aphic
seven
land
oads,
ne of
nding
four
ints),
amon
ng to
A
iched
3 and
gram
zing.
d for
line
after
d by
y of
etric
rmed
cally
been
ision
of a
data
991;
c of
k for
sion”
the
Vays,
tent,
, Our
data
tural
The discussion in this section is based on Ramirez (199523,
1995b, 1996). The obvious reason why spatially referenced data
(or maps) are revised is because they represent a dynamic
surface: the surface of the earth. The surface of the earth is
subject to the action of natural forces and man-made actions.
Both produce changes on the earth's surface. Only the subset of
changes in elements traditionally represented in topographic
maps (including relief) are of interest here.
Natural forces, in general, generate two types of changes:
systematic and abrupt. Systematic changes are those continuous
changes on the surface of the earth generated by the forces of
gravity, wind, life-cycle, and others. Systematic changes are
predictable (we know that they will happen and affect the surface
of the earth) and require a time interval (t? - tj) to alter the
currency of the spatial data representation. Abrupt changes
caused by the forces of nature immediately affect the currency of
spatial data. Examples of these changes are those caused by
earthquakes, flooding, forest fires, and landslides. Abrupt
changes are unpredictable, and affect the currency of the spatial
data representation in a very short time interval (t4 - t3).
Human actions also modify the surface of the earth in two ways:
by predictable and unpredictable changes. Again, only those
changes that affect the currency of spatial digital data are
considered here. Predictable changes are those whose outcome
will be known in advance and are evident by a time (ts).
Examples of these include construction of roads, shopping malls,
sport fields, and parks. Unpredictable changes are those changes,
such as open-field mining and logging, whose outcome is
unknown at time (ts) and are evident only later at time (t7).
All of the above changes are local in nature. They altera specific
geographic zone and, in most cases, the relief and the
representation of the features on the terrain. Features of interest
here are those contained in conventional topographic maps.
These features can be classified in a set of layers or coverages.
There is not a universal classification for map features. However,
a typical example of classification is the one used by the USGS
(see Table No. 1). In this classification, features are grouped in
nine layers.
Table No. 1
Cartographic Elements: Major Coverages
T Med Vegetative Features
Public Land Survey S stem
.. Transportation ystem
Systematic changes due to natural forces are apparent only over
long periods of time. For example, hypsographic changes become
significant only when they reach the magnitude of about half the
contour interval of the cartographic product. Abrupt changes are
impossible to predict and can affect the terrain representation
679
immediately. They have the potential of changing the terrain
representation in the most radical way; however, it may be a long
time between abrupt terrain changes.
Terrain changes due to human actions, especially predictable
changes, are the most common. The terrain is constantly
changing, due to new constructions, particularly of transportation
features (all kinds of roads, airports, etc.), and miscellaneous
cultural features (buildings, shopping malls, and so forth).
Unpredictable changes because of human actions also affect the
terrain representation -- perhaps more radically, but usually less
frequently. Some unpredictable changes are only temporal (at
least in the USA). For example, open-field mining changes the
relief substantially. However, once mining is completed, by law,
the relief must be reconstructed to its original shape. Based on
this discussion, the need for terrain revision could be classified
and summarized as shown in Table No. 2.
Table No. 2
Topographic Map Revision: Change Factors
4.3 The Cartographic Language
The analytical study of spatial data (maps) provides another part
of the framework for spatial data revision. As part of the study of
spatial data, Ramirez (1991) has identified a cartographic
language to represent spatial features. The cartographic language
is composed of the alphabet and grammar. The alphabet is the set
of primitive signs from which all spatial features (cartographic
elements) can be generated. It is equivalent to the alphabet of any
natural language (for example, a, b, c, d, etc., for the English
language). The grammar is the set of operations, rules, and
writing mechanisms that allows (and constrains) the generation
of spatial features from the cartographic alphabet. In this context,
spatial features or cartographic elements are the terrain features
represented on a spatial database or map (for example, the
outline of a house). In the next paragraphs, a brief description of
the cartographic alphabet and grammar are given.
The Ramirez alphabet is composed of four signs: point, line,
curve, and blank space. Point is the sign that occupies no area
and has no length. The alphabetic sign point is different from the
cartographic point which occupies an area (for example, the
cartographic point representing an individual tree). As a matter
of fact, the alphabetic sign point is the skeletal representation of
cartographic points. It carries positional and representational
information. The alphabetic sign line has length but occupies no
area. It joins two points on the plane or in the space (the shortest
distance). It is different from the cartographic line, which has
length and occupies an area (for example, the cartographic line
representing a street in a map). The alphabetic sign line is part of
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996