For many years surveyors have been remote from the GIS industry as the levels of precision they maintain was not represented in cadastral databases and the business case to rectify that is only now being recognised. The spatial accuracy of cadastral databases is now seriously on the agenda as technology provides high accuracy across all data. The time has come for cadastral management in GIS to be driven by surveyors, their surveying methods and their high level of spatial awareness.
SURVEY BOUNDARY DEFINITION PROCESSES
Survey information is preserved on public record for surveyors to accurately redefine property boundaries in the field and serves underpin land title security guaranteed by government. Surveyors create the cadastre with the highest precision but that spatial quality is lost in creating cadastral databases. Surveys are now undertaken with the highest accuracy in measurement and geodetic spatial referencing.
However when defining boundaries they often have to interpret accurate spatial location on the ground from survey plans that may be over 100 years old with dubious accuracy or occasionally contain incorrect or misleading information. It becomes an exercise in survey experience to fit that record information with survey marks or monuments found on the ground to effect a decision on the spatial location of a property boundary.
In Australia and many other jurisdictions, all the evidence to make that decision is represented on public record and provides confirmation of the title measurements to the authority registering the plan. That data also provides current (more accurate) title measurements of surrounding properties. Up till now this supporting survey measurement information used in the surveyor's decision making process was never utilised in GIS cadastral management.
A CADASTRAL DATABASE BUILT SOLELY FROM SURVEY MEASUREMENTS
20 years ago experienced land surveyor turned academic Dr Michael Elfick and Cadastre Survey New GIS based cadastral precision efficiences Fig. IT Systems Analyst Michael Fletcher recognised that position based systems (GPS) would supersede measurement based systems that historically had defined land boundaries. A process based on survey methods was developed to meet that challenge.
Property ownership underpins the capitalist economy and defining that ownership accurately can be critical. Changing the survey method of defining those boundaries will not happen in the near future as surveyors are required to verify definitions in comparison with historical records and mark the new location on the ground with survey marks.
When a surveyor places survey marks to redefine parcel corners, the measurements between those marks often differ from the dimensions shown on the title. Usually older titles have the greater differences. This difference means trying to fit together shapes only defined by the title dimensions will not replicate the continuous fabric of the cadastre in the real world. Slithers and overlaps occur when only the raw title data is modeled and adjustment of that data is required to create the continuous fabric. In the model, the parcel dimensions are adjusted, however, the original title dimensions legally define the property and should always be available as the title attributes.
Elfick has developed an electronic process that follows survey boundary definition protocols to provide a cadastral database system that combines historical survey measurement information with modern measurement and computing technology. This creates a SURVEY DATA MODEL as the basis for a geodetic cadastral database.
Fig. 1: DATA ENTRY – Survey Information collected
A high level of financial responsibility rests on a surveyor's boundary definition work and as such, normal practice is to incorporate as much checking of their work as possible. This level of rigour is inherent in the Elfick process.
GIS CADASTRAL DATABASE HISTORY
Up till the 1990's cadastral databases consisted of paper or plastic film charting maps maintained by hand. Spatial accuracy for charting maps was not important as they provided the link to the registered survey title plans which provided the survey measurements and background monument information for spatial accuracy in a local reference frame.
These charting maps were the only reference database of the cadastre and were generally digitised to provide the first electronic cadastral databases for GIS. This was adequate at the time as the GIS representation was 'pictorial' and was a means of linking to other feature datasets relating to the property. Also the computing technology limited the storage & computations of accurate Geodetic coordination. These databases were of no benefit to surveyors other than as a mapping tool for 'big picture' representations.
Fig. 2: PARCEL JOINING
As GPS technology highlighted the spatial deficiencies of cadastral databases, improvements to the precision of cadastral databases involved 'rubber sheeting' the model to fit surveyed coordinates. Increasing the density of coordinated points then increased model precision. This method provides no spatial validity to intermediate points between accurately coordinated points and can become 'survey' intensive.
More recently, spatial precision has been increased by adding varying degrees of survey measurement information. This is a progression towards the Elfick model but will always remain an improved mapping outcome rather than a tool of survey merit.
SURVEYORS MEASUREMENT DATABASES
Surveyors are renowned for not throwing away anything. In their records every job they have worked on contains some elements of survey information that may be of benefit to them in the future. Thus for years, these file records on paper have been retained as their personal database.
Since the 1980's Geodetic referencing of surveys was required, but the majority of titles were still referenced by local survey monuments such as pegs, buried pipes or concrete blocks. Surveyors found these marks to orientate their surveys and stored references to them in their records. Some of this data would be used in the preparation of subdivision plans and would then be available on public record. The surveyor's public database consists of the survey plans of subdivisions done by other surveyors and stored on paper in local or State Offices. The advent of computers and coordinate geometry software then allowed them to electronically store their personal survey boundary definitions, firstly on a local datum and more recently on a standard geodetic one.
CREATING A SURVEY DATA MODEL AND CADASTRAL DATABASE
The measurements from survey plans are used to model the parcels and those parcels are then joined to form a mesh of survey geometry known as the SURVEY DATA MODEL. The SURVEY DATA MODEL is then used by the least squares adjustment to fit the geometry mesh to geodetic survey control marks and then generate coordinates to create the cadastral database. The stages of the process are:
The options to create a survey data model include: Manual data entry of survey plans – Survey measurements are used to define parcels. Other survey information from plans is entered to provide updated dimensions and connectivity to other parcels. Connections across roads are also generated to ensure those roads maintain their nominated fixed width. Survey bearings and measurements are entered and held fixed in tables. Other cadastral parcel attributes are included. i.e. Plan Number, Age, Area, survey status (historical, compiled). Weightings are generated to nominate survey quality based on the age of the plan or can be nominated manually.
Manual data entry is the preferred method as 'parcel' miscloses are checked before joining to the model which checks data entry and plan quality.
Importing electronic data from other survey geometry software – Survey software packages can usually output relevant cadastral data (with parcel attributes) in a format which can be imported to the survey data model.
Migration of other existing GIS cadastral databases – Blocks of some mapping databases are able to be imported by inversing the existing mapping boundaries. This method is used to quickly populate a cadastral database but has no survey merit and is often too inaccurate to be adjusted. In such circumstances, pockets of good data can be used to improve large areas of relatively poor data as an iterative process.
The geometry model of each parcel is then joined to the Survey Data Model which invokes another level of survey rigour.
Each point of the joining parcel is joined directly to the correct node on the model. If a point sits somewhere along an existing line that point is designated as a 'line point' to maintain the existing line as a straight line.
As each point is connected, the precision of the 'fit' of the new geometry to the survey data model is measured so the spatial quality of the new geometry is reported. A parcel is not joined until the user is satisfied with the joining residuals. Drafting or other problems with plans are quickly found in this process.
The joining process is currently being investigated by the Land Titles Office of New South Wales (Australia) as a means of checking the survey integrity of new subdivision plans relative to the existing cadastral survey data model and as a format for future electronic lodgment.
The survey data model stores the original measurement data and always returns to that original data to form the least square equations for every adjustment. ACTIVE control points are held fixed and the geometry mesh is adjusted to them. NON-ACTIVE control points are not held fixed and new coordinates are calculated through the model for them. They provide an independent check on the spatial quality of the database created. After each adjustment is run by the operator, a report is available which compares the computed database distance with the original surveyed distance for every line in the model.
Fig. 3: The Adjustment Report The GIS can
THE OUTCOMES FOR A SURVEY DATA MODEL
- Curve geometry is mathematically defined
- The survey data model produces the most complete and accurate representation of the legal cadastre.
- The precision of the model is determined by the accuracy of the survey information used. i.e. for modern survey information the model accuracy should be less than 20- 30mm
- Once entered, all the survey information is stored as part of the model and is efficiently updated with new information.
- As the model uses survey measurements to compute the database coordinates between control points, the amount of survey control is greatly reduced.
THE SURVEY DATA MODEL AS A FIELD SURVEY TOOL
Using laptop computers, surveyors are taking the survey data model in the field to locate remote or buried survey marks by handheld or RTK GPS with sub-meter accuracy. Once a survey mark is found the survey coordinates are held fixed for that point in the model and the adjustment is rerun. Within 15 minutes the surveyor is then searching for the next mark with a much higher model precision. This has been of great benefit in rural areas or rugged terrain or to quickly confirm a mark has probably been destroyed. Surveyors have reported benefits by being able to resolve problems with old or poor standard plans in the field.
As surveys are completed, a coordinated survey database is built. This information is available as a basis for future surveys in the same or adjoining areas.
THE OUTCOMES FOR GIS
If an existing database is spatially accurate, there is a greater awareness of all users to ensure new information meets the same level of spatial quality, so all data being added to an asset database should retain its spatial integrity. The rules for an accurate database can then ensure that no data is adjusted away from an accurate coordinated spatial location to match a poor quality cadastre
A NEW RELATIONSHIP BETWEEN
SURVEY AND GIS Survey accuracy in a cadastral database encourages a mutually beneficial environment for both surveyors and GIS professionals In local government and utilities, the surveyors are now working more closely with the GIS staff for a mutually beneficial environment. Surveyors are now more interested in contributing accurate information for the GIS. The GIS can assist the surveyor with additional information (aerial photography) that may assist in locating survey marks or other assets in the field.
THE BUSINESS CASE
The benefits of an accurate survey data model are efficient operations and no problems caused by a cadastral layer of unknown spatial quality.
Currently there is no appreciation of the real economies of the issues and no true accounting of the real cost of inefficiencies or problems.
Spatially accurate data is more efficiently migrated internally & externally and risk management issues and potential liabilities are reduced.
In New South Wales a Continuous Operating Reference Stations (CORS) network of GPS stations will soon provide centimetre accurate geodetic coordinates to surveyors across the State and the general accessibility of modern spatial technology means a spatially accurate cadastral database is critical to effective data management.
The survey data model is the key to meeting the future spatial challenges of Surveyors and GIS Professionals so the cadastral layer has survey and spatial merit.
Fig. 4: CONTROL REPORT – Review of model precision through non-active control points