Estuaries are among the earth's most valuable and productive environmental resources. To further our understanding of the impact of human activities on estuaries, there is a need for appropriate statistical methods for analyzing estuarine data. Estuaries possess a number of features that must be considered during spatial data analyses. Estuaries are irregularly shaped non-convex regions. Therefore, Euclidean distance may not be an appropriate distance metric for spatial analyses of estuaries, especially if the line segment connecting two sites intercepts land. Furthermore, some environmental variables may take deterministic values at estuarine boundaries. For example, shorelines are saturated with dissolved oxygen, and the salinity at estuarine mouths should be close to that of the ocean. This paper considers methods for spatial modelling and prediction using dierent distance metrics, and under ®xed boundary conditions. These methods are illustrated using data from Charleston Harbor, an estuary on the coast of South Carolina, USA # 1998 John Wiley & Sons, Ltd. KEY WORDS distance metric; boundary conditions; heteroscedasticity
1. Introduction
Estuaries are among the earth's most valuable and productive environmental resources. They provide critical feeding, spawning, and nursery habitats for many ®sh, shell®sh, bird, and mammal species. However, runo from development along estuarine shorelines, as well as pollution of rivers feeding estuaries, threaten the integrity of this important resource. To understand the impact of these human activities, investigations of the estuarine environment are required. Hence, there is a need for appropriate statistical methods for analyzing the resulting data.
Estuaries are characterized by environmental gradients, the most important of which is a gradient of decreasing salinity with increasing distance to the estuarine mouth. Dierent species of aquatic organisms tolerate dierent ranges of salinity. Consequently, species composition and diversity will also vary with distance to the estuarine mouth. Additional environmental variables, such as dissolved oxygen and contaminant concentrations, are also expected to show patterns of spatial variation across an estuary. The quanti®cation of such patterns of spatial variation