Article
Environmental Sciences
Fanggang Li, Erzhu Li, Ce Zhang, Alim Samat, Wei Liu, Chunmei Li, Peter M. Atkinson
Summary: This study proposed the use of multi-temporal MODIS and VIIRS data to construct a general feature set for predicting large-area impervious surfaces. The random forest model based on these features showed improved prediction accuracy in China and across Asia. Comparisons with existing impervious products indicated superior performance in certain regions for the proposed method.
Article
Geography, Physical
Xiang Wang, Zhidan Wen, Ge Liu, Hui Tao, Kaishan Song
Summary: This study proposes a stable hybrid model using local regression and WRF to estimate TSM concentrations in different estuaries in China. Different band indices exhibit varying sensitivity to TSM concentrations, with the RF model outperforming other models in prediction. The WRF model provides a more reasonable approach for TSM concentration mapping compared to the original RF model.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2022)
Article
Soil Science
Xiang Wang, Liping Wang, Sijia Li, Zongming Wang, Miao Zheng, Kaishan Song
Summary: This study proposes a probability hybrid model for estimating and mapping soil organic carbon (SOC) content and distribution, and compares it with other models. The results show that the probability hybrid model achieves higher prediction accuracy and spatial continuity in digital SOC mapping.
Article
Environmental Sciences
Xueli Peng, Guojin He, Wenqing She, Xiaomei Zhang, Guizhou Wang, Ranyu Yin, Tengfei Long
Summary: This paper analyzes and compares the performance of imagery from GF-1 WFV, Landsat 8, and Sentinel-2 satellites in forest/non-forest classification tasks, showing that Sentinel-2 data have the highest accuracy, GF-1 WFV the second highest, and Landsat 8 the lowest.
Article
Geosciences, Multidisciplinary
Sen Zhang, Jia Tian, Xia Lu, Qingjiu Tian
Summary: In this study, the dynamic changes of soil organic carbon (SOC) content in coastal wetlands over the past 23 years were investigated in the third core area of the Dafeng Elk Nature Reserve in Jiangsu Province, China. By using Landsat images, a normalized difference soil spectral index (NDSOC) and six auxiliary environmental variables related to SOC were constructed and extracted. The inversion model for coastal wetland soil surface SOC was constructed using multiple linear regression, support vector machine, and particle swarm optimization-based random forest regression (PSO-RFR). The results showed that the NDSOC index, NDVI, soil clay content, surface temperature, and soil salinity were strongly linked to SOC content in coastal wetlands.
Article
Environmental Sciences
Katsuto Shimizu, Wataru Murakami, Takahisa Furuichi, Ronald C. Estoque
Summary: In this study, the applicability of Landsat time series temporal segmentation and random forest classifiers for mapping LULCC and forest disturbances in Vietnam was examined. The results showed that although there were some accuracy issues, this method is still useful for consistently mapping LULCC and forest disturbances.
Article
Computer Science, Information Systems
Hebing Zhang, Hongyi Yuan, Weibing Du, Xiaoxuan Lyu
Summary: This study focused on crop identification using multi-source features fused from Sentinel-1 and Sentinel-2 data in Wen County, Henan Province, China. The results showed that fusion data improved the accuracy of crop identification, and multi-temporal data effectively characterized the phenological stages of crop growth.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2022)
Article
Computer Science, Artificial Intelligence
Hideyuki Doi, Tomoki Hirai
Summary: This study used machine learning methods to classify Landsat 8 satellite images and estimate the area of uprooted trees in forests. Support vector machines and random forests performed better in two-class classification, with all methods achieving up to 93% correct classification rates. CNN performed significantly better in four- and two-class classification, aiding in estimating the area of uprooted trees.
PEERJ COMPUTER SCIENCE
(2021)
Article
Environmental Sciences
Junwei Wang, Zhongping Lee, Daosheng Wang, Shaoling Shang, Jianwei Wei, Alex Gilerson
Summary: The innovative MPACA approach for atmospheric correction in coastal waters, based on a revised POLYMER model, showed promising results in accurately retrieving remote sensing reflectance (R-rs(lambda)) from high-spatial-resolution satellite measurements. The performance of MPACA, which assumed uniform aerosol types and varying aerosol loads and water properties, was validated using Landsat-8 OLI images and in-situ matchups, outperforming other methods like SeaDAS and Acolite.
REMOTE SENSING OF ENVIRONMENT
(2021)
Article
Environmental Sciences
Yansi Chen, Jinliang Hou, Chunlin Huang, Ying Zhang, Xianghua Li
Summary: Accurate estimation of crop area is crucial for adjusting regional crop planting structure and water resource planning. High-resolution multi-temporal Sentinel-1 (S1) radar backscatter and Sentinel-2 (S2) optical reflectance images were combined to map maize in complex landscapes. The new two-step method of vegetation extraction and maize extraction based on RF classifiers improved accuracy compared to using S1 or S2 images alone.
Article
Environmental Sciences
Federico Santini, Angelo Palombo
Summary: This paper investigates the impact of topographic correction on remote sensing images and highlights the necessity of atmospheric correction in mountainous environments. By using data acquired under different illumination conditions with different sensors and conducting a statistical analysis, the importance of topographic correction is demonstrated.
Article
Environmental Sciences
Adeniyi Adeyemi, Abel Ramoelo, Moses Azong Cho, Jacobus Strydom
Summary: This study assessed the impact of rapid urban growth on natural lands and the expansion of impervious surface area in 1995, 2005, and 2015. The findings showed that most areas in Pretoria experienced ISA expansion, and the results from Landsat TIRS bands calculations were also accurate.
GEOCARTO INTERNATIONAL
(2022)
Article
Environmental Sciences
Zhaohua Liu, Zilin Ye, Xiaodong Xu, Hui Lin, Tingchen Zhang, Jiangping Long
Summary: This study aims to improve the accuracy of forest stock volume (FSV) estimation in planted eucalyptus forests. Time series Landsat 8 OLI (LC8) images and ZY-3 stereo images were utilized to extract spectral variables and corrected canopy height model (CCHM) for FSV estimation using four models. The results indicate that spectral variables based on crown growth characteristics and forest height play a significant role in improving FSV mapping accuracy, while composite images and CCHM have the potential to delay saturation in FSV mapping for planted eucalyptus forests.
Article
Remote Sensing
Yanxi Li, Rui Chen, Binbin He, Sander Veraverbeke
Summary: This study integrated optical data and synthetic aperture radar (SAR) data to estimate foliage fuel load (FFL) by analyzing spatiotemporal features. The results showed that both SAR and optical data contributed significantly to FFL estimation, with the best performance achieved when the two data sources were combined. Additionally, temporal features were found to be more important predictors of FFL than spatial features.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2022)
Article
Environmental Sciences
Wen Liu, Yoshihisa Maruyama, Fumio Yamazaki
Summary: The study focuses on identifying washed-away or collapsed bridges using multi-temporal SAR intensity images, with a random forest model trained on mixed events showing the highest capability to extract collapsed bridges. After improvement by introducing an oversampling technique, the model achieved high accuracy with an F-score of 0.87 and a kappa coefficient of 0.82.
Article
Environmental Sciences
Su Ye, John Rogan, Zhe Zhu, J. Ronald Eastman
Summary: This study introduces a new approach for monitoring forest disturbances using Landsat data time series, which incorporates a mathematical tool and state space model for accurate analysis of forest disturbances. The results show that the method performs well in both quantitative and qualitative analyses, with slightly higher accuracy compared to current methods, and provides better real-time monitoring capabilities.
REMOTE SENSING OF ENVIRONMENT
(2021)
Article
Geography, Physical
Xiao Huang, Zhenlong Li, Yuqin Jiang, Xinyue Ye, Chengbin Deng, Jiajia Zhang, Xiaoming Li
Summary: This study reveals that higher-income counties in the U.S tend to react more aggressively in reducing mobility during the COVID-19 pandemic, showing significant disparities with lower-income counties. The unique characteristics of mobility data from different sources demonstrate the multifaceted nature of human mobility.
INTERNATIONAL JOURNAL OF DIGITAL EARTH
(2021)
Article
Geography, Physical
Qiang Zhou, Zhe Zhu, George Xian, Congcong Li
Summary: Harmonic analysis of time series is important for revealing seasonal land surface dynamics using remote sensing information, but frequency selection can be difficult. The Harmonic Adaptive Penalty Operator (HAPO) is a novel regression method that addresses this issue.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Amal H. Aljaddani, Xiao-Peng Song, Zhe Zhu
Summary: Development and population growth in Saudi Arabia have led to significant urban expansion. This study generated a long-term dataset and analyzed the urban expansion patterns and trends in all regional capitals in Saudi Arabia, providing high accuracy classification results and urban change maps.
Article
Environmental Sciences
Chadwick D. Rittenhouse, Elana H. Berlin, Nathaniel Mikle, Shi Qiu, Dustin Riordan, Zhe Zhu
Summary: This study aimed to develop a multi-source, object-based approach using Landsat, LiDAR, and aerial imagery to map young forest and shrubland vegetation in a temperate forest. The estimated young forest and shrubland vegetation in Connecticut accounted for 3.3% of total land cover and 6.3% of forest cover, highlighting the need for conservation and management efforts.
Article
Geography, Physical
Qiang Zhou, George Xian, Josephine Horton, Danika Wellington, Grant Domke, Roger Auch, Congcong Li, Zhe Zhu
Summary: Forests cover about one-third of the land area of the conterminous United States (CONUS) and play a crucial role in offsetting carbon emissions and supporting local economies. The demand for information on forest regrowth and recovery following disturbances has increased, particularly for cost-effective nature-based climate solutions. However, mapping the tree regrowth duration at an annual time interval and high resolution remains challenging.
GISCIENCE & REMOTE SENSING
(2022)
Review
Environmental Sciences
Zhe Zhu, Shi Qiu, Su Ye
Summary: This paper discusses the rapid evolution of land change science and the significant role played by remote sensing in observing, monitoring, and characterizing land change. A new framework is proposed to understand land change from multiple perspectives using remote sensing, and five aspects of land change are recommended. The impacts of different domains of remote sensed data on observing, monitoring, and characterizing land change are evaluated. The paper also explores current land change products.
REMOTE SENSING OF ENVIRONMENT
(2022)
Article
Engineering, Electrical & Electronic
Xinrong Huang, Wenjie Liu, Jinhao Meng, Yuanyuan Li, Siyu Jin, Remus Teodorescu, Daniel-Ioan Stroe
Summary: This article investigates the effect of low-frequency positive pulsed current (PPC) charging on the lifetime and charging performance of Li-ion batteries. Experimental results show that compared to traditional constant current (CC) charging, Li-ion batteries cycled by PPC charging have improved lifetime, maximum rising temperature, and energy efficiency by 81.6%, 60.5%, and 9.1% respectively. Therefore, low-frequency PPC charging should be considered as a promising charging strategy for Li-ion batteries.
IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS
(2023)
Article
Environmental Sciences
Zhe Zhu, Yu Li, Yan Sun, Zhihong Liu, Chi Zhang
Summary: Due to the uncertainty in sensor data and low model accuracy in water quality modeling, pollution source detection often leads to non-uniqueness effect. This study proposes a decision support framework that uses a consensus-based multiple information fusion method to reduce uncertainty and identify unique solutions. Multiple water quality information sources are fused via spatial clustering and temporal Bayesian updating. The framework is validated using real and hypothetical case studies, and the results demonstrate its effectiveness in water quality control.
WATER RESOURCES RESEARCH
(2023)
Article
Environmental Sciences
Jie Hu, Alfred E. Hartemink, Ankur R. Desai, Philip A. Townsend, Rose Z. Abramoff, Zhe Zhu, Debjani Sihi, Jingyi Huang
Summary: Current carbon cycle models have focused on the effects of climate and land-use change on primary productivity and microbial-mineral dependent carbon turnover in the topsoil, but have overlooked the importance of vertical soil processes and soil response to land-use change along the profile. In this study, spatial-temporal analysis was used to estimate soil organic carbon (SOC) change at NEON sites in the USA over 30 years. The study found that different soil types and land-use practices had significant impacts on SOC accumulation or loss, and identified runoff/erosion, leaching potential, vertical translocation, and mineral sorption as the key factors controlling SOC variation.
JOURNAL OF GEOPHYSICAL RESEARCH-BIOGEOSCIENCES
(2023)
Article
Remote Sensing
Yongquan Zhao, Zhe Zhu
Summary: A new Artificial Surface Index (ASI) was proposed in this study to extract artificial surfaces using multispectral Landsat 8 imagery. ASI showed advantages in suppressing non-artificial surfaces with similar spectral signatures to artificial surfaces, and improved the separability between artificial and non-artificial surfaces across eight study areas.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2022)
Article
Environmental Studies
Roger F. Auch, Danika F. Wellington, Janis L. Taylor, Stephen V. Stehman, Heather J. Tollerud, Jesslyn F. Brown, Thomas R. Loveland, Bruce W. Pengra, Josephine A. Horton, Zhe Zhu, Alemayehu A. Midekisa, Kristi L. Sayler, George Xian, Christopher P. Barber, Ryan R. Reker
Summary: This study used sample-based estimates and complete coverage land-cover maps to analyze and describe the patterns of annual land-cover change in the contiguous United States (CONUS) from 1985 to 2016. The results showed that while most of the land cover remained stable over the period, there were significant changes in natural resource cycles, urbanization, and surface-water dynamics. The study also revealed a reduction in the rate of urban expansion after 2006, new growth in cropland after 2007, a net decline in cropland since 1985, and two periods of net tree cover loss.
Article
Geosciences, Multidisciplinary
George Z. Xian, Kelcy Smith, Danika Wellington, Josephine Horton, Qiang Zhou, Congcong Li, Roger Auch, Jesslyn F. Brown, Zhe Zhu, Ryan R. Reker
Summary: This paper introduces a new dataset of land cover and land surface change created by the Land Change Monitoring, Assessment, and Projection (LCMAP) program in the conterminous United States. The dataset consists of annual land cover and land cover change products from 1985 to 2017 at a high resolution. It provides valuable information for land resource management and the study of Earth system dynamics.
EARTH SYSTEM SCIENCE DATA
(2022)
Article
Environmental Sciences
Shi Qiu, Zhe Zhu, Rong Shang, Christopher J. Crawford
Summary: The study shows that Landsat 7's orbit drift has led to a general decrease in surface reflectance and TOA reflectance, with more significant impacts on the NIR and SWIR bands. Future effects are expected to be more pronounced, with varying impacts on different land cover types.
SCIENCE OF REMOTE SENSING
(2021)
Article
Geography
Bo Zhao, Shaozeng Zhang, Chunxue Xu, Yifan Sun, Chengbin Deng
Summary: This paper examines the emergence and potential impacts of deep fake geography, and proposes methods to counteract deep fakes. The study serves as a warning about the potential applications of deep fakes in geography, emphasizing the importance of detecting and mitigating deep fakes.
CARTOGRAPHY AND GEOGRAPHIC INFORMATION SCIENCE
(2021)
