Network tracing in ArcMap
Network analysis involves network trace, or tracing. The term tracing is used here to describe building a set of network elements according to some procedure. You can think of tracing as placing a transparency on top of a map of your network and tracing all the network elements that you want to include in your result onto the transparency.
When working with networks, tracing involves connectivity. A network element can only be included in a trace result if it is in some way connected to other elements in the trace result. The trace result is the set of network features that is found by the trace operation.
For example, suppose you want to find all the features upstream of a particular point in a river network. Using a transparency placed over the map of the river network, you could trace over all the branches of the river that were upstream of that point. What is drawn on the transparency after this would be your desired result.
Similarly, when you perform a trace operation in ArcMap, your result is a set of the network elements included in the trace. In ArcMap, your trace results can either be drawings on top of your map or a selection.
Flags and barriers
In ArcMap, flags define the starting points for traces. For example, if you are performing an upstream trace, you use a flag to specify where the upstream trace will begin. Flags can be placed anywhere along edges or on junctions. When performing the trace operation, ArcMap uses the underlying edge or junction feature as the starting point. Network elements connected to these edges or junctions are considered for inclusion in the trace result.
Barriers define places in the network past which traces cannot continue. If you are only interested in tracing on a particular part of your network, you can use barriers to isolate that part of the network. Like flags, barriers can be placed anywhere along edges or on junctions. When performing trace operations, ArcMap treats the underlying network features as if they are disabled, thus preventing the trace from continuing beyond these features.
Disabling features and feature layers
Disabling features is a more permanent method of creating a barrier at a particular location. In a municipal water network, for example, if a water main has been opened and capped due to a road construction project, water cannot flow through that section of the water main. Disabling the network feature representing this water main would stop a trace at this feature.
In some cases, disabling entire layers may be necessary. For example, by disabling the switches layer in an electrical distribution network and tracing from some point in the network, you can find the switches that need to be thrown to isolate this point from the network; these will be the features at which the trace operation is stopped.
Edges and junctions can have any number of weights associated with them. A weight is a property of a network feature typically used to represent a cost for traversing an edge or moving through a junction. An example of an edge weight is the length of the edge. In a shortest path analysis, you would choose this weight if you wanted the resulting path to be of the shortest length. Another example is the resistance to traversing an edge in an electrical network. Using a resistance weight, the shortest path would be the path of least resistance.
When you build a network, you specify which attributes of edge and junction feature classes will become weights. You can use these weights to specify the cost of including a feature in the results of a trace operation. Of the trace tasks included with ArcGIS, only Find Path, Find Path Upstream, and Find Upstream Accumulation use weights to calculate the cost of the trace.
To find the cost using these trace tasks, you must specify which weights to use. For junction features, a single weight is used. For edge features, two weights can be used: one along the digitized direction of the edge feature (the from-to weight) and one against the digitized direction of the edge feature (the to-from weight). The digitized direction of an edge feature refers to the order in which the shape nodes of the feature are stored in the geodatabase. You can specify a different weight for each direction of an edge for cases where tracing an edge in one direction has a different cost from tracing it in the other direction.
You can use a weight filter to limit the set of network features that may be traced. A weight filter specifies which network features can be traced based on their weight values. A weight filter serves the same purpose as creating a selection of network elements based on a simple SQL query, except that the performance of the weight filter is much faster.
Using a weight filter, you specify valid or invalid ranges of weight values for network features that may be traced. As with using weights to represent the cost of including a feature in trace results, a single weight is used for junction features and two weights may be used for edge features.
Traced features versus features stopping the trace
When tracing using the Find Connected, Trace Downstream, or Trace Upstream trace tasks, you can return either the features that are traced or the features that stop the trace. Features that are traced are those that are actually traced over by the operation, while features that stop the trace are those past which the trace cannot continue. Features that stop the trace include the following:
- Disabled features
- Features on which barriers are placed
- Traced features that are only connected to one other feature (dead ends)
- Features that have been filtered out with a weight filter
Using selections to modify trace tasks
When tracing, ArcMap lets you use selections in three main ways:
- Using the Analysis Options dialog box, you can specify whether the trace operation is performed on all features in the network, on the selected features only, or on the unselected features only. Tracing on just the selected features means that unselected features act as barriers, while tracing on just the unselected features means that selected features act as barriers. By using selections in this manner, you could perform a trace operation to produce a set of barriers for a subsequent operation, or you could build a selection query to produce a set of network features on which to perform a trace operation.
- You can specify which layers are selected when performing a trace operation. From the Selection menu in ArcMap, you can specify which layers can and cannot be selected. When ArcMap returns the results of a trace operation as a selection set, the settings you specify in the Selection menu are used to determine which features should be included in the selection set returned by the trace.
- You can use the interactive selection method—set through the Selection menu—to specify the behavior of the resulting selection set. You can create a new selection, add the results of your trace operation to the current selection, select the results of your trace operation from the current selection, or remove the results of your trace operation from the current selection.
By using the power of selections in ArcMap, you can use the simple trace tasks included with ArcMap to perform compound and complex trace operations.
Putting it all together
Tracing operation with the Utility Network Analyst toolbar
With the Utility Network Analyst toolbar, you can find the following:
- All network elements that lie upstream of a given point in your network (trace upstream).
- All network elements that lie downstream of a given point in your network (trace downstream).
- The total cost of all network elements that lie upstream of a given point in your network (upstream accumulation).
- An upstream path from a point in your network (find path upstream).
- The common features that are upstream of a set of points in your network (find common ancestors).
- All the features that are connected to a given point through your network (find connected).
- All the features that are not connected to a given point through your network (find disconnected).
- Loops that can result in multiple paths between points in a network (find loops).
- A path between two points in the network. The path found can be just one of a number of paths between these two points, depending on whether your network contains loops (find path).