Article
Environmental Studies
Xiang Yan, Wencui Yang, Xiaojian Zhang, Yiming Xu, Ilir Bejleri, Xilei Zhao
Summary: E-scooters have both competing and complementary effects on public transit and bikeshare. During COVID-19, travelers paid a higher price premium but saved less travel time when choosing e-scooters, indicating that public health considerations were the main determinant of travel behavior.
TRANSPORTATION RESEARCH PART D-TRANSPORT AND ENVIRONMENT
(2021)
Article
Computer Science, Information Systems
Pyae Phyoe Kyaw, David W. Macdonald, Ugyen Penjor, Saw Htun, Hla Naing, Dawn Burnham, Zaneta Kaszta, Samuel A. Cushman
Summary: This study used camera-trap data to investigate the co-occurrence of five felid species in Htamanthi Wildlife Sanctuary, Myanmar, finding that they were mainly partitioned on a spatial rather than temporal dimension, with no evidence of mesopredator release. The largest niche differences in the use of space and time occurred between the three smallest species.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2021)
Article
Computer Science, Artificial Intelligence
Ming Tong, Kaibo Yan, Lei Jin, Xing Yue, Mingyang Li
Summary: This study introduces a discriminative multi-focused and complementary temporal/spatial attention framework which overcomes the three main limitations in current works on human action recognition, achieving state-of-the-art results.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Nian Liu, Kepan Nan, Wangbo Zhao, Xiwen Yao, Junwei Han
Summary: In video salient object detection, it is crucial to utilize spatial-temporal (ST) knowledge, including long-short temporal cues and global-local spatial context. Existing methods have only explored parts of this knowledge and neglected their complementarity. This article proposes a novel Complementary ST Transformer (CoSTFormer) that combines short-global and long-local branches to aggregate complementary ST contexts. The CoSTFormer effectively models the context relationship and enables the combination of appearance, motion, and ST factors, achieving state-of-the-art results on benchmark datasets.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Bolong Zheng, Qi Hu, Lingfeng Ming, Jilin Hu, Lu Chen, Kai Zheng, Christian S. Jensen
Summary: This study focuses on a setting where there is a changing set of transportation requests from an origin to a destination before a deadline, and a group of agents capable of fulfilling the requests. The goal is to minimize the average idle time of the agents by assigning them to the requests. The study proposes a solution to the problem by developing a spatial-temporal prediction model for the requests and a route planning algorithm that takes into account both the predictions and the supply-demand state.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Engineering, Civil
Richard D. Crago, Russell Qualls, Jozsef Szilagyi
Summary: This study aims to compare multiple versions of the Complementary Relationship (CR) between actual regional evaporation and apparent potential evaporation, and investigate their response to changes in spatial and temporal scaling. By using data from seven eddy-covariance flux stations in Australia and global ERA5 reanalysis data, the performance and parameter values of these versions were assessed.
JOURNAL OF HYDROLOGY
(2022)
Article
Physics, Multidisciplinary
Hatem Abdelaty, Moataz Mohamed, Mohamed Ezzeldin, Wael El-Dakhakhni
Summary: This study evaluates the robustness of bus transit networks considering different hourly operations. The findings show significant variation in the hourly performance compared to the daily performance when disruptions occur. The network is more robust during on-peak and mid-peak hours, but is sensitive to losing service frequency compared to passenger flow during disruptions.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2022)
Article
Forestry
Jiaxin Wei, Ang Hu, Xiaoyu Gan, Xiaodan Zhao, Ying Huang
Summary: This study explores the relationships and driving mechanisms of ecosystem services (ESs) in the western Sichuan Plateau region of China, aiming to support regional ES and ecological sustainability. The results show that water supply decreased, habitat quality decreased in some areas, carbon storage improved, soil conservation increased and then decreased, and food supply services increased. The study also reveals trade-offs and synergistic relationships among ESs, and spatial heterogeneity in the relationships between ecosystem services.
Article
Transportation
Hadi Gholi, Mohammad Kermanshah, Amir Reza Mamdoohi
Summary: Improving the service quality of public transport is crucial for maintaining the loyalty of current users and attracting new travelers. This study collected stated preference data from 360 commuters in Tehran to evaluate the sources of heterogeneity in passengers' preferences for bus travel attributes. The results showed that observed characteristics can to some extent explain the heterogeneity, providing valuable insights for transit operators to improve service aspects.
JOURNAL OF PUBLIC TRANSPORTATION
(2022)
Article
Chemistry, Multidisciplinary
Junhao Ming, Dongmei Zhang, Wei Han
Summary: Spatial-temporal prediction is crucial for various applications, such as urban traffic control, management, and planning. However, predicting real-world spatial-temporal data accurately is still challenging due to their complex patterns. Most existing models lack effective aggregation of spatial features and comprehensive time series analysis for intricate dependencies. This paper proposes a novel multi-scale spatial-temporal transformer network (MSSTTN) that addresses these issues and outperforms conventional techniques in publicly available datasets.
APPLIED SCIENCES-BASEL
(2023)
Article
Economics
Young Gwan Lee, Gengping Zhu, Bijay P. Sharma, Burton C. English, Seong-Hoon Cho
Summary: The study suggests that introducing a new economic objective with competitive relationships may have negative effects on existing complementary objectives in conservation investment decisions, affecting distributional equity and potentially increasing sacrifices in existing complementary objectives.
FOREST POLICY AND ECONOMICS
(2021)
Article
Geosciences, Multidisciplinary
Silvia De Angeli, Bruce D. Malamud, Lauro Rossi, Faith E. Taylor, Eva Trasforini, Roberto Rudari
Summary: This paper presents a five-step conceptual framework for analyzing the impacts of multi-hazard interactions on the built environment. The framework is based on the systematic analysis of spatial and temporal evolution of hazards to identify potential impact interactions. It is applicable to a wide range of hazards and considers both hazard and impact interactions, as well as residual damage and recovery processes.
INTERNATIONAL JOURNAL OF DISASTER RISK REDUCTION
(2022)
Article
Environmental Sciences
A. Musolff, Q. Zhan, R. Dupas, C. Minaudo, J. H. Fleckenstein, M. Rode, J. Dehaspe, K. Rinke
Summary: The study analyzed high-frequency measurements over a two year period in four neighboring catchments in Germany. It found that SAC(254) showed the most pronounced variability in C-Q hysteresis and slope in all catchments, while NO3-N variability was significant in forested catchments. Event-scale C-Q analysis provided key insights into catchment functioning, helping to disentangle scattered C-Q patterns.
WATER RESOURCES RESEARCH
(2021)
Article
Biodiversity Conservation
Mingli Qiu, Dianfeng Liu
Summary: Sustainable land use should balance development and protection, as intensified land use can affect the trade-offs between ecosystem services. However, there is limited understanding of how land-use intensity affects the spatial heterogeneity of ecosystem service relationships.
ECOLOGICAL INDICATORS
(2023)
Article
Computer Science, Artificial Intelligence
Xuehu Liu, Chenyang Yu, Pingping Zhang, Huchuan Lu
Summary: This research proposes a spatial-temporal complementary learning framework called deeply coupled convolution-transformer (DCCT) for high-performance video-based person re-identification. It couples CNNs and Transformers to extract two kinds of visual features and achieves complementary learning in both spatial and temporal domains, resulting in better performance.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)