| This research is focused on the study of network design that uses a mix of exploratory techniques, analytical tools and human judgment to determine potential particulate matter monitoring sites. The design problem is to determine the mix of locations, time periods, sample durations, sample frequencies and variables measured that achieve the objectives within available technical and resource constraints. A comprehensive network design protocol is presented that follows the philosophy that a judgmental sampling strategy approach the most practical but not necessarily the most optimal from a statistical viewpoint. The design strategy advocated here is based on an approach that involves a combination of expert human judgment, exploratory statistical methods and physical and statistical models.; The role of expert judgment remains one of the most critical parts of the decision making process. Sound decisions are proportional to the amount of data that is available for the design. The results from this study show that several analytical and graphical techniques have the potential to be integrated into routine network design decisions. One of the most beneficial tools for network design involves the use spatial data analysis and a geographic information system (GIS). GIS have matured from being an obscure, hard to use and awkward software to being a commonplace and easy to use tool.; While this research is specifically applied to particulate matter networks in an arid region dominated by fugitive dust sources, the principles can be transferred to other regions and pollutants. |
