The variogram is a graph of semivariance vs. separation distance. Where autocorrelation is present, semivariance is lower at smaller separation distances (autocorrelation is greater). This typically yields a curve such as that described below, which can be modeled using three terms – a nugget variance, a sill, and a range.
Four types of isotropic variogram models are provided in GS+: Linear, Spherical, Exponential, and Gaussian. Each model can be described using the terms above:
• Nugget Variance or Co – the y-intercept of the model;
• Sill or Co+C – the model asymptote; and
• Range or A – the separation distance over which spatial dependence is apparent. In some texts this is called the effective range in order to distinguish range A from a model range parameter A0. In GS+ the Range A is calculated from A0 as noted in the model descriptions.
GS+ calculates default values for each parameter of the four models. You may change any of these three model parameters from the Isotropic Variogram Model window.