Environmental optimization: Applications of genetic algorithms

The genetic algorithm (GA) has found wide acceptance in many fields, ranging from economics through engineering. In the environmental sciences, some disciplines are using GAs regularly as a tool to solve typical problems; while in other areas, they have hardly been assessed for use in research projects. The key to using GAs in environmental sciences is to pose the problem as one in optimization. Many problems are quite naturally optimization problems, such as the many uses of inverse models in the environmental sciences. Other problems can be manipulated into optimization form by careful definition of the cost function, so that even nonlinear differential equations can be approached using GAs (Karr et al. 2001; Haupt 2006). Although optimization is usually accomplished with more traditional techniques, using a genetic algorithm allows cost functions that are not necessarily differentiable or continuous (Marzban and Haupt 2005). Chapter 5 of this volume provides an introduction to genetic algorithms and some basic examples that demonstrate their use. Chapter 14 describes a specific problem in optimization in which a GA proved useful-using field monitored contaminant concentration data coupled with a transport and dispersion model to backcalculate source and meteorological information. The purpose of this current chapter is to review some of the broad range of applications of genetic algorithms to environmental science problems, to present a few examples of how a GA might be applied to some problems, and to suggest how they may be useful in future research directions.