search

opengeospatial / HY_Features

http://opengeospatial.github.io/HY_Features/
Other
8 stars 1 forks source link

non-dendritic network support note #199

Closed dblodgett-usgs closed 7 years ago

dblodgett-usgs commented 7 years ago

Note that non-dendritic networks are supported by HY_Features prior to talking about dendritic simplification.

Emphasis on dendritic structures. Modeling catchments and flowpaths as dendritic structures leads to some simplifications that can be exploited. However, it can also increase complexity - some jurisdictions implementing this model force secondary streams or braids to be disconnected from upstream waterbodies (which you might say is not necessary, but this is what they do nonetheless). Figure 12 and the related discussion describe how aggregating reach catchments can lead to dendritic catchments: “Using this encapsulation approach, catchments can be represented using a simple tree structure with clear upstream-downstream relations.” (page 26) This is certainly so, but there are cases where the removal of cycles is detrimental to a hydrologic understanding of the area in question. Montreal is bounded by rivers, as is Manhattan. A section of Winnipeg is particularly interesting in that a major river is on one side and a controlled channel (the Red River Floodway) on the other that may or may not contain water. Some larger lakes exist with multiple outlets; these are sometimes referred to as bifurcation lakes (with the assumption of specifically two outlets). The same kinds of statements can be made about distributaries on deltas. For example, consider the city of Richmond, British Columbia, which is bounded by channels of the Fraser River. Approaches to network modeling that do not need to leverage dendritic behaviour may be worth noting. Please see the comment on directed graphs below.

dblodgett-usgs commented 7 years ago

The group agreed to this at the Delft TC. Resolves an issue in #182

dblodgett-usgs commented 7 years ago

Draft Response:
The emphasis put on dendritic structures was not intended to indicate that non-dendritic networks are not supported. A summary of how non-dendritic networks arr supported by the catchment model will be added prior to discussing dendritic simplification of complex hydrography.

dblodgett-usgs commented 7 years ago

Added material to section 6.3 and 6.4.

dblodgett-usgs commented 7 years ago

related to #185