Review
Environmental Sciences
Milos Rusnak, Tomas Goga, Lukas Michaleje, Monika Sulc Michalkova, Zdenek Macka, Laszlo Bertalan, Anna Kidova
Summary: Riparian zones are important ecosystems that are shaped by interactions between river systems and their surrounding environments. This paper provides an overview of studies that have used remote sensing techniques to understand riparian form, function, and change over time. The majority of studies used aerial and satellite imagery, with unmanned aerial vehicles (UAVs) being increasingly used for low-cost monitoring. However, the challenge remains in effectively transferring remote sensing data to managers and stakeholders for decision making and successful management of riparian zones.
Article
Environmental Sciences
Masoud Mahdianpari, Jean Elizabeth Granger, Fariba Mohammadimanesh, Sherry Warren, Thomas Puestow, Bahram Salehi, Brian Brisco
Summary: This study aims to produce the first high-resolution wetland map of the City of St. John's in Canada using advanced machine learning algorithms, very high-resolution satellite imagery, and airborne LiDAR technology. By applying an object-based random forest algorithm to features extracted from WorldView-4, GeoEye-1, and LiDAR data, the study characterizes five wetland classes within an urban area with an overall accuracy of 91.12% and produces wetland surface water flow connectivity using LiDAR data.
JOURNAL OF ENVIRONMENTAL MANAGEMENT
(2021)
Article
Green & Sustainable Science & Technology
Wanyuan Tang, Fan He, Ali Kashif Bashir, Xun Shao, Yanfen Cheng, Keping Yu
Summary: This paper proposes a remote sensing image rotation object detection method based on dynamic position information Transformer to address the problems of low accuracy and slow detection speed in existing algorithms. The method improves detection accuracy by enhancing the cross-attention operation of the decoder and iteratively updating the position information of object queries. It also improves the network's robustness for remote sensing image object detection using an image pyramid data processing method and introduces a rotating IoU matching loss function for oriented object detection to improve the accuracy of matching predicted boxes to true boxes. Experimental results on DOTA and SSDD datasets show that the proposed algorithm achieves an average detection accuracy of 73.70% and 90.3%, respectively, effectively improving the average detection accuracy of Transformer-based rotating object detection algorithms in aerial remote sensing images and providing better real-time detection performance.
SUSTAINABLE ENERGY TECHNOLOGIES AND ASSESSMENTS
(2023)
Article
Engineering, Electrical & Electronic
Yingchao Han, Weixiao Meng, Wei Tang
Summary: Frequent and accurate object detection based on remote sensing images is important. The DETR model falls short in complex remote sensing scenes where entity information and relative positions between objects are critical. In this article, we propose CI_DETR, a detection model that uses capsule inference to improve remote sensing object detection. Our approach incorporates a multilevel feature fusion module, a capsule reasoning module, and a sausage model, resulting in superior detection performance compared to current detectors.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Environmental Sciences
Spencer Dakin Kuiper, Nicholas C. Coops, Piotr Tompalski, Scott G. Hinch, Alyssa Nonis, Joanne C. White, Jeffery Hamilton, Donald J. Davis
Summary: Understanding changes in salmonid populations and their habitat is crucial due to climate change and their importance as a keystone species. Airborne Laser Scanning (ALS) data can be used to assess the quality and quantity of salmonid habitat, as well as characterize detailed stream attributes. ALS data provides detailed Digital Elevation Models (DEMs) and can be utilized for sustainable forest management decision making and advanced salmonid habitat modeling.
REMOTE SENSING OF ENVIRONMENT
(2022)
Article
Remote Sensing
Spencer Dakin Kuiper, Nicholas C. Coops, Lukas R. Jarron, Piotr Tompalski, Joanne C. White
Summary: The accurate detection and mapping of instream wood is important for sustainable forest management. This study developed and tested a novel framework to use Airborne Laser Scanning (ALS) data to automatically detect and map instream wood. The results showed that the method had moderate overall accuracy and could be used for fish habitat modeling and assessing management practices.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2023)
Review
Environmental Sciences
Bhargavi Janga, Gokul Prathin Asamani, Ziheng Sun, Nicoleta Cristea
Summary: Integrating AI techniques with remote sensing has the potential to revolutionize data analysis and applications in Earth sciences. This review paper synthesizes existing literature on AI applications in remote sensing, analyzing methodologies, outcomes, and limitations. The primary objectives are to identify research gaps, assess the effectiveness of AI approaches, and highlight emerging trends and challenges. The paper explores diverse applications of AI in remote sensing, presents an overview of technologies and methods employed, and discusses challenges and potential solutions. It provides a comprehensive overview for researchers, practitioners, and decision makers in the AI and remote sensing intersection.
Article
Environmental Sciences
Joan Grau, Kang Liang, Jae Ogilvie, Paul Arp, Sheng Li, Bonnie Robertson, Fan-Rui Meng
Summary: The study used UAV and photogrammetry method to map boundaries of riparian zones, showing that high-resolution UAV-derived DEM had high agreement with field-measured riparian zones, making it a suitable approach in agricultural watersheds.
Article
Geochemistry & Geophysics
Jiaqing Zhang, Jie Lei, Weiying Xie, Yunsong Li, Geng Yang, Xiuping Jia
Summary: This paper proposes a guided hybrid quantization with one-to-one self-teaching (GHOST) framework, which combines the synergy of quantization and distillation to achieve a lightweight model. The framework introduces a guided quantization self-distillation (GQSD) structure, a hybrid quantization (HQ) module, and a one-to-one self-teaching (OST) module. Experimental results demonstrate the superiority of the GHOST framework in terms of object detection and lightweight design.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Environmental Sciences
Jiaxin Yang, Miaomiao Yu, Shuohao Li, Jun Zhang, Shengze Hu
Summary: With the development of remote sensing technology, a novel long-tailed object detection method for multimodal remote sensing images is proposed in this paper, which effectively fuses complementary information and adapts to the imbalance between positive and negative samples. Experimental results show that the proposed method achieves state-of-the-art performance on three public benchmark datasets.
Article
Environmental Sciences
Lipeng Gao, Wenzhong Shi, Jun Zhu, Pan Shao, Sitong Sun, Yuanyang Li, Fei Wang, Fukuan Gao
Summary: This paper proposes a novel method for 3D road extraction by integrating LiDAR and remote sensing imagery, which first performs road probability estimation, automatic stratification, multifactor filtering, and elevation interpolation to restore road elevation information, and the experimental results demonstrate the effectiveness of the method.
Article
Environmental Sciences
Baodi Liu, Lifei Zhao, Jiaoyue Li, Hengle Zhao, Weifeng Liu, Ye Li, Yanjiang Wang, Honglong Chen, Weijia Cao
Summary: This paper introduces a saliency-guided remote sensing image super-resolution (SG-GAN) method that utilizes saliency maps to guide the recovery process, achieving the generation of high-resolution salient objects in remote sensing images while maintaining competitive PSNR and SSIM values. Experimental results show the superiority of SG-GAN in restoring structures and generating remote sensing super-resolution images.
Article
Environmental Sciences
Xuesong Zhang, Zhihui Gong, Haitao Guo, Xiangyun Liu, Lei Ding, Kun Zhu, Jiaqi Wang
Summary: This study proposes an adaptive adjacent layer feature fusion (AALFF) method to address the challenges of object detection in remote sensing images. The method incorporates an adjacent layer feature fusion enhancement (ALFFE) module and an adaptive spatial feature fusion (ASFF) module to accurately locate objects and improve adaptability. Experimental results show that the proposed method achieves high mAP values on multiple datasets.
Article
Environmental Sciences
Dongdong Ma, Tanzeel U. Rehman, Libo Zhang, Hideki Maki, Mitchell R. Tuinstra, Jian Jin
Summary: This study examined diurnal variations in remote sensing data in crop phenotyping, and improved prediction accuracy of plant features through modeling.
Article
Computer Science, Information Systems
Zahra Hossein-Nejad, Mehdi Nasri
Summary: This paper proposes a new approach for object recognition in remote-sensing images, using Scale Invariant Feature Transform (SIFT) for matching the object in the template and test images. An adaptive Random sample consensus (RANSAC) algorithm is used to reduce false matches of SIFT, and the extended region-growing algorithm is used to extract the exact object boundary.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Environmental Sciences
Eva M. Kovacs, Chris Roelfsema, James Udy, Simon Baltais, Mitchell Lyons, Stuart Phinn
Summary: The authors developed a machine learning and cloud processing protocol to map seagrass meadows in Moreton Bay, Australia. By incorporating field-survey data, satellite imagery, and a bathymetric layer, they trained a random forest classifier to accurately map seagrass presence/absence. The method proved effective and generated a bay-wide map.
Article
Ecology
Phillip B. McKenna, Alex M. Lechner, Lorna Hernandez Santin, Stuart Phinn, Peter D. Erskine
Summary: This paper evaluates the capability of remote sensing data for monitoring ecosystem restoration and proposes a combination of remote sensing with the ecological recovery wheel (ERW) for improved restoration outcomes.
RESTORATION ECOLOGY
(2023)
Article
Engineering, Environmental
Kasper Johansen, Aislinn F. Dunne, Yu-Hsuan Tu, Burton H. Jones, Matthew F. McCabe
Summary: This study demonstrates the use of high spatial resolution satellite imagery for sub-daily monitoring of coastal water flows and mapping dye plume concentration with the assistance of machine learning methods and high-frequency UAV data. The results show that sub-daily SkySat imagery can effectively track dye plume extent and model dye concentration with low errors, while also showcasing the potential of UAV imagery for scaling between ground data and satellite data for tracking coastal water flow dynamics.
Article
Environmental Sciences
Noam Levin, Stuart Phinn
Summary: In this study, a combination of optical and radar sensors, as well as nighttime and daytime sensors, was used to monitor and evaluate the impact of a flood event in Australia during the summer of 2022. By utilizing various imagery data and a new spectral index, the extent of flooding was accurately mapped, and changes in electricity usage were analyzed as a proxy for flood impact.
Article
Environmental Sciences
Aaron Aeberli, Andrew Robson, Stuart Phinn, David W. Lamb, Kasper Johansen
Summary: This research successfully distinguished banana plants infested with spider mites from those unaffected using field-based spectroscopy, demonstrating that remote sensing approaches can accurately detect mite infestations. Multispectral sensors have the potential to provide a more commercially accessible means of detecting outbreaks, showcasing the importance of utilizing advanced technology in agriculture.
Article
Environmental Sciences
Ting Li, Oliver Miguel Lopez Valencia, Kasper Johansen, Matthew F. F. McCabe
Summary: Agricultural intensification has led to groundwater depletion in many regions. Saudi Arabia, in particular, has experienced significant agricultural expansion. To evaluate the impact of water resource management policies, accurate information on the number and acreage of center-pivot fields is necessary.
Article
Environmental Sciences
Aaron Aeberli, Stuart Phinn, Kasper Johansen, Andrew Robson, David W. Lamb
Summary: Traditional methods of phenological monitoring for banana plantations are labor-intensive and time-inefficient. Unmanned aerial vehicle (UAV) remote sensing technology can provide high-resolution individual plant monitoring, facilitating the determination of banana plant growth stages and yield forecasting.
Article
Astronomy & Astrophysics
James R. R. Kellner, John Armston, Laura Duncanson
Summary: GEDI is a laser altimeter on the ISS designed to measure vegetation height and quantify carbon stocks. The algorithm uses height metrics and linear models to predict aboveground biomass density. The predictions provide globally comprehensive estimates of AGBD.
EARTH AND SPACE SCIENCE
(2023)
Article
Environmental Sciences
K. C. Cushman, John Armston, Ralph Dubayah, Laura Duncanson, Steven Hancock, David Janik, Kamil Kral, Martin Krucek, David M. Minor, Hao Tang, James R. Kellner
Summary: In this study, the sensitivity of Global Ecosystem Dynamics Investigation (GEDI) data and aboveground biomass density (AGBD) predictions to leaf phenology was tested. The results suggest that, with consideration of model choice, GEDI data without considering leaf status can be used for AGBD prediction, which increases data availability and reduces sampling error in some forests.
ENVIRONMENTAL RESEARCH LETTERS
(2023)
Article
Computer Science, Information Systems
Bagus Setiabudi Wiwoho, Stuart Phinn, Neil McIntyre
Summary: This study investigates the land-use changes in Java's Brantas River Basin, showing that major transitions occurred from forest to shrubs, dryland agriculture, and urban areas during 1995-2015. The drivers of these changes include economic, social, technological, and biophysical attributes, as indicated by land-user questionnaires. The heterogeneity and scale-dependence of the land-use change process are highlighted by the combination of these two approaches.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2023)
Article
Ecology
Kim Calders, Benjamin Brede, Glenn Newnham, Darius Culvenor, John Armston, Harm Bartholomeus, Anne Griebel, Jodie Hayward, Samuli Junttila, Alvaro Lau, Shaun Levick, Rosalinda Morrone, Niall Origo, Marion Pfeifer, Jan Verbesselt, Martin Herold
Summary: Climate change and human activities are affecting ecosystems and biodiversity. Quantitative measurements of essential biodiversity variables and climate variables are used to monitor and evaluate interventions. Spaceborne measurements lack detailed information on three-dimensional vegetation structure at local scales, but ground-based laser scanning shows potential for systematic monitoring.
REMOTE SENSING IN ECOLOGY AND CONSERVATION
(2023)
Article
Environmental Studies
Andrew Clark, Stuart Phinn, Peter Scarth
Summary: Data pre-processing is important for developing a generalised land use and land cover (LULC) deep learning model using earth observation data. This paper trialled different methods of data preparation for Convolutional Neural Network (CNN) training and achieved accurate classification of LULC features in aerial photography. The results suggest that stratified random sampling, smaller batch sizes, data augmentations, scaling, and averaging multiple grids of patches improved the model accuracy and aesthetic result.
Editorial Material
Environmental Sciences
Nathan Thomas, Mikhail Urbazaev, Atticus E. L. Stovall, Laura Hess, John Armston, Amy Neuenschwander, Lola Fatoyinbo, Laura Duncanson
ENVIRONMENTAL RESEARCH LETTERS
(2023)
Article
Remote Sensing
Jamal Elfarkh, Kasper Johansen, Victor Angulo, Omar Lopez Camargo, Matthew F. Mccabe
Summary: Land Surface Temperature (LST) is a crucial variable used in various applications, and while satellites offer moderate-resolution LST data, unmanned aerial vehicles (UAVs) can provide high-resolution thermal infrared measurements. However, the continuous and rapid variation in LST poses challenges in producing orthomosaics from UAV-based image collections. This research examines LST variations during standard 15-20 min UAV flights over diverse surfaces and identifies factors such as wind speed, solar radiation, irrigation, and atmospheric conditions that contribute to temperature variations. Understanding these factors is essential for developing correction procedures and interpreting UAV-based thermal infrared data and orthomosaics.
Article
Biodiversity Conservation
Yongcui Lan, Jinliang Wang, Qianwei Liu, Fang Liu, Lanfang Liu, Jie Li, Mengjia Luo
Summary: This study focuses on the five major plateau lake basins in central Yunnan, China, and constructs an ecological security pattern using the source-resistance surface-corridor-pinch point framework. The study simulates land use/cover change in the region and identifies early warning regions where future urban expansion poses a threat to current ecological source areas and corridors.
ECOLOGICAL INDICATORS
(2024)
Article
Biodiversity Conservation
Pingping Huang, Feng Zhao, Bailing Zhou, Kuidong Xu
Summary: This study investigates the distribution of benthic microeukaryotes in the China Seas and finds that they can stride over the ecological barrier of 32 degrees N. The study also highlights the significant influence of depth, temperature, and latitude on communities in the China Seas.
ECOLOGICAL INDICATORS
(2024)
Article
Biodiversity Conservation
Federico Morelli, Yanina Benedetti, Jesse Stanford, Leszek Jerzak, Piotr Tryjanowski, Paolo Perna, Riccardo Santolini
Summary: Species distribution models (SDMs) are numerical tools used for predicting species' spatial distribution. This study found that ecological characteristics, such as habitat specialization, play a role in improving the accuracy of SDMs.
ECOLOGICAL INDICATORS
(2024)
Article
Biodiversity Conservation
Xiaoxuan Wu, Hang Liu, Wei Liu
Summary: Global climate change, urbanization, and economic development have increased the need for sustainable human development, urban ecological governance, and low-carbon energy transformation. This study analyzes the green ecological transition in Chengdu based on panel data from 2010 to 2020, exploring its spatiotemporal evolution and key factors. The results show an overall upward trend in Chengdu's green ecological development and positive spatial autocorrelation in certain districts.
ECOLOGICAL INDICATORS
(2024)
Article
Biodiversity Conservation
Castaldi Simona, Formicola Nicola, Mastrocicco Micol, Morales Rodriguez Carmen, Morelli Raffaella, Prodorutti Daniele, Vannini Andrea, Zanzotti Roberto
Summary: Sustainable agricultural practices are increasingly important for global and national environmental policies and economy. This study compared the sustainability of grape production under integrated and organic management using multiple indicators. The results showed that organic management was more beneficial for most environmental aspects of the agroecosystem compared to integrated management, without affecting grape yield.
ECOLOGICAL INDICATORS
(2024)
Article
Biodiversity Conservation
Gaia Vaglio Laurin, Alexander Cotrina-Sanchez, Luca Belelli-Marchesini, Enrico Tomelleri, Giovanna Battipaglia, Claudia Cocozza, Francesco Niccoli, Jerzy Piotr Kabala, Damiano Gianelle, Loris Vescovo, Luca Da Ros, Riccardo Valentini
Summary: Phenology monitoring is important for understanding forest functioning and climate impacts. This research compares the phenological behavior of European beech forests using Tree-Talker (TT+) and Sentinel 2 satellite data. The study finds differences in the information derived by the two sensor types, particularly in terms of season length, phenology changepoints, and leaf period variability. TT+ with its higher temporal resolution demonstrates precision in capturing the phenological changepoints, especially when satellite image availability is limited.
ECOLOGICAL INDICATORS
(2024)
Article
Biodiversity Conservation
Huanhuan Pan, Ziqiang Du, Zhitao Wu, Hong Zhang, Keming Ma
Summary: The land use and cover changes resulting from coal mining activities and ecological restoration have had a significant impact on ecosystem services in mining areas. This study investigates the relationship between ecosystem services and land use intensity in coal mining areas, emphasizing the importance of understanding this interdependence for balanced human-land system development. The research examines the evolving relationship across different reclamation stages in Shanxi, China, using a coupling coordination degree model. The findings suggest the need for timely and judicious reclamation of coalfields, considering the land's bearing capacity.
ECOLOGICAL INDICATORS
(2024)
Article
Biodiversity Conservation
Jingjuan He, Yijun Shi, Lihua Xu, Zhangwei Lu, Mao Feng
Summary: This study examines the spatial interplay between changes in the blue-green spatial distribution and modifications in land surface temperature grades in Shanghai. The findings reveal that the transformation of the blue-green spatial pattern differs between different sectors of the city, and the impact on the thermal environment varies spatially.
ECOLOGICAL INDICATORS
(2024)
Article
Biodiversity Conservation
Yi Xu, Di Zhang, Junqiang Lin, Qidong Peng, Xiaohui Lei, Tiantian Jin, Jia Wang, Ruifang Yuan
Summary: This study analyzed the response relationship between phytoplankton growth and water environmental parameters in the Middle Route of the South-to-North Water Diversion Project in China using long-term monitoring data and machine learning models. The results revealed the differences between monitoring sites and identified the key parameters that affect phytoplankton growth.
ECOLOGICAL INDICATORS
(2024)