Figure 1
The basic building block of the EPA River Reach File is the REACH, which is a distinctly identified lineal element A reach is defined as either: 1) the lineal hydrographic feature between two stream confluences, or 2) a unique hydrographic feature as denoted by the source data (USGS DLG). The distinction between these two definitions is that the first reach type is normally connected hydrologically with other reaches, while the second reach type is not usually connected.
RF3 assigns a unique reach code to each stream arc in DLG hydrography and determines the upsteam/downstream relationships of each reach. The structure of the reach id code (RF3RCHID) is analogous to a postal address, where the Cataloging Unit (CU) is like a ZIP code, or area code, the segment (SEG) is like a street within that zip code, and finally, a Marker Index (MI) is like a house number along the street, with numbers increasing in the upstream direction. For example, the RF3RCHID (in CU-SEG-MI sequence) for the reach at the mouth of the hypothetical watercouse in the figure is 18010106 1 0.00. The 0.00 would indicates that this reach is the downstream-most reach in the series of arcs composing the SEG. The upstream-most reach has the RF3RCHID of 18010106 1 10.06. The database record for a given reach contains the RF3RCHID of that reach, as well as the RF3RCHIDs of other reaches (in separate fields) that the given reach is connected to, i.e. the reaches downstream and upstream of the given reach. This design makes it possible to logically connect reaches together to model hydrologic transport networks. The strength of this approach is the simplicity of fixed fields and records in a table, however there is no guarantee of accurate relationships between RF3 logic and ArcInfo topology.