For the optimum exploitation of commercial forests, it is necessary know the volume of the standing trees as precisely as possible. The problem: some morphologically 'extreme’ trunks aren’t easy to measure using traditional techniques. Therefore, more sophisticated tools are required, capable of creating three-dimensional models of the trees. Thanks to the work of a young PhD student in the Management of Forest Resources and Natural Environments Unit (University of Liège), one of these tools – the ‘terrestrial LiDAR’ – is being tried out for the first time in a afrotropical rainforest. A more than promising advance, at a time when the ecosystem services provided by our forests (climate, biodiversity, etc.) are continuously being promoted.

We can’t prepare a feast without having first noted what’s in the larder... While keeping everything in proportion, what is valid in the culinary domain is also valid in forestry. It is indeed impossible for any self-respecting forest manager to assess the value of their property if they haven't first identified the species present in their forest, as well as the age, thickness (diameter), height and volume of the trees in it. To carry out this type of approach (specialists speak of an ‘inventory’), the foresters in our regions have classic dendrometric tools such as a compass or – more basic still - a simple tape. After having determined the circumference of a tree with a standard height (most often 1.3 metres), suffice it to refer to ‘volume equations’ (also called ‘volume tables’) which allow them to instantaneously obtain the volume of the tree measured.

And that’s it? Not quite! These ‘volume tables’ may have played the role of ‘bible’ for entire generations of foresters, but they are nevertheless limited, especially in the case of morphologically irregular trees. Try measuring the diameter at breast  when it is covered in flutes or buttresses! In forestry, flutes can extend across the entire length of a trunk, from the ground to the first main branches.  Sometimes, far more spectacular buttresses appear as oblique ‘roots’ coming straight out of the ground. Seldom present on the majority of trees in our temperate forests, these morphological singularities can be very marked on tropical rainforest trees, to the point of making it impossible – the error rate can be up to 30 %! - to take a classic dendrometric measurement.

Herein lies the interest of having more sophisticated tools, such as the ‘terrestrial LiDAR’. The LiDAR (Light Detection and Ranging System) is a small electronic device – about the size of a shoe box – mounted on a tripod. Thanks to a mirror that turns at high speed, it emits a laser beam in all directions. Based on the echoes emitted by the objects encountered, it is possible to form a point cloud in 3D representing a tree of an imposing size or a group of trees in the same forest plot (i.e. a surface of approximately 15 to 20 metres long and wide).  “Originally used in the field of architecture and industrial geometry, LiDAR started being fully used in the domain of forest research only about ten years ago", explains , research assistant in the Management of Forest Resources and Natural Environments Unit, Gembloux Agro Bio Tech (ULg). “But while it is starting to be successfully used in Scandinavian forests, the same isn’t true of tropical rainforests where its use has yet to be proven”.

Sébastien Bauwens is one of the first researchers in the world – if not the first – to attempt to use the LiDAR on species in the African tropical rainforest. One of the first steps of his doctorate, begun last year, consisted of testing the device on species listed under the vernacular names ‘Ayous’ and 'Fraké', well known in Central Africa. “The interest of LiDAR is that it can give a very precise estimation of the tree trunk’s volume without the tree being felled”, the young researcher points out. “This accuracy, down to the nearest millimetre, extends up to the first big branches. Above that, i.e. at the level of the crown, it’s more difficult because several ‘obstructions', associated with the presence of foliage, prevent us from getting as much information as for the trunk”.

The precise estimation of the trunk’s volume is of course what interests forest managers above all. Multiplied by the number of trees in their forest, this type of information will allow them to optimally plan its exploitation. Another parameter that interests foresters is the estimation of the trees’ growth: in other words, a sort of anticipation of the way in which they will grow over time. LiDAR is helping to develop ‘taper functions that can be useful for both forest research and to plan logging, because they will allow users to correct the estimation of the growth of trees with buttresses. “In general, a trunk bears more resemblance to a cone or the trunk of a cone than a cylinder. The taper shows the reduction in the thickness of the trunk when you move away from the ground towards the branches”. In the case of trees supported by considerable buttresses, foresters are forced to measure the circumferences at a height well above 1.3 metres, sometimes several metres high! “The second objective of my thesis, with regards to species with particular morphologies, therefore consists of ‘correcting’ the diameter measurements taken at different heights to correctly estimate the growth in diameter of the trunk at a given height. This correction is done using several taper functions that take the buttresses into account”.

Since the beginning of his PhD, Sébastien Bauwens has been testing the LiDAR’s positioning and handling techniques in situ. A first trip to the logging company Alpicam in South-East Cameroon took place in April 2013, and another to the area surrounding Kisangani (Democratic Republic of Congo) in August 2013. But the majority of his work is done in the office. It primarily consists of linking the different point ‘clouds’ relating to each positioning of the scanner around the tree being examined, in order to obtain a single point cloud. Then the crucial step of determining the volumes begins, through the elaboration of sophisticated algorithms. “This comprises the majority of my work. For instance, I have to be able to make virtual cuts at different heights using these algorithmic tools, in order to determine the basal area (i.e. the surface of the section of the trunk) for these heights”. 

The terrestrial LiDAR has great potential in the area of forest research and forest management. Its implementation in the field may be more laborious than some traditional measurements taken in forests. But the amount and the precision of the parameters that can be extracted from these LiDAR scans can compensate for the time required to scan a tree or a plot composed of several trees. “The processing of the LiDAR data remains the sticking point in this technology. But, once the processes have been automated, it will be easier to do the processing on a computer and make it directly accessible to foresters”.

To foresters, yes, but not only to estimate the commercial value of the log. “In the last few years, there has been growing interest in the scientific and political world concerning new forest stakes, essentially linked to climate change. LiDAR could also be used to provide a better assessment of the total biomass of the trees: not just the trunk, but also the branches and the foliage. Using the values obtained, we could more easily assess the role of tropical trees as carbon ‘sinks’ thanks to photosynthesis. And, from there, better acknowledge certain forest management models within the framework of the carbon credits markets, attributed to countries that make an effort to fight deforestation”.