RPAS-based photogrammetry to support tree stability assessment: Longing for precision arboriculture

•The RPAS-based photogrammetry is used for single tree parameters extraction.•Tree stability assessment is performed with reference to biomechanical model.•A quantitative index permits to balance costs and improving economic efficiency.•A new Precision Arboriculture approach fits extensive contexts requirements is proposed. Tree stability evaluation is an important issue with great practical implications. In the recent years, tree potential to cause harm has been increasing in consequence of climate change effects, mainly related to windstorms and tree diseases that represent the main tree failure causes. A tree owner has a duty of safety, imposed by civil and penal laws; consequently, he must operate an appropriate tree management to avoid foreseeable injuries or harms. A relevant problem arises when tree monitoring concern wide areas (extensive contexts), like natural park or urban forest; in these situations a variety of management factors have to be taken into account: the spatial size of the monitored areas; the great heterogeneity of trees vegetative conditions; the relevant number of trees; the balance between environmental protection and safe use of the area; the conspicuous cost of controls and technical interventions. With these premises an efficient planning tool is mandatory to manage this complex resource. Geomatics can support these requirements by integrating different techniques like survey, spatialization and modelling of territorial/environmental variables. In this work authors propose a new approach, hereinafter called “Precision Arboriculture” (PA), for tree management, fitting extensive contexts requirements. The proposed workflow is mainly based on RPAS photogrammetry technique and is specifically aimed at (i) accurately estimating single tree parameters; (ii) developing a robust algorithm to assess tree stability with the aim of reducing costs by better addressing ground controls through a spatially based management tool. This technology proved to generate estimates of the main dendrometric parameters with accuracies consistent (sometime higher) than the one ordinary required in the arboricultural context. Nevertheless, some ground data are however needed to calibrate models and testing accuracy of estimates. The proposed methodology proved to be able to generate an easy to use tool (Tree Safety Factor map) for better address ground controls aimed at testing tree stability and reducing the correlated hazard. Safety Factor map enhance