ABSTRACT
Increasingly in spatial data settings there is need for analyzing multivariate measurements obtained at spatial locations. Such data settings arise when several spatially dependent response variables are recorded at each spatial location. A primary example is data taken at environmental monitoring stations where measurements on levels of several pollutants (e.g., ozone, PM2.5, nitric oxide, carbon monoxide, etc.) would typically be measured. In atmospheric modeling, at a given site we may observe surface temperature, precipitation, and wind speed. In a study of ground level effects of nuclear explosives, soil and vegetation contamination in the form of plutonium and americium concentrations at sites have been collected. In examining commercial real estate markets, for an individual property at a given location, data include both selling price and total rental income. In forestry, investigators seek to produce spatially explicit predictions of multiple forest attributes using a multisource forest inventory approach. In each of these illustrations, we anticipate both dependence between measurements at a particular location, and association between measurements across locations.
