The premise of the course is motivated by the recent advancements in geoinformation data acquisition and
storage and their intended use for evidence-based planning and monitoring. The spatial references of geoinformation data may be attributed to the exact locations of measurements or over a fixed set of contiguous
regions or lattices. This course seeks to handle the three main classes of spatial data/processes:
geostatistical data/spatially continuous process, lattice data/discrete process, and point pattern data/point
process. Such data appear common in diverse application fields like environmental science, agriculture,
natural resources, environmental epidemiology, and so on. The aim is to present methods that can be used
to explore and model such data. Naturally, data vary in space and in time; hence data close to each other
(either in space or time) are more similar than those farther. Geostatistical modeling based on the
semivariance and/or covariances and interpolation (kriging) in space and time will therefore be introduced.
The methods will be extended and applied to data aggregated over contagious regions. The uncertainty is
quantified, and attention will be given to making maps showing the probabilities that thresholds are
exceeded. Attention is also given to optimal sampling and monitoring. Further, data that arise out of the
occurrences of events; thus point pattern data will be considered. The significance of exploring the first
and second-order properties of point patterns in diverse application domains like environmental and
disaster (like earthquakes) modeling will be explained and applied. The last focus will be on lattice data; in
principle, this kind of data consists of observed values over a set of fixed contiguous regions. This kind of
data is rather easy to acquire and is mostly applied in health studies where data aggregation is a standard
form of protecting locational privacy.