An Optimal Spatial Sampling Design for Intra-Urban Population Exposure Assessment

Area of Science:

• Environmental Science
• Spatial Analysis
• Geographic Information Systems

Background:

• Previous spatial sampling designs for air pollution exposure monitoring have limitations.
• The design by Kanaroglou et al. (2005) for NO(2) monitoring in Toronto failed to capture spatial variability effectively due to issues with demand surface and location-allocation models.
• Spatially autocorrelated data and population weighting can lead to redundant sample site selection.

Purpose of the Study:

• To propose an optimal spatial sampling design that maximizes captured variance while minimizing sample size.
• To address the weaknesses identified in previous spatial sampling methodologies for air pollution monitoring.
• To develop and implement a novel sampling design for identifying optimal sites for monitoring airborne particulates in diverse urban environments.

Main Methods:

• Development of a computer application using a spatial search algorithm (in C++) for implementing the proposed sampling design.
• Utilizing a novel approach to spatial sampling that accounts for spatial autocorrelation and aims for maximum variance with minimum sample size.
• Implementation and testing of the design in three distinct urban settings: Cleveland, OH; Delhi, India; and Iowa City, IA.

Main Results:

• The proposed spatial sampling design effectively captures maximum variance with a reduced sample size.
• The developed computer application successfully implemented the sampling design in diverse urban environments.
• The study identified optimal sample sites for monitoring airborne particulates in Cleveland, Delhi, and Iowa City.

Conclusions:

• The proposed optimal spatial sampling design offers a more effective approach to monitoring air pollution compared to previous methods.
• The developed methodology and application provide a valuable tool for environmental scientists and urban planners.
• This design is adaptable for identifying optimal monitoring sites for various environmental parameters in different urban contexts.