3D image of an area of the Litchfield Savanna Supersite. Image generated by the Lidar and supplied by Dr Peter Scarth.
There have been some interesting goings-on at Litchfield National Park over the last fortnight. Just off the strictly 4WD track to the Park’s eerily beautiful “Lost City” people can be observed engaging in a diverse range of unusual activities; from slingshotting down tree branches, to partaking in odd circling dances with pirouetting, laser-emitting machines, to sweeping the landscape with “pushbroom” hyperspectral scanners in high-tech-laden aircraft. These people are part of a group of 23 researchers from institutions and organisations across the country, currently descended on the area for one loosely linked, jointly coordinated field trip. They are taking measurements for a variety of projects and purposes, but all contribute to developing the area as a new supersite (the Litchfield Savannah Supersite, or SSS) in the Australian Supersite Network.
RIEL’s Associate Professor Stefan Maier is at the helm of one of these projects for his current research in conjunction with AusCover. Stefan endeavours to understand the impact fires have on the carbon sequestration potential of the high rainfall tropical savanna forest – the dominant ecosystem found in northern Australia. While Stefan has been monitoring the site for some time as part of his research, the aim of the current field trip is to take extensive on-ground field measurements to calibrate and validate remote sensing of the site’s biomass. In other words, to help accurately interpret aerial and satellite imagery of the area’s vegetation.
“There’s a lot of variation between the ground and the remote sensor, due mainly to the influence of atmosphere,” RIEL Research Fellow and CDU remote sensing lecturer Dr Karen Joyce is one of the researchers assisting in Stefan’s project – and incidentally excellent at delivering impromptu crash courses in the technical art of remote sensing.
“To cater for this variability we go out on the day there’s something flying over – satellite or aircraft,” she explains. “We make the measurements on the ground at the same time that the satellite or aircraft are making theirs, so we can look at the difference between the two measurements and identify anomalies in the remote sensing data, putting them down to things like clouds or haze or gases in the atmosphere. Then we can calibrate based on that. The end result is that your remote sensing image represents exactly what’s on the ground.”
There is a veritable swathe of field measurements occurring for the purposes of calibration during this field trip. Dr Andrew Edwards of RIEL’s new Darwin Bushfires Research Centre is involved in measuring the area’s biomass:
“We’re measuring things like leaf area index, projected foliage cover, canopy cover, ground cover, the biomass of trees in the area,” Andrew lists, before hopping on a quad bike to get to the more difficult-to-reach areas of the SSS. “Such data is directly tied to carbon sequestration potential. For example the density of foliage relates to the level of uptake of carbon dioxide from the atmosphere, and a tree’s biomass indicates how much carbon it can store.”
Meanwhile, in a more accessible neck of the savanna forest, a machine that looks rather like R2-D2 affixed to a pole is chirping and whirring. In fact the device is called a Lidar, and is currently recording a holistic 3D image of the area being measured by various other devices, among them a TruPulse laser rangefinder and a humble roll of measuring tape. The scientists currently manning the Lidar are CSIRO’s Mick Schaefer, QLD government’s Dr Peter Scarth and Stefan’s research associate Miguel Tovar-Valencia. They gesture at more hapless onlookers to move out of the way of the Lidar’s laser gaze so they don’t obstruct the targeted features of the scan.
“The Lidar takes a 360 degree laser scan as it rotates around, then we flip it on its side to do another rotation to fill up the hole in the top of the scan,” Mick explains. “Then when we get back to the lab we put the two scans together to get the whole hemisphere of the area in one 3D picture. The Lidar is good because it gives you a whole site characterisation which is generally much better than photographs because you can measure certain things like the basal area of trees in the site.”
Back at nearby Litchfield Super Site HQ Stefan is looking at some of the aerial images taken that morning. With him are pilots Dr Andrew McGrath and Wolfgang Lieff of Flinders University's Airborne Research Australia. The three have just returned from a three-and-a-half hour flyby shooting of the SSS, flying directly above the area along 41 parallel routes spaced 125m apart, taking both “normal” photographs (showing images in the 3 bands of light – the red, green and blue bands visible to the human eye) as well as collecting hyperspectral images that depict objects in 250 wavelength bands of light reflecting off them.
“From the hyperspectral image you can look at an area’s spectral signature,” Karen explains. “Spectral signatures are shaped by the different amounts of light reflecting off a physical target in each of the 250 different wavelength bands. From a spectral signature you can work out the health and density of vegetation in an area, which subsequently gives you some view into the potential that area has for sequestering carbon.”
Over at the Batchelor airstrip, Andrew and Wolfgang provide a little more insight into remote sensing, demonstrating how their sophisticated camera system works up in the air.
“We normally fly with somebody driving the aeroplane and the other person driving the instruments that operate the cameras,” Andrew explains. “Each of the big pods under the aeroplane’s wing contains a hyperspectral camera with its own computer, and we control them by running a remote desktop from those computers in the pods to this one in the cockpit, in much the same way that the IT guy fixes your computer problems from his desk in a totally different building.”
Andrew and Wolfgang will be using two separate aircraft to contribute data to Stefan’s research. The second plane is almost identical to the current one, except for the equipment installed in its wing pods. Instead of a hyperspectral camera there will be another Lidar building up a 3D picture of the vegetation in the SSS, helping to tie together field measurements and hyperspectral scans.
“The measurements taken from the field are indispensible,” Stefan says of the field trip. “They will help ensure the accuracy of interpretations of future remote sensing of this type of high rainfall tropical savanna forest, which is in turn critical to designing future methodologies to manage our natural resources effectively.
“If we can accurately gauge and monitor the carbon sequestration potential of the SSS via remote sensing, we can then use this technology to inform or recommend current and future policies and practices for such forests across northern Australia. For example, the carbon farming initiative will benefit from our assessments into what extent such forest is a good carbon sequestration option. Similarly, resource management strategies like burning regimes can be developed from findings, namely the impact of fire on the forest and the role it plays in the area as a carbon source or sink.”
The next stage of Stefan’s project will see infrastructure in the form of 40m flux towers being installed at the SSS. The towers will be measuring exchanges of gases, water and energy in the area and will tie into the rest of the data currently being collected.
To find out more about Stefan's project, have a read of this story from TERN e-news »