Wakool-Murray Lidar, Imagery and Derived Products 2015  

Authors / CoAuthors

Abstract

High resolution elevation data covering an area of approximately 12,600 sq km in the Wakool-Murray catchment area. The survey is situated along the Victoria and NSW border, located between Yarrawong, Echuca, Robinvale and Moulamein and Deniliquin. A set of products were produced including classified lidar point clouds, hydro-flattened bare earth DEMs, DSMs, Canopy Height Models (CHM) and Foliage Cover Models (FCM). The outputs of the project are compliant with National ICSM LiDAR Product Specifications and the NEDF.

Product Type

dataset

eCat Id

110282

Contact for the resource

Keywords

( {1} )

• HVC_144648

( Product )

• Lidar

( Product )

• Elevation

( {1} )

• Published_Internal

Publication Date

Creation Date

2015-10-01T00:00:00

Security Constraints

Australian Government Security ClassificationSystem    

Classification - unclassified

Legal Constraints

Creative Commons Attribution 4.0 International Licence    

Access - license

Use - license

Status

Purpose

High resolution elevation data for the Wakool-Murray region.

Maintenance Information

notPlanned

Topic Category

geoscientificInformation

Series Information

Lineage

Unclassified LAS: The IMU and post processed airborne GPS logs were used to generate the raw LiDAR point cloud from the waveform instrument data; one LAS v1.4 file per swath. Each unclassified LAS file is associated with a waveform data packet file, having the same name and the extension .wdp. No adjustment has been applied (automatic or manual) to this data. Classified LAS: Raw LiDAR swaths were levelled to establish internal consistency, merged and 1km x 1km tiles in LAS v1.2 format were created. An automatic classification algorithm was applied in TerraScan software to produce an initial classification of ground (2) and unclassified (1). Low and high noise points were automatically classified and allocated to class 7 and 18, respectively. The ground classification was improved manually by visually scanning the ground surface and reassigning points from ground to unclassified to remove spikes and by assigning unclassified points to ground where the ground surface lacked sufficient detail to describe the terrain (i.e. large TIN triangles). The classification of ground points was to the ICSM level 3 standard (99% accuracy with respect to ground points). On completion of the ground classification, automatic algorithms were used to classify unclassified above ground points to ICSM Level 1 specification using classes low vegetation (3), medium vegetation (4), high vegetation (5), buildings (6), water (9), transmission towers (15), and bridge (17). No transmission towers were identified in the project area; however, power poles were included in this class where identified. Water and bridge points were then manually reviewed further. Water polygons were mapped by reference to the LiDAR and then reviewed with the imagery. Within each water polygon, all points classified as ground were reclassified to water. Islands were also mapped, ground points being confirmed and retained. During this interactive process, ground points on bridges were reassigned to class 17. Classification of water and bridges was performed in accordance with ICSM Level 4 specification (99% accuracy). The classification accuracy was not measured empirically. For AHD format, las data were transformed from ellipsoid to orthometric using the Australian Gravimetric Quasi geoid model 2009 (AGQG 2009) which was provided as a surface by Geoscience Australia. The data was then updated to Las v1.4 format. The classification accuracy was not measured empirically. Hydroflattened Digital Elevation Model: A DEM with a ground pixel spacing of one metre was generated from the LiDAR ground points (Class 2 only). The method of deriving the height for each grid point is by triangulation between the three nearest LiDAR ground points, and utilising a TIN to Raster process with Natural Neighbour interpolation. Hydroflattening was then applied using waterlines (break lines) that were manually digitised using the LiDAR intensity image and orthophotography as reference for all visible water in any part of a connected river network or recognised waterway (regardless of size) and all other waterbodies greater than 625 m2. The hydroflattening process considers the heights of the waterline at each vertex along each bankline, assigns the elevation from the DEM to the vertex and levels the water height perpendicular across the river. The process enforces hydrologically correct flow to ensure the river always slopes downstream.The DEM was exported to ESRI Binary grid as 2km by 2km tiles. The DEM accuracy will be limited by the spatial accuracy of the LiDAR point data and will contain some additional error due to interpolation, particularly in areas of dense vegetation where ground points are sparse. There may also be some minor error due to ground point misclassification. Digital Surface Model: A DSM with a ground pixel spacing of one metre was generated from the LiDAR mass point data. Height is assigned as the highest non-ground point falling on each pixel utilising the Classes 2, 3, 4, 5, 6, 9 and 17. Pixels which have no LiDAR points falling on them are assigned a height value from the interpolated DEM. The DSM generation employed a TIM to Raster process with Natural Neighbour interpolation. The DSM was exported to ESRI Binary grid as 1km by 1km tiles. The DSM accuracy will be limited by the spatial accuracy of the LiDAR point data. Pixels upon which no LiDAR points fall are assigned from the interpolated DEM and therefore will include some interpolation error. Canopy Height Model: The CHM (canopy height above ground) surface was produced by selecting the highest LiDAR point present on each two metre cell with classification of low, medium, or high vegetation (4, 5, or 6), and assigning elevation. The DEM was resampled to two metre pixels using bilinear resampling and subtracted from the vegetation elevation surface to form the CHM. The CHM was exported to ESRI grid as 1km by 1km tiles in accordance with ICSM specifications. The CHM was checked visually and the CHM, DSM and DEM were compared to verify the arithmetic. Fractional Cover Model: FCM with a ground pixel spacing of ten metres was generated as one minus the gap fraction probability as defined by the proportion of counts from all vegetation returns at least 2m above the ground expressed as a percentage (%). The FCM was exported to ESRI Binary grid as 1km by 1km tiles in accordance with ICSM specifications. The accuracy of the FCM reflects the accuracy of the LiDAR vegetation classification. Intensity Images: Intensity image with a ground pixel spacing of 50cm was generated from the LiDAR point cloud using an RPS algorithm that calculates the intensity image as the mean intensity of the ‘first’ and 'last' return points falling on each pixel. The first intensity image was provided as an ESRI new raster mosaic dataset with 16 bit greyscale 1km x 1km tiles. The intensity image accuracy will be limited by the spatial accuracy of the LiDAR point data. Aerial Imagery: High end commercial photogrammetric software and Aerometrex proprietary software was utilised in the orthorectification and mosaicing process. This methodology assisted the production of a seamless georeferenced orthophotography product. The imagery was provided as an ECW mosaic and as 1km x 1km geotiff tiles.

Parent Information

Extents

[-36.18157857839976, -34.587338150423655, 143.17382812499994, 146.35986328124994]

Reference System

GDA94 / MGA zone 54 (EPSG:28354)

Spatial Resolution

1 Metre grids, 4 points per square metre lidar

Service Information

Associations

Downloads and Links

Source Information