average kriging variance and sum of squares for error between observed and estimated values. The results indicate a significant improvement in the average kriging variance and sum of squares by using co-kriging estimators. EC was used to estimate SAR because of the high correlation between them. This was not true for clay content. A saving of two-thirds of the cost and time was achieved by using electrical conductivity as an auxiliary variable to estimate sodium absorption ratio. The nonlinear estimator, disjunctive kriging, was an improvement over co-kriging in terms of the variances. More information at the estimation site is a more important consideration than when the estimator is linear. Disjunctive kriging was used to produce an estimate of the conditional probability that the value at an unsampled location is greater than an arbitrary cutoff level. This feature of disjunctive kriging aids salinity assessment and reclamation management. A solute transport model was used to show how saptially variable initial conditions influenced the amount of water required to reclaim a saline soil at each sampling point in a simulated leaching of the area. / Graduation date: 1992