In December 2004 Geoscience Australia received the scheduled cadastral update from Tony Yeomans of the Department of Land Information (DLI, formerly DoLA). This update consisted of two ESRI shapefiles and two access databases.

One of the ESRI shapefiles contains current property boundary information and the other contains lodged changes to property boundaries. Both ESRI shapefiles arrived at Geoscience Australia pre-projected in UTM (GDA94).

The reason for differences between the lodged cadastre and the current cadastre is that current data could have been calculated or digitized, etc. When lodged data is integrated into current, the area is spatially upgraded with the new survey data (lodged data) as this would be more accurate.

One access database contained data relating to tenure information including address. The other contained data on reserves.

In January 2005 Geoscience Australia also obtained ground audit data from Peter Brockmulen of the Department for Planning and Infrastructure (DPI, formerly DoLA). This ground audit was conducted in 2004 at the request of the Department of Territories and Rural Services in Western Australia. It contains data describing land use and reserve usage.

To produce the final cadastral information the files were modified and combined as follows.

Due to initial consultation with users of the GIS it was decided to keep the cadastre containing current property boundaries and the cadastre containing lodged changes to property boundaries separate. As such the following changes were made to each file separately. Additionally the cadastral boundary for North Keeling Island, taken from the last version of the cadastre, was joined to the current cadastre shapefile.

From discussion with users of the GIS it became apparent that the field LOT_TYPE was causing confusion with it's meaning. This is because it is a computer generated field that does not reflect the tenure of the land. Instead land tenure is derived from the PI_PARCEL field. To avoid this problem Geoscience Australia removed the LOT_TYPE field and replaced it with a TENURE field derived from the PI_PARCEL field and based on the definitions from the NORM manual as supplied by DLI.

Using tenure information from the database supplied by DLI and from the land audit information supplied by Peter Brockmulen, Geoscience Australia joined on the fields ADDRESS, ZONING, RESERVE, PURPOSE and MANAGEMENT were joined on lot number. For multiple instances of lot numbers information was entered manually based on the land audit information supplied by Peter Brockmulen. From reserve data supplied by DLI the fields RSV_VESTIN, RESERVE_US and LGA were joined using the PIN field.

Finally Geoscience Australia made the decision to move the cadastral data to be in line with previous data such as orthophotography. As such the following xy shift was used on both the cadastre and the lodged shape files:

• X+3.26m
• Y-1.04m

In the 'cadastre_dli' folder of the cadastre directory is included the lodged and cadastre shape files as projected by DLI but with all data joined.

The following steps were carried out to merge the cadastre with previously collected ground audit data:

1. The cadastre was supplied to Geoscience Australia in two halves. One contained leaseholds that were current and the other contained leaseholds that had been lodged to be changed. A new field was added to each cadastre called 'legal_stat' standing for Legal Status. In the cadastre of current leases this was set to 'final' and in the cadastre of lodged leases this was set to 'lodged'.
2. The lodged and current cadastres were combined to form a full cadastre of the Cocos (Keeling) Islands.
3. This new cadastre and the previous version's cadastre were linked via polygon pin number. This allowed GA to separate those polygons which were had different pin numbers in the new cadastre.
4. Those polygons which had not changed were joined to the additional data on pin number.
5. Polygons which had new pin numbers were joined to the ground audit data with a spatial join.
6. The tables containing the unchanged polygons and the new polygons were then merged to create the current cadastre.
7. Duplicate fields in the cadastre were removed preserving the latest data available.
8. This new cadastre was then incorporated into the Arc View and Arc Explorer projects.

These changes were made using Arc View.