DCDB Upgrades

The purpose of the DCDB surveys is to upgrade the accuracy of the existing Digital Cadastral DataBase (DCDB) onto the Map Grid of Australia (MGA94) Survey Datum.

This involves entering land parcel information by numerical entry methods into a software package and establishing on-ground accurate points to control the data set in a least squares adjustment.

METHOD OF SURVEY

Stage 1 - Data Input
The existing DCDB information and digital plan information (.TIF format) was provided to DSS by City Council.

Plan information was entered for each lot, starting with the newest plans first. This was to ensure the most recent (and hopefully most accurate) dimensions were entered for each line segment.

Dimensions were entered into the Spatial Adjustment Engine software against each registered plan number in order to provide easier error checking and statistics against each plan.

The simple and user friendly data input box is resembled by the box on the following page.

The bearing and distance information was entered for each line and, in this example, we can see that the values entered agree with both the current DCDB data (in the yellow highlighted area) and with the previously entered line data from another plan (in red half way down). This provides an on-screen check of the data and improves the least squares adjustment strength at adjustment time by having redundant measurements.

At the time each line is entered, and once all the plan information has been entered, the user can check if there are any large discrepancies between the entered dimensions and the DCDB dimension. This is a good tool for finding gross errors.

The information box for a typical plan is represented as:

We can see for this plan that the digitised points are not exact when we look at the DiffBear column against the DPBrng column. This shows that, for this rectangular block, the existing digitised coordinates are slightly in error. However, when we compare the length of each boundary (DiffDist), we find that the data entered is very accurate, and more correct than the original DCDB data.

Jumping a few steps ahead, once the adjustment is completed, we can now check the adjusted coordinates against the input coordinates for this same plan. Comparing against the box shown above, the following box shows us that after the final adjustment, the bearings of each line are now within about 30 minutes of the estimated accuracy of a rectangle (that is the DiffBear for each line against the average Rotation of the plan). Also, the DiffDist for each line is well within the required 0.1 metres of accuracy.

A statistical tool within the SAE allows us to query the whole database at the end of the adjustment and determine where any errors may lie.

Stage 2 - Survey Control
DSS utilise the GPS technology in various modes of operation; static, kinematic and real-time differential techniques. On this project, real-time and static methods were used for installation of survey control.

The real-time method of surveying involves setting a GPS receiver over a known survey station and emitting a positional correction via VHF radio link. This correction is received by all roving receivers and after a few minutes initialisation, the remote receiver‘s location is computed to an accuracy of better than 0.05 metres, based on the control station used.

The real-time method of GPS requires correctional data to be emitted via VHF radio from a known base station with accurate coordinates. A roving receiver is placed at the new station to be surveyed for a period of a few seconds up to a few minutes, until an accurate position is computed.

The roving receiver is moved, usually by vehicle, to the next station to be surveyed and the process is repeated.

During the days observations, numerous checks to existing control stations are made and verified in real-time to ensure data integrity.

The static method of observations requires numerous GPS units to log data for a common time period between stations. This time is usually 15 minutes plus 1 minute per kilometre of baseline length, but is almost always longer in time.

Post-processing of the GPS data is required and numerous checks are made within each network to ensure data integrity. In this case the data was processed and analysed using Waypoint Consulting’s “GRAFNET” software.

Checking, Adjustment and Final Data Delivery
By linking accurate ground control points to boundary points in the DCDB layer, a least squares adjustment can be performed. In an ideal adjustment, control points on the extremity of the area would be held fixed and an even spread of control points would be present throughout the data set.

For a minimal, first adjustment, key ground control points (ie permanent markers) are connected and an adjustment is run. By comparing the misclosures to other ground control points (iron pins and pegs) we can determine if the adjustment in that area is correct and accurate, or if there is an error in a portion of the dataset.

Once each area is proved to be correct, the remaining ground points are held fixed for a final adjustment, thus providing the most accurate representation of the DCDB. Riparian boundaries are adjusted by holding all final data fixed as control. This provides a ‘best fit’ solution based on the original digitised data for these boundaries. After the final adjustment is accepted by the Surveyor, the original data set (DCDB) is upgraded by writing adjusted coordinates for each point in the original data set.

There are commonly a few graphical errors in each data set of the DCDB. These could be any of the following:

Extra points
These points are located along a boundary line, usually representing easement pegs or line pegs, and need not be there for this project.

Incorrect Topology
These are parcel topology inconsistencies as compared to the registered plan lots. Also, where a corner has been amended, the DCDB may not have been updated to represent the correct information as stated on the plan.

Incorrect Parcel Naming
Where a parcel has the wrong “lot_plan” identifier attached.

In each of the these cases, some user editing is required in order to provide a clean and accurate data set. Information noted at the data entry stage process, on standard quality check forms, enables the data set to be modified after a final adjustment has been computed, using DSS In-house software PolyEdit Version 1.0.
Using MapInfo Version 7.8, there are three more quality checks that are run to ensure data integrity.

The first of these is a self-intersection test which identifies if any line in a parcel crosses itself. This is simply represented by the following diagram.

The problem most likely occurs when the Surveyor is editing the data, deleting points and adding new points. This is simply fixed by swapping the order of the points in the polygon.

The second quality check is to check for any overlaps in polygon data. Again, this is easily resolved by adding or removing points in one of the affected parcels.

The third quality check is to check for any gaps between polygons. Again, this is resolved by adding or removing points in certain polygons.

The output of a final data set is now produced.

DSS provides final data in MapInfo format, containing both the DCDB and SCDB layers.

Recent DCDB upgrades include:

> Rockhampton City Council > view images