Aimed at various kinds of original gaps on zoned heterogeneous spatial data in distributed environment

this paper p roposes a three-relational matrix （ 3RM） model. The model is divided into two phases. The first phase partitions the spatial computed unit using mapping rule or unified code and clusters the features of the zoned heterogeneous spatial data into groups. The s econd phase deals with the gaps in each spatial computed unit. The 3RM model focuses on the description and discrimination methods of the physical gaps on the zoned boundary. Based on the feature topology relation

attributive relation

and precision

the 3RM processing matrices are established. Meanwhile

38 kinds of potential gaps in the zoned boundary of the linear and regional spatial objects are discovered

and respective processing methods are put forward. Finally

the actual distributed and zoned heterogeneous land use data are used in the gap-processing experiment using the 3RM model. The result indicates that the 3RM gap-p rocessing model can deal effectively with various gaps in the actual zoned data and strongly support the seamless spatial analysis in distributed environment.