This study addresses the critical need for accurate mapping of submerged terrain, which is essential for hydraulic modeling, environmental monitoring, and water resource management. Traditional bathymetric techniques, such as topographic surveys and acoustic soundings, face spatial continuity and usability challenges in shallow or vegetated waters. Recent advances, including Uncrewed Surface Vessels (USVs) equipped with GNSS and acoustic sensors, along with UAV-based photogrammetry for 3D modeling in clear waters, have expanded capabilities. However, optical methods suffer from depth underestimation due to light refraction, requiring geometric corrections. To address these limitations, the paper proposes a multi-sensor fusion workflow that integrates high-precision topographic data from total stations and GNSS, depth measurements from a USV equipped with a single-beam echo sounder, and UAV-derived optical bathymetry corrected for refraction using Structure from Motion (SfM) techniques. The goal is to combine each method's strengths to overcome their weaknesses and produce an accurate, high-resolution bathymetric model. Validation against ground truth data demonstrated significant improvements in data quality, aligning with standards for shallow-water mapping. Notably, the use of corrected UAV photogrammetry extended effective depth measurements to 4–5 meters, exceeding typical optical limits. The combined methodology ensures robust spatial coverage, precise georeferencing, and transparent independent measurements, making it particularly well-suited for complex lacustrine (lake) environments. The results highlight the operational benefits of using complementary technologies and suggest potential for further enhancement through Machine Learning and Deep Learning techniques to refine data integration and analysis.

The sustainable preservation of cultural heritage, encompassing both tangible assets like monuments and intangible expressions such as traditions, necessitates proactive, evidence-based conservation strategies. This paper underscores the importance of rigorous, continuously updated documentation as the cornerstone of planned maintenance practices, an approach that ensures early detection of vulnerabilities, minimizes invasive interventions, and promotes the longevity and integrity of heritage assets. Within this context, UNESCO's World Heritage framework emphasizes the need for detailed documentation and systematic monitoring, particularly for complex sites like Italy’s Sacri Monti, a group of devotional complexes recognized for their integration of landscape, art, and architecture. Focusing on the Sacro Monte di Crea in Piedmont, this study details a case-specific strategy for documenting Chapel XVI, which faces critical conservation issues such as biological growth, salt efflorescence, and material degradation. A multimodal geomatics survey was conducted using terrestrial laser scanning, mobile mapping systems, drone-based photogrammetry, and smartphone LiDAR, enabling the creation of a comprehensive, high-resolution 3D dataset. These datasets were processed, aligned, and fused to produce an accurate and scalable digital representation of the site. This robust foundation supports not only traditional maintenance workflows but also emerging applications of artificial intelligence for 3D reconstruction and interpretation. The integration of such technologies marks a decisive step forward in enhancing both the documentation process and long-term conservation planning for vulnerable cultural heritage sites.

The SUNRISE Summer School (Seashore and UNderwater documentation of aRchaeological herItage palimpSests and Environment) has reached its second edition. The school was co-organised by the Politecnico di Torino and the Italian Society of Photogrammetry and Topography (SIFET) and supported by the International Society of Photogrammetry and Remote Sensing (ISPRS), CIPA HD and several other Italian institutions. The second edition of the summer school took place in the municipality of Porto Cesareo (LE - Italy) from September 8 to 14, 2024, and involved 24 students – architects, engineers, archaeologists, and marine ecologists from Europe, the United States, Latin America, and Asia. The school started from the experience built in the first edition, and throughout the entire duration of the event, the students were guided by 23 tutors, ensuring an almost one-to-one student-tutor ratio. The school offered theoretical activities with lectures and fieldwork, applying a learning-by-doing teaching approach. Two sites were the focus of the fieldwork: Torre Chianca (a 16th-century defensive building) and the Roman marble columns (a submerged archaeological site). Both sites are part of the protected marine area of Porto Cesareo. Field data collection was supported by representatives from various companies: Images, Microgeo, Stonex, Pix4D, Geomax, Leica, Dynatech, and 3DTarget (the companies were also sponsors of the initiative), who demonstrated the latest technologies for integrated multi-sensor surveying of cultural heritage, such as drones, terrestrial and aerial laser scanners, and mobile mapping systems.

This study investigates the classification of individual tree and shrub species within a complex Mediterranean ecosystem using UAV-acquired multispectral imagery and machine learning techniques. Conducted on Culuccia Island (Sardinia, Italy), the research integrates high-resolution photogrammetric products with Object-Based Image Analysis (OBIA) and a Random Forest classifier to delineate and identify vegetation species at the crown level. A total of 272 geo-referenced samples from 16 species were used to train and validate the model, which achieved an overall accuracy of 0.71. The workflow demonstrates the effectiveness of single tree segmentation in highly heterogeneous environments and highlights the potential of phenology-informed feature sets for improving classification. The results underscore the value of UAV-based methods in conservation monitoring, ecological assessment, and habitat management. Future research directions include the integration of LiDAR data, deep learning architectures, and multi-temporal observations to enhance scalability and model interpretability.