Amici, Stefania

The exacerbation of wildfires, attributed to the effects of climate change, presents substantial risks to ecological systems, infrastructure, and human well-being. In the context of the Sustainable Development Goals (SDGs), particularly those related to climate action, prioritizing the assessment and management of the occurrence and intensity of extensive wildfires is of utmost importance. In recent times, there has been a significant increase in the frequency and severity of widespread wildfires worldwide, affecting several locations, including Australia, Italy, and the United States of America. The presence of complex phenomena marked by limited predictability leads to significant negative impacts on biodiversity and human lives. The utilization of satellite-derived data with neural networks, such as convolutional neural networks (CNNs), is a potentially advantageous approach for augmenting the monitoring capabilities of wildfires. This research examines the generalization capability of four neural network models, namely the fully connected (FC), one-dimensional (1D) CNN, two-dimensional (2D) CNN, and three-dimensional (3D) CNN model. Each model’s performance, as measured by accuracy, recall, and F1 scores, is assessed through K-fold cross-validation. Subsequently, T-statistics and p-values are computed based on these metrics to conduct a statistical comparison among the different models, allowing us to quantify the degree of similarity or dissimilarity between them. By using training data from Australia and Sicily, the performances of the trained model are evaluated on the test dataset from Oregon. The results are promising, with cross-validation on the training dataset producing mean precision, recall, and F1 scores ranging between approximately 0.97 and 0.98. Especially, the fully connected model has superior generalization capabilities, whilst the 3D CNN offers more refined and less distorted classifications. However, certain issues, such as false fire detection and confusion between smoke and shadows, persist. The aforementioned methodologies offer significant perspectives on the capabilities of neural network technologies in supporting the detection and management of wildfires. These approaches address the crucial matter of domain transferability and the associated dependability of predictions in new regions. This study makes a valuable contribution to the ongoing efforts in climate change by assisting in monitoring and managing wildfires.

The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER ) has operated since 19 December 1999 from NASA’s Terra Earth-orbiting, sun synchronous satellite. Emissivity and temperature standard products are based on the TES algorithms and require periodical validation campaign. In the frame of the EC project PREVIEW (http://www.preview-risk.com/) a field campaign on Volcano Teide was carried on, from the 16th to 24th of September 2007, to validate and to integrate the satellite derived products services.

Fixed or rotary wing manned aircraft are currently the most commonly used platforms for airborne reconnaissance in response to natural hazards, such as volcanic eruptions, oil spills, wild fires, earthquakes. Such flights are very often undertaken in hazardous flying conditions (e.g., turbulence, downdrafts, reduced visibility, close proximity to dangerous terrain) and can be expensive. To mitigate these two fundamental issues--safety and cost--we are exploring the use of small (<100kg), relatively inexpensive, but effective, unmanned aerial vehicles (UAVs) for this purpose. As an operational test, in 2004 we flew a small autonomous UAV in the airspace above and around Stromboli Volcano. Based in part on this experience, we are adapting the RAVEN- INGV system for such natural hazard surveillance missions. RAVEN- INGV has a 50km range, with a 3.5m wingspan, main fuselage length of 4.60m, and maximum weight of 56kg. It has autonomous flight capability and a ground control station for mission planning and control. It will carry a variety of imaging devices, including a visible camera, and an IR camera. Such flexible, capable, and easy-to-deploy UAV systems may significantly shorten the time necessary to characterize the nature and scale of the natural hazard threats if used from the outset of, and systematically during, natural hazard events. When appropriately utilized, such UAVs can provide a powerful new hazard mitigation and documentation tool for civil protection hazard responders. This research was carried out under the auspices of the Italian government, and, in part, under contract to NASA at the Jet Propulsion Laboratory.

The Canarian Arcipelago is made up of seven islands that represent different stages of geologic evolution. Tenerife is the central island of archipelago and has developed within the complex formed by the rifts associated with the Teide-Pico Viejo (T-PV-Lat 28° 16’ 30” Lon 16°38’ 42”) stratovolcanoes that reach a height of 3718m, 7500 above the ocean floor. It is an active, though currently quiescent shield volcano, which last erupted in 1909. In this study we have geologically characterized Volcano Teide by using multispectral and hyperspectral satellite imaging data. Radiance data were preprocessed and calibrated into reflectance, following which unsupervised and supervised classification methods were applied. The supervised classification primarily utilized in situ ground truth obtained during 2007 field campaign (EC project PREVIEW FP6). In this work we compare results obtained by using several methods.

Wildfire have a significance at both global and local scale. For example at global scale fires affect global climate through processes such as trace gas and aerosol production. At local scale fires have health impact, poor air quality, landscape / watershed scale effect, water quality degradation. Global warming and hotter summer contribute to the shift of fire regime, and also Arctic pole countries are experiencing wildfires in significant number and at both high scale and intensity. INGV (Italy) has a long record of wildfire studies by using observation from space and by using sophistic sensors on airborne within international cooperation

One of the United Nations (UN) Sustainable Development Goals is climate action (SDG-13), and wildfire is among the catastrophic events that both impact climate change and are aggravated by it. In Australia and other countries, large-scale wildfires have dramatically grown in frequency and size in recent years. These fires threaten the world's forests and urban woods, cause enormous environmental and property damage, and quite often result in fatalities. As a result of their increasing frequency, there is an ongoing debate over how to handle catastrophic wildfires and mitigate their social, economic, and environmental repercussions. Effective prevention, early warning, and response strategies must be well-planned and carefully coordinated to minimise harmful consequences to people and the environment. Rapid advancements in remote sensing technologies such as ground-based, aerial surveillance vehicle-based, and satellite-based systems have been used for efficient wildfire surveillance. This study focuses on the application of space-borne technology for very accurate fire detection under challenging conditions. Due to the significant advances in artificial intelligence (AI) techniques in recent years, numerous studies have previously been conducted to examine how AI might be applied in various situations. As a result of its special physical and operational requirements, spaceflight has emerged as one of the most challenging application fields. This work contains a feasibility study as well as a model and scenario prototype for a satellite AI system. With the intention of swiftly generating alerts and enabling immediate actions, the detection of wildfires has been studied with reference to the Australian events that occurred in December 2019. Convolutional neural networks (CNNs) were developed, trained, and used from the ground up to detect wildfires while also adjusting their complexity to meet onboard implementation requirements for trusted autonomous satellite operations (TASO). The capability of a 1-dimensional convolution neural network (1-DCNN) to classify wildfires is demonstrated in this research and the results are assessed against those reported in the literature. In order to enable autonomous onboard data processing, various hardware accelerators were considered and evaluated for onboard implementation. The trained model was then implemented in the following: Intel Movidius NCS-2 and Nvidia Jetson Nano and Nvidia Jetson TX2. Using the selected onboard hardware , the developed model was then put into practice and analysis was carried out. The results were positive and in favour of using the technology that has been proposed for onboard data processing to enable TASO on future missions. The findings indicate that data processing onboard can be very beneficial in disaster management and climate change mitigation by facilitating the generation of timely alerts for users and by enabling rapid and appropriate responses.

In this paper, we exploited satellite remote sensing data, acquired by SAR and optical sensors to map the lava emplacement during the eruption of Pi co do Fogo volcano, in Cape Verde. The eruption took place in November 2014, and lasted for about 2 months. The event was imaged by several satellite missions. In particular, the ESA Sentinel-IA platform operated in that area, collecting several images with its novel acquisition mode, the so-called TOPSAR. SAR images have been processed to extract changes automatically and to infer the advancement of the lava emitted from November 23, 2014 to January 2017, by using an adaptive parametric thresholding and a hierarchical split based approach. This automatic procedure allowed mapping the evolution of the lava coverage. The results obtained thanks to this method were compared to the ones derived by using the optical images collected by Landsat-8 and EO-l optical sensors

Gli incendi sono un fenomeno che colpisce ogni anno il nostro pianeta ed in particolare anche l’Italia. Oltre all’effetto immediato sul territorio, è stato riconosciuto un effetto a livello di impatto climatico. I fuochi cambiano lo stato fisico della vegetazione rilasciando nell’atmosfera gas che giocano un ruolo importante nell’effetto serra. E’ stato stimato che la biomassa bruciata in un anno, contribuisce al 38% dell’Ozono in troposfera, al 32% di monossido di Carbonio, al 20% degli altri gas (Levine, 1991; Andreae, 1991; Kaufman et al., 1998a,b). L’uso dei canali termici (8-14 micron) o relativi al vicino infrarosso (1.0 -2.5 micron) sono tradizionalmente utilizzati per la detection e lo studio di parametri fisici come il potere radiante, l’NDBR (Normalised Difference Burn Ratio), o il tasso di combustione della biomassa. Nel visibile una banda di emissione diagnostica dello stato di fiamma, è quella del Potassio (K-method) che fino ad ora non è stata molto studiata in quanto limitata dalle prestazioni degli strumenti. Nell’estate 2006, con il progetto AIRFIRE finanziato da ESA, una campagna aerea è stata effettuata su incendi non controllati utilizzanodo un prototipo di sensore iperspettrale denominato SIM-GA di Selex Galileo. Il SIM-GA è un sensore iperspettrale ad altissima risoluzione spettrale 1.2nm e 2.5 nm rispettivamente nel VISIBILE (350-1200nm) e nel vicino infrarosso(1200-2500nm) che opera in modalità pushbroom. I dati ottenuti hanno permesso di verificare e testare il metodo della panda del Potassio. I risultati ottenuti hanno mostrato come la combinazione del K-method con le analisi nel termico possa completare l’analisi dell’incendio.

A substantial proportion of Italian students (60/%)- are unaware of the connection between what they learn at school and their work opportunities. This proportion would most likely increase if data were collected today, given the generation of a broad range of new jobs that have arisen due to advancements in technology. This gap between students’ understanding of what they learn at school and its application to the broader world – the society, the economy and the political sphere – suggests there needs to be a rethinking of how teaching and learning at school is conceived and positioned. To help students to approach ongoing social and economic transformations, the Italian Educational Ministry (MIUR) has endorsed a school–work interchange program which, aligned with the principle of open schools, aims to provide students with work experience. It is within the scope of this initiative that we have tested high school students with Remote Sensing (RS) from space projects. The experience-based approach aimed to verify students’ openness to the use of satellite data as a means to learn new interdisciplinary skills, to familiarize themselves with methodological knowledge, and finally, to inspire them when choosing a university or areas of future work. We engaged three cohorts, from 2017, 2018 and 2019 respectively, for a total of 40 hours each year, including contact and non-contact time. The framework of each project was the same for the three cohorts and focused on the observation of Earth from space with a specific focus on wildfires. However, the initiative went beyond this, with diverse activities and tasks being assigned. This paper reports the pedagogical methods utilized with the three cohorts and how these methods were transformed and adapted in order to improve and enhance the learning outcomes. It also explores the outcomes for the students, teachers and family members, with respect to their learning and general appreciation

Hyperspectral sensors have become a standard technology used in the techniques of observation by satellite and aerial platform for observing the terrestrial ecosystem with particular interest in the detection and identification of minerals, vegetation, materials and artificial environments. The detection of real materials depends on the coverage spectral resolution and signal to noise ratio of the spectrometer itself, as well as the density of the material and the absorption characteristics for the material in the region of wavelength measured. The signal to noise ratio in particular is one of the parameters that need to be estimated to establish the quality of images acquired by these systems. In this contribution a method to estimate the Signal to Noise Ratio (SNR) for unsupervised hyperspectral images has been investigated. The method uses the computation of local means and local standard deviations of small homogeneous blocks in order to define respectively the average signal and the mean noise of the images. If the noise may be considered mainly addictive the local standard deviation may be considered as the mean noise of image. This method uses all the spatial information contained in the image scene giving a representative SNR of entire image. The technique has been engineered in IDL environment and applied to hyperspectral data of HYPER-SIMGA sensor, developed in the frame of AIRFIRE Project for wildfire detection by airborne remote sensing data. The SNR results point out that HYPER-SIMGA SWIR images are quite noisy and the spectral range that has to be taken into account for data analysis is from 1000 to 1700 nm.