The original mapping information that constitutes the DTDB was initially created from 1970 at 1:25 000, 1:50 000 and 1:100 000 scales from stereoscopic aerial photography supported by ground truth, field data capture and updating. These scales have an inherent accuracy as per the National Mapping Council of Australia, Standards of Map Accuracy (1975). Therefore, the position of the feature instance will be within 0.5mm at map scale for 90% of the well-defined points. That is, 1:25000 = 12.5 m, 1:50000 = 25 m and 1:100000 = 50 m. The original manuscripts were scanned and vectorised and translated to Genamap files. Beginning in 2004, the data was migrated to an ESRI geodatabase. Since then the database has been continuously updated and upgraded through a program of external data integration from custodial and other sources, data matching and data verification of other internal and external databases, field investigation and research, as well as through Land and Property Information (LPI)’s regular cycle of updating aerial imagery of 10 cm, 50 cm and 2.5 m GSD. The programs address the content correctness, currency and accuracy of different Feature Classes on a continuous or cyclical basis. The specific programs utilised for the maintenance of each feature are largely defined by the Capture Source Attribute attached to the feature. Data is updated via multiple source programs and captured using various methods; Registered Image Trace, Ortho Image Trace of 10 , 20 and 50 cm ground sample distance imagery from ADS and 2.5 m from SPOT 5 imagery , Table Digitisation , DCDB , Scan / Vectorise , Differential GPS , Non-corrected , Comparative estimate , Survey Controlled , Mobile GPS , Stereo Image Trace Aggregation or replication of existing line segments , Engineering Design File (CAD) Scanned Map Image Trace , Positioned within features cadastral or ortho image extent. Some capture methods are unknown which may occur where data was supplied by an external source where the data capture method is undefined. Depending on the capture source, capture method, digital update and control point upgrade, every feature instance reportedly has a positional accuracy within the range of 1 m-100 m. It is important to note that not all features may be captured within the database for each feature class, and may not be an accurate representation of the current state.