2 Provision of base maps for Torbay Catchment Restoration Plan & other purposes

2.1 Introduction

The required datasets were identified by the tender brief, but many required additional development. These are detailed in this section. A list of base datasets used is provided in Appendix 1.

The base maps produced are listed in Appendix 2.

Catchment hydrology

Previous work had prepared a complete hydrological coverage for Torbay, including stream ordering, some stream naming, and separation of streams from other features such as lakes, wetlands, areas subject to inundation etc. This was based on the 1:25,000 topographic mapping of the area.

Significant additional work was required to amalgamate three separate stream surveys for Five Mile Creek, Seven Mile Creek, and Torbay Main Drain catchments. This created a single shapefile out of the three, with a single consistent classification of stream foreshore condition, and existing and recommended fencing. In addition, the line-work was cleaned, connections corrected and a consistent stream order classification carried out.

1 PEWOC - Policy Evaluation in Wilson and Oyster Catchments - Agriculture WA project finished in 2002.

Subcatchment boundaries had previously been drawn as part of the PEWOC¹ project, but these were based on a regional-scale DEM. Due to the significant modification of the waterways in the lower catchment, plus drainage works, much of the PEWOC routing was inaccurate or even quite incorrect. Re-drawing and significant changes to reflect the actual stream/subcatchment routing were required. The basic subcatchments were therefore re-classified - in consultation with DoE staff - and in doing so we used a classification that also matched existing gauging stations. Final subcatchments were as follows:

Note that the Cosy Corner subcatchment (previously called 'West Torbay') actually drains to the sea, not to Torbay Inlet as assumed in the PEWOC modelling.

An additional classification was made to divide the two largest subcatchments - Marbellup Brook and Torbay Main Drain - into smaller subcatchments for the purpose of modelling nutrient assimilation in the creation of nutrient risk maps. Draft names have been given to these, as show in Table 2.

Catchment soils and geomorphology with stream network overlay

The Department of Agriculture has created a soils/geomorphology dataset which covers the entire south-west, through their land resource mapping (Van Gool et al. 2005). This is used for a range of land assessment purposes, including nutrient loss risk and land capability assessment. It was felt that this is the best available soil/geomorphology mapping, and was used as is.

Catchment contour map stream network

Contours were extracted from 1:25,000 topographic map sheet datasets.

Remnant vegetation

A map showing the estimated original vegetation dataset was available for the entire catchment, as well as a remnant coverage based on this. This provided a basic vegetation classification for the catchment as shown in the table below.

Land use - map shows general land-use categories with stream network and roads

The only landuse map provided for this project was produced for the PEWOC project from a range of data sources, including satellite image interpretation and the National Land and Water Resource Audit (NLWRA). Both of the latter were produced at large scales, and consequently contain significant errors when applied at the small catchment scale. In addition the data sources dated back to at least 1998, so were up to 8 years out-of-date. Hence the PEWOC landuse map contained many significant inaccuracies, and required updating or replacement.

A new aerial photo mosaic (flown in 2004) was supplied from SCRIPT, and after checking the PEWOC Landuse map against this we decided to completely re-do the land use mapping. A number of sources of information were used:

• Aerial photographic interpretation;
• Existing vegetation (remnant vegetation map)
• Known land uses for specific parcels (sourced from previous surveys, survey of landholders, and the cadastral coverage);
• Cadastral parcel size matching (for urban/peri-urban land use); and
• Discussion with contractors/staff of DoE who have spent a lot of time on the ground.

The initial GIS process involved a union of the cadastre and vegetation coverages, to provide a base division of the catchment. Unless errors were found, the remnant vegetation classification was left as was - called 'remnant vegetation'.

Land use easily identified from the aerial photographs included plantations, annual and perennial horticulture, and some special uses such as the irrigated wastewater plantations.

Very small parcels (generally less than 0.15ha) containing a house were classified as 'urban', while small parcels (generally less than 5ha) containing a house, pasture, and plantings of trees etc were termed 'peri-urban'.

Known land uses included existing dairy farms and piggery, as well as some grazing use from a landholder survey (Duxbury 2005).

Cadastre - base map

An existing file was used for cadastral coverage. Note that some of the available cadastral shapefiles contain duplication of many of the parcels due to dual ownership, which precluded their use.