Search

Modeling human mobility helps to understand how people are accessing resources and physically contacting with each other in cities, and thus contributes to various applications such as urban planning, epidemic control, and location-based advertisement. Next location prediction is one decisive task in individual human mobility modeling and is usually viewed as sequence modeling, solved with Markov or RNN-based methods. However, the existing models paid little attention to the logic of individual travel decisions and the reproducibility of the collective behavior of population. To this end, we propose a Causal and Spatial-constrained Long and Short-term Learner (CSLSL) for next location prediction. CSLSL utilizes a causal structure based on multi-task learning to explicitly model the "when→what→where", a.k.a. "time→activity→location" decision logic. We next propose a spatial-constrained loss function as an auxiliary task, to ensure the consistency between the predicted and actual spatial distribution of travelers' destinations. Moreover, CSLSL adopts modules named Long and Short-term Capturer (LSC) to learn the transition regularities across different time spans. Extensive experiments on three real-world datasets show promising performance improvements of CSLSL over baselines and confirm the effectiveness of introducing the causality and consistency constraints. The implementation is available at https://github.com/urbanmobility/CSLSL. [ABSTRACT FROM AUTHOR]

Non-point source (NPS) pollution has become a major source of water pollution. A combination of models would provide the necessary direction and approaches designed to control NPS pollution through land use planning. In this study, NPS pollution load was simulated in urban planning, historic trends and ecological protection land use scenarios based on the Conversion of Land Use and its Effect at Small regional extent (CLUE-S) and Soil and Water Assessment Tool (SWAT) models applied to Hunhe-Taizi River Watershed, Liaoning Province, China. Total nitrogen (TN) and total phosphorus (TP) were chosen as NPS pollution indices. The results of models validation showed that CLUE-S and SWAT models were suitable in the study area. NPS pollution mainly came from dry farmland, paddy, rural and urban areas. The spatial distribution of TN and TP exhibited the same trend in 57 sub-catchments. The TN and TP had the highest NPS pollution load in the western and central plains, which concentrated the urban area and farm land. The NPS pollution load would increase in the urban planning and historic trends scenarios, and would be even higher in the urban planning scenario. However, the NPS pollution load decreased in the ecological protection scenario. The differences observed in the three scenarios indicated that land use had a degree of impact on NPS pollution, which showed that scientific and ecologically sound construction could effectively reduce the NPS pollution load in a watershed. This study provides a scientific method for conducting NPS pollution research at the watershed scale, a scientific basis for non-point source pollution control, and a reference for related policy making. [ABSTRACT FROM AUTHOR]

Analyzing spatiotemporal characteristics of the historical urbanization process is essential in understanding the dynamics of urbanization and scientifically planned urban development. Based on historical urban area maps and remote sensing images, this study examined the urban expansion of Shenyang from 1910 to 2010 using area statistics, typology identification, and landscape metrics approaches. The population and gross domestic product were analyzed as driving factors. The results showed that the urban area of Shenyang increased 43.39-fold during the study period and that the growth rate has accelerated since the 1980s. Three urban growth types were distinguished: infilling, edge-expansion, and spontaneous growth. Edge-expansion was the primary growth type. Infilling growth became the main growth type in the periods 1946–70, 1988–97, and 2004–10. Spontaneous growth was concentrated in the period of 1997 to 2000. The results of landscape metrics indicate that the urban landscape of Shenyang originally was highly aggregated, but has become increasingly fragmented. The urban fringe area was the traditional hot zone of urbanization. Shenyang was mainly located north of the Hun River before 1980; however, the south side of the river has been the hot zone of urbanization since the 1980s. The increase of urban area strongly correlated with the growth of GDP and population. Over a long time scale, the urbanization process has been affected by major historical events. [ABSTRACT FROM AUTHOR]