Search

Prediction of fine particulate matter with particle size less than 2.5 µm (PM2.5) is an important component of atmospheric pollution warning and control management. In this study, we propose a deep learning model, namely, a spatiotemporal weighted neural network (STWNN), to address the challenge of poor long-term PM2.5 prediction in areas with sparse and uneven stations. The model, which is based on convolutional neural network–bidirectional long short-term memory (CNN–Bi-LSTM) and attention mechanisms and uses a geospatial data-driven approach, considers the spatiotemporal heterogeneity effec It is correct.ts of PM2.5. This approach effectively overcomes instability caused by sparse station data in forecasting daily average PM2.5 concentrations over the next week. The effectiveness of the STWNN model was evaluated using the Xinjiang Uygur Autonomous Region as the study area. Experimental results demonstrate that the STWNN exhibits higher performance (RMSE = 10.29, MAE = 6.4, R2 = 0.96, and IA = 0.81) than other models in overall prediction and seasonal clustering. Furthermore, the SHapley Additive exPlanations (SHAP) method was introduced to calculate the contribution and spatiotemporal variation of feature variables after the STWNN prediction model. The SHAP results indicate that the STWNN has significant potential in improving the performance of long-term PM2.5 prediction at the regional station level. Analyzing spatiotemporal differences in key feature variables that influence PM2.5 provides a scientific foundation for long-term pollution control and supports emergency response planning for heavy pollution events. [ABSTRACT FROM AUTHOR]

The urban agglomeration plays a significant role in enhancing integrated regional development. Nevertheless, the expansion of urban agglomerations has demonstrated a lackluster ability to attract cities. Presently, finding solutions to stabilize the existing urban strength and effectively extend attraction to neighboring cities has become a crucial matter. This study adopts the enhanced comprehensive attraction model, fracture point model, and radiation radius model to examine the level of city attraction, intensity of radiation, and range of radiation in the northern and southern Xinjiang city clusters between 2010 and 2020. Based on the analysis, the following conclusions are drawn: (1) the comprehensive strength and city attraction of cities in the northern Xinjiang region is higher than that of the southern Xinjiang region; (2) the intensity of spatial connection between cities in the northern and southern Xinjiang regions is gradually increasing, and the intensity of spatial connection of cities in the northern Xinjiang region is significantly greater than that in the southern Xinjiang region. The intensity of spatial connection between cities in the northern and southern Xinjiang regions is significantly greater than that in the southern Xinjiang region; (3) the central role of the central cities in the northern and southern Xinjiang regions is weakening, the development of cities in the region is gradually becoming unified and coordinated, and regional integration is gradually being strengthened. This study reveals the similarities and differences in urban development in the north and south of Xinjiang and provides important theoretical reference value for regional urban development. [ABSTRACT FROM AUTHOR]

Glaciers are susceptible indicators of climate change and crucial parts of the world's water cycle. In the context of global warming, we took the Urumqi Glacier No.1 (UG1) as an example, which is situated at the source of the Urumqi River on the northern slope of the Tianshan Mountains, Xinjiang, combined with the climate data of Daxigou Meteorological Station from 1980 to 2020, and the change of glacier mass balance and its response to extreme climate are discussed. The results suggest that the glacier mass balance of UG1 showed a downward trend over the studied 41-year period, and the mass loss increased. The cumulative glacier mass balance value was −19,776 mm w.e., and the average annual value was −482 mm w.e.a−1. The Mann-Kendall trend test showed that the change point occurred around 1994, and the mass balance of UG1 became more negative after 1994. In the same period, the changing mass balance trend of UG1 was not the same in different seasons. The inter-annual variation of summer mass balance was drastic, showing a marked downward trend; the inter-annual change of winter mass balance was small, showing a slight uptrend. The changing of extreme climate indices where UG1 is located showed that only TX90p and TX10p changed observably from 1980 to 2020, and the extreme precipitation indices changed evidently and had been on the rise. The changing trend of extreme climate indices indicated that the temperature was rising, the warming was significant, and the precipitation was increasing. During 1980–2020, the glacier mass balance was substantially correlated with the extreme temperature indices (TX90p, TXx) but not with the extreme precipitation indices. Analyzing on a seasonal scale, the summer mass balance was memorably correlated with the extreme temperature indices (TX90p, TX10p, TXx), and the correlation coefficient between winter mass balance and the extreme precipitation index R95p and winter precipitation was in the range 0.36~0.40 (p < 0.05). According to the correlation between glacier mass balance and extreme climate indices, the summer mass balance was mainly affected by temperature, and the winter mass balance was affected primarily by precipitation. [ABSTRACT FROM AUTHOR]

The Xinjiang Uyghur Autonomous Region of China is the largest arid region in Central Asia, and is heavily dependent on glacier melt in high mountains for water supplies. In this paper, glacier and climate changes in Xinjiang during the past decades were comprehensively discussed based on glacier inventory data, individual monitored glacier observations, recent publications, as well as meteorological records. The results show that glaciers have been in continuous mass loss and dimensional shrinkage since the 1960s, although there are spatial differences between mountains and sub-regions, and the significant temperature increase is the dominant controlling factor of glacier change. The mass loss of monitored glaciers in the Tien Shan has accelerated since the late 1990s, but has a slight slowing after 2010. Remote sensing results also show a more negative mass balance in the 2000s and mass loss slowing in the latest decade (2010s) in most regions. This needs further investigation on whether the slowing is general and continuing. In addition, glacier surging occurs more frequently in the Karakoram and Kunlun Mountains. [ABSTRACT FROM AUTHOR]

[Display omitted] • Integrated ESI and landscape connectivity to determine the ecological sources. • Combined MCR and gravity models to extract and divide the ecological corridors. • Provide a reference for constructing ESP in Xinjiang. Constructing ecological security patterns (ESP) effectively maintains ecosystems' green, healthy, and sustainable development. The InVEST model and Conefor software were utilized in this study to pinpoint ecological sources (eco-sources) through a blend of landscape connectivity and the importance of ecosystem services (IES); the ecological resistance surface was constructed by synthesizing the gravity model and minimum cumulative resistance model (MCRM), extracted ecological corridors (eco-corridors) and nodes. Ultimately, the research in this study built the ESP for Xinjiang. The findings showed that: (1) 14 eco-sources were identified in Xinjiang, spanning an area of 2.13 × 105 km2; 32 eco-corridors were extracted, including 9 important and 23 general eco-corridors, comprising a total length of 1.26 × 104 km; identified 46 ecological nodes(eco-nodes); (2) the size of important and more important patches of ecosystem services in Xinjiang was 2.82 × 105 km2, which accounted for a relatively small size, and the value of ecosystem services was generally average. The overall level of landscape connectivity in the primary eco-sources was high, but the spatial distribution was uneven; (3) the ecological resistance values in Xinjiang vary significantly, with high-level resistance surfaces primarily located in the Junggar Basin and Tarim Basin and low-level resistance covering most of Xinjiang's natural ecological reserves and other areas. The study's findings can provide some references for constructing ESP at the provincial scale and in arid and semiarid regions and for planning spatial ecological restoration in Xinjiang. [ABSTRACT FROM AUTHOR]