Remote Sensing Research Group

Southern Earth Observatory

April 18, 2018
by e29293
0 comments

Estimating wildfire fuel hazard in dry sclerophyll eucalypt forests from UAV-borne sensing technology  

Classifying and quantifying fuel is complex and includes assessing a range of characteristics.

Fine fuel load has been considered one of the most significant fuel variables affecting the behaviour of fire and has been used to predict the rate of fire spread.

Tasmanian Ridgeway Site – dry schlerophyll forest with recent burn history.

 

Limited research has been conducted using UAV mounted laser scanners and cameras for the purpose of Structure from Motion (SfM) to measure below-canopy structure. Preliminary studies in native Australian forest stands have further demonstrated the use of UAV laser scanning and SfM to estimate various structural properties of a native stands. A research gap exists in assessing the accuracy of point clouds generated from UAVs in measuring surface and near-surface vegetation characteristics.

In order to address this literature gap, fieldwork (sponsored by the Bushfire and Natural Hazards CRC), was conducted at a patch of native dry sclerophyll eucalypt forest located southeast of Hobart, Tasmania, Australia. The objective of this work was to compare the accuracy of UAV SfM and UAV laser scanning in assessing height and cover. Direct measurements were taken within the plot as reference data. Terrestrial point clouds were also captured using terrestrial laser scanning and SfM as a further point of comparison. Analysis will commence in comparing these different forms of measurement to investigate the viability of UAV SfM and laser scanning in assessing fine fuel load.

 

Setting up direct measurement instrumentation.

Recording direct measurements of vegetation height and cover.

Research Fellow operating Terrestrial Laser Scanner in Tasmania

 

 

 

 

 

 

 

 

 

 

 

 

 

Researchers were given a tour of the TerraLuma – UAS Remote Sensing lab at the University of Tasmania

The collaboration used the UTAS UAS with airborne LiDAR sensor.

 

 

Text by: Samuel Hillman

April 6, 2018
by e29293
0 comments

Mapping fuel hazard in Chile

A group of academics and HDR students from Geospatial science recently returned from field work in Southern Chile, exploring the use of Fuels 3D techniques in Chilean forests.

This, RMIT – University of Chile, collaboration extends the Bushfire and Natural hazards CRC project Fuels 3D, led by Professor Jones and Dr. Reinke, into the international realm. The Fuels 3D project has mapped and characterized fire landscapes in S

Pre-fire point cloud

E Australia, with these techniques being actively trialed by several state fire authorities and land management agencies in Australia. Chile faces similar wildfire threats to Australia and these techniques are of great interest to Chilean forestry managers and fire authorities as they develop risk assessment and management plans for large fires.

Post-fire point cloud

 

 

 

 

 

 

 

 

The group used terrestrial and UAV (RPAS / drone) based photography to capture 3D point clouds of plantations and native forests in Southern Chile. These techniques, developed by research fellow Dr. Luke Wallace, allow the reconstruction of 3D point clouds that can be used to extract fuel and vegetation structural information. As part of the research, the integrity and completeness of the point clouds is validated through intensive and extensive physical measurements of the forest structure using laser scanners and sample transects.

The field team: Franco Magni (U-Chile), Sam Hillman (RMIT), Liliana Guzmán (U-Chile), Chris Bellman – Simon Jones – Bryan Hally – Daisy San Martin (all RMIT), Jaime Hernandez (U-Chile)

Whilst in Chile the group presented at an international seminar “Percepción remota en manejo de bosques, conservación biológica e incendios forestales” at the University of Chile.

 

Drone view from Estación Experimental Justo Pastor León, mega fire February 2017 destroying the town of Santa Olga

 

Field site Pa-1, pine plantation burnt in recent fires

Site Pa-3 (unburnt pine plantation)

 

 

 

 

 

 

 

 

 

Native forest site Fr-1. Up to 100 overlapping downward looking images captured from a drone, such as this over a Chilean Native forest, can be processed to generate key fuel hazard metrics such as canopy base height, canopy cover and tree height.

 

 

Vulcan Osorno lava field site (Os-2)

 

 

Vulcan Osorno with details to the field location

 

 

 

Vulcan Osorno lava field site (Os-2)
The lava flows can be accurately dated to 1835 as observed by Charles Darwin on the 2nd voyage of the Beagle

 

Further information: http://www.bnhcrc.com.au/resources/poster/2021

Text by: Simon Jones
Skip to toolbar