Scale Is Introduced in Spatial Datasets by Observation Processes
Andrew U. Frank · 2009
An ontological investigation of data quality reveals that the quality of the data must be the result of the observation processes and the imperfections of these. Real observation processes are always imperfect. The imperfections are caused by (1) random perturbations, and (2) the physical size of the sensor. Random effects are well-known and typically included in data quality descriptions. The effects of the physical size of the sensor limit the detail observable and introduce a scale to the observations. The traditional description of maps by scale took such scale effects into account, and must be carried forward to the data quality description of modern digital geographic data. If a sensor system is well-balanced, the random perturbations, size of the sensor and optical blur (if present) are of the same order of magnitude and a summary of data quality as a ‘scale’ of a digital data set is therefore theoretically justifiable.