In the current context of spatial planning of national land, delineating the urban development boundary objectively and scientifically is a key and difficult task in planning work. However, most existing methods about the delineation of urban development boundary is existing some problems such as data selection, method build and result analysis. In view of natural environment, social economy and policy orientation, a method of delineating urban development boundary automatically was been proposing based on multi-source data fusion and deep learning. Furthermore, the proposed method has been used to delimit the urban development boundary of Huadu District, Guangzhou City and analysis of influencing factors. The results show that: 1) This method can delimit the urban development boundary automatically; 2) The model’s results are highly consistent with the planning results in terms of spatial distribution trend, with a high degree of land intensive and economical use, which is more in line with the requirements of future land development; 3) Urban development is the result of a combination of multiple factors, among which transportation and population are the primary factors affecting the urban development. All in all, the proposed method can delimit the urban development boundary automatically, objectively and scientifically. What’s more, the proposed method’s results are in line with the future trend of land use development, thus can provide better guidance for China’s spatial planning of national land.
In China, investment banks play an important role as intermediaries in companies’ initial public offerings. The economic geography of investment banks is one of the keys to understanding economic dynamics. However, the existing literature pays little attention to the geographical pattern and influencing factors of China’s investment banks. Therefore, this paper takes Chinese investment banks as the research object and studies the spatial pattern and influencing factors of the service capability of investment banking departments of Chinese securities companies through their initial public offering (IPO) business. In the study of spatial pattern, this paper constructs a database of Chinese A-share IPOs from 1993 to 2020. Based on the significance of IPO amount to investment banking services, this paper puts forward a method to calculate the total value of urban investment banking services, and increases the weight of urban investment banking services according to IPO amount. In this paper, the consumer price index (CPI) is used to unify the purchasing power of the initial public offering amount of A-share listed companies in the past 30 years, and the data are standardized. In terms of the research on influencing factors, this paper collected the data of all 35 cities with investment banking service capability from 2006 to 2020, and established a two-way fixed effect regression model for analysis. The explained variable is the total value of city investment banking services, and the explanatory variables are human capital factor, city level and city innovation power. The main findings of the study are as follows: 1) China’s investment banking service pattern is primarily centered in Beijing, Shenzhen, and Shanghai, forming a dense distribution pattern in the southeastern region and a sparse distribution in the northeast, northwest, and southwest China. Overall, the allocation of investment banking service resources in China is uneven, and there is a trend of spatial agglomeration; 2) Beijing is China’s investment banking service center. In addition to Shanghai and Shenzhen, cities such as Guangzhou, Nanjing, Hangzhou, and Fuzhou have performed well. The mutual service levels between “Beijing-Shenzhen” and “Beijing-Shanghai” are high, and the cooperation is close. Beijing has the highest self-service value, while Shenzhen, Guangzhou, and Shanghai have relatively low self-service ratios; 3) Urban human capital level, administrative level, and urban innovation capability have significant positive impacts on urban investment banking service capabilities. Heterogeneity analysis reveals that they have significant positive impacts on the investment banking service capabilities in the eastern region, while the impacts in the central, western, and northeastern China are not obvious or negative. This paper focuses on the geographical pattern of investment banks in China, and provides informative evidence on the spatial differences and influencing factors of the service capabilities of the industry, which makes up for the lack of geographical attention to investment banks in existing studies. On the other hand, from the perspective of financial geography, this paper deepens the understanding of regional development imbalance, and provides reference for further strengthening inter-regional financial cooperation, building financial center cities and studying the pattern of regional economic development in China.
The construction and improvement of the new energy industry technology transfer system is of great significance for promoting the high-quality transformation of the energy structure, the industrialization of scientific and technological achievements, and the close integration of economic development and scientific and technological innovation, and provides strong technical support for achieving carbon peaking and carbon neutrality goals. Based on the patent data of China’s new energy technology from the State Intellectual Property Office from 2001 to 2019, the study analyzes the spatio-temporal pattern of technology transfer network complexity in China’s new energy industry by constructing a directed weighted network of technology transfer, through the methods of complex network analysis, module mining, directed alternative centrality and directed alternative power model analysis. The research found that: 1) The technology transfer of China’s new energy industry has developed rapidly, and the participation of cities has been continuously increased, and the regional restrictions have been continuously broken. 2) The overall network structure shows that the scale of China’s new energy technology transfer network continues to expand, the agglomeration situation increases, and the correlation improves as a whole. But there are disadvantages such as the stability of network correlation is not strong and the network interoperability is weak. National central cities such as Beijing, Shanghai, and Shenzhen play a spatial diffusion/agglomeration effect. The new energy technology transfer diffusion follows the technology gradient transfer law. 3) Module analysis finds that the network continuously forms a highly localized clustering characteristic of triadic closure, and the transfer effect of cities occupying structural hole locations similarly accelerates technology transfer. 4) The urban network power level can be classified as either core cities, with high centrality and high power, or peripheral cities, with low centrality and low power. The damping effect of distance influences the degree connection within a city. Other relationships exist, such as center city clusters, with high center and low power, and powerful gateway cities, with low centrality and high power cities.
China serves as the origin of numerous international rivers. Collaboration between China and the neighboring countries along these international rivers materializes primarily through international river treaties. Building on a new dataset, we map the geographical distribution of these treaties and employ qualitative comparative analysis (csQCA) to investigate the combination of key factors leading to the signing of the treaties. Grounded in the theoretical framework of transboundary political ecology, this study emphasizes nature and political factors encompassing the inherent ecological attributes of rivers, the characteristics of the countries through which these rivers flow, and the interrelationships among these nations. Findings include: 1) Northeastern and southwestern regions have signed more treaties. Focal points of cooperation in these treaties encompass river management, comprehensive development and utilization, and economic collaboration. 2) Factors influencing China’s involvement in international river treaties underscore the size of the river and the diplomatic relations between riparian nations. 3) Diverse influencing factors are at play with regard to different facets of cooperation, reflecting the multifaceted and intricate nature of international river collaboration. This research contributes to a deeper understanding of the determinants influencing international river cooperation and offers a reference for China to demonstrate its image as a responsible power in international cooperation.
The transport system, serving as the foundational backbone for global trade movements, significantly influences the trading relationships among domestic cities and their integration into global production networks. The existing literature has analyzed the global trade networks carried by a particular transport mode or investigated the resilience of the trade network regardless of its transportation modes, both of which neglect the transportation resilience in the global trade networks. In light of this, this study seeks to investigate the spatial evolution and structural resilience of trade networks facilitated by various transportation means through social network analysis coupled with scenario simulation techniques, based on city-level trade data extracted from General Administration of Customs. P. R. China (2000—2015). The marginal contributions of this study can be summarized as: 1) Systematically explores the evolution and characteristics of trade networks from the perspective of transportation modes, filling the gap in the existing economic geography literature where transportation networks and trade networks are disconnected; 2) Clarifies the importance and changing process of specific cities in the trade network from the perspective of transportation mode complexity; 3) Measures the resilience level of trade networks under specific transportation modes and identifies alternative city transportation networks. The key findings include: 1) Across all four transport modes, there is a consistent rise in trade network density, with waterways showing the highest density and railroads demonstrating the highest degree of network modularity. Such results indicate the maturity of waterway-based trade network and the limitations of track-dependent railway-based trade networks. 2) The development of the domestic transport infrastructure helps to mitigate the intricacy of trade connections between cities across different regions, particularly in terms of access to a variety of transportation options. Inland cities now have more access to waterway-based trade networks, and coastal cities have developed complicated trade connections with Asian or European countries carried by railways or trucks. 3) Among the transport systems, the waterway-based trade network demonstrates the highest level of resilience, followed by air-based and road-based transport networks. Over the past 15 years, the trade networks have been less dependent on top 15 ports for each transport mode, suggesting a higher level of resilience of each mode. 4) The swift progression of airport infrastructure augments the availability of alternative cities that can respond to temporary disruptions in air cargo-based trade activities, by either reduce the distance from nearest available backup cities or increase the number of nearby backup options, especially for cities in inland area. In summary, these findings offer valuable insights for national strategies aimed at refining major production configurations and bolstering the growth of strategic hinterlands. Firstly, the domestic transportation system plays a critical role offering various regions the possibility to develop multiple trade transport modes. Secondly, although railway transport along the Eurasian Land Bridge has developed rapidly in recent years, it still has limitations and falls short of the scale and system of waterway, air, and highway transport. Thirdly, when countries plan for productive force safety layouts, they need to explore potential alternative city locations based on existing foreign trade connections to ensure the resilience of foreign trade at the transportation level and smooth the transfer and handover processes between cities.
This study aims to delve into the intricate relationship between the service-oriented transformation of manufacturing industries and their integration within the global value chain (GVC) and national value chain (NVC) across Chinese provinces. Utilizing a unified measurement framework, this research incorporates data from Chinese provinces to analyze service-oriented transformation, GVC embedding, and NVC embedding. The findings reveal a distinct positive association pattern of high-high and low-low clustering between the service-oriented transformation of manufacturing industries and their positions within the GVC and NVC across Chinese provinces. The service-oriented transformation significantly enhances the GVC and NVC of manufacturing industries in Chinese provinces, primarily through the mediation effects of innovation efficiency and production costs. Notably, the enhancement effect on the NVC is higher than that on the GVC in most regions, suggesting a stronger internal dynamic within China’s economic structure. The service-oriented enhancement effect is most pronounced in economic pole areas, such as the Beijing-Tianjin-Hebei region and the northern and eastern parts of China. Additionally, service-oriented transformation originating from other Asian countries has a more significant enhancement effect on manufacturing industries compared to that from other regions.By considering the spatial and temporal effects, this study provides a comprehensive understanding of the service-oriented transformation's impact on the GVC and NVC of Chinese manufacturing industries.
With the rapid development of suburban new towns in China, new towns have attracted a large number of employment inflows, but the residence of many employees has not moved with them, and there has been a general phenomenon of separation of work and residence. However, at present, the reasons for the separation of employment and residence in suburban new towns are mostly qualitative discussions, and the key factors and their impact effects are still unclear. To this effect, the study takes suburban new towns in Wuhan as an example and utilizes multiple sources of spatiotemporal big data, such as mobile phone signaling, to investigate the impact characteristics of commuting distance, personal attributes, and built environment on the residential relocation of workers experiencing work-residence separation in these new towns. The study employs a multilayer MLR-BLR (Binary Logistic Regression-Multiple Linear Regression) model to analyze the influencing factors, and the questionnaire survey is used to supplement the core influencing factors of migration missing in big data. The research findings are as follows: Quality built environment, long distance commuting, high-quality educational resources, appropriate housing prices and stable jobs can promote the residential migration of the new town’s job-housing separation workers to varying degrees; on the contrary, it will hinder its migration. Family factors play a key role in the relocation of non-single workers, especially women. This study quantitatively investigates the influencing factors of residential relocation among workers experiencing work-residence separation in suburban new towns. On one hand, it expands the research perspective on subgroups of urban work-residence separation. On the other hand, it can also provide reference for the practice of new town location, the attraction of suburban new towns, and the promotion of job-housing balance in suburban new towns.
Based on sample data from 2013 to 2022, this study selects Heilongjiang, Jilin, Liaoning, Beijing, Tianjin, Hebei, Inner Mongolia, and Xinjiang as typical regions for snow-ice tourism in China. Using the Super-EBM model, the efficiency of snow-ice tourism development in these regions was measured, and the modified DEA model was applied to more scientifically assess the current status of snow-ice tourism development in these regions. The study analyzes the temporal variations, spatial disparities, and dynamic evolution characteristics of efficiency and investigates the impact of snow-ice tourism policies and snow-ice events on the efficiency of snow-ice tourism development through the Tobit Model, as well as the mediating effects of snow-ice tourism attention and the number of snow-ice tourism enterprises. The research findings are as follows: 1) From 2013 to 2022, the average efficiency of snow-ice tourism development did not reach an effective state. The efficiency level showed a clear correlation with the snow-ice resource endowment, presenting a generally fluctuating upward trend. 2) There was a significant disparity in the efficiency of snow-ice tourism development across regions, and the spatial pattern changed significantly during the study period. 3) The type transfer of snow-ice tourism development efficiency in the typical regions exhibited overall inertia, and regions with lower efficiency levels exhibited stronger path dependence. There is also a certain probability of cross-type transition in the efficiency of snow-ice tourism development across regions. 4) Snow-ice tourism policies and snow-ice events had a significant positive impact on the efficiency of snow-ice tourism development. The introduction of snow-ice tourism policies and the hosting of snow-ice events can improve the attention to snow-ice tourism and promote an increase in the number of snow-ice tourism enterprises, thus boost the efficiency of snow-ice tourism development. Based on these findings, this paper proposes strategies to enhance the snow-ice tourism policy support system, expand the synergistic effect between snow-ice tourism and snow-ice events, optimize the supply capacity of snow-ice tourism enterprises, and improve regional snow-ice tourism brand attention.
Accelerating the transformation of the Bohai Rim Area into a true growth pole in northern China had profound significance for building a new era of regional development pattern, based on this, this paper systematically analyzed the spatiotemporal evolution characteristics and driving mechanisms of the Ecological- Economic-Social (EES) system coupling coordination in the Bohai Rim Area from 2001 to 2020. The results indicate that: 1) The overall development of the coupling degree of the EES system was better than that of coordination, and the absolute and relative differences showed a slight upward trend; in terms of space, there was an evolutionary pattern of “agglomeration+differentiation”, forming three relatively high value areas: the mid-southern Liaoning urban agglomeration, the Beijing-Tianjin-Tangshan district, and the Shandong coastal urban agglomeration, and the coupling coordination spatial correspondence was obvious. 2) The coupling and coordination spatial agglomeration characteristics of the EES system were obvious, with a circular distribution of hot and cold spots from coastal to inland areas; the scale pattern showed inter provincial differences>intra provincial differences>interval differences. 3) The coupling and coordination spatial differentiation characteristics of the EES system were obvious, showing core edge characteristics, and the central city of the urban agglomeration was the regional growth pole; featuring functionalization, comprehensive functional cities were superior to resource-based cities; It had a hierarchical characteristic and was positively correlated with the administrative level of the city; coastal cities develop better than inland cities. 4) The coupling and coordinated spatiotemporal evolution of the EES system was driven by various factors such as natural geographical location, subsystem interaction, spatial proximity effect, strategy and policy, and related factors; the direction and region of the driving factors of the two were basically consistent.
Based on the connotation of urban-rural integrated development, 78 prefecture-level cities in the Yellow River Basin were taken as the case study areas, the evaluation index system of urban-rural integrated development in the Yellow River Basin was scientifically constructed from 5 dimensions, and multiple measurement methods were used to comprehensively study the spatio-temporal differentiation and driving mechanism of urban-rural integrated development in the Yellow River Basin in 2011—2021. The results are obtained as follows: 1) The level of urban-rural integration in the Yellow River Basin is generally on an upward trend, with low-level areas decreasing and high-level areas continuing to increase. The level of urban-rural integration in the downstream areas is generally higher than that in the middle and upstream areas, and generally shows a pattern of increasing from west to east; 2) Spatial clustering characteristics are significant, the overall existence of a more obvious spatial dependence, high value areas are mainly concentrated in the eastern and northern areas of the Yellow River Basin; 3) The integrated development of urban and rural areas is the result of multi-scale and multi-factor interactive and integrated development driven by the interaction of many factors, and the driving factors have obvious spatial and temporal differences, showing spatial heterogeneity in spatial bands or pieces of the distribution pattern.
Based on comprehensive population census data spanning from 2000 to 2020, alongside relevant socioeconomic indicators, a comprehensive analysis has been undertaken to explore the intricate spatial and temporal evolution of the floating population structure across various counties within Fujian Province, China. This analysis delves deep into the complex factors that influence this population dynamic and finds that: The trend of intra-provincialization of the floating population structure in Fujian Province has increased, and rapidly intra-provincialized areas are spatially connected into three horizontal axes in the “southeast-northwest” direction. A closer examination reveals that, compared to urban centers or municipal districts, the shift towards intra-provincialization is even more pronounced in counties and county-level cities. In addition, this study conducted research on economic development level, industrial structure, administrative division level, air quality, public service conditions, and regional differentiation in coastal and inland areas, and found that: the growth of employment scale in the secondary industry are directly linked to encourages the inter-provincialization of floating population. Conversely, the growth of employment scale in the tertiary industry encourages intra-provincialization of floating population. Counties encourages intra-provincialization of floating population too. Additionally, inland regions, unlike their coastal counterparts, demonstrate a heightened sensitivity to changes in environmental factors such as air quality and public service provisions, which further shape the floating population structure. The significance of these findings lies not only in enhancing our grasp of population mobility patterns within China but also in their practical applications. The research findings have significantly broadened our comprehension of the intricate transformation patterns observed in China’s population mobility structure. Moreover, these insights can serve as a valuable reference for policymakers and planners aiming to foster high-quality development and the rational allocation of public service resources in the southeastern coastal regions of the country. By considering the subtle interactions between economic, environmental, social factors, and floating population structure, more responsive and sustainable policies can be designed and implemented. Ultimately, this will help improve the livelihoods and social welfare of floating population in Fujian and other regions.
The quality of tourism resources is the material basis for the high-quality development of tourism, providing a strong attraction for the convergence of tourism flow. Based on the current real problems such as the difficulty of realizing the economic value of tourism in ancient villages and towns in Yunnan Province and the imbalance of development, this paper analyzes the spatio-temporal evolution characteristics of the coupled and coordinated development of the resource quality of ancient villages and towns and the tourism flow of 125 county-level administrative units in Yunnan Province in 2019 to 2022 by using the coupling coordination degree model, the Dagum Gini coefficient, and the gray correlation degree model. The study found that: 1) The spatial distribution of coupling coordination in Yunnan Province is based on the axis of “Diqing-Lijiang-Dali-Chuxiong-Yuxi”, and the development levels on both sides of the axis are mostly lower than those in the axis. 2) During the study period, the overall level of coupling coordination in Yunnan Province has gone through an evolutionary process of “low level coupling-antagonistic stage”, but it still belongs to a relatively low coupling level. 3) During the study period, the overall level of coupling coordination in Yunnan Province has gone through an evolutionary process of “low level coupling-antagonistic stage”, but it still belongs to a relatively low coupling level. 4) The proportion of tertiary industry income in GDP and the number of tourist trips are the core factors affecting the coupled and coordinated development of the resource quality of ancient villages and towns and tourism flow in Yunnan Province. Based on the analysis, Yunnan Province should pay attention to the dual balanced development of state, city and county-level administrative units, promote the development of regional characteristics, multi-point radiation industrial layout, break through the current development dilemma, and achieve high-quality development.
Tourism is highly susceptible to climate and weather variations, with favorable climatic conditions being recognized as a crucial tourism resource. In climate comfort research, the Universal Thermal Climate Index (UTCI) has emerged as a widely applied and highly validated evaluation metric in recent years. This study aims to estimate UTCI temperatures for tourist attractions of Jinan City using satellite remote sensing data, providing tourists with more accurate thermal comfort information. This research integrates ground meteorological station data and satellite remote sensing data using quantitative inversion techniques. The methodology first reviews quantitative remote sensing applications for retrieving key meteorological parameters (air temperature, land surface temperature, wind speed, and relative humidity), followed by the development and validation of a UTCI quantitative retrieval model. Validation using 122 data pairs from 2021 Landsat 8 and MODIS satellite imagery against 22 automatic weather stations in the Jinan metropolitan area demonstrates the efficacy of the Random Forest-based UTCI model, achieving a Mean Absolute Error (MAE) of 0.89°C and Root Mean Square Error (RMSE) of 1.37°C. Application to Jinan’s tourist attractions reveals that: 1) Remote sensing-based UTCI retrieval is feasible and reliable for tourist destinations; 2) UTCI distributions show distinct temporal and spatial patterns; 3) Substantial disparities exist between UTCI and conventional temperature forecasts; and 4) Optimal tourism conditions in Jinan vary seasonally—encompassing all regions in spring, high-altitude shaded slopes and valleys in summer, and sun-facing slopes in autumn. The study concludes by addressing improvements in model accuracy, data quality, landscape feature identification, tourist characteristic differentiation, and mitigation of subjective variations in tourist responses.
Urban surface water is increasingly important to urban ecology and development, but complex urban surface environment, shadow, and other noise interferences always restrict the extraction of urban surface water, and long time series urban surface water data sets are especially scarce. In order to reveal the real changes of surface water in major cities in China, this study selected 33 built-up areas of provincial capitals in China, inclding Hong Kong Special Administrative Region, as the study area. Based on Google Earth Engine (GEE) cloud computing platform for remote sensing big data, long time series Landsat remote sensing images were used. Combining traditional water index and urban shadow index, an automatic water extraction method with Land Surface Temperature (LST) was constructed. Based on 24 894 Landsat remote sensing images, annual surface water data with 30 m spatial resolution in major cities in China during 1990—2020 were automatically produced. The results showed that the overall accuracy of urban surface water extracted in this study was above 93%. During this study period, surface water showed an overall increasing trend, increasing from
Multi-source remote sensing data were selected as the primary data sources, including MODIS, Landsat, and land use cover change production dataset. From these data, a total of 21 characteristic variables were extracted encompassing reflectance properties, vegetation indices, climatic factors, as well as soil texture and nutrient content. Following the optimization of these characteristic variables, 5 distinct machine learning methods were employed to assess their individual advantages across different climate types. Based on the model with the highest accuracy The spatiotemporal characteristics of forest and grassland aboveground biomass during the growing season in Inner Mongolia from 2000 to 2020 were analyzed. The results show that: 1) The number of variables after feature selection varies from 4 to 21, among which reflectance, vegetation index and climatic factors are sensitive characteristic variables of aboveground biomass in all ecosystems and climatic zones; 2) Random forest is the model with the highest inversion accuracy, and the precision is significantly better after zoning; 3) The annual average aboveground biomass during the forest growing season fluctuates modestly around the average value (3.68 kg/m2) with a slight increase over 21 years, while grassland shows a trend of first decreasing and then increasing, with an overall increase, with the annual average value increasing from 46.36 g/m2 in 2000 to 56.19 g/m2 in 2020; 4) The spatial distribution of aboveground biomass during the forest growing season shows a “low-high-low-high” trend from north to south, while grassland gradually increases from west to east. Over 21 years, the area of low biomass decreased and the area of high biomass increased. This study helps to understand the dynamics of local natural resources and provides ideas for large-scale multi-ecosystem biomass inversion.
In this study, combining with field investigation and experimental analysis, we describe the distribution and types of peatlands and peat properties in the Altai Mountains, and analyze the differences in peat accumulation changes between different peatland and its main influencing factors. The results show that peatlands in the Altai Mountains are mainly distributed in intermountain depressions at an altitude of
Snow characteristics are the basis for studying snowmelt laws, snowmelt runoff characteristics, and snowmelt erosion processes. The response of snow and soil to climate warming is extremely sensitive, and climate warming will significantly affect the changes of snow characteristics and soil temperature in winter. However, the exact impact of climate warming on this process is currently unclear. Therefore, in this study, an infrared radiometer was used to increase the temperature of the snow-covered soil and the 45-day stable snow of natural deposition to simulate the changing characteristics of this process under the background of future climate warming. The results show that: under the background of climate warming, the influence of snow cover thickness and air temperature on soil temperature is significant, and after passing the test, binary linear regression equations were established for H treatment surface layer, H treatment 10 cm soil layer, L treatment surface layer and N treatment surface layer; Regarding snow characteristics, the snow in this test is mainly dry snow and slightly wet snow, and the moisture content of snow is always <3% during the whole test period,the moisture content of snow samples in the high temperature heating area (H treatment) is higher than that in the medium heating area (L treatment); The density of snow samples in the high temperature heating area (H treatment) showed a trend of first decreasing and then increasing, and the snow density in the medium heating area (L treatment) decreased slightly; The porosity of the snow remained between 75%-85% in the whole layer, and the porosity of the upper layer was significantly larger than that of the lower layer in each treatment. The particle size of snow is mainly between 1.2-2.5 mm, and the snow particles in the upper layer are relatively independent and the particle size is small, and the ice crystal particles in the lower layer are seriously connected, and the ice crystal particles are larger. This experiment provides a scientific basis for future research on the effects of climate warming on the physical and chemical properties of snow and snow-covered soil.
At present, heat waves (HWs) are gradually becoming the norm from extreme events. HWs have serious adverse effects on human health, natural ecological environment, and socio-economic systems. So, accurately predicting the duration of heatwaves is an urgent problem to be solved. The middle and lower reaches of the Yangtze River and the Huaihe River Basin are one of the three major frequent HWs areas in the world. Our study based on both the observed meteorological data from 1984 to 2020 in the Dongting Lake Basin, which located in the middle reaches of the Yangtze River, and four types of HWs influencing factors data (i.e., global warming, large scale atmospheric circulation, human activity and land-atmosphere coupling), the significant factors affecting the duration of HWs in the Dongting Lake Basin were screened by full subset regression, and then the duration model of HWs in the Dongting Lake Basin was established by full subset regression and BP neural network algorithm. The results show that: 1) The duration of HWs had a decreasing trend from 1960s to early 1970s, and remained stable from early 1970s to the mid-1990s, then shown a significant increase since the mid-1990s. 2) The duration of HWs is positively correlated with air temperature, surface solar radiation, east extension ridge of South Asian High, vegetation growth, aerosol, ground hardening and urbanization, and negatively correlated with precipitation, relative humidity, west Pacific subtropical high west extension ridge and ENSO. The west Pacific subtropical high ridge, ENSO change, aerosol, ground hardening and urbanization were identified as the significant factors of the duration of HWs with the full subset regression model. 3) A prediction model for the duration of HWs is established based on the observed data of significant factors during the same period. BP neural network model has better performance than the whole subset model. Thus it could be used as the model for the duration of HWs in the Dongting Lake Basin from 1984 to 2020.
