In ecological research, the sustainability of data collection and biodiversity assessment plays a key role in gaining a comprehensive understanding of our natural environment. Torresani was recently awarded the prize for the best presentation by a young researcher at the International Conference on Community Ecology in Trieste. Torresani's presentation at the ComEc conference, entitled 'Exploring Biodiversity Patterns in Forest Ecosystems Using LiDAR Remote Sensing Data: Unraveling the Influence of Forest Height Heterogeneity', focused on the sustainable and innovative approach to measuring biodiversity by combining remote sensing information and artificial intelligence, with a particular focus on forests and woodlands. In today's era of technological advancement, the field of ecology has witnessed a significant change in data collection techniques. Traditional methods were often challenged by external factors that could affect the accuracy and reliability of the data. However, with the advent of remote sensing technologies, such as satellite-based systems and drones, and the integration of artificial intelligence, researchers now have access to more consistent, less vulnerable and highly advanced data collection methods.
From pixels to preservation: Michele Torresani's LiDAR vision for biodiversity
Michele Torresani's award-winning presentation presented an innovative approach to biodiversity assessment using LiDAR remote sensing data. LiDAR, a remote sensing technology, relies on a sensor to measure distance. When mounted on an aircraft or drone, it calculates the distances between the sensor and objects and uses laser pulses to create detailed 3D models of terrain and vegetation. This advance greatly improves the accuracy of estimating structures in forests and grasslands. Torresani's novel "height variation hypothesis" is based on the strong correlation between ecosystem height heterogeneity, as measured by 3D LiDAR data, and biodiversity. Put simply, areas with significant height variation, as determined by 3D LiDAR insights, exhibit greater environmental diversity, leading to higher species richness and diversity. The height variation hypothesis marks a significant departure from traditional biodiversity assessment methods, which often rely on indirect and imprecise measurements, making it difficult to understand the intricacies of ecosystems. Torresani's methodology goes further by incorporating artificial intelligence algorithms into LiDAR data analysis, allowing for finer resolution in biodiversity assessment. This approach represents a transformative shift in how ecologists and researchers approach biodiversity assessment, ultimately contributing to a more comprehensive understanding of ecosystems.
From canopy to space
The height variation hypothesis offers significant potential for advancing our understanding of biodiversity and the ecological complexity of diverse landscapes. Torresani tested this approach not only with LiDAR data derived from drones or aircraft (which limits the analysis to the flight of the drone/aircraft) but also with GEDI LiDAR data, a LiDAR sensor installed on the International Space Station that provides an open-source estimate of forest height on a global scale. In addition, he extended the analysis beyond assessing vegetation biodiversity to include bird biodiversity, with results showing that greater structural variation in forests, as estimated by LiDAR data, correlates with increased forest structural heterogeneity and bird diversity.
In a local context, Torresani also demonstrated the practical application of the proposed approach to the role of forest stability during the VAIA storm. The research showed that the forest areas affected by VAIA had similar elevational heterogeneity to the areas unaffected by the storm. This underlines that the forest loss caused by the storm wasn't due to weak forest structure, but rather to the extreme nature of the event itself. Michele Torresani's work is an example of the evolving landscape of ecological research, highlighting the importance of sustainable and accurate data collection methods that contribute to more effective conservation and management strategies for our natural world.