Geographic Information Systems

Intro to GIS

Objects

Spatial phenomena, both thematic and geometric data linked to a location or a set of locations

Geometric data

Spatial data: one or multpile co-ordinate points that define the Location of a spatial phenomenon (object)

Thematic data

Non-spatial data: Information that describes attributes of spatial phenomena, often represented in categories such as land use or population density.

→ manipulated through querying

The field approach

An approach to spatial data modeling where terrain characteristics are continuous and can be measured at any point, often used in mapping and analyzing surface features. This method is used for understanding natural phenomena (elevation, depth,…) and environmental changes (temperature, precipitation) over a geographic area.

Interpolation method

A statistical technique used to estimate unknown values at specific locations based on known values from surrounding points. This method is commonly applied in the field approach to spatial data models to create continuous surfaces from discrete data.

The object approach

An approach to spatial data modelling where discrete objects are given geometric and attribute data within a geographic area. Usually this method is used in vector-analysis, where data is represented as points, lines or polygots

Model transformation

Transfering field data into an object data model, facilitating the treatment & interpretation of data.
e.g: the production of a soil map based on soil profile analysis

The raster approach

A way of visualising spatial structures into a regular grid of cells, that completely covers the area of study

→ definded by origin coordinates, resolution (size of cell) & dimensions

→ one cell: one attribute: overlay is needed to combine themes

Central point method

Method whereby the cell in a raster model receives the value of the center of said cell

Dominant unit method

Method whereby a cell in a raster model receives the value of the largest area covered in said cell

Map algebra

Way(s) of combining themes in a raster model; has its origins in model planning & landscape architecture

= a standardized tool for overlay analysis (see logical AND / OR functions)

The vector approach

A way of visualizing spatial structures by means of objects, with each type of object (object class) representing one vector layer

The spaghetti model

Also dubbed ring model; a way of representing the object geometry in a vector approach where all objects are represented seperately: no relational topology

→ need for on the fly analysis of spatial relations & more data redundancy

The topological model

way of representing the object geometry in a vector approach wher objects’ relations are incorporated, less redundancy

→ less computationally demanding, but more complex storage & creation

Triangulated Irregular Network model

A special kind of vector model ideal in representing 3D models built through a network of triangles based on elevation measures

→ adaptable level of detail (=\= raster GIS)
→ higher precision in representing elevation changes
→ less storage needed in flat areas (larger triangles)

Delauney triangulation method

Method used to triangulate irregular networks through the

• cirlce criterion: when constructing a circle through the three points of the triangle, not other point of triangulation may be included (avoids narrow angles)

• Min-max criterion: When considering the six internal points of two adj. triangles, the minimum angle is maximized

Topography

The observed physical model that accurately describes the size and shape of earth

Geodesy

The best approximation of the earth: the earth as a geoid, approached by a mathemathical model

Ellipsoid

Fitted onto the geoid approximation of earth to obtain a geodetic datum, with three-dimensional geodetic coordinates

Geodetic (coordinates)

coordinates that are not fixed but depend on the geodetic datum
→ Longitude (lambda); Latitude (phi); Ellipsoidal height (he)

→ reflect the angles of the geodetic datum

Geocentric (coordinates)

Coordinates that are fixed, independent of the geodetic datum

eg: ETRS89

→ x, y, z

WGS

Global geodetic datum used by GPS systems; found by the US defence departement

Cylindrical (projection)

Map projection that has no distortion around the equator; usually used for areas extending along a great circle (eg. Chile uses a transversal … projection)

Conical (projection)

Map projection that has no distortion on a chosen parallel: usually used for areas extending along a small circle (eg Belgium)

Azimuthal (projection)

Map projection that has no distortion on the pole(s); usually used for areas with compact shape, with no clear extension in a certain direction (eg; netherlands)

Standard points (or lines)

The point of touch in a map projection that has no distortion

Equal-area

Maintenance of the area (surface) in a map projection

Conformal

Maintenance of the angles in a map projection

Equidistant

Maintenance of distance in a map projection (only possible in specific directions)

Mercator projection

A conformal cylindrical map projection that is widely used for navigating purposes, but has significant distortion towards the poles

Gall-peters projection

An equal-area cylindrical map projection

Normal (orientation)

Orientation of projection bodies along the parallels of the earth

Transverse (orientation)

Orientation of projection bodies along the meridians of the earth

Oblique (orientation)

Diagonal orientation of projection bodies

Belgian Lambert (coordinate system)

A large-scale conformal reference system that uses a secant conical projection body with two standard lines within Belgium; generally usefull for countries with a dominant East-West projection

Zonal systems

Slicing up a large country into smaller parts with each their own standard parallels for large-scale reference systems; used in eg. France, Germany and the US

Universal Transverse Mercator (UTM)

Us department of defence projection that defined 60 meridian zones with a 6° longitudinal width that allows only 1m of distortion within each zone: used between 84° north and 80° south (not for poles → use UPS)

Bursa-wolff 7 (transformation)

Datum transformation used when performing coordinate tranformation with two different geodetic data

→ moves the origin, rotation & scale

Coordinate reference system

A definition of both the geodetic datum (the 3D positioning of the mathemathical model of the earth)
and the map projection (The transformation used for datum coordinates to map coordinates)

→ each has its own EPSG code, described in GIS metadata, that also describes parameters to transform between global and local data

→ in vector gis, this is saved under .prj

Geographic coordinates

Coordinates, usually using degrees as unit of measurment, that uses a three-dimensional model of Earth

Projected coordinates

Coordinates, usually using numeric values as unit of measurment, that uses a two-dimensional model of Earth

Local operation

A spatial operation in raster GIS where new values are determined based on the attributes (thematic data) specific to said location

→ looking vertically through layers: one or several input layers

Focal Operation

A spatial operation in raster GIS where new values are determined based on the immediate or wider neighborhood

→ immediate neighborhood (moving window): calculation of slope or orientation based on elevation matrix

→ Wider neighborhood approach: clumping regions (parcelling), defining buffers or visibility, calculating transportation costs/ shortest distance

Zonal operation

A spatial operation in raster GIS where new values are determined by the zone characteristics to which a cell belongs (zone = collection of cells with the same value)

→ calculating the area or perimeter; or summary value (avg slope)

→ possible on vector layers (stored in attribute table, esp usefull for administrative zones)

→ need for an input value (area, slope, …)

Cartographic modeling

Defining a flow chart by logically combining local, focal and
zonal operations, in such a way that the output of one operation becomes the input of another one in Raster GIS

→ multi-criteria analysis for eg land evaluation or allocation

→ Criteria (factors & constraints) are formed into an overall suitability index

In Vector GIS: Combination spatial querying (selecting) and Topological operations (creating)

→ only Boolean ranking (yes-no), no possiblity for continuous values

Single-objective (cartographic modeling)

Finding locations suited for a single use

→ eg: finding suitable habitat for a species to live; best land to grow a certain crop; …

Multiple-objective (cartographic modeling)

Finding locations based on multiple alternatives, i.e. conflicting interests

→ eg: finding suitable location for carpet & agriculture industry

→ hierarchical or conflict approach

Factors

one type of criteria in cartographic modeling

One aspect for the degree of suitability → a discrete or continuous (fuzzy) scale

→ all … should have the same scale (eg 0 - 5)

Constraints

one type of criteria in cartographic modeling

A location where the objective cannot be reached in any case (null value)

Method of Saaty

a method of determining the weights in a trade-off scenario, where weighted linear combinations of factor scores decide the overall suitability index

→ based on 1 to 1 comparisons of the factors in a crosstab with fiinal weights as eigenvector of the matrices

Hierarchical approach (to multiple-objective cartographic modeling)

Objective 1 gains priority over objective 2: iterative increase or decrease of treshold value for suitability map untill sufficient area is allocated to objective 1, afterwards do the same for objective 2 with remaining area

Conflict approach (to multiple-objective cartographic modeling)

Two objectives have equal (or weighted) priority; based on two suitability maps, allocate area to both x and y; then need to partition the “conflict area”: iterative decrease in treshold value for both suitability maps untill both objectives are reached

The object-relational (data) model

The integration of geographic and attribute data into one model, due thanks to spatial database solutions

→ two steps to define said model (storage & treatment of non-spatial data)

1. The conceptual phase

2. The logical phase

feature class

The description of spatial objects in GIS as basic geometry types: Points, lines or polygons (superclasses → subclasses)

One layer consists of one ….

Conceptual phase

The first step in relational data models; creating a conceptual scheme that identifies object classes, features of object classes and relational classes

→ Entity-relationship modeling

Entity class

Represent real-world objects or phenomena that are modeled in the database; represented by a rectangle in ER modeling.

Examples: "Buildings," "Roads," or "Municipalities."

the main components of the data model, with attributes describing their properties = instances (e.g., a building may have a height, address, and type).

Relationship class

Represent associations or relationships between entity classes.

Examples: "Contains" (relationship between Municipalities and Buildings) or "Crosses" (relationship between Roads and Rivers

They define how entities are linked, often including rules or cardinalities (e.g., one municipality contains many buildings).

cardinality

A value between [0;n] that describes the dependency between two enitity classes

→ if maximum of one entity class is 1 there is Functional dependency

Logical phase

Second phase in relational data modeling; transforming the conceptual scheme into a logical model for a specific Database Management system (for GIS, using tables)

Primary key

The underlined attribute of an entity class;

→ composed … for relationship classes: both entity classes

Foreign key

In a one-to-many cardinality, we can directly put the primary key and attributes of the “owner” entity table into the second entity table, making the relation table obsolete

Characterized by a #

Reflexive relational model

In the logical phase of relational data modeling, when there is a relation between instances of a same entity class

→ For a one-to-many cardinality (functional dependency): no relation class, primary keys are duplicated and re-used in the entity class with a different name

→ eg: Owners & tenants; with ID-number as primary key → add “ID-number_owner” as new attribute in table

→ For other cardinalities: Relationship class with primary key and duplicated primary key as attributes, each receiving a different name

→ eg: evolution of land use: relation class with Parcel-number_before and Parcel-number_after as attributes

Topological operations

Spatial operations in vector GIS that modify the geometry of individual objects based on spatial relationships

• Buffers

• Clipping

• Differencing

• Dissolving

Overlay operations

Spatial operations in vector GIS that create new datasets & geometries based on attributes of multiple objects

• Unions

• Intersections

• Symmetrical difference