Lecture 5 - Vector Data Flashcards
What is vector representation?
- complex rep based on primitive objects
- represents spatial location of features explicitly
What are the different ways that vector data can be put together into a vector data model?
- Spaghetti model
- Vertex dictionary
- Dual independent map encoding (DIME)
- Arc/node topological model
What is a TIN model?
A triangular irregular network is a vector model used to represent continuous surfaces
Explain the spaghetti model
- simple (one point and polygon)
- easy to manage & understand
- no topology
- lots of duplication (large storage space needed
- data model is map expressed as a list of cartesian (x, y) coordinates)
- no inherent structure (messy like spaghetti)
Explain what a point, line, and polygon would look like in spaghetti data model
Point: identifier and single x,y coordinate pair
Line: identifier and list of x,y coordinate pairs
Polygon: identifier and list of x,y coordinate pair identical to first pair (closed loop)
What are the weaknesses of the spaghetti model?
- redundancy: common boundaries recorded twice
- no topological info (connectivity): no automatic error checking; spatial relationships must be derived by computation
- nested polygons: polygon order important for graphic display; area measures may be incorrect
- slivers/gaps: logical errors occur if boundaries are not listed the same
Explain the vertex dictionary/ whole polygon model
- polygons described by listing coordinates of points in order as you “walk around” outside boundary of polygon
- vertices stored in one file while lines/polygons stored in another
- no duplication but model still does not use topology
What are the weaknesses of the vertex dictionary model?
- coordinates/borders for adjacent polygons stored twice (can result in slivers/gaps)
- no topological info about polygons
Explain the DIME model
- nodes (line intersections) are identified with codes
- assigns a direction code (a “from node” and a “to node”)
- street addresses and UTM coordinates explicitly defined for each link
- polygons described by listing ID numbers of points in order as you “walk around” outside boundary
- second file lists all points and their coordinates
- solves duplicate coordinate/border problem
- still no topological info
Explain the Arc/Node topological model
- uses topology
- overcomes major weaknesses of spaghetti model (allowing for GIS analysis, like overlaying, network, etc.)
- requires all lines to be connected, polygons closed, and loose ends removed
What is topology?
- relationships in spatial terms b/w connected or adjacent geographical objects
- connectivity: point-line relationship; line-line connection
- area definition: line-polygon relationship
- contiguity: polygon-polygon adjacency
- mathematical model used to define spatial relationships
Why is topology important?
- connections/relationships b/w objects that are independent of their coordinates (geometry changes under transformations)
Explain point, arc, vertices, node, and polygon in terms of vector primitives/features
- point: defined by single x,y pair
- arc: line defined by ordered set of x,y pairs (can be straight of curved)
- vertices: points on an arc that are not nodes
- node: end points of an arc, or intersection of two arcs
- polygon: area defined by arcs making up its boundary
What are the advantages of the topological model?
- spatial analysis: performed without using coordinate data but using topological data alone (more efficient)
- data validation: ensures that associated data form a consistent and clean topological fabric (reduces data set size and additional items can be added to attribute tables)
What are the disadvantages of the topological model?
- when data later is inputted or edited, the topology must be updated (can be time-consuming, but ArcGIS does for you)
- originally did not deal well with continuous surfaces (TIN model helps with)
What are the stages of database creation?
- Input of spatial data
- Input of attribute data
- Linking spatial and attribute data
Explain what a digitizing table is
It is like a printer scanner (has electrodes) and control points with a mouse to trace to digitize paper maps
- automatic editing helps get rid of errors
- edge-matching process to combine several maps
Describe attribute tables
- handled separately from spatial data (non-spatial)
- resides in database tables (rows for spatial entity and columns for variable)
- GIS provides linkage b/w spatial and non-spatial data
(ex. location - what is there?)
What is a DBSM
Database Management Systems
- computer programs that control creation, maintenance, and use of database in comp platform
- sits b/w data files and applications
- programs interact directly with data files
What are the 3 types of database models?
- Flat file/spreadsheet
- Hierarchial
- Relational
What is an ideal database model?
- easy to access
- versatile (maximizes uniqueness of every feature)
- minimizing quantity of data
Explain terminology meaning of data item and record
- data item (attribute): reps occurrence of particular characteristic pertaining to an entity (ex. person/thing/event); smallest unit of stored data
- record: group of related data items; occurrences of diff characteristics pertaining to same person/thing/event
Explain the flat file database model
- simple method
- fast data retrieval
- data stored in single table (all records have same number of fields; individual records have diff data in each filed; one field serves as key to locate particular record)
What are the problems with the flat file database model?
- cumbersome to search
- can only search by key field
- adding fields means reprogramming entire database
- adding new records time consuming
- slow data retrieval w/o key
Explain the hierarchical file database model
- data stored in more than one type of record
- like a family tree (parent-child/ one-to-many) relationship
- one field is key to all records, but data in one record does not have to be repeated in another (records with similar attributes associated together)
- records linked to each other by a key field
What are the advantages of the hierarchical file database model?
- relationship clearly defined and queries follow standard routine
- efficient data structure (access to diff records readily available)
- adding/deleting records easy
What are the disadvantages of the hierarchical file database model?
- uni-directional: must access master record in order to link downward to access other records
- repetitive data
- large amount of storage
Explain the relational file database model
- data files organized as tables (matrices of tables used to store info)
- connect diff files without using internal pointers/keys
- common link of data is used to join/associate records (not hierarchical)
- enables user to form new inquiries and data outputs
- simple data organization into set of relations or tables
Explain relational model terminology: tuple, attribute/item, domain, relation, cardinality, degree, key
- tuple: row
- attribute/item: column
- domain: range of possible values defined for each attribute
- relation: entire table
- cardinality: uniqueness; number of tuples it contains at any point in time
- degree of relation: number of attributes in table
- key of relation: subset of attributes which uniquely identify each tuple in the relation (ex. ID number)
What is a relational join?
- operation of relating/physically merging 2 attribute tables using common column
- required to do relational query (ex. what are taxes on property x? links property id with tax rates)
What are the advantages of the relational file database model?
- flexibility: data can be manipulated by joining tables
- simplicity: organization and modelling
- efficiency of storage: redundant data minimized
- non-procedural nature: queries do not need to take into account internal data organization
- easy to use
What are the disadvantages of the hierarchical file database model?
- new relations require considerable processing
- sequential access slow
- easy to make logical mistakes due to flexibility
What are the advantages of the vector data structure?
- good for transportation, utilities, marketing (many purposes
- provides precise positioning of features in space
- intuitive representation
- good rep of entity data models
- compact data structure
- topology explicitly described
- coordinate transformation easy
- accurate graphic rep at all scales
What are the disadvantages of the vector data structure?
- complex data structures
- some analytical functions are complex/difficult
- display and plotting time consuming and expensive
- spatial analysis within units like polygons impossible b/c internally homogenous
