Transformations - Spatial Mathematics: Theory and Practice through Mapping - page 63

different ways. Establishing relationships between attribute tables is useful
so that the relationships can be mapped and analyzed. For example, in Esri's
ArcGIS for Desktop, most map layers contain an attribute table. The map can
be thought of as the “G” part of GIS, while the attribute table can be thought
of as the “I” part of GIS. The “S” part is the relationship that links each fea-
ture (such as a river, a county, a hurricane track) with its associated attributes
(such as whether the river is perennial or intermittent, the name of the county
and its 2010 population, and the intensity of the hurricane, respectively). This
relationship is usually maintained via a common feature ID linking the fea-
ture to its attributes. If this relationship is broken, not only will the features
not have attributes, but the database is broken and the features will not even
be able to be mapped.
However, additional relationships can also be established within the geodata-
base. A one-to-one or many-to-one relationship can be established between a
map layer's attribute table and another, separate, external table. An example
of this happening via a one-to-one relationship is for weather stations. Each
station is represented by a point on a map. Each point is represented by a
record, or row, in its associated Table A. Table B contains the weather data,
such as temperature, humidity, wind speed, and other variables for each sta-
tion. Table B contains multiple columns, or fields, for each station, but each
station is still represented by one row in the table. A common ID establishes
a one-to-one relationship between the two tables, allowing the analyst to map
the weather stations and analyze the spatial pattern of the weather variables
on that map.
At other times, a many-to-one relationship is needed. For example, suppose
that a particular polygon map layer shows types of land use in an area. The
land use layer's attribute table only stores a land use code. A separate table
stores the full description of each land use type. Joining these two tables
together establishes a many-to-one relationship because many records in the
layer's attribute table are related to the same record in the table of land use
descriptions. The analyst can use the more descriptive text when generating
the legend for the map. This descriptive table is often referred to as a “lookup
table.”
In a GIS, tables can be joined together to perform the activities above, but
tables can also be related without performing the join. Relating tables in
ArcGIS for Desktop is done with a “relate” function, which simply defines a
relationship between two tables. The associated data are not appended to the
layer's attribute table as they are with the process of “joining.” Instead, the
related data can be simply accessed when the analyst works with the layer's
attributes.
For example, if you as the analyst select a building, you can find all of the
tenants that occupy that building. Similarly, by selecting a tenant, you can
find what building the tenant resides in. The tenant may reside in several
buildings, as in the case of a chain of stores in multiple cities and in multiple