Abstract
The construction of spatial databases often requires considerable computing and storage resources, due to the inherent complexity of spatial data and their manipulation. Thus, it would be desirable to devise methods enabling a designer to estimate performances of a spatial database since from its early design stages. We present a method for estimating both the size of data and the cost of operations based on the conceptual schema of the spatial database. We also show the application of the method to the design of a spatial database concerning botanic data.