The Weed Manager’s Guide To Remote Detection
The Weed Manager’s Guide To Remote Detection
Now searching for:
The potential of remote sensing for weed management is underexplored. The existing applications related to weed detection and mapping have been significantly limited to passive optical remote sensing in which visible and infrared light from the sun is utilised.
Although optical remote sensing systems are relatively cheaper and produce a more intuitive imagery, they can be significantly limited by cloud cover and are rarely useful for retrieving the structural attributes of plants.
The height, foliage volume, foliage orientation, leaf area and other morphological attributes of the invasive and native plants can readily be obtained using active remote sensing systems such as radio detection and ranging (RADAR) and light detection and ranging (LiDAR).
The RADAR system is unhindered by cloud cover and can provide regular wall-to-wall datasets for uninterrupted monitoring of our environment for early detection of weeds. The spaceborne RADAR remote sensing technology is growing, but the majority of the existing missions are commercial with the European Space Agency’s Sentinel-1 the only synthetic aperture radar data available at no charges.
Australia, through a collaborative program, owns a share in the operation of the NOVASAR-1 (a new satellite RADAR mission). A LiDAR remote sensing system is able to produce a 3D image of the target, making it possible to structurally discriminate between native and invasive plants in a heterogeneous landscape.
Satellite-based LiDAR missions are currently rare. The NASA Global Ecosystem Dynamics Investigation (GEDI) LiDAR is the first and only satellite LiDAR data available, but this may open up opportunities for similar missions in the future in which very high resolution RADAR and LiDAR datasets will be available for weed management.
The integration of plant multispectral characteristics with the structural information for discriminating between native and invasive plant species is emerging and expected to improve the current remote sensing methodology for weed detection. The availability of SAR and LiDAR systems efficiently functional on UAV platforms, too, is expected to widen the uptake of remote sensing for weed management.