What is Geoscape

Geoscape is a suite of digital datasets which represent buildings, surface cover, trees, provided by PSMA Australia. The organisation’s goal is to facilitate broad and sustainable access to high-quality location data, and we’ve partnered with their resell partner Omnilink, to bring this data for research use.

Geoscape includes the spatial coverage of Australia’s buildings using remotely sensed information. Detailed description below is from the Geoscape® Product Description by PSMA Australia (2018).

A building is a digital representation of the roof outline of buildings which have been classified from remotely sensed information using a combination of automated and manual processes to extract and orthogonalise building roof outline and identifies every structure greater than 9m2. These processes use electromagnetic radiation reflectance in the red, green and blue bands (visible bands) to classify pixels based on known patterns of signal combinations from various building roof materials.

The Trees dataset is a digital representation of areas which are classified using remotely sensed information as ‘trees’, generally using additional spectral bands. Where information is classified as trees, various digital surface model and digital elevation model sources have been used to derive each cell values height.

Surface Cover is a digital representation of the location of different types of surface cover which have been classified from satellite imagery. This theme contains two variations; one a high resolution dataset representing surface cover within urban areas, the other a national 30m resolution dataset representing surface coverage nationally.

Data quality and potential capture timelines will vary across Australia based on three categories: Urban, Remote Communities, and Rural Balance. Each category has been developed based on a number of factors defined by the population distribution, industrial/commercial activities and the probability of natural events (e.g. flooding). Population distribution are categorised based on population size. Areas with a population greater than 200, or with significant industrial/commercial activity in a visual assessment are defined as ‘Urban’.

For the purpose of CRC for Water Sensitive City’s tool, we have purchased data in the Urban category across Australia’s metropolitan regions - Adelaide, Brisbane, Melbourne, Perth, Sydney. This data was released on August 2018 (Release 7), and our team will incorporate new releases according to customer feedback. To find out more about Geoscape’s Product or roll out schedule, check out https://www.geoscape.com.au/rollout/.

How is Geoscape data applied

The Scenario Tool uses 2m Geoscape raster data to inform the baseline proportions of land cover assigned to each microclimate grid cell. The raster exists in a GeoTIFF format, and is employed within the metropolitan region of every major Australian city. The data is generated through classification of satellite imagery using automatic feature extraction routines, which has been assessed as being greater than 90% accurate.

Geoscape’s land cover data is categorised into eleven different groups. In order for the Scenario Tool to utilise the raster information, the groups have been translated based on what they represent so they conform to the model’s land cover categories. The table details what the raster values represent and how they were translated.

Table 1 Table 1. Geoscape land cover categories and their Scenario Tool equivalents.

Geoscape raster value

Geoscape category

Scenario Tool value

Scenario Tool category

Notes

2

Bare earth

5

Dry grass

3

Road and path

13

Road

4

Grass

7

Irrigated grass

5

Trees

1

Trees

6

Unspecified vegetation

7

Irrigated grass

Inspection of georeferenced satellite imagery showed that this category intersected irrigated open space e.g. recreational reserves

7

Built-up area

15

Concrete

8

Water

2

Water

9

Buildings

12

Roof

10

Cloud

N/A

N/A

Not considered by model; this does not effect output as grid cell fractions are made up of the remaining intersected land cover

11

Shadow

N/A

N/A

Not considered by model; this does not effect output as grid cell fractions are made up of the remaining intersected land cover

12

Swimming pool

2

Water

Access for commercial use

Our current license with Geoscape is limited. That being said, the Geoscape set has links to other PSMA products: G-NAF, CadLite (Cadastre and Property) and Administrative Boundaries (Suburb/Locality and Mesh Blocks). So if your organisation already have a license with PSMA Australia or their partners, you may have access to Geoscape too.

If you are interested in using the tool’s embedded Geoscape data for a commercial project, we can arrange for a quick quote with our partner who arranges the Geoscape data purchase, license, and support. The benefit of this is trustworthy data underpinning our models, and resource time saved on classifying the land cover. For a quote, please reach out to TAP team [mailto:admin@crcwsc.org.au] with the boundary file.

Alternatives to Geoscape

Classify land cover with QGIS: Follow our land cover prepration guide [https://www.wsc-design.org/doc/landcover_preperation.html] to create your own geotiff land cover classification file. Experience with QGIS is highly recommended.

Spatial GIS team for land cover: Try reaching out internally to teams responsible for GIS spatial data in your territory. It could be a quick conversion or editing existing data sets in your organisation to the land cover file specifications - geotiff file type set to WGS84 coordinate system.