Implementing remote sensing strategies to support environmental compliance assessment: A neural network application

Environmental compliance assessment and enforcement are complex, labor intensive tasks that would benefit from some form of decision support. Despite the well-documented application of remote sensing in environmental analysis and assessment, the operational use of this technology for environmental compliance enforcement remains in its infancy. Because the majority of environmental laws and regulations in the United States were enacted prior to the emergence of a mature science of remote sensing, there exists a degree of skepticism regarding the operational utility and reliability of remotely sensed data as an environmental compliance tool. This paper provides a review of the role of remote sensing in environmental compliance assessment and introduces a framework to guide the application of this technology based on the use of artificial neural networks. The customized neural networks developed in this study serve as a simplifying strategy to assist the process of compliance monitoring. Functioning as “intelligent” classifiers, the network models provide a spatial characterization of activities likely to cause damage; achieving mapping accuracies that ranged from 68 to 90% for the test case samples selected for investigation.