Integrated Stormwater Mapping Across Local Government Areas in Adelaide, South Australia

A GIS was identified as the best way to manage the stormwater infrastructure and to help them to comply with the AAS27 requirement (A quality standard for the financial valuation and reporting of assets) for asset valuation and reporting.  The original data was collected by various means including GPS, aerial photography, heads-up digitising, and CAD.  This data contained a vast list of attributes many of which were not required for the project.  Many of the original fields related to asset management requirements, a lot of data was missing and attribute fields left blank. Errors were found in both the spatial and attribute data.  The process of consolidating the data was made difficult by replicated unique identifiers, missing links, missing data in fields, ambiguous data and conflicting fields. The data contained a number of errors.  Some were obvious, such as drains with pit attributes and box culverts with diameters.  The data entry errors were not so obvious but still had to be identified and corrected.   Our study suggested that data entry rules should be developed and applied to data entry to avoid some of these problems.

The data needed for this project consisted of the following:

• Nodes (Points) - Side Entry Pits, Grated Inlet Pits, Junction Boxes, Headwalls, Inlet and Outlet Structures, etc.

• Drains (Arcs or Lines) - Pipes, Box Culverts, Open Channels, etc.

• Features (Polygons) - Wetland, Gross Pollutant Trap, Lake, Weir, Dam, Trash Rack, Detention Basin, etc.

The data available from the councils at the start of the project can be summarised in the table below.

The first step in creating an integrated database was to collect all the attribute data and process it into single common format. The second stage was to then capture all the spatial information about the stormwater network of pipes and open drains. The spatial data then had to be integrated to create a contiguous system that allowed water to flow through it.

Uniform Data Dictionary

Because each council had their own method for storing information a single database that could store all the original data and the new attributes had to be created.  This ensured that no original data was lost and made provision for all the new attributes of the pipes to be used throughout the system.  Some of the components that needed to be captured were, size, construction date and the material used for the pipe. The system is also recorded features such as side entry pits, junction boxes and gratings, which have more of a point location, rather than the linear features of the pipes. Some council’s recorded both point and linear features in the same database creating difficulties with attributes like width and diameter. A standardised approach was adopted to handle all the different types and provide a useful final product.  The data dictionary developed for the project was an amalgamation of a number of different approaches to storing such data.  This dictionary format is now being requested by other councils to use as a tool for standardising their asset data collection.

Spatial Data Integration

The spatial data was then taken from the range of different digital systems and combined into ARC/INFO. The range of data types included the GIS packages ARC/INFO, ArcView, MapInfo, Maestro (Formida) and Latitude and also included an asset management package called Paramap.

The main exercise was to convert each data format into an ARC/INFO coverage and to get the attribute data into the INFO tables.

A new field called ‘accuracy’ was added for spatial and attribute data based on the capture method and the time since checking. Additional spatial data had to be generated to comply with the project topology and make the system flow.

Once the data was aggregated the topology had to be built to provide the modelling and tracing functionality described below. The process involved the following steps:

Allocation of points to network - Original data in two separate data sets. Arc/Info data model;

• Transfer Attributes - Several alternatives. Chose to transfer attribute data from old points to new nodes. Automated procedure;

• Perform Query to check transfer - Microsoft Access;

• Rectify Errors - Manual procedure;

• Development of Points Coverage - Some points couldn’t be allocated to network;

• Transfer data sets onto final CD.

Edgematching - Connect data at council boundaries;

• Identify Problems - Overlap (obvious replication). Gaps;

• Establish Rules - To ensure consistency - Data from native council. Join data at nodes. Migrate all elements to final data set. Maintain optimum attribute data;

• Join and Edit - Final step. Flag edits and additions.

Connectivity - Connect data within council boundaries;

• Identify Gaps - Two types of gaps: Missing elements and Data not snapped properly;

• Establish Limits and Rules - Where element was missing, new arc generated. Where snapping problem existed, original data edited;

• Edit Data

Flow Direction - Make sure nothing runs uphill. Arc/Info Network module

• Identify Sink - eg. Mouth of Patawalonga;

• Trace Source - What flows into it?

• Rectify Connectivity Errors;

• Many iterations

Figure 3. Example of Data Processing showing original data without Node Snapping

Figure 4. Final Stormwater Network Coverage for the Patawalonga

Figure 5. Detail of Stormwater Network Coverage

Applications of an Integrated Stormwater GIS

Once the database had been created and a single spatial system developed a number of different uses were possible.

For example modelling can be undertaken to:

• Assess the Stormwater System Performance / Capabilities,

• Determine Design Flows,

• Identify Critical Points in the System,

• Prepare Flood Mapping,

• Assess the effect of development within the catchment.

• Flood mapping (this would require - 3D Surface Model. Areal Extent vs Frequency. Flow Depths / Velocities. Travel Times.)

• The dataset could also be used for:

• Asset management

• Preventative maintenance

• Modelling and managing systems across council boundaries.

The following data was supplied on CD to the various stakeholders at the completion of the project.

• Drains - line coverage of the pipes and open channels etc

• Nodes - point coverage of the junction boxes and inlets etc

• Metadata - text description of the data

• ArcExplorer - Data Viewer

Figure 6. Trace Example through the network from a site of an Oil Spill

Figure 7. Trace Example through the system showing locations of two pollutants and their possible source

Figure 8. An Example of the Stormwater Network overlaid on an Aerial Photograph for Asset Management and Maintenance

9. Another Example of the Stormwater Network added to a 3D model for Flood Management

Conclusion

The construction of the Integrated Stormwater GIS has created an opportunity for

• Local councils to better manage their stormwater infrastructure.

• Emergency Services to respond faster and more effectively

• The EPA to trace the sources of pollution and model impacts on catchment changes and

• Demonstrated the value of an integrated data set and the importance of data standards

Acknowledgments

The project was funded by the South Australian Environmental Protection Agency, Planning SA, Catchment Management Subsidy Scheme, Transport SA and the Patawalonga Catchment Water Management Board.

The project was undertaken in collaboration with BC Tonkin Pty Ltd who were involved in the hydrology and data collection, and the City of West Torrens who performed data digitising and field data pick-up. The author wishes to acknowledge the assistance of Megan Tutty and Karen Hansen, at Mapping and Beyond, for their work in constructing the GIS database. Key people from BC Tonkin included Paul Lightbody and David McDonald along with Stephen Walker from the City of West Torrens.

References

Integrated Stormwater GIS for the Patawalonga Catchment, Adelaide, South Australia - Final Report. (2000). BC Tonkin Pty Ltd, Mapping and Beyond Pty Ltd and the City of West Torrens. Presented to the Environmental Protection Agency, South Australia.

Ian Overton
Managing Director
Mapping and Beyond Pty. Ltd.
33 Queen Street, Thebarton, South Australia, 5031
+61 8 8303 3841
+61 8 8303 5719
ioverton@mappingandbeyond.com
http://www.mappinganbeyond.com