Article
Water Resources
Daniel Wonde Mebrie, Tewodros T. Assefa, Abdu Y. Yimam, Sisay A. Belay
Summary: Proper estimation of evapotranspiration is important for the management of irrigation systems. However, the lack of climatic stations in Ethiopia makes it difficult to estimate reference evapotranspiration (ETo) and crop coefficient (Kc) spatially. This study aimed to estimate crop evapotranspiration (ETc) by deriving crop coefficients using remote sensing products. The correlation between MODIS potential evapotranspiration and Penman-Monteith estimates was good, and a strong correlation was found between Sentinel-based NDVI and FAO crop coefficient. Calibrating and integrating MODIS with Sentinel 2B can provide a feasible approach for estimating Kc and ETc. The findings highlight the importance of proper estimation of crop water needs for better productivity in the region.
APPLIED WATER SCIENCE
(2023)
Article
Multidisciplinary Sciences
Xin-Xing Zhou, Yang-Yang Li, Yuan-Kai Luo, Ya-Wei Sun, Yi-Jun Su, Chang-Wei Tan, Ya-Ju Liu
Summary: This study used remote sensing images and vegetation indices to construct a decision tree classification model for accurately obtaining the spatial distribution information of fruit tree planting. The results showed that this method can effectively monitor large-scale fruit tree planting areas.
SCIENTIFIC REPORTS
(2022)
Article
Environmental Sciences
Stefano Gobbo, Alessandro Ghiraldini, Andrea Dramis, Nicola Dal Ferro, Francesco Morari
Summary: The study demonstrates that integrating remote sensing and crop modeling offers a reliable, objective, and less labor-intensive method to estimate crop hail damage, providing a unique opportunity for the crop insurance market.
Article
Agriculture, Multidisciplinary
Giulia Ronchetti, Giacinto Manfron, Christof J. Weissteiner, Lorenzo Seguini, Luigi Nisini Scacchiafichi, Lorenzo Panarello, Bettina Baruth
Summary: Operational crop yield forecasting services use regression models to predict crop yields based on agro-environmental variables such as meteorological data, crop simulation models, or satellite-derived indicators. This study examines the impact of using different crop masks on the correlation between yield data and the Normalized Difference Vegetation Index (NDVI) in Europe. The results show that using annual crop group-specific masks improves yield estimation accuracy and timeliness, particularly for soft wheat and grain maize.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2023)
Article
Environmental Sciences
Shangharsha Thapa, Virginia E. Garcia Millan, Lars Eklundh
Summary: This research explores the potential of near-surface sensors to track forest phenology and validate satellite-derived phenology against observations from UAVs, PhenoCams, and Spectral Reflectance Sensors. The study shows significant differences in phenology between different sensors, with PhenoCams and UAVs demonstrating potential for satellite data validation and upscaling. The combination of these near-surface vegetation metrics provides a promising foundation for analyzing the interoperability of different sensors for vegetation dynamics and change analysis.
Article
Remote Sensing
Bing-Bing Goh, Peter King, Rebecca L. Whetton, Sheida Z. Sattari, Nicholas M. Holden
Summary: This study aims to evaluate winter wheat development by comparing predicted biophysical properties from Sentinel-2 data with target growth benchmarks. The results show that phenology-specific models perform better and can be reliably used for crop monitoring.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2022)
Article
Engineering, Electrical & Electronic
Krishna Karthik Gadiraju, Ranga Raju Vatsavai
Summary: Machine learning methods using aerial imagery have been widely used for crop classification. Traditional per-pixel-based, object-based, and patch-based methods have been used, but deep learning-based systems are becoming popular. However, building complex deep neural networks for aerial imagery is challenging due to limited labeled data and the variability associated with agricultural data. This article discusses these challenges and evaluates transfer learning methodologies for improving remote sensing image classification performance.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Environmental Sciences
Asier Uribeetxebarria, Ander Castellon, Ana Aizpurua
Summary: Accurately estimating wheat yield is essential for precision agriculture and crop management. This study combines Sentinel-1 and Sentinel-2 data with the CatBoost algorithm to predict wheat yield in 39 fields. The results show that the combination of S1 and S2 data with CatBoost algorithm can achieve high accuracy and reduce yield errors.
Article
Environmental Sciences
Amal Chakhar, David Hernandez-Lopez, Rocio Ballesteros, Miguel A. Moreno
Summary: This study assessed the potential of integrating Sentinel-1 and 2A data to perform crop classification and identified the most important input data for accurate results. The best performing scenario integrated VH and VV with NDVI using a cubic support vector machine (SVM) as the classifier.
Article
Multidisciplinary Sciences
Santosh Hiremath, Samantha Wittke, Taru Palosuo, Jere Kaivosoja, Fulu Tao, Maximilian Proll, Eetu Puttonen, Pirjo Peltonen-Sainio, Pekka Marttinen, Hiroshi Mamitsuka
Summary: This study investigates the feasibility of using satellite images and machine learning models to classify agricultural field parcels into those with and without crop loss. Despite the poor quality of data, the random forest model shows promising results in identifying new crop-loss fields based on reference fields of the same year. There is potential for various applications in efficient agricultural monitoring and verifying crop-loss claims.
Article
Remote Sensing
Oumayma Bounouh, Houcine Essid, Ana Maria Tarquis, Imed Riadh Farah
Summary: Different studies on predicting future green cover changes exist with various levels of success, each focusing on a different story about which model is more appropriate. The experimentation of forecasting vegetation indices employing two univariate time series models and new accuracy metrics is discussed, highlighting the importance of integration of the decomposition step and the performance of different forecasting models.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2021)
Article
Green & Sustainable Science & Technology
Wonga Masiza, Johannes George Chirima, Hamisai Hamandawana, Ahmed Mukalazi Kalumba, Hezekiel Bheki Magagula
Summary: Adverse weather is a major source of risk in agriculture, and the lack of effective risk management mechanisms exacerbates its impacts. Weather-index-based insurance is gaining popularity, particularly in developing countries, but its uptake is still low. This study addresses this issue by evaluating the suitability of different datasets for weather-index-based insurance, finding that CHIRPS data has higher correlations with in situ rainfall measurements. The study also highlights that the relationship between crop yield and weather and vegetation indices is weak, and water is not always the main limiting factor for smallholder farming systems.
Article
Environmental Sciences
Thuan Ha, Yanben Shen, Hema Duddu, Eric Johnson, Steven J. Shirtliffe
Summary: This study estimates hail damage to crops in the Canadian Prairies using vegetation indices calculated from Sentinel-2 images. The temporal changes in vegetation indices were found to correlate well with ground estimates of hail damage.
Article
Environmental Studies
Meenakshi Rajeev, Pranav Nagendran
Summary: Small-scale agriculturists in developing countries face weather-related risks exacerbated by climate change. Despite efforts to introduce formal crop insurance, insurance penetration rates remain low in less developed countries. Reliance on informal credit as a risk-coping mechanism and socioeconomic factors influence the adoption of crop insurance. Informal interest rates are positively correlated with the probability of taking up crop insurance, and disparities exist based on economic and social classes. The study provides policy implications for enhancing insurance penetration among poor farmers.
Article
Environmental Sciences
Binbin Song, Songhan Min, Hui Yang, Yongchuang Wu, Biao Wang
Summary: This study applies frequency-domain deep learning to classify crops in remote sensing images, enhancing interclass differences and reducing intraclass variations by adjusting different frequency components, leading to improved classification accuracy and robustness.
Article
Entomology
Enrico Borgogno Mondino, Federico Lessio, Alessandro Bianchi, Mariangela Ciampitti, Beniamino Cavagna, Alberto Alma
Summary: An iterative spatially based model was developed to predict the spreading dynamics of the Japanese beetle. The model performed well, with validation showing a determination coefficient ranging from 0.39 to 0.87. The predicted spread of the Japanese beetle is mainly towards the south and southeast.
ENTOMOLOGIA GENERALIS
(2022)
Article
Forestry
Samuele De Petris, Filippo Sarvia, Enrico Borgogno-Mondino
Summary: Forest height is an important parameter in forestry, used to assess productivity and estimate carbon storage. However, there are measurement errors that affect the accuracy of estimates. This study proposes a statistically based method to estimate height measurement uncertainty, finding that angles are the main factor influencing height uncertainty. The study also maps tree height uncertainty globally and summarizes it by forest biomes using Google Earth Engine.
Article
Environmental Sciences
Samuele De Petris, Evelyn Joan Momo, Filippo Sarvia, Enrico Borgogno-Mondino
Summary: This study proposes a multi-temporal approach based on Sentinel-1 SAR data to monitor forest canopy and describe forest damage before and after a significant fire event in Piemonte Region, Italy. The analysis includes trend analysis and cluster analysis to characterize burned areas and their ecological behavior.
GEOCARTO INTERNATIONAL
(2022)
Article
Environmental Sciences
Francesco Palazzi, Marcella Biddoccu, Enrico Corrado Borgogno Mondino, Eugenio Cavallo
Summary: This study used Copernicus Sentinel-2 (S2) satellite optical data to evaluate the spatial and temporal variations of vineyard ground cover. The results suggest that different inter-row soil management in vineyards can be classified using NDVI and NDWI indices. Further research can provide important data support for erosion risk management and crop modeling.
Article
Environmental Sciences
Tommaso Orusa, Duke Cammareri, Enrico Borgogno Mondino
Summary: This study proposed an approach to map land cover in mountain areas by overcoming remote sensing limitations and following the newest EAGLE guidelines. Copernicus Sentinel-1 and 2 data were used and tested in Aosta Valley, Italy. The results showed that K-Nearest-Neighbor and Minimum Distance classification methods can improve accuracy and reduce errors.
Article
Chemistry, Multidisciplinary
Tommaso Orusa, Duke Cammareri, Enrico Borgogno Mondino
Summary: The Earth Observation services play a crucial role in continuous land cover mapping and have gained significant attention worldwide. The Google Earth Engine Dynamic World serves as a global example in this field. This study focuses on developing a land cover mapping service in the geologically complex areas of Aosta Valley in NW Italy, following the latest European EAGLE legend starting from 2020. The research utilizes Sentinel-2 data processed in the Google Earth Engine, combining multispectral indexes and k-nearest neighbor classification for accurate mapping of various land cover classes. Deep learning and GIS updated datasets, along with SAR Sentinel-1 SLC data, are also employed for mapping urban and water surfaces. The effectiveness of the implemented service and methodology is tested by comparing the overall accuracy with other approaches, and the mixed hierarchical approach proves to be the most effective for mapping geologically complex areas.
APPLIED SCIENCES-BASEL
(2023)
Article
Environmental Sciences
Vanina Fissore, Lorenza Bovio, Luigi Perotti, Piero Boccardo, Enrico Borgogno-Mondino
Summary: The DestinE European initiative has introduced the concept of the Digital Twin (DT) in Earth Sciences, aiming to create a high precision digital model of the Earth by 2030. This digital model will be continuously updated with Earth Observation data and will provide digital replicas for different domains within the earth sciences. This study focuses on defining a theoretical framework for a future DT of the Italian Alpine glaciers and explores the necessary information and relationships that need to be included. The study considers the spatial dynamics and health status of glaciers and evaluates the availability of EO data for constructing the DT at both local and regional scales.
Article
Environmental Sciences
Tommaso Orusa, Annalisa Viani, Boineelo Moyo, Duke Cammareri, Enrico Borgogno-Mondino
Summary: Earth observation data play a crucial role in environmental monitoring and risk assessment. This study focused on assessing population exposure to rising temperatures and heat waves in the Aosta Valley Region, Italy, using population distribution and land surface temperature trend data.
Article
Biology
Annalisa Viani, Tommaso Orusa, Enrico Borgogno-Mondino, Riccardo Orusa
Summary: The widespread diffusion of wild boars in Italy and their use for hunting have allowed for studies on various diseases in this animal, although only a few diseases have received substantial funding and attention. This study focused on sarcoptic mange in the wild boar population, including sympatric species, in Aosta Valley. Empirical evidence suggested a potential role of snow metrics in the spread of this parasite. By performing remote sensing analysis on snow metrics, new tools were provided for veterinarians, foresters, biologists, and ecologists to better understand wild boar dynamics and enhance management and planning strategies. The results showed that sarcoptic mange is present in an endemic form with low prevalence values, and favorable conditions for spreading were observed under specific snow metrics values.
Article
Environmental Studies
Federica Ghilardi, Andrea Virano, Marco Prandi, Enrico Borgogno-Mondino
Summary: The production and quality of grapes for wine depend heavily on specific site features such as topography, soil, and climate. Mapping the local specificities of a wine-production area, known as terroirs, is important for understanding the environmental conditions that drive wine production. This study proposes an approach to map territorial differences and identify distinct zones within a wine-production area in Piemonte, Italy based on free and open data and GIS. The results demonstrate significant variations in the study area, with each zone showing different preferences for vine varieties.
Article
Plant Sciences
Giulia Squillacioti, Samuele De Petris, Valeria Bellisario, Enrico Corrado Borgogno Mondino, Roberto Bono
Summary: The urban environment has an influence on sedentary behavior in school-aged children, with children living in less urbanized and more vegetated areas spending less time sedentary and being less likely to exceed recommended screen time. Maternal education level may act as an effect modifier. These findings emphasize the importance of considering environmental characteristics in public health strategies to prevent sedentary behavior and promote sustainable and healthier cities.
URBAN FORESTRY & URBAN GREENING
(2023)
Article
Remote Sensing
S. De Petris, F. Sarvia, E. Borgogno-Mondino
Summary: In this study, the authors presented an approach to map the theoretical uncertainty of widely used vegetation spectral indices (VIs) in both time and space domains, using the variance propagation law (VPL). The uncertainty of VIs was mapped over Europe for the entire year 2020, revealing seasonal trends. Additionally, the uncertainty of VI differences resulting from change detection analyses was tested, indicating that some differences were not significant.
EUROPEAN JOURNAL OF REMOTE SENSING
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
F. Ghilardi, S. De Petris, A. Farbo, F. Sarvia, E. Borgogno-Mondino
Summary: This study investigates the stability and spatial variability of crops in a specific area in the western part of the Piemonte Region, Italy, and proposes a method to assess land use intensity. The results show a high rate of crop variation in the area, indicating the need for yearly updates of crop type maps for robust flood damage estimation.
COMPUTATIONAL SCIENCE AND ITS APPLICATIONS - ICCSA 2022 WORKSHOPS, PART III
(2022)
