Present day fluvial systems display well established scaling relationships between channel width to depth, channel belt width to depth and point bar width to channel depth. These relationships are important as they provide a potential mechanism for predicting channel body dimensions from one-dimensional subsurface datasets such as wireline logs or core. Scaling relationships can also aid in reducing uncertainty associated with estimating fluvial channel body width in the rock record, which, for a given thickness can vary by between 2 and 5 orders of magnitude.
Linking present day channel and channel belt dimensions to preserved channel belt sandstone bodies is difficult, as preserved channel belt stacking patterns are not easily observed in modern day basins. Stacking patterns are related to spatial location in basins and the type of fluvial depositional systems (distributive, axial or valley). I will show how dimensional data can vary depending on the type of different fluvial depositional systems. Whilst dimensional data from modern day channels and channel belts are not directly transferable to the preserved rock record, this approach does provide generic relationships that could potentially form the basis for more accurately predicting sandstone body dimensions for subsurface fluid extraction and/or storage.