Dynamic Spatial Sampling

This paper pretends to give new tools for dynamic spatial sampling designs to find the optimal estimation and the optimal spatial prediction, based on the variation of spatial dependence structure in both cases, discrete and continuous time. In order to model the time series of the spatial covariance parameters, the measurement error and the bias caused by the estimation are included in the formulation of state space models. A discussion of useful properties and techniques to estimation and forecasts in several scenarios is presented. The methodology is applied to a network of quality air in the Bogotá city.