GNAT

Stream Network Preparation

Step 1 - Stream Network Preparation

One of the primary reasons a user may want to use GNAT will be to produce a stream network dataset that is segmented using the Segment Stream Network tool. In order for a segmentation process to be successful, that tool requires a stream network that is free from topological errors, with all subnetwork identified. For GNAT, we define a topologically clean network based on the following criteria:

  • No duplicate reaches
  • Only one outflow per subnetwork
  • All digitized directions flow downstream

The first tool found in GNAT under Analyze Network Attributes > Step 1 - Stream Network Preperation is the Find Subnetworks tool. This tools is intended to be run iteratively, ultimately producing a topologically-clean stream network that will serve as an input for the Generate Network Attributes tool. The Generate Network Attributes tool will calculate and add network edge type attributes to the network, and will also generate a network node dataset. This attributed network will then be supplied to the Generate Strahler Stream Order tool, and the output for that tool can serve as input for the last step in the stream network preparation process - the Generate Stream Branches tool. The end goal of the workflow described here is a topologically clean, attributed shapefile that can serve as an input for the Segment Stream Network tool. The following steps are a suggested workflow for producing such a network.

Suggested Workflow for Step 1

Identifying subnetworks and fixing topology errors
  1. Create a workspace (folder or file geodatabase) that will store all interim data products.
    • Please Note: When creating or using a Riverscapes project, please ensure that all interim, working data products are stored separately from where Riverscapes project files are stored. Do NOT use datasets stored in Riverscape project folders for active processing in the tools.
  2. If the stream network is a USGS NHD dataset, which includes an attribute field called FType, we recommend removing all features where FType = 336'. These features represent canals within the stream network, and will likely cause problems with network topology.
  3. Add the stream network (which must be in a shapefile format) as an input to the Find Subnetworks tool. Ideally, this should be an NHD flowline dataset, and include a stream name attribute field such as GNIS_Name.
  4. Do NOT choose the Find topology errors option.
  5. Run the Find Subnetworks tool, which produces:
    • a new shapefile, with the following attribute fields added:
    • _FID_ : unique ID code for each line feature
    • _netid_ : unique ID code for each subnetwork
  6. The user should now symbolize the output shapefile in ArcMap using the _netid_ attribute field, by assigning a unique random color for each _netid_ value. This will highlight all of the disconnected subnetworks found by the tool. The user must now decide which subnetworks are disconnected erroneously, and which subnetworks should stay disconnected (or be discarded entirely). Erroneously disconnected networks should be connected manually using ArcMap editing tools.

find_subnetworks_output.PNG

  1. Rerun the Find Subnetworks tool again with the edited shapefile as input, but this time select the Find topology errors option.
  2. If the Find topology errors option is selected, the resulting output shapefile will include three additional attribute fields:
    • _err_dupe_ : boolean value indicates if the line feature is a duplicate of another feature (based on calculate length).
    • _err_conf_ : boolean value indicates if the line feature has an upstream flow direction.
    • _err_out_ : boolean indicates either a problem with flow direction (multiple reachs flowing to the same node) or a subnetwork where the number of outflow features > 0.
  3. Add the output to ArcMap, but this time symbolize using the error fields. By highlighting the location of the errors, the user can then use ArcMap editing tools to manually correct these features.
  4. At this point, the user can choose to continue running the Find Subnetworks tool until no errors are found.

Please Note: the errors found by the Find Subnetworks tool should only be viewed as potential errors. Whether or not a stream feature actually has a topological error, and how it should be corrected, is up to the user and should be determined on a case-by-case basis.

Generate network attributes, stream order, and stream branch IDs.

  1. Using a shapefile representing a topologically clean network with subnetworks identified, run the Generate Network Attributes tool. This tool iterates through each subnetwork within the shapefile, and labels each feature with an ‘edge type’. Edge types are based on the topological context of each feature in relation to the rest of the features in subnetwork.

    Edge types include:

    • headwaters
    • connectors
    • mainflow
    • braids
    • outflow

    In addition, the Generate Network Attributes tool also generates a new point shapefile, which represents network nodes.

network_attrb_edges.PNG

  1. After processing is finished, the edge type classification of output shapefile should be reviewed in ArcMap, with symbolization based on the _edgetype_ attribute field.
  2. The output from the Generate Network Attributes tool should then serve as the input for the Generate Strahler Stream Order tool. Stream order will be calculated separately for each subnetwork found within the network shapefile.
  3. Symbolize the output shapefile using the _strmordr_ attribute field, and review the strahler stream order values. Each subnetwork within the shapefile will have it’s own separate stream order classification.
  4. Finally, run the Generate Stream Branches, and then symbolize and review the stream branch ID values, using the BranchID attribute field.

At this point, the network shapefile produced by the Generate Stream Branches tool can be used as input for the Segment Stream Network tool.