Article
Geography, Physical
Mengmeng Cao, Xuyan Li, Tingwei Cui, Xinliang Pan, Yan Li, Yanlong Chen, Ning Wang, Yanfang Xiao, Xingai Song, Yuzhu Xu, A. Runa, Bing Mu, Song Qing, Rongjie Liu, Wenjing Zhao, Yuhai Bao, Jie Zhang, Lan Wei
Summary: Green macroalgae bloom (GMB), dominated by Ulva prolifera, has been occurring regularly along the China coast since 2007. The satellite-observed GMB annual maximum coverage (AMC) rebounded sharply in 2021 to an unprecedented level, which raises questions about the reasons for this rebound and the significant interannual variability. Through the analysis of long-term satellite observations, meteorological data, and water quality statistics, two key determinants for AMC were identified as the macroalgae distribution in a key area and nutrient availability. A novel model for AMC prediction was proposed and validated, which can explain the significant interannual variability and align well with the latest observation in 2022.
INTERNATIONAL JOURNAL OF DIGITAL EARTH
(2023)
Article
Geography, Physical
Zifu Wang, Yun Li, Kevin Wang, Jacob Cain, Mary Salami, Daniel Q. Q. Duffy, Michael M. M. Little, Chaowei Yang
Summary: With the advancement of AI technologies and big Earth data, DL has become an important method in Earth science. However, computational challenges still exist for DL-based applications. This study aims to address these challenges by revising DL models/algorithms and testing their performance on multiple GPU computing platforms.
INTERNATIONAL JOURNAL OF DIGITAL EARTH
(2023)
Article
Geography, Physical
Ao Pang, Qi Liang, Weijia Li, Lei Zheng, Teng Li, Xiao Cheng
Summary: In this study, a ResUNet model was used to identify the spatial distribution of surface fractures on Antarctic ice shelves using MODIS data. It was found that the total area of surface fractures decreased from 2004 to 2014, primarily due to changes in fractures within 20 km of the ice front in the Amundsen and Wilkes sectors. High concentrations of fractures were particularly observed in the Thwaites Ice Shelf, Crosson Ice Shelf, and Getz Ice Shelf in the Amundsen sector. This study provides comprehensive and detailed information about surface fractures on Antarctic ice shelves and has implications for evaluating their vulnerability.
INTERNATIONAL JOURNAL OF DIGITAL EARTH
(2023)
Article
Geography, Physical
Xiuwan Yin, Bo Jiang, Shunlin Liang, Shaopeng Li, Xiang Zhao, Qian Wang, Jianglei Xu, Jiakun Han, Hui Liang, Xiaotong Zhang, Qiang Liu, Yunjun Yao, Kun Jia, Xianhong Xie
Summary: This study evaluated ten long-term land surface net radiation products under different spatial scales, spatial and temporal variations, and different conditions. The results showed that GLASS-MODIS performed the best during 2000-2018, followed by ERA5 and GLASS-AVHRR. During 1983-2018, GLASS-AVHRR and ERA5 ranked top and performed similarly. The differences in net radiation between satellite and reanalysis products may be attributed to radiative components, meteorological variables, and algorithm applicability.
INTERNATIONAL JOURNAL OF DIGITAL EARTH
(2023)
Article
Geography, Physical
Maximilian C. Hartmann, Ross S. Purves
Summary: City Walking Tour Videos (CWTVs) are a unique source of Volunteered Geographic Information that offer detailed street-level imagery. Through a mobility study focused on the City of Paris, we found that these videos contain rich information for analyzing urban transportation. By detecting transport modes and examining changes in mobility mix, we demonstrated the potential of CWTVs for urban analytics and identified shifts in transportation patterns during the Covid-19 pandemic.
INTERNATIONAL JOURNAL OF DIGITAL EARTH
(2023)
Article
Geography, Physical
Jinlian Liu, Xinyao Zhou, Hanya Tang, Fengqin Yan, Shiwei Liu, Xuguang Tang, Zhi Ding, Ke Jiang, Pujia Yu
Summary: Land surface temperature (LST) and its day-night difference (LSTd - LSTn) are influenced by vegetation and climate change. Quantifying their contribution and feedback is crucial for mitigation strategies. This study used partial correlation and spatial analysis to investigate the response of LST to vegetation and climate variables. Results showed negative responses and feedback from vegetation on LST, with vegetation being a major contributor to LSTd - LSTn decline. Air temperature (AT) played a decisive role in LST warming trend, partially mitigated by vegetation and precipitation (Pre).
INTERNATIONAL JOURNAL OF DIGITAL EARTH
(2023)
Article
Geography, Physical
Yi Han, Guangyun Zhang, Rongting Zhang
Summary: This paper proposes a grid graph-based point cloud registration algorithm to align unordered point clouds. The algorithm divides the point cloud into a set of 3D grids and uses a voting strategy based on feature descriptors to measure the similarity between two grids. A graph matching method is then proposed to capture spatial consistency and refine the corresponding grids hierarchically to obtain point-to-point correspondences.
INTERNATIONAL JOURNAL OF DIGITAL EARTH
(2023)
Article
Geography, Physical
Hao Sun, Qian Xu, Yunjia Wang, Zhiyu Zhao, Xiaohan Zhang, Hao Liu, Jinhua Gao
Summary: Agricultural drought is a serious threat to global food security. A new soil moisture dataset was created in this study by enhancing satellite remote sensing data with machine learning. The study revealed the spatial and temporal dynamics of agricultural drought in China and showed a decreasing trend in drought severity from 1982 to 2020. The findings contribute to a better understanding of agricultural drought and highlight the need for improved satellite soil moisture datasets with faster updates and higher resolution.
GISCIENCE & REMOTE SENSING
(2023)
Article
Geography, Physical
Di Wu, Ruopu Li, Banafsheh Rekabdar, Claire Talbert, Michael Edidem, Guangxing Wang
Summary: The study aims to develop deep learning models for classifying images with flow barrier locations. Different Convolutional Neural Network (CNN) models were trained and compared using High-Resolution Digital Elevation Models (HRDEMs) and aerial orthophotos in four different watersheds in the U.S. Midwest. Results show that most deep learning models achieved over 90% accuracies consistently. The CNN model with HRDEMs as the sole input feature was found to be the best-fit model, while the addition of aerial orthophotos did not significantly improve the model's accuracy.
GISCIENCE & REMOTE SENSING
(2023)
Article
Geography, Physical
Marco Assiri, Anna Sartori, Antonio Persichetti, Cristiano Miele, Regine Anne Faelga, Tegan Blount, Sonia Silvestri
Summary: This study estimates aboveground biomass (AGB) in alpine peatlands in the Veneto Region, Italy, using a combination of in situ vegetation samples and datasets from UAV-based LiDAR, hyperspectral, and RGB sensors. The results indicate that UAV LiDAR data provides the most reliable solution for estimating AGB in alpine peatlands, while the inclusion of hyperspectral data provides only a minor improvement in accuracy.
GISCIENCE & REMOTE SENSING
(2023)
Article
Geography, Physical
Shili Meng, Yong Pang, Kebiao Huang, Zengyuan Li
Summary: Accurate forest cover mapping is crucial for monitoring forest extent in Southeast Asia. A novel method for mapping forest cover in the presence of cloud cover was presented, resulting in more accurate and reliable information than the initial maps. This approach provides a framework for improving the spatial continuity of existing thematic maps.
GISCIENCE & REMOTE SENSING
(2023)
Article
Geography, Physical
He Cai, Bo Zhong, Huilin Liu, Bailin Du, Qinhuo Liu, Shanlong Wu, Li Li, Aixia Yang, Junjun Wu, Xingfa Gu, Jinxiong Jiang
Summary: This study proposes an improved deep learning network called SPAODnet for retrieving AOD from satellite imagery, focusing on sub-pixel clouds. By incorporating a spatial adaptive bilateral filter, a channel attention mechanism, and a composite loss function, the SPAODnet significantly enhances the accuracy, spatial resolution, and coverage of AOD, particularly in the presence of sub-pixel clouds and cloud shadows.
GISCIENCE & REMOTE SENSING
(2023)
Article
Remote Sensing
Bing Liu, Yifan Sun, Ruirui Wang, Anzhu Yu, Zhixiang Xue, Yusong Wang
Summary: This paper proposes a global-local U-Net augmented by transformers (TrmGLU-Net) for the classification of hyperspectral images. The model captures long-distance dependencies in both spatial and spectral dimensions and performs well under the condition of small samples.
EUROPEAN JOURNAL OF REMOTE SENSING
(2023)
Article
Remote Sensing
Huma Hayat, Bruce Raup, Sher Muhammad, Shiyin Liu, Romana Khan, Siddique Ullah Baig, Adnan Ahmad Tahir
Summary: This study estimated the Equilibrium Line Altitude (ELA) and geodetic mass balance of fifteen glaciers in the HKH using satellite images and DEMs. Results showed that the ELA of most glaciers shifted upward and the mass balance exhibited a heterogeneous pattern. The study also found significant or slight increasing or decreasing trends in temperature, precipitation, and discharge in the study basins.
EUROPEAN JOURNAL OF REMOTE SENSING
(2023)
Article
Environmental Sciences
Hao Tan, Xiangjian Xie, Junjun Sun, Yuqian Wang, Yuhong Jiang, Shuaishuai Huang
Summary: A multi-scale classification framework based on spectral-spatial features was proposed in this paper for monitoring red bed rocky desertification. Spectral indices and spatial features were used at pixel and patch scales, respectively, and validated using an OLI image in northwestern Jiangxi. The experimental results were satisfactory, providing a methodological supplement for monitoring red bed rocky desertification.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Khilola Amankulova, Nizom Farmonov, Uzbekkhon Mukhtorov, Laszlo Mucsi
Summary: In order to manage agricultural land and ensure food security, timely crop yield information is essential. This study explored the use of remote sensing data from Sentinel-2 to monitor sunflower crop phenology and predict crop yield at the field scale. Ten sunflower fields in Mezohegyes, southeastern Hungary, were studied in 2021, and Sentinel-2 images were collected throughout the monitoring period. Vegetation indices (VIs) were extracted to monitor crop growth. Multiple linear regression and two different machine learning approaches were used to predict crop yield, with random forest regression (RFR) showing the best performance. The study provides valuable insights for developing a robust and timely prediction method for sunflower crop yields to support decision-making regarding food security.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Kaushlendra Verma, J. Indu
Summary: The SWOT satellite mission, scheduled for launch in December 2022, is expected to effectively monitor freshwater resources. However, the infrequent temporal sampling of the SWOT orbit will lead to inconsistent estimation of river discharge. This study investigates the influence of unique temporal sampling on the calibration of a hydrological model, and suggests that using SWOT data for calibration can provide similar results to daily in-situ discharge measurements.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Projo Danoedoro, Prima Widayani, Iswari Nur Hidayati, Sanjiwana Arjasakusuma, Diwyacitta Dirda Gupita, Huwaida Nur Salsabila
Summary: This study compared the capabilities of forest cover density (FCD) transformation and random decision forest (RDF) classification in complex tropical landscapes. Landsat-8 OLI imagery was used for vegetation structural composition mapping in Central Java, Indonesia. The FCD transformation achieved an accuracy of 69.32%, while the RDF classification achieved accuracies ranging from 70.76% to 75.19% depending on the parameter setting.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Trung Nguyen Nhu, Nam Bui Van, Kha Tran Van
Summary: This paper uses three-dimensional direct gravity inversion to determine the Moho and basement depths of Tuchinh-Vungmay basin offshore southeastern Vietnam. The Moho depth is predicted from the mantle residual gravity anomaly with lithosphere thermal gravity correction, while the basement depth is determined by enhancing the resolution of the basement topography through the downward continuation of the basement residual gravity anomaly. The depths of the Moho and basement surfaces are constrained by the power density spectrum of the residual gravity anomalies and oceanic bottom seismic data.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Ruiqi Yang, Chen Zheng, Leiguang Wang, Yili Zhao, Zhitao Fu, Qinling Dai
Summary: In this study, a dual-stream network MAE-BG was proposed, consisting of an edge detection (ED) branch and a smooth branch with boundary guidance (BG). The ED branch enhances weak edges and suppresses false responses caused by local texture, while the MAE networks extract multiscale edge information to complement detail loss. The segmentation results with improved boundaries are obtained by stacking the output of the ED and smooth branches. The proposed method achieves precise object boundary location and improved segmentation performance.
GEOCARTO INTERNATIONAL
(2023)
