Cadastral Data Modelling

1
Cadastral Data Modelling

• Mohsen Kalantari
• Land Administration - 2008

2
Lecture Overview

• What is data modelling?
• Importance of data modelling
• Data modelling for data capture
• Isolated vs Coordinated (Integrated) Surveys
• Data modelling for database design
• Digital Cadastral Databases (DCDBs)
• Data modelling for cataloging purposes
• What is recorded in a DCDB?
• Data modelling for exchange of DCDBs
• Exchange language for maintenance of the DCDB
• Victorian DCDB
• Other DCDB examples
• Current core cadastral data model
• A new cadastral data model

3
What is data modelling?

• The database is specified by a data model
  describing what sort of data will be held, and
  how it will be organized.

Land Parcel Property
Rights
Person
4
What is data modelling?

• Data modelling is a Design activity like
  architecture. Do not expect to find a single
  correct answer.
• The requirement of data modelling process are
  flexible enough to accommodate a variety of
  different solutions, try to be Creative, then you
  will always have Choices.
• Data modelling is NOT data analysis, the
  difference is like prescription versus
  description.

5
Why is the data model so Important?

• A small change to the data model may have a major
  impact on the system
• Example Spatial and non-spatial identifiers
• The program design heavily depends on data
  modelling
• Example Network analysis needs topology
• A well designed data model can make programming
  simpler and cheaper
• Example Automatic adjustment of parcel
  boundaries facilitates calculation of a parcel
  area
• Poor data organization is often expensive to fix
• Automatic accuracy check
• A very powerful tool for expressing and
  communicating business requirements. It can take
  us more directly to the heart of the business
  requirements.

6
Data Modelling Approaches

• Function Driven
• Focuses on function specification
• Example e-conveyancing , digital signature,
  financial settlement
• Data Driven
• Emphases on developing data model before detailed
  function
• Example digital cadastral database
• Prototyping
• Cut and try approach dont waste the time build
  , show, modify, show it again

7
Data Modelling

• The most common alternative approaches to data
  modelling are the
• entity relationship (E-R) and
• the object oriented approach .

8
Entity-Relationship

• The E-R data model views the real world as a set
  of basic entities and relationships among these
  objects.
• An entity is an object that exists and is
  distinguishable from other objects and a
  relationship is an association between several
  entities.

Area
Land Parcel
Belongs
Owner
ID
9
Object Oriented

• A core object-oriented data model consists of
  object and object identifier, object attributes
  and methods and classes.
• In the object oriented approach any real world
  entity is uniformly modelled as an object (eg.
  land parcel), every object has a state (eg.
  rights associated) and behaviour (eg. area).
• Objects are then organised through classes as a
  means of grouping all the objects which share the
  same set of attributes and methods (eg. building
  belongs to the property class).

10
Object Oriented
Owner Name
Property Value Use
Land Parcel Rights in Land Area
Building Type
11
Role of Cadastral Data Model
12
Data modelling and Data Capture

• Isolated Survey Approach
• Isolated parcel surveys, connected but not fully
  integrated to neighboring parcels
• Graphical Accuracy only
• Usually a mix of survey accuracy to map accuracy
  as a result of digitisation of original base maps
• Most common form of cadastre across Australian
  States and Territories (with exception ACT)

13
Data modelling and Data Capture

• Isolated survey data model

14
Data modelling and Data Capture

• Coordinated Survey Approach
• Integrated parcel surveys support a coordinated
  cadastre
• Survey accuracy, where a parcel survey
  incorporates surrounding parcels and are fully
  tied into each other
• a Survey Accurate DCDB is based on coordinates
  determined by ground surveys which are used to
  define, describe and re-define parcel boundaries.
  For all practical purposes, the coordinates in
  this DCDB are the true coordinates result is a
  fully coordinated survey system

15
Data modelling and Data Capture
Coordinated survey data model
16
Graphical vs. Survey Accurate Cadastre

• Level of accuracy is usually survey accuracy
• Isolated Survey Approach leads to a non-survey
  accurate DCDB
• Coordinated Survey Approach leads to a survey
  accurate DCDB which in turn supports a
  coordinated cadastre
• What are the Advantages and Disadvantages of
  non-survey accurate DCDB?
• What are the Advantages and Disadvantages of a
  survey accurate DCDB?

17
Graphical vs. Survey Accurate Cadastre

• Non-Survey Accurate (Graphical) DCDB
• Common approach in Australia
• Less required to setup/maintain DCDB
• More time to complete surveys (less connection to
  neighbouring parcels) but need to re-establish
  parcel nearly every time and connect to geodetic
  control
• Inappropriate use of non-accurate cadastral
  dataset
• Survey Accurate DCDB
• More required to setup/maintain DCDB (up to
  Billions)
• More time to perform initial survey, however very
  little work required to re-do same parcel
• Value-added potential of accurate cadastral
  dataset

18
Data Modelling and Cadastral Databases

• Representation of the geometric component of a
  cadastre in electronic format is termed Digital
  Cadastral Database (DCDB)

19
Data Modelling and Cadastral Databases
The physical designer sometimes is not able to
meet performance requirements, so logical
database design is not direct equivalent of the
data model
Selected Relations between Data
Logical Database Design
Program
data
Data Model
Physical Database Design
20
Data Modelling and Cadastral Databases

• The performance of the information system is
  primarily the responsibility of physical
  designer
• But the alternative to improve the performance
  might be to modify the data model

21
Data Modelling and Data Catalogue

• Consists
• Parcel boundaries
• Easements
• Other Additional Contents
• Property boundaries
• Building footprints
• Street addresses
• Administrative boundaries
• Valuation
• Other land use features (garden, roads, railway
  lines, forest, etc.)

22
Data modelling and Data Exchange
Data model in UML
Data model in exchange format
Convert
Switzerland cadastral data model - Interlis
23
Data modelling and Data Exchange
Step 3 Validation and locking. Step 4
Making new state active after decision
Step 1 Selecting parcels
Export to XML
Step 2 Import from XML
software
Modify
24
Updating of DCDBs

• Subdivisions are charted in LTO's (Land Registry)
  on an index map
• Update of the DCDB refers to those processes
  that ensure that all new and existing legal
  subdivisions are recorded, ie. the cadastral map
  or DCDB is up-to-date
• Specifically this should include
• Recording all new legal subdivisions
• Ensuring map completeness

25
Upgrading of DCDBs

• Upgrade activities (an improvement, not
  maintenance)
• increase accuracy
• inclusion of survey measurements
• alignment of cadastral features with topographic
  features
• changes in data model
• generation of topological structures
• inclusion of historical layer
• unique identifiers for spatial entities

26
Historical Perspective of Australian
State/Territory DCDBs

• Digistised from best available paper maps
• Urban, Rural and Metro
• varying accuracy
• /- 0.5m - 2m in Urban Metro areas
• /- 20m or more in Rural areas
• sourced from different bodies
• different data emphasis

27
Observations of Australian State/Territory Data
Model

• Australia wide cadastral coverage, varying
  accuracy
• Data models are mostly based on graphical
  cadastre
• Data models High GIS focus
• Individual State and Territory solutions
• Meta data
• Incremental updates
• Accuracy upgrade (graphical to survey)

28
Victorian DCDB VicMap Property

• Combination of digitised paper maps and datasets
  held by Water authorities (MMBW)
• Vicmap Property, previously known as the State
  Digital Cadastral Map Base
• Isolated Survey system, non-survey accurate DCDB
• Higher accuracy in Urban and Metro regions, lower
  accuracy in regional areas
• Very good cadastral coverage, over 90 of
  property recorded
• Geodetic datum AMG GDA94, not yet fully adopted

29
Current cadastral data models

• Land parcels constitute the basic building block
  in land administration systems. Parcels were
  employed to identify the areas related to
  interests in land.
• Besides that, land parcels have been used as
  indices for organising land information in
  various land administration subsystems or within
  other related agencies.

30
Current cadastral data models

• Interests recorded in land administration systems
  are traditionally those associated with private
  ownership.
• The most important Interest in land is the
  ownership of land.
• Ownership is usually recorded along with
  restrictions, most commonly caveats, mortgages
  and rights of way.

31
Current cadastral data models

• Land administration systems have so far focused
  on private interests rather than public
  interests.
• Although, some government agencies are
  responsible for management of public lands, their
  systems are usually not as mature as those used
  for managing records of private land, and are
  usually separately maintained.

32
Current cadastral data models
33
The issues associated with the current model

• How new interests and RRRs might be incorporated
  into a cadastral fabric?
• How to facilitate interoperability ?

34
The issues associated with the current model

• Spatial dimension of the interests does not
  necessarily fit to a land parcel
• A biota right exists as an interest that often
  appears to be attached to land parcels , but the
  commercial exploitation of the opportunities
  arising from biota may not neatly align to
  individual land parcels.
• A water catchments right.
• A noise restriction.

(Bennett et al., 2005)
35
The issues associated with the current model

• Current cadastral data models are not really
  spatially-enabled and accordingly do not support
  interoperability
• Grantor/Grantee (Deed based)
• Title numbers (Title based system)
• Block-and-plot numbering system
• Street address
• Spatial identifiers??????????????

36
An alternative model - principle

• Land is not a legal entity until an interest is
  attached to it
• Land parcel is registered when an ownership is
  attached it.
• Right of way is registered when there is a
  restriction in a specific dimension of a land
  parcel.

37
An alternative model - principle

• The very close relationship between each interest
  and its spatial dimension in the real world
  should be also recognised in information systems.
  
• They should be maintained together as a unique
  entity in a cadastral information system.
• This unique entity must define both the interest
  and its spatial dimension.

38
An alternative model core model
Incorporating Interests
Integrating Interests and their spatial dimensions
39
An alternative model conceptual model
40
An alternative model- challenges

• The conversion of RRRs into their respective
  spatial dimensions
• The relationships among legal property object
  layers.

41
Topological relationship

• The topological relationship can quickly find
  neighbouring legal property objects.
• Object sharing makes maintenance for consistency
  and updates easier.

42
Spatial relationship

• To determine whether one legal property object
  touches, coincides with, overlaps, is inside or
  is out side of another legal property object
• One might want to determine which car space
  footprints fall inside a particular land parcel.

43
General relationship

• The general relationship is not physically
  explicit for example, the relationship between
  the owner(s) with an apartment
• For instance the same position may involve
  complex relationships among persons, each
  interested in a different way, for example as a
  car space, a water catchment, or as owner of the
  parcel.

44
Vertical relationship

• Vertical integrity is the ability to relate legal
  property objects from one data set with legal
  property objects from another.
• For example, automatic realignment of the gas
  line through its topological link to the property
  boundary is possible.

45
New Cadastral data model
46
Any Question