Shannon's entropy concept in optimum air monitoring network design

Abstract

The selection of an optimum number of stations and their optimum locations in a region are the main objectives in air monitoring network design. There are a number of approaches to the achievements of these objectives. Among these, the criterion based on the estimation performance measurements, and the identification of adversely affected subregions are the most commonly used approaches. This article introduces a new concept of air monitoring network design using Shannon's entropy concept. The multivariate discrete entropy concept is proposed with the objectives of optimum estimation performance measurements. However, this concept does not guarantee the location of stations in adversely affected subregions. A generalized entropy concept using multivariate lognormal distribution is developed to consider both criteria in optimum air monitoring network design. The methodology is then applied to design an optimally air monitoring network scheme for Al-Khobar city in Saudi Arabia.