Space-to-tree: Architectural framework for real-time monitoring of pines in natural and historical park

Abstract

The conservation and promotion of natural and cultural heritage, including landscapes, constitutes a subject of great economic and social importance. In recent times, there has been an increasing emphasis on the debate around strategies for developing an integrated approach to environmental and cultural heritage. The United Nations Educational, Scientific and Cultural Organization (UNESCO) has highlighted these challenges in its conventions, emphasizing the need for proactive measures to protect cultural heritage. The European Space Agency project “From space to Tree” (S23) aimed to develop an alert system for real-time monitoring of trees stability. The experimental site was the Archaeological Park of Colosseum in Rome. Inside the park there are countless pine trees of historical-cultural interest. The System and Service Architecture (SSA) describes the structure of the pilot system, detailing the high-level architecture and its constituent components. The monitoring system follows an integrated multi-scale approach and combines the health status of trees monitored by multi-temporal Sentinel-2 remote sensing images, the movements of the trees in response to wind stress, monitored by four inertial measurement units (IMUs) installed at different heights on each individual tree to detect its movement in response to wind stress, a weather station equipped with an ultrasonic anemometer and the in-field surveys and analyzes carried out by expert forestry agronomists. The acquired data is transmitted in real time to a dedicated server, by using the 5G network. Results show that data transmission system becomes more complex as the number of monitored trees increases, consequently the transmission system was designed taking this criticality into account. The data are displayed and analyzed in a dedicated Web-GIS platform. The experimentation is still ongoing, during the first experimental year, the analysis algorithm was trained in the first 6 months. During the test period, an alert is generated when changes are found in the behavior of the pines, based on remote sensing images and trees' response to wind stress analysis, compared to the training period.