Geoscience Reference
In-Depth Information
Reaches with one coincident node are further scanned to check if both
endnodes are coincident with the master list. Reaches with both nodes co-
incident are added to the centerline master list and removed from the can-
didate list.
Figure 3
illustrates what this process looks like for a subsection
of the Pomme de Terre subbasin where two reaches fulfill the criteria for
coincidence.
Figure 3:
A portion of Pomme de Terre flowline geometry shown at 1:10,000. The circles
represent nodes in the stream network. Gaps in the discontinuous centerline in the left
panel are filled in the right panel by flagging individual reaches having both nodes coinci-
dent with centerlines.
The next step works with reaches in the candidate list sharing at least one
node with the centerline and one node with each other. Therefore, this step
finds series of two reaches which also connect to nodes bounding remain-
ing gaps in centerlines. These reaches are added to the master centerline
list and removed from the list of candidate reaches.
After completing the second scan, some gaps may still remain in the cen-
terline, where a gap spans more than two reaches (
Figure 4
). The candidate
list is once again scanned for reaches that share at least one node with a
centerline. Reaches that share nodes will occur at the edge of centerline
gaps but also will occur at confluences where minor tributaries connect to
the centerline (see circles in
Figure 4
), creating false candidates. The test
to remove false candidates considers two alternatives. In one case, at least
one reach will not share its second node with any other flowline, that is, it
is a terminal reach (upper red circle in
Figure 4
). These reaches can be
eliminated from consideration. In the other case, one of the reaches will
share a node with another reach which does
not
eventually connect to the
centerline (lower circle). In this case, an estimate of upstream drainage
