Quality considerations for optimal positioning when integrating spatial data
As spatial data producers are entering an era of data maintenance, new problems are emerging with respect to data quality. The availability of high accuracy positioning technologies, such as Global Navigation Satellite Systems GNSS, has facilitated the rapid collection of new data. This data is typically of higher quality than the legacy database into which it is to be integrated and, as such, will often not fit. The question has arisen as to how this new data can be best integrated into an existing dataset and also, as to what is the accuracy of the resulting data. This paper considers the positional accuracy and topological consistency aspects of spatial data quality as new, higher accuracy data is integrated into an existing dataset. A methodology is developed that provides optimal positioning solutions by resolving the best fit between the new data and the legacy database, whilst preserving spatial relationships that exist among features. The method uses positional information, together with its associated accuracy, in combination with geometric and topological constraints in a rigorous process based on least squares. In addition, quality information at the point level is provided, enabling the spatial variation in positional accuracy of the resultant dataset to be portrayed. The developed method is applied in a case study to upgrade a subset of the Victorian cadastral database using data from an extensive survey in the area.