Bushfire Prone Land Analysis


George Wong

Date, Revisions

Created on 6 January, 2017


ESRI ArcMap 10.1


Snapshot of analytical tools used to identify land parcels that are bush fire prone.


This is a snapshot of analytical tools used to identify land parcels that are bush fire prone.

The NSW Rural Fire Service produces a bush fire prone land map layer that is publicly accessible through it's web portal.

The purpose of the map layer is to inform the public on whether a property is designated bush fire prone. This may have a range of impacts to the planning, design, engineering, maintenance and monitoring of the land.

The mapping tool compliments other tools like the Section 149 Planning Certificate, which can be obtained from a Local Council. The Section 149 Planning Certificate contains all of the relevant statutory planning controls or development rules that apply to the land, of which the Council is aware of, at the date of publication.

Spatial Analysis

From a GIS perspective identifying land parcels affected by Bush Fire Prone Land layer consist of using a sequence of logical selections (add and removal). Depending on the geographic extent and number of parcels it may take a bit of time for the geoprocessing. ArcGIS Desktop can dedicate a CPU core and two threads for each background geoprocess. So having a beefy workstation will help speed up the calculations.

Start importing data layers to the workspace, symbolise as appropriate and more importantly order and use descriptive labels if the file names are generic. This is because you'll be referring to them in your selections window and it is confusing after a while with generic labels!

In the above workspace I have imported four basemap layers: land parcels, land parcel annotations, the LGA boundary and the pending subdivisions layer (additional requirement for the client). Two target layers have been imported. One is the original dataset with different categories of bushfire prone land. The "149 Certificates" layer is a single blob of the original dataset (features have been merged).

Start the analysis by finding the Select By Location... tool under ArcGIS Desktop's menu.

The selection method is Select Features From...

The target layer (being selected) is the Land Parcels layer.

The source layer (layer used to do something to the target layer) is the Bushfire Prone Land - 149 Certificates. In this scenario I wanted to use the single blob layer instead of the original source named Bush Fire Prone Land as the client is only interested in whether a land parcel is completely affected, partially affected or not affected.

As you can see from the above there are a ton of different functions we can use under spatial selection methods. Each one has its own section in the ArcGIS technical manual because each one is different in how it performs the query. If you require clarification RTFM. For this scenario, the first selection I'm using is Completely Within. So essentially I'm asking ArcGIS Desktop to identify all the land parcels Completely Within the Bushfire Prone Land - 149 Certificates blob layer.

There are over 10,000 parcels that meet the above criteria.

Next task is to identify parcels that are partially affected by the Bush Fire Prone Land layer.

Instead of Completely Within, we will go with Intersect. This instruction is to identify all parcels that are partially affected by Bushfire Prone Land - 149 Certificates blob layer. Intersect will actually get us both parcels that are Completely Within and parcels that only partly contain the source layer, so I would have to remove the previous selection.

With the parcels under the Intersect already selected. We can opt to remove from the currently selected features, thus eliminating the ones Completely Within the source layer. Below: Red are parcels completely within the source layer, orange are parcels only partly within the source layer.

The last can be obtained by mathematical deduction. If there are 20,000 records, and 11000 are completely within, and 5000 are partially within, then the remainder are not affected. Give or take 5% for statistical tolerances.

To do this, just run the second operation (above) again. Open up the Land Parcels table and run an inverse selection.

In this task I exported several tables containing land parcel information for the stakeholder. Maps could also be produced if it was required.


back to