There are two primary data types that are used to implement linear referencing in ArcGIS:
- Route feature classes
- Event tables
Using dynamic segmentation, events from event tables are located on line features in a route feature class.
Route feature classes
A route feature class is a line feature class that has a defined measurement system. These measurement values can be used to locate events, assets, and conditions along its set of linear features.
In ArcGIS, the term route refers to any linear feature, such as a city street, highway, river, or pipe, that has a unique identifier and a common measurement system along each linear feature.
In simple terms, feature vertices in route feature classes include m-values (x,y,m or x,y,z,m). These measured coordinates form the building blocks for route features. In route feature classes, line features have their x,y (or x,y,z) coordinates that describe location as well as a measurement (m) value along the line.
A collection of routes with a common measurement system is a route feature class. Each route in the feature class will also have a unique identifier. Line features with the same unique identifier are considered to be part of the same route:
Route feature classes are created and managed as line feature classes in the geodatabase. You can also use route feature classes from coverages and polyline shapefiles that include route identifiers and measured features.
Route feature geometry
Route features have a measurement system stored with their geometry. Each measured line's segments have x-, y-, and m- (measure) or x-, y-, z-, and m-values. When a measure value is unknown for a particular vertex, its m-value is recorded as NaN (not a number) as illustrated in this example:
Simple linear features are represented by lines with one path. Complex linear features are represented by lines with many paths.
It is important to note that although most applications use measures to represent increasing distances along a linear feature, measure values can arbitrarily increase, remain constant, or decrease along line features.
Measure values are independent of the horizontal coordinate system of a feature class (and the vertical coordinate system as well if one is defined). That is, the measure values are not required to be in the same units as the feature class's x,y,z coordinates. For example, features stored in a feature class whose coordinate system is Universal Transverse Mercator (UTM) meters might have their measure values stored in feet or miles or time.
Event tables contain information about assets, conditions, and events that can be located along route features. Each row in the event table references an event and its location is expressed as measurements along named (identifiable) linear features.
There are two types of events: point events and line events. A point event describes a discrete location along a route (a point) whereas a line event describes a portion of a route (a line).
- A point event location uses only a single measure value to describe a discrete location—such as Mile 3.2 on I-91.
- A line event uses both from- and to-measure values to describe a portion of a route—for example, Mile 2 to mile 4 on I-91.
Because there are two types of route events, there are two types of route event tables: point event tables and line event tables. All event tables must contain a route identifier and measure location fields containing measure information. A point event table uses a single measure field to describe their discrete location. A line event table requires two measure fields (a from- and to- measure) to describe their location.
Route locations and their associated attributes are stored in an event table based on a common theme. For example, four event tables containing information on speed limits, year of resurfacing, present condition, and accidents could be included and used to dynamically locate events on a route feature class.
An event table can be any type of table that ArcGIS supports. This includes INFO, dBASE, geodatabase tables, delimited text files, and database management system (DBMS) tables accessed via an Object Linking and Embedding database (OLE DB) connection.
Event table example
Hydrologists and ecologists use linear referencing on stream networks to locate various types of events such as is illustrated in the example below. The route feature class for streams provides measures along the streams using river reach mile. Point and line event tables record the route ID and location along each river reach. These event tables can be used to locate point and line events.
Storing features using relative locations
With linear referencing, locations along linear features are referred to in terms of their route measure, or distance from a known point. For example, it often makes sense to describe the location of an accident as occurring at 12 miles from the beginning of the interstate rather than at a GPS coordinate such as 1,659,060.25, 1,525,238.97.
To determine a location along a linear feature, a system of measurement is required. When a measurement system is stored along with a linear feature, any location along that linear feature can be expressed in terms of the measure values.
In addition to making data more intuitive, storing data as a relative location along a linear feature has the added benefit of ensuring that spatial phenomena you know to fall on a linear feature is located on the feature. For example, in the absence of a very accurate basemap, locating accidents using x,y coordinates may end up displaying accidents that do not fall on the road network as they should. This will not happen if the accidents are located using linear referencing.