Geoscience Reference
In-Depth Information
original data. The errors that result from coordinate transformation must be considered to ensure
the quality of each data layer. Most surveys performed on small areas are based on the assumption
that the earth is a planar surface. For large-area surveys, it is necessary to consider the earth's cur-
vature by computing an
x
,
y
position in terms of geodetic latitude and longitude. These computations
can be quite complicated and lengthy, so a plane rectangular coordinate system is preferable.
Due to the constant improvement of geodetic surveys, the lack of agreement for a standard
coordinate system among agencies, and the slow update of the older geodata layers, GIS analysts
often encounter spatial data in a variety of projections and coordinate systems. For example, one
may obtain a soils map in an Alber's equal area projection, Digital Elevation Model (DEM) data
in a UTM coordinate system using the NAD 1927 datum, Global Positioning System (GPS) data of
specific points located in a geographic coordinate system, a Landsat Thematic Mapper (TM) image
in a UTM coordinate system using the WGS 1984 datum, or aerial photographs with a local coor-
dinate system. Familiarity with the concepts of map projection and datums is necessary to convert
data layers with different coordinate systems into a unified standard coordinate system.
10.5 SpAtIAl dAtAbASeS
GIS databases are used to organize and process geospatial data layers. The relational database
model is the most widely used model in a GIS. The relational database is a collection of tables,
known also as relations, which are connected to each other by keys. The row of the table is called a
tuple (or record), and the column of the table is called an attribute (or a field) that describes the spa-
tial entity. The attributes can also be called items or variables. The tables are related to each other
by using keys. The keys represent one or more attributes whose values uniquely identify a record in
a table. Data in the table can be logically joined together using a key that is the common attribute
data value between two tables. An example of a relational data model is shown in Figure 10.5. The
key connecting zoning and parcel is the zone code, and the key connecting parcel and owner is the
PID (parcel ID number). When used together, the keys can relate zoning and owner.
The relational database—simple relational tables—is queried using Structured Query Language
(SQL). This is an English-like language that uses logical commands to manipulate the data stored in
the relational tables. Examples of logical commands would include restrict, project, union, intersect,
difference, product, join, and divide (Bolstad, 2005).
A comprehensive database used in soil surveys is the Soil Survey Geographic (SSURGO) data-
base. The Natural Resources Conservation Service (NRCS) collects SSURGO data from field map-
ping activities. The data are archived in 7.5 min quadrangles for a soil survey area. Linked to
SSURGO is the Map Unit Interpretations Record (MUIR) attribute database, which contains a
series of tables of around eighty-eight estimated soil physical and chemical properties, interpreta-
tions, and performance data. There are three important tables in the MUIR database: the soil survey
Zoning
Owner
Zone Code
PID
Parcel
fIGURe 10.5
An example of a relational data model.
