Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Vasil Yordanov has a total of 85 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Improving Sentinel-5P Imagery Usability Through Machine Learning Gap-Filling

    by Zhanbin Wu, Vasil Yordanov, Maria Antonia Brovelli

    Satellite-based air quality monitoring, particularly using Sentinel-5P (S5P) TROPOMI observations of NO2 and SO2, provides high-resolution global coverage but is substantially affected by spatio-temporal data gaps. These discontinuities, caused by cloud cover, surface reflectance, viewing geometry constraints, and strict quality assurance filtering, limit the reliability of downstream applications such as emission inversion, exposure assessment, and time-series analysis. This study investigates missing data patterns and develops a robust gap-filling framework for the Po Valley (Northern Italy) over 2019–2023, a region characterized by complex basin topography and frequent winter inversions that exacerbate both pollution accumulation and data loss. A statistical assessment revealed average missing rates of 45.4% for NO2 and 77.4% for SO2, with strong seasonality and extreme winter sparsity (up to 93.1% missing for SO2). Missingness was strongly correlated with elevation (r ≈ 0.77 for NO2, r ≈ 0.92 for SO2), highlighting terrain-related influences on retrieval quality. To reconstruct missing observations, we implemented two complementary machine learning models: a LightGBM baseline and a 3D Convolutional Neural Network (3D CNN). Both models were trained on a harmonized multi-source feature stack including lagged pollutant fields, spatial neighborhood statistics, ERA5 meteorology, static geographical variables, and cyclic temporal encodings. Synthetic masking simulated realistic gap scenarios during training (2019–2022), with evaluation on 2023 data. The 3D CNN outperformed LightGBM, achieving R2 values of 0.9469 (NO2) and 0.7387 (SO2), indicating strong spatio-temporal modeling capability. This framework enhances the continuity of S5P observations and supports their use for air quality analysis in data-sparse conditions.

  • Open Access

    INTEGRATED EARTH OBSERVATION AND DATA-DRIVEN APPROACHES FOR URBAN AIR QUALITY AND THERMAL ENVIRONMENT MONITORING

    by Vasil Yordanov, Alberto Vavassori, Sadra Zahed Kachaee, Claudia Isabela Saud Mino, Matej Zgela, Afshin Moazzam, Maria Antonia Brovelli

    Urban environments are increasingly affected by air pollution and urban heat island (UHI) effects, requiring integrated monitoring approaches. Within the Space It Up! project, this work presents coordinated research lines combining Earth Observation (EO) data and data-driven methods. The air quality component includes methane screening, multi-source data integration, and Sentinel-5P data gap analysis, while the thermal comfort component focuses on LCZ mapping and high-resolution temperature modelling. Together, these activities form a multi-scale framework linking atmospheric conditions, urban morphology, and thermal response, supporting improved urban environmental monitoring.

  • Open Access

    Harnessing EO and Natural Experiments for Urban Development: The UDENE Approach

    by Vasil Yordanov, Daniele Oxoli, Nikola Obrenović, Branislav Pejak, Onur Lenk, Nebiye Musaoğlu, Bekir Kapukaya, Yücel Erbay, Seda Kara, Tacettin Yücel İpEk, et al.

    Urban Development Explorations using Natural Experiments (UDENE) is a forward-looking initiative under the Horizon Europe program that merges Earth Observation (EO) technologies with urban planning to tackle pressing urban challenges. By utilizing Copernicus satellite imagery and organizing local in-situ data into interoperable data cubes, UDENE provides a comprehensive framework for data-driven decision-making. Further, it is applied the concept of “natural experiments” - real-life changes analyzed with the rigor of controlled studies - to uncover causal relationships in urban development. A primary goal is to incorporate structured urban data into the broader Copernicus data cube federation, enabling consistent analysis of urban impacts across different times and locations. To support this, UDENE develops advanced sensitivity analysis methods for validating and applying multivariate causal models, enhancing predictions on factors such as air pollution, urban heat, mobility, and disaster resilience. To close the gap between high-level EO technologies and real-world planning needs, we are introducing the three core tools: the UDENE’s Data Cube, which populates in-situ data EO based analysis-ready data and datasets; the Exploration Tool, which empowers planners and policymakers to simulate, assess, and visualize urban interventions; and a matchmaking tool connecting users with EO-based services. Together, these tools foster informed urban strategies grounded in EO data and causal inference.

  • Open Access

    Estimating Ground-Level Carbon Monoxide Concentrations Using Machine Learning Techniques: The Metropolitan City of Milan Case Study

    by Zhongyou Liang, Jesus Rodrigo Cedeno Jimenez, Vasil Yordanov, Maria A. Brovelli

    This work presents a structured data-driven framework for estimating ground-level carbon monoxide (CO) concentrations in the Metropolitan City of Milan (MCM) by integrating Sentinel-5P satellite observations, Copernicus Atmosphere Monitoring Service reanalysis data, and ERA5 meteorological variables with advanced machine learning techniques. The methodology employs unified data preprocessing, systematic feature engineering (e.g., boundary layer height-adjusted CO, lagged meteorological variables), Bayesian optimization for hyperparameter tuning, SHAP-based feature selection, model ensembling, and robust statistical validation. Eight regression models, including a custom Dense Attention Network (DAN), were evaluated across multiple temporal aggregation windows (4–24 hours before 15:00 GMT+1) to identify optimal configurations for CO estimation. Using data from January 2019 to November 2024, the framework identified the 21:00–15:00 GMT+1 window as most effective for capturing atmospheric dynamics such as nighttime accumulation, morning emission peaks, and daytime dilution. The DAN achieved the best performance, with a mean normalized root mean squared error of 0.4879 ± 0.0252 on the test set, outperforming ensemble and traditional regression models, offering a scalable, interpretable, and cost-effective approach to urban CO monitoring in data-scarce environments with potential adaptation to other pollutants and regions.


  • Open Access

    Remote Sensing and Machine Learning for Urban Air Quality and Heat Island Monitoring

    by Maria Antonia Brovelli, Jesus Rodrigo Cedeno Jimenez, Afshin Moazzam, Vasil Yordanov, Alberto Vavassori

    This study presents a dual-strategy approach to monitor urban environmental stressors, conducted within the ASI-MUR-funded Space It Up! project, focusing on atmospheric pollution and the urban heat island (UHI) effect. First, we developed a scalable machine learning (ML) framework for estimating ground-level concentrations of NO2, SO2, and CO in Milan using Sentinel-5P satellite data, ERA5 reanalysis, CAMS forecasts, and ARPA Lombardia ground measurements. Data preprocessing pipelines were optimized by switching to Google Earth Engine, reducing retrieval times and enabling operational scalability. Despite known satellite retrieval limitations in winter months for SO2, model performance remained robust, with normalized RMSE values consistently below 0.85. For CO, a Deep Attention Network achieved the best results (NRMSE = 0.4879), demonstrating the adaptability of the framework across pollutants. Additionally, a comparative analysis of low-cost air quality sensors showed high performance from AirGradient devices, particularly for PM2.5 and temperature, though significant inter-brand discrepancies were observed for CO2. Second, we implemented an advanced LCZ classification method integrating hyperspectral PRISMA imagery, Sentinel-2 data, and urban canopy parameters (UCPs). Applied to the Metropolitan City of Milan, the proposed workflow achieved substantial improvements over existing methods, with an overall accuracy increase up to 16% when utilizing PRISMA data compared to the state-of-art LCZ Generator approach. We also presented ongoing efforts to further improve the proposed methodology, including the automation of data retrieval and training and test sample creation. The methodology is being applied across multiple urban areas worldwide by also testing other ML techniques. Together, these methodologies provide a comprehensive and reproducible framework for urban environmental monitoring.

  • Open Access

    High-Resolution Global Land Cover Maps and Their Assessment Strategies

    by Qiongjie Xu, Vasil Yordanov, Lorenzo Bruzzone, Maria Antonia Brovelli

    Global High-Resolution Land Cover Maps (GHRLCs), characterized by spatial resolutions higher than 30 m per pixel, have become essential tools for environmental monitoring, urban planning, and climate modeling. Over the past two decades, new GHRLCs have emerged, offering increasingly detailed and timely representations of Earth’s surface. This review provides an in-depth analysis of recent developments by examining the data sources, methodologies, and validation techniques utilized in 19 global binary and multi-class land cover products. The evolution of GHRLC production techniques is analyzed, starting from the use of singular source input data, such as multi-temporal remotely sensed optical imagery, to the integration of satellite radar and other geospatial data. The article highlights significant advances in data pre-processing and processing, showcasing a shift from classical methods to modern approaches, including machine learning (ML) and deep learning techniques (e.g., neural networks and transformers), and their direct application on powerful cloud-computing platforms. A comprehensive analysis of the temporal dimension of land cover products, where available, is conducted, highlighting a shift from decadal intervals to production intervals of less than a month. This review also addresses the ongoing challenge of land cover legend harmonization, a topic that remains crucial for ensuring consistency and comparability across datasets. Validation remains another critical aspect of GHRLC production. The methods used to assess map accuracy and reliability, including statistical techniques and visual inspections, are briefly discussed. The validation approaches adopted in recent studies are summarized, with an emphasis on their importance in maintaining data integrity and addressing emerging needs, such as the development of common validation datasets. Ultimately, this review aims to provide a comprehensive overview of the current state and future directions of GHRLC production and validation, highlighting the advancements that have shaped this rapidly evolving field.


  • Open Access

    An interpretable and transferable model for shallow landslides detachment combining spatial Poisson point processes and generalized additive models

    by Giulia Patanè, Teresa Bortolotti, Vasil Yordanov, Ludovico Giorgio Aldo Biagi, Maria Antonia Brovelli, Xuan Quang Truong, Simone Vantini

    Shallow landslides are rapidly moving and strongly dangerous slides. In the present work, the probabilistic distribution of the landslide detachment points within a valley is modelled as a spatial Poisson point process, whose intensity depends on geophysical covariates according to a generalized additive model. This jointly allows to obtain good predictive performance and to preserve the interpretability of the effects of the geophysical predictors on the intensity of the process. We propose a novel workflow, based on Random Forests, to select the geophysical predictors entering the model for the intensity. The statistically significant effects are interpreted as activating or stabilizing factors for landslide detachment. The transferability of the resulting model is guaranteed by training, validating and testing the algorithms on mutually disjoint valleys in the Alps of Lombardy (Italy). Finally, the uncertainty around the estimated intensity of the process is quantified via semiparametric bootstrap.


  • Open Access

    An interpretable and transferable model for shallow landslides detachment combining spatial Poisson point processes and generalized additive models

    by Giulia Patanè, Teresa Bortolotti, Vasil Yordanov, Ludovico Giorgio Aldo Biagi, Maria Antonia Brovelli, Xuan Quang Truong, Simone Vantini

    Less than 10 meters deep, shallow landslides are rapidly moving and strongly dangerous slides. In the present work, the probabilistic distribution of the landslide detachment points within a valley is modelled as a spatial Poisson point process, whose intensity depends on geophysical predictors according to a generalized additive model. Modelling the intensity with a generalized additive model jointly allows to obtain good predictive performance and to preserve the interpretability of the effects of the geophysical predictors on the intensity of the process. We propose a novel workflow, based on Random Forests, to select the geophysical predictors entering the model for the intensity. In this context, the statistically significant effects are interpreted as activating or stabilizing factors for landslide detachment. In order to guarantee the transferability of the resulting model, training, validation, and test of the algorithm are performed on mutually disjoint valleys in the Alps of Lombardy (Italy). Finally, the uncertainty around the estimated intensity of the process is quantified via semiparametric bootstrap.

  • Co-registration of PRISMA Hyperspectral Imagery for Accurate Land Cover Classification

    by Qiongjie Xu, Vasil Yordanov, Xuan Quang Truong, Ludovico Biagi, Maria Antonia Brovelli


  • Open Access

    An interpretable and transferable model for shallow landslides detachment combining spatial Poisson point processes and generalized additive models

    by Giulia Patanè, Teresa Bortolotti, Vasil Yordanov, Ludovico Giorgio Aldo Biagi, Maria Antonia Brovelli, Xuan Quang Truong, Simone Vantini

    Less than 10 meters deep, shallow landslides are rapidly moving and strongly dangerous slides. In the present work, the probabilistic distribution of the landslide detachment points within a valley is modelled as a spatial Poisson point process, whose intensity depends on geophysical predictors according to a generalized additive model. Modelling the intensity with a generalized additive model jointly allows to obtain good predictive performance and to preserve the interpretability of the effects of the geophysical predictors on the intensity of the process. We propose a novel workflow, based on Random Forests, to select the geophysical predictors entering the model for the intensity. In this context, the statistically significant effects are interpreted as activating or stabilizing factors for landslide detachment. In order to guarantee the transferability of the resulting model, training, validation, and test of the algorithm are performed on mutually disjoint valleys in the Alps of Lombardy (Italy). Finally, the uncertainty around the estimated intensity of the process is quantified via semiparametric bootstrap.