Article
Environmental Sciences
Andrey Shikhov, Polina Ilyushina, Olga Makarieva, Anastasiia Zemlianskova, Maria Mozgina
Summary: This study provides a detailed overview of land-cover changes associated with gold mining in the Magadan region of Russia's northeastern part. It found that the floodplains of the Berelekh and Debin Rivers are most heavily impacted by gold mining, with a removed vegetation canopy occupying 16.0% and 11.2% of their area.
Article
Environmental Studies
Athos Agapiou
Summary: This study explores the potential of using colourised images for land cover characterisation from archive greyscale satellite sources, achieving promising results of over 85% classification accuracy on various land cover classes through the application of the support vector machine (SVM) classifier.
Article
Environmental Sciences
Fabien H. Wagner, Ricardo Dalagnol, Celso H. L. Silva-Junior, Griffin Carter, Alison L. Ritz, Mayumi C. M. Hirye, Jean P. H. B. Ometto, Sassan Saatchi
Summary: Monitoring changes in tropical tree cover using a U-net deep learning model with high-resolution satellite images showed high accuracy in mapping monthly tree cover and identifying deforestation in Mato Grosso, Brazil. The model's performance was validated by LiDAR data, and the deforestation map had consistent agreement with the official map from Brazil. The study estimated that 14.8% of Mato Grosso's area experienced clear-cut logging between 2015 and 2021.
Article
Geography, Physical
Xin-Yi Tong, Gui-Song Xia, Xiao Xiang Zhu
Summary: High-resolution satellite images are valuable for land cover classification, but their application in detailed mapping at large scale is limited. To address this, we present a large-scale land cover dataset called Five-Billion-Pixels, with over 5 billion labeled pixels from 150 high-resolution Gaofen-2 satellite images. We also propose a deep-learning-based unsupervised domain adaptation approach for large-scale land cover mapping. Experimental results show promising performance even with entirely unlabeled images.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2023)
Article
Environmental Sciences
Mojtaba Saboori, Saeid Homayouni, Reza Shah-Hosseini, Ying Zhang
Summary: This paper proposes a comprehensive analysis for optimum feature selection and the most efficient classifier for accurate urban area mapping. The experiments reveal that RF, PSO, and NCA are the most efficient classifiers, and wrapper-based and filter-based methods are the most efficient feature selection methods. Dissimilarity, contrast, and correlation features play the greatest contributing role in the classification performance.
Article
Environmental Sciences
Vincent B. Verhoeven, Irene C. Dedoussi
Summary: Land cover plays a crucial role in the Earth's climate, food security, and biodiversity. By using classification trees to generate annual land cover maps, it is possible to accurately classify and monitor changes in land cover on the European continent.
JOURNAL OF ENVIRONMENTAL MANAGEMENT
(2022)
Article
Environmental Sciences
Yibo Wang, Xia Zhang, Changping Huang, Wenchao Qi, Jinnian Wang, Xiankun Yang, Songtao Ding, Shiyu Tao
Summary: A Spatial-Convolution Spectral-Transformer Interactive Network (SCSTIN) model was proposed to enhance feature extraction capabilities and address the challenges in satellite hyperspectral imagery classification and mapping. The model achieved satisfactory performance in accuracy and efficiency, making it reliable for large-scale fast refined land cover classification and mapping.
Article
Geography, Physical
Fei Xu, Ben Somers
Summary: The study introduces an image fusion algorithm UnFuSen2 that improves accuracy in urban landscapes, and applies Multiple Endmember Spectral Mixture Analysis (MESMA) to generate urban land cover fractions. The results show a decrease in RMSE for impervious surface and vegetation fractions compared to traditional methods.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2021)
Article
Engineering, Electrical & Electronic
Xining Zhang, Yong Ge, Feng Ling, Jin Chen, Yuehong Chen, Yuanxin Jia
Summary: The study introduces a super-resolution mapping method called SRMGCN, which aims to improve SRM results by capturing structure information on the graph, and demonstrates its superiority through experimental results.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Article
Environmental Sciences
Da He, Qian Shi, Jingqian Xue, Peter M. Atkinson, Xiaoping Liu, Marie Weiss
Summary: In this research, a learnable correlation-based sub-pixel mapping (LECOS) method is developed to tackle the mixed pixel effect in urban land use/land cover classification. The method effectively models teleconnections and diverse global correlation patterns, resulting in accurate sub-pixel reconstruction of complex urban scenes. The derived results demonstrate rich urban spatial heterogeneity and suggest the potential for greater understanding of urban issues.
REMOTE SENSING OF ENVIRONMENT
(2023)
Article
Environmental Sciences
Zhongxiang Fang, Martin Brandt, Lanhui Wang, Rasmus Fensholt
Summary: Plant phenology provides crucial information on the seasonal dynamics of plants, with changes reflecting the impact of climate change and human management on the biosphere. This study focuses on the impact of changes in tree cover on satellite observed land surface phenology globally over the past three decades, revealing that areas where tree cover increased experienced an extension of the growing season length in 36.6% of cases, compared to only 20.1% in areas with decreased tree cover. Additionally, the ratio between tree cover and short vegetation cover plays a role in influencing the length of the growing season, with denser tree cover showing a more pronounced extension (especially in boreal forests).
SCIENCE OF THE TOTAL ENVIRONMENT
(2022)
Article
Environmental Sciences
Wanli Ma, Oktay Karaku, Paul L. Rosin
Summary: Land cover mapping is a widely used technique in remote sensing computational imaging that provides spatial information on various classes of physical properties on the Earth's surface. It plays a crucial role in developing solutions to environmental problems and faces challenges in integrating complementary information from multi-modal remote sensing imagery.
Article
Environmental Studies
Guste Metrikaityte, Jurate Suziedelyte Visockiene, Kestutis Papsys
Summary: The aim of this article is to choose the most appropriate method for identifying and managing land cover changes over time. The study is based on remote sensing techniques and different data and methods were used. The results of SAR and MSI image segmentation differed, with lower accuracy in identifying urban areas. Improving the reliability and accuracy of the results requires combining field surveys and calculating additional indices to enhance the visibility of the segmentation results.
Article
Engineering, Electrical & Electronic
Emmanuel Capliez, Dino Ienco, Raffaele Gaetano, Nicolas Baghdadi, Adrien Hadj Salah
Summary: Nowadays, satellite image time series (SITS) are commonly used to derive land-cover maps (LCM) to support land management applications. However, the availability of ground truth data for supervised machine learning models is often limited. This study proposes a spatially aligned domain-adversarial neural network framework to transfer a classification model from one time period to a successive one in a specific study area, overcoming distribution shifts between source and target domains. Experimental results in Burkina Faso demonstrate the superiority of this approach in three different transfer tasks, outperforming other unsupervised domain adaptation methods by 7 to 12 points of F1-score.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Environmental Sciences
Elif Sertel, Burak Ekim, Paria Ettehadi Osgouei, M. Erdem Kabadayi
Summary: This research focuses on the deep learning-based segmentation of VHR satellite images, specifically for LULC mapping. The study shows that the DeepLabv3+ architecture with a ResNeXt50 encoder achieves the best performance, providing highly accurate LULC maps.
Article
Biology
Yuzi Zhang, Howard H. Chang, Qu Cheng, Philip A. Collender, Ting Li, Jinge He, Justin Remais
Summary: Passive surveillance systems are cost-effective and widely used in monitoring disease occurrence, but are associated with imperfect case ascertainment and heterogeneous capture probabilities. This study proposes a hierarchical modeling framework that combines multiple surveillance systems to improve estimation of incident cases, and applies the model to pulmonary tuberculosis surveillance in China, leading to bias-corrected estimates and identification of risk factors.
Article
Environmental Sciences
Tao Liu, Le Yu, Xin Chen, Hui Wu, Hui Lin, Chengxiu Li, Jiaru Hou
Summary: In recent decades, China has implemented ecological restoration projects to improve biodiversity and ecosystem services, and environmental laws have been issued for guidance. However, the quantitative evaluation of these projects' effectiveness remains unclear. In response to the UN Decade on Ecosystem Restoration, a meta-analysis was conducted to assess the effectiveness of these projects based on China's environmental protection and land administration laws. The findings showed that the projects enhanced ecosystem services by 15-58%, with varying effects in different regions and climate factors being crucial.
JOURNAL OF ENVIRONMENTAL MANAGEMENT
(2023)
Article
Remote Sensing
Fengjie Ren, Hui Lu, Bin Wang, Kun Yang, Le Yu, Weiwei Gan, Tianjie Zhao
Summary: In this study, SMAP36 and SMAP09 products from the SMAP mission and the AMSR2 FT product were compared against surface soil temperature measurements in Russia. The results show that AMSR2 FT performs better than the SMAP products, particularly in the overall accuracy of the descending orbit. AMSR2 FT has high accuracy in inland areas but lower accuracy along coastal zones due to the presence of water bodies or lake ice. The study also highlights the need to improve the algorithm for wet snow detection in mixed forest areas.
REMOTE SENSING LETTERS
(2023)
Article
Environmental Sciences
Xueliang Feng, Shen Tan, Yun Dong, Xin Zhang, Jiaming Xu, Liheng Zhong, Le Yu
Summary: Bamboo forest is a unique forest landscape mainly composed of herbal plants. It has a stronger capability to increase terrestrial carbon sinks than woody forests, playing a special role in absorbing atmospheric CO2. Accurate and timely bamboo forest maps are necessary for understanding and quantifying their contribution to the carbon and hydrological cycles.
Article
Environmental Sciences
Xianda Huang, Fu Xuan, Yi Dong, Wei Su, Xinsheng Wang, Jianxi Huang, Xuecao Li, Yelu Zeng, Shuangxi Miao, Jiayu Li
Summary: This study explores the potential of Chinese GF-1 PMS high-spatial-resolution images for corn lodging monitoring and finds a robust and efficient way to accurately identify corn lodging. The combination of spectral bands, optimized vegetation indexes, and texture features achieves an overall accuracy of 93.81% and a Kappa coefficient of 0.91 in classifying corn lodging. The random forest is an efficient, robust, and easy classifier, achieving high accuracy in identifying non-lodged, moderately lodged, and severely lodged areas.
Article
Environmental Sciences
Zhiying Yao, Yuanyuan Zhao, Hengbin Wang, Hongdong Li, Xinqun Yuan, Tianwei Ren, Le Yu, Zhe Liu, Xiaodong Zhang, Shaoming Li
Summary: As orchards have high economic, ecological, and cultural value, accurate and timely mapping of orchards is highly demanded. Selecting a remote-sensing data source is crucial, and a trade-off between spatial and temporal resolution must be made. In this study, different spatial and temporal resolution images were tested for orchard mapping, and their performance was assessed. Results showed that increasing spatial resolution and the number of images used improved overall accuracy, with temporal information having a higher classification ability than spatial information. Combining spatial and temporal features improved accuracies compared to using only temporal features or single-source data.
Article
Environmental Sciences
Wei Wu, Qinchuan Xin
Summary: Monitoring and analyzing the dynamics of land surface spring phenology is crucial for quantifying the impact of climate change on terrestrial ecosystems. This study analyzed the spatial and temporal patterns of spring phenology in different land cover types in the conterminous United States using MODIS data from 2001 to 2021, and evaluated the performance of satellite-derived data. The results showed a delayed trend of spring phenology with increasing latitude, and a positive correlation between satellite-derived data and in-situ data.
Article
Urban Studies
Wanru He, Xuecao Li, Yuyu Zhou, Xiaoping Liu, Peng Gong, Tengyun Hu, Peiyi Yin, Jianxi Huang, Jianyu Yang, Shuangxi Miao, Xi Wang, Tinghai Wu
Summary: Cellular automata (CA) based models are widely used in urban sprawl modeling for sustainable urban planning. However, most existing urban CA models only consider abrupt conversion, ignoring the difference in urbanization levels among grids and the gradual increase in urban densities. In this study, we proposed an impervious surface area (ISA) based urban CA model that can simulate urban fractional change within each grid. The model was implemented in Beijing and evaluated through comparison and scenario analyses. Results showed that the ISA-based urban CA model captures the dynamics of urban sprawl better than the traditional urban CA model and has great potential in supporting sustainable urban development.
Article
Environmental Sciences
Hui Wu, Shiming Fang, Le Yu, Shougeng Hu, Xin Chen, Yue Cao, Zhenrong Du, Xiaoli Shen, Xuehua Liu, Keping Ma
Summary: To maximize co-benefits after 2020, the existing protected areas network needs to be optimized and conservation targets need to be identified. A study in Southwest China found inconsistencies between biodiversity, climate vulnerability, and wilderness hotspots, and identified significant gaps in the protected areas network. To address this, conservation targets for 2025, 2030, and 2050 are proposed to expand the coverage of protected areas.
JOURNAL OF ENVIRONMENTAL MANAGEMENT
(2023)
Article
Biochemistry & Molecular Biology
Jiayuan Xu, Nana Liu, Elli Polemiti, Liliana Garcia-Mondragon, Jie Tang, Xiaoxuan Liu, Tristram Lett, Le Yu, Markus M. Noethen, Jianfeng Feng, Chunshui Yu, Andre Marquand, Gunter Schumann
Summary: Analyses of data from the UK Biobank reveal that different urban living environments are associated with affective, anxiety, and emotional instability symptom groups in adults. These associations are mediated by distinct neurological and genetic pathways. Using data from 156,075 participants, researchers found that social deprivation, air pollution, street network, and urban land-use density were positively correlated with affective symptoms. On the other hand, greenness and generous destination accessibility were negatively correlated with anxiety symptoms. The study suggests that urban living environments may influence specific psychiatric symptom groups through different neurobiological pathways.
Article
Environmental Sciences
Victoria L. Woltz, Camille LaFosse Stagg, Kristin B. Byrd, Lisamarie Windham-Myers, Andre S. Rovai, Zhiliang Zhu
Summary: Accurate assessments of greenhouse gas emissions and carbon sequestration in natural ecosystems are important for climate mitigation strategies. Remote sensing products are needed to quantify carbon stocks and fluxes for large areas. This study developed spatially explicit models for tidal herbaceous marshes to estimate aboveground biomass carbon stock and net primary productivity (NPP) using remote sensing data.
Article
Green & Sustainable Science & Technology
Yongjian Ruan, Baozhen Ruan, Xinchang Zhang, Zurui Ao, Qinchuan Xin, Ying Sun, Fengrui Jing
Summary: This paper proposes a framework for extracting fine-resolution land surface phenology using the supercomputer Tianhe-2. The framework combines MODIS and Landsat images to generate a dense two-band enhanced vegetation index (EVI2) time series, and then extracts the long-term and fine-resolution phenology using the fused EVI2 dataset. The results show that the proposed framework can obtain robust fine-resolution land surface phenology and has potential applications in ecological environmental studies.
Article
Food Science & Technology
Faqin Lin, Xuecao Li, Ningyuan Jia, Fan Feng, Hai Huang, Jianxi Huang, Shenggen Fan, Philippe Ciais, Xiao-Peng Song
Summary: Ukraine and Russia, as major grain producers and exporters, have a significant impact on the global wheat market. The conflict between the two countries has led to reduced wheat production in Ukraine and disruptions in wheat trade. Analysis shows that the conflict will result in a drop in global wheat trade, soaring prices, and severe food insecurity, particularly for countries heavily reliant on Ukrainian wheat imports. Additionally, the trade blockade caused by the conflict will lead to price increases and welfare declines for affected countries.
GLOBAL FOOD SECURITY-AGRICULTURE POLICY ECONOMICS AND ENVIRONMENT
(2023)
Article
Multidisciplinary Sciences
Rui Wang, Wenjia Cai, Le Yu, Wei Li, Lei Zhu, Bowen Cao, Jin Li, Jianxiang Shen, Shihui Zhang, Yaoyu Nie, Can Wang
Summary: Assessing biomass resource potential is crucial for China's goals of carbon neutrality, rural revitalization, and poverty eradication. This study estimates the biomass resource potential for various types of biomass feedstock in China at a high spatial resolution of 1 km. The assessment framework developed in this study combines statistical accounting and GIS-based methods, ensuring transparency and compliance with principles of food security, land and biodiversity protection. The reliability of the dataset is verified through comparisons with existing literature.
Article
Remote Sensing
Fu Xuan, Yi Dong, Jiayu Li, Xuecao Li, Wei Su, Xianda Huang, Jianxi Huang, Zixuan Xie, Ziqian Li, Hui Liu, Wancheng Tao, Yanan Wen, Ying Zhang
Summary: Northeast China, as a major grain bank in China, has a significant impact on food security. In order to address the challenges posed by increasing food demands and soil protection, crop rotation and fallowing policies have been implemented. This study utilizes time-series Landsat-8 OLI images and applies an automatic sampling approach using a hexagon strategy and tile-based classification using the random forest algorithm to monitor crop types and their changes in Northeast China. The resulting crop maps exhibit high credibility with overall accuracies ranging from 0.89 to 0.97 and demonstrate good agreement with city-level statistical data. The dataset generated from this study provides reliable long-term crop maps, which can be valuable for food security and regional agricultural production management.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2023)
