THE UNIVERSITY OF NEW SOUTH WALES


“Optimising Survey Control using GNSS RTK & CORS Networks

for Spatially Upgrading the DCDB – A case study in the Penrith Area”

by Dean Watkins

Supervisor: Dr. Craig Roberts

October 2007


Abstract     Introduction     Site of DCDB Upgrade     Fieldwork/Data Processing     Adjustment Results     Conclusions     Further Information


Introduction

A digital cadastral database is basically a digital dataset of information pertaining to the cadastre.  The DCDB graphically shows the position of cadastral lot boundaries as well as their relative position to other topographic features such as roads, rivers and lakes. While the DCDB shows the location of cadastral boundaries graphically, it does not define the position of the boundaries legally. Figure 1, below is an example of a DCDB of a residential area in New South Wales.

Figure 1: Example of a DCDB showing parcel boundaries and roads
(source: Williamson & Enemark, 1996)


The issue with the current DCDB is related to its accuaracy. At present, the DCDB is only graphically accurate, i.e., they are not spatially accurate. the reason for this lack of spatial accuracy is a direct result of the process used to initially create the DCDB.

In the mid 1980’s, a state wide DCDB was created for New South Wales by manually digitising the best available hardcopy cadastral maps.  The accuracy of the DCDB is rather varied due to the many different scales of maps used in the digitising process, with errors of a couple hundred metres not uncommon in rural areas. Figure 2, below, shows a typical example of a hardcopy plan commonly used to upgrade the DCDB.

 

Figure 2: An old cadastral map similar to those used in creating the digital cadastre.
(source: RTA, 2007)

 

Due to the digitising of hardcopy maps, inaccuracies were directly related to the scale of the plan and the skill of the digitiser and accuracies of 1mm of map scale were typically achieved. This inaccuracy represents errors of between 0.5m-4.0m in urban areas where plans with scales between 1:500 and 1:4,000 were often used, and in the range of 25m-100m in rural areas where plans of scales 1:25,000 – 1:100,000 were commonly used for digitising.


The required accuracy of an upgraded DCDB is largely dependant upon its intended application. There are several users of DCDB including local councils, government bodies and utility management organisations for many applications including:

However, these users are now requiring a spatial accuracy of their DCDB's of up to 0.3m. As a result, these users are seeking an efficient and cost-effective method for upgrading the DCDB and other DCDB-dependant data layers such as utilities.

Gardner, 2006, outlines the desired spatial accuracy for both Wagga Wagga and Richmond Valley Shire Councils DCDB data. Tables 1(a) and 1(b), show the accuracy specifications for Wagga Wagga and Richmond Valley Shire Council’s DCDB. Note that they have varied desired accuracies dependant on the use and size of the parcel.

Land Use

Area

Accuracy

Urban

<5,000m2

0.1m

Rural Residential

5,000 – 25,000m2

0.25m

Rural Living

25,000 – 500,000m2

1.0m

Rural

>50 Ha

5.0m

Table 1(a): Table showing desired DCDB accuracies for Wagga Wagga Council (source: Gardner, 2006)

 

Land Use

Area

Accuracy

Urban

<5,000m2

0.1m

Semi Urban

>5,000m2

0.3m

Table 1(b): Table showing desired DCDB accuracies for Richmond Valley Council (source: Gardner, 2006)

 


Although there are a few methods of upgrading the spatial integrity of DCDB including transformation of blocks of DCDB data, also known as a block shift, the most common method is a parametric least squares adjustment.  This method is commonly used in surveying to adjust complex traverse networks (Merritt, 2005).

There are several steps involved in upgrading the spatial accuracy of the DCDB ranging from data collection (if necessary) right through to adjusting the DCDB using the method of least squares.  The process for upgrading the DCDB using a parametric least squares adjustment can be summarised in the steps below:

  1. Existing survey data is obtained - Deposited plans (DPs) and coordinates of any State Survey Control with connections to the cadastre can be obtained from the NSW Department of Lands.
  2. Coordinates of CRMs are obtained - if necessary, coordination of cadastral reference marks (CRMs), such as drill hole & wings, concrete blocks and GI pipes.
  3. Data Entry - both coordinate data of CRMs and the title dimensions obtained from the DPs are entered into a least squares input file.
  4. Adjustment - the parcel fabric is adjusted with the control from the coordinates of the CRMs.

The result is a cadastral fabric which has been upgraded to a particular coordinate system. This means that each boundary corner has an accurate coordinate which can be used to merge this data with other datasets on the same coordinate system.