Multi-source data is an important foundation and support for city research in the current information age. It is also a key research direction of geography and urban planning. City Health Examination is a new proposition under the background of high-quality city development. The organic fusion and scientific application of multi-source data in city health examination are crucial for the in-depth analysis of city problems and the exploration of their impact mechanisms. Starting from combing the application of subjective and objective data in existing city research, it combines the national city health examination in 2019 and 2020 and the practical work of Beijing city health examination in 2018, 2019 and 2020. And explores the application of multi-source data in different spatial scales (city-district-street-community), different time scales (year-month-day-hour) and different city health examination dimensions (convenient, inclusive, healthy, orderly, characteristic, safe, livable and dynamic). It points out the scientificity, timeliness, dynamics and refinement of objective data, and recognizes the advantages of strong sample representativeness, strong pertinence and easy processing of subjective data. Finally, this paper proposes the organic integration of multi-source data represented by objective data and subjective data, and shows the case of "urban parking problem in Beijing", which empirically shows the complementarity and combination of different types of data in city health examination research. This paper hopes to provide suggestions for the improvement of data application methods in city health examination, and provide ideas for the research and development of city health examination.
Big cities abroad such as London, New York and Tokyo have also established relatively complete system of planning implementation evaluation, which can provide reference for improving urban physical examination and evaluation in China. With the advancement of urbanization, "big city disease" has become a common challenge for all big cities. Urban physical examination and planning implementation evaluation are the response to periodic problems in urban development and also the product of policy formulation and institutional arrangement. Large cities in developed countries, such as New York and London, have deeper development needs due to their high level of social development, and pay more attention to new issues such as economic vitality, social equity and inclusiveness, transportation resilience, and community construction. At present, urban physical examination and evaluation is a new model of urban planning evaluation in China. It’s also an important basis for dynamic maintenance of planning. Influenced by social system, historical culture and other factors, big cities at home and abroad have formed different evaluation frameworks. From pre-implementation consultation, data collection, analysis and demonstration, problem diagnosis to post-implementation feedback, there are great differences among big cities at home and abroad. From the perspective of the evaluation index system, China’s indexes of urban physical examination and evaluation are mainly macroeconomic and social indicators. However, big cities abroad such as London also carry out dynamic monitoring of new projects and new facilities. From the perspective of evaluation methods, the development of big data and intelligent digital platforms has greatly enhanced the accuracy and scientificity of evaluation. Furthermore, big cities abroad have made more explorations in micro-level evaluation and multi-subject participation. Based on this, this paper puts forward the following suggestions to optimize urban physical examination and evaluation in China, including dynamically optimizing the index system of urban physical examination and evaluation, in-depth mining and utilization of multi-source urban monitoring data, strengthening the connection and coordination of multi-level evaluation, and improving the regular guarantee mechanism for urban physical examination and evaluation.
The study of the nonlinear relationship between population density and human settlement environment perception is of great practical significance for the rational control of the size of big cities, the formulation of appropriate population density control standards and the scientific development of urbanization. Taking Hangzhou as a case city, based on the results of the social satisfaction questionnaire of “city health examination” of the Ministry of housing and urban rural development in 2020 and the spatial data of Hangzhou urban built environment, this paper analyzes the nonlinear relationship between population density and residents’ perception of ecological livability, health and comfort, and transportation convenience by using three-stage stepwise regression model and intermediary effect test, It also analyzes the influence path of population density on the perception of human settlements in different dimensions. The results show that: there is a significant inverted U-shaped correlation between population density and residents’ perception of ecological livability, health and comfort, and transportation convenience. When the population density is between 15 000 persons/km2 and 20 000 persons /km2, residents’ perception of these three dimensions is the best; In the aspect of impact path, the impact path of population density on Residents’ perception of ecological livability is partial intermediary, and the impact path of population density on Residents’ perception of health and comfort and traffic convenience is complete intermediary. The research can not only deepen the understanding of compact city theory, but also provide the corresponding decision-making basis for Hangzhou urban planning, and promote the healthy and orderly development of new urbanization.
Based on 337 prefecture-level above cities’ housing prices/rent and household disposable income data, this article objectively evaluates housing affordability and its changes for Chinese cities from perspectives of both homeowners and tenants, then identifies spatial types of housing affordability, and finally summarizes the influence mechanism and control measures of housing affordability. The results show that price-to-income ratio for Chinese cities is slightly higher and has increased, with an average being 7.01 and 7.76 in 2015 and 2019 respectively, but rent-to-income ratio for Chinese cities is more reasonable and has declined, with an average of 25.04% in 2015 and 22.01% in 2019. Moreover, there is distinct spatial differentiation of housing affordability in urban China. Price-to-income ratio is witnessed with higher value in the eastern region, while lower value in the central and western regions and rent-to-income ratio with higher value in the southern region and lower value in the northern region. Following the joint distribution of price-to-income ratio and rent-to-income ratio, housing affordability in China can be divided into five spatial types: type of weak affordability for both renters and purchasers, type of slight weak affordability for both renters and purchasers, type of weak affordability for only purchasers, type of strong affordability for both renters and purchasers, and type of weak affordability for renters. Distinguished control strategies have been raised for different spatial types of housing affordability. Last, the influencing mechanism of housing affordability in urban China includes housing supply and demand, urban quality of life, social expectation and housing preference, financial and real estate policies, and household income level.
The high-quality development of China’s urbanization should be based on the sharing of beautiful city life by all residents. The research on the perception of urban inclusiveness by disadvantaged groups could provide a scientific basis for the construction of city for all. This paper attempted to evaluate urban inclusiveness from the perspective of relatively disadvantaged groups, and verify the influence of individual attribute characteristics, subjective perception variables, and objective urban characteristics on residents’ inclusiveness perception. Based on the connotation of urban inclusiveness, this paper defined the evaluation elements and influence factors of inclusiveness perception. Then using a questionnaire survey data of city health examination in 2020, this research evaluated the inclusiveness perception of four types of disadvantaged groups, here are the elderly, the low-income groups, the low-educated groups and the migrants without local Hukou. Then based on multilevel model method, this paper empirically examined the effect of the individual attributes, objective characteristics of cities and residents’ subjective perception on the inclusiveness perception. The results are: 1) Urban inclusiveness in this study is a comprehensive concept covering 26 elements of 7 dimensions, including equal housing, shared service facilities, equal travel conditions, adequate social insurance, equal safety, equal health and friendly atmosphere. 2) Compared with their referenced advantaged groups, the elderly, low-income groups and migrants are more dissatisfied with the inclusiveness factors, and these factors mainly reflect the unique needs of the relatively disadvantaged groups. On the other hand, there is little difference in the inclusiveness perception between the low-educated group and the high-educated group. 3) The inclusiveness perception of relatively disadvantaged groups differs significantly among cities. The satisfaction of elderly group is low in Taiyuan. The satisfaction of low education group is low in Changsha. The satisfaction of low-income group is low in Luoyang and Lanzhou. Migrants’ satisfaction is low in Suining. 4) The inclusiveness perception varies with individual attributes and urban objective characteristics. Residents’ inclusiveness perception in small and medium-sized cities is significantly lower. The proportion of permanent residents, housing affordability, per capita housing area and population density have positive effects on the inclusiveness perception. 5) Social security, friendliness to disadvantaged groups, neighborhood relationship and general hospital satisfaction are the common factors affecting the inclusiveness perception of the four relatively disadvantaged groups. The inclusion factors that relatively disadvantaged groups pay the most attention to reflect the attributes of this group. The conclusions include: Construction of inclusive city is a never-ending process. City health examination provides a powerful tool for urban governance and a way for extensive public participation. The implementation and in-depth application of city health examination work will guarantee strong support for the construction of inclusive city.
Affected by the epidemic of COVID-19, China’s economy is facing severe challenges in 2020.This article studies the impact of the epidemic situation of COVID-19 on the economic development of Beijing through multi scenario settings by applying system dynamics model. The results show that: 1) by industry, the direct impact of COVID-19 epidemic on the transportation industry comes from travel control. The sharp decline in demand and relevant restrictive measures lead to weak growth of the industry in the short term; The demand for leasing and business services will not increase immediately with the lifting of epidemic control measures, and there is a certain time lag in the recovery of the industry. Affected by the delayed recovery of other industries, the downturn period of the development of the financial industry will be longer. The orderly promotion of resumption of work and production has accelerated the development of industry and construction industry, which is conducive to enhancing its contribution to maintaining economic stability. 2) Overall, novel coronavirus pneumonia outbreaks downtown pressure on the economy in Beijing in the short term. The introduction of various regulatory measures and the shrinking demand will further reduce economic growth. However, as the spread of the epidemic is effectively contained and the pace of returning to work and production is accelerated, the adverse impact of COVID-19 epidemic on Beijing’s economic development will be greatly reduced. 3) From the perspective of the effect of policy intervention, the improvement of urban economic development level caused by lifting the restrictions on enterprise production and business activities is greater than the financial subsidies to relevant industries. The combined intervention of the two policies will further reduce the adverse impact on the economy caused by the outbreak of COVID-19 epidemic and avoid economic decline.
In order to investigate the network characteristics and spatial pattern of inter-region value chain coordination in China, this article uses inter-province across industries input-output tables in 2007, 2010 and 2012, calculates the value chain coordination indexes across 30 provincial level regions and constructs networks of inter-region value chain coordination, based on the method of value added decomposition and the prospective of complex network. It turns out that the inter-region value chain coordination in China demonstrates remarkable small-world network feature, according to the higher value of average clustering coefficients and lower value of average shortest length of paths across 30 nodal regions. With the help of PageRank nodal centralities, it is also found that there are remarkable regional differences among the 30 provincial level regions and Jiangsu Province has the highest nodal centralities and the most significant influences in the network of inter-region value chain coordination in China. Henan Province and Guangdong Province also have the higher nodal centralities while Qinhai, Hainan and Ningxia have the lowest nodal centralities in the network. The result from Dagum Gini coefficients of regional nodal centralities demonstrates that spatial differences among regions and spatial disparities among eastern-middle-western areas narrow down. Moreover, according to the Moran’ I index which is the highest in 2012, there is significantly positive spatial association among the network centralities of regional nodes in the network. The econometrical analysis verifies the theoretic hypothesis that the levels of industrialization, the involvements in international trade and the sizes of economy in the regions positively affect the nodal centralities in the network of inter-region value chain coordination. In other words, ceteris paribus, the higher level of industrialization, the deeper involvement in international trade and the larger size of economy, the stronger the nodal centrality of a region. This research also examines the node clustering and community classification of the 30 nodal regions in the network of inter-region value chain coordination, according to the approach of optimal community clustering based on modularity. The results of node clustering and community classification show that there are several network communities in the network of inter-region value chain coordination and local network interactions among regions are also remarkable. There are notable spatial adjacencies and geographical continuities among the nodal regions in the same network community. Especially, 8 provincial level regions with relatively higher economic development levels including Shanghai, Zhejiang, Anhui, Fujian, Shandong, Hunan, Guangdong and Hainan, belong to the same network community from 2007 to 2012 and have the most stable local interactions with each other in the network of inter-region value chain coordination. The conclusions above can benefit public policy making. Since there are significant network characteristics in the inter-region value chain coordination in China, the government really needs to take network effect into consideration when putting forward the regional development policies. The government also should take advantage of leading role of some developed regions to promote the development of regions which belong to the same network community in the network of inter-region value chain coordination.
Existing research on global production networks has mainly focused on industrial sectors that produce tangible products, such as manufacturing. This paper presents a first analysis of the production network of a typical cultural industry: the entertainment and media industry based on a case study of China. Through collecting data of the trajectories of the entertainment celebrities included in the 2012 and 2017 ‘Forbes China Celebrity List’, the paper explores the spatial organization of the global production network of China’s entertainment and media industry using the method of geospatial analysis and social network analysis. The results show that: 1) The globalization of China’s entertainment and media industry has gone through the three stages of ‘product introduction’ ‘product export’, and ‘outbound investment, mergers and acquisitions and transnational production’. It exhibits a distinctive feature of domestic-demand-driven globalization. 2) The production network of China’s entertainment and media industry has formed an obvious ‘core-periphery’ structure. The evolution of the production network of entertainment and media industry is more dynamic than that of typical manufacturing industries. 3) The geographic distribution of the different types of value activities of entertainment and media industry shows remarkable diversities, which reflect the patterns of strategic coupling between the demand of each value segment and the specific platforms, resources, or market conditions of different regions. 4) The overseas activities of China’s entertainment and media industry tend to agglomerate in Western developed countries and Asian neighboring countries, demonstrating the dominance of advanced economies in the current global entertainment and media market, as well as the impact of ‘geo-cultural proximity effect’. This study reveals the developmental status and unique characteristics of the globalization of China’s entertainment and media industry. It also extends the theoretical and empirical research on global production networks through a case study of the entertainment and media industry.
Urban vitality reflects human activities and their interactions with spatial entities, which exhibits both spatial and temporal variations. However, the spatio-temporal variations have not been well addressed in existing research on the impacts of the built environment on urban vitality. Based on a collection of Sina Weibo check-in records and POI-based built environment data in central Guangzhou, this research visualizes the urban vitality at the scale of 1 km×1 km grid and 2 h period over one day. Geographically and temporally weighted regression (GTWR) are used to unveil the influence of location, density, and diversity on the urban vitality and their spatiotemporal heterogeneity. The difference between urban vitality and the impacts of the built environment on weekdays and on weekends have also been examined. The results show that: 1) the urban vitality in central Guangzhou extends along the east-west strip, with White Goose Pool, Beijing Road, Shangxiajiu Road, Zhuajing Newtown, Pazhou as the agglomerations. In one day, the urban vitality experiences dynamic changes from relatively scattered to agglomeration during the daytime and keep further agglomeration till midnight. 2) Although the location, density, and diversity significantly influence urban vitality, their impacts vary across space and time over a 24-h period. Generally, distance to the city center is negatively associated with urban vitality, while diversity and the density of leisure, housing, transport, and other facilities are positively associated with urban vitality. 3) Due to a lower activity constraint of space and time on weekdays and on weekends, their urban vitality and the spatio-temporal impacts of the built environment are obviously different.
The coordinated development of healthcare resources allocation and service utilization is closely related to the happiness of residents, the future of the country and the social progress. In particular, the research on the spatial pattern of the coordinated relationship is of great significance to narrow the interregional gaps in development and ease the contradiction between supply and demand of healthcare resources. At present, Chinese scholars pay more attention to the single aspect of healthcare resources allocation and service utilization, but the research on the relationship between them is still relatively shallow, especially from the perspective of spatial geography. Based on the hierarchical analysis framework of institutions, this paper focuses on the community healthcare institutions and hospitals in China by analyzing their spatiotemporal evolution characteristics and driving factors of healthcare resources and utilization in methods of statistical and spatial analysis. This article found that the type of coupling coordination of the two types of institutions has changed from the type of on the verge of maladjustment and recession to the type of barely coordinated development from 2010 to 2018. The coupled and coordinated development generally shows the spatial distribution pattern that the level of eastern economically developed region is higher, that of the central region is in the middle, while the northwest, northeast, southwest and other economically underdeveloped border areas are lower. The spatial aggregation characteristics of the hot and cold spot areas centered on the middle and lower reaches of the Yangtze River and the Qinghai-Tibet Plateau,and there are differences in the degree of improvement of the spatial pattern, the intensity of the local spatial agglomeration effect, and the range of changes in the cold and hot spots. Besides, the coupled and coordinated development are driven by population distribution and structure, economic development level, and geographic and spatial factors, and it also has obvious characteristics of spatio-temporal heterogeneity.
Coal and water resources are important strategic natural resources in China. How to evaluate the evolution of coal and water consumption and its influence mechanism is great significantly for the coordinated management of coal and water consumption. The coal-water footprint is calculated based on ISO standard method to measure the space-time evolution trend of coal-water footprint in Yellow River basin in 2000-2017, using coal- water footprint pressure index evaluation to analysis the matching relation between coal and water, and using Kaya identities and LMDI model to analyze the driving effect of coal-water footprint. The conclusions are as follows: 1) The total coal-water footprint showed an increasing trend during the study period, and the coal-water footprint was mainly dominated by raw coal and thermal power generation. The coal-water footprint in Shanxi and Inner Mongolia was the highest, while that in Shaanxi was the lowest. 2) The coal-water footprint pressure index gradually increased, and gradually changed from the moderating type of coal-water relationship to the moderating type of coal-water relationship. From the perspective of spatial variation, Shanxi was the largest, with the intense type of coal-water relationship. Shaanxi followed closely, and Inner Mongolia was the smallest, both of which changed from the moderating type to the moderating type of coal-water relationship. 3) The main factors influencing the coal-water footprint pressure were economic effect and technological effect. The former had a gradually enhanced influence on the coal-water footprint pressure, while the latter had a gradually weakened influence on the coal-water footprint pressure. The spatial differentiation of each driving factor was obvious. The research results can provide reference for the sustainable utilization of coal and water resources in the coal-rich areas, and provide decision support for the resource management of ecological protection and high-quality development in the Yellow River Basin.
Based on the recent largest war “The Riots of Hui” on the Chinese Loess Plateau (CLP), the evolution of environment-human interaction by this event was rebuilt by combining with the erosion flux in the Heshui and Jingbian landslide-dammed reservoirs at the different geomorphic regions, population number and ecology-social status recorded by the related documents. A weakened soil erosion was presented under “The Riots of Hui” event in the Heshui catchment at the plateau-gully region, mainly due to the rapidly natural recovery of the local eco-environment while the population decreased sharply. However, although the population increased following the ending of this war, the local ecological balance was not significantly broken, showing a relatively higher ecological carrying capacity and resilience. On the other hand, an accelerated soil erosion occurred by this war in the Jingbian catchment at the hilly-gully region. Natural vegetation in this area recovered slowly, and climate (precipitation) and soil erosion reached a balanced state until the “Reclamation-forbidden Period”. Nevertheless, soil erosion and ecological pressure were intensified again once a population explosion, displaying lower ecological resilience. All above insights suggest that it should be combined with development and protection for contemporary ecological protection and sustainable development at the plateau-gully region with near 600 mm precipitation on the CLP, while protection should be priority supplementing with land use within the check-dams system at the hilly-gully region with near 400 mm precipitation. Besides, more attention should be paid to soil and water management and scientific utilization of abandoned lands in the “hollowing villages” on the CLP.
Chengdu (102°54′-104°53′E, 30°05′-31°26′N) is located in the region of southwest, and on the east edge of Tibetan Plateau, where the precipitation is affected by the South Asia monsoon and East Asia monsoon, and rainy season is concentrated from May to October. The meteorological observation was started from 1951, which is too short for revealing the decadal to centennial climate variation. Thus, it is important to reconstruct long-term precipitation series in Chengdu, and also it has great significance for not only recognizing teleconnections of precipitation with climatic modes, but also analyzing the long-term variation of Asian monsoon subsystems and their impacts on precipitation. There are abundant Yu (rainfall)-Xue (snowfall)-Fen (Chinese length unit, 0.32 cm)-Cun (10-Fen, 3.2 cm) archives reported to the Emperors by officials in the Qing Dynasty in Southwest China, which recorded precipitation dates or event frequencies in Chengdu in detail, providing important data for the reconstruction of the rainy season precipitation in Chengdu since the middle and late Qing Dynasty. Based on Yu-Xue-Fen-Cun archives and modern instrumental data, this paper reconstructed the rainy season (May to October) precipitation in Chengdu from 1796 to 2015 with annual resolution, using the correlationship between monthly rainy days and precipitation. The results show that the average rainy season precipitation during the past 220 years in Chengdu is 838±99.3 mm (95% confidence interval). The precipitation is higher than normal in 1820s-1840s, 1880s-1910s, 1930s-1940s, and lower in 1800s-1820s, 1850s-1870s, 1920s, and 1950s-2000s, and there is an obvious abrupt change from less rainfall to more in 1879-1880. The ten years of 1832, 1896, 1898, 1899, 1903, 1907, 1915, 1921, 1937 and 1947 have highest precipitation, while 1814, 1838, 1865, 1868, 1869, 1872, 1930, 1939, 1970 and 2002 have lowest precipitation. On the decadal time scale, the rapid increasing precipitation occurs at 1939-1948, and rapid decreasing precipitation occurs at 1829-1838. On the centennial time scale, the rapid increasing in 100-year occurs at 1808-1907 with a rate of 2.1 mm/a, and the rapid decreasing in 100-year occurs at 1896-1995 with a rate of -2.5mm/a. The reconstructed series in Chengdu show significant oscillation signal with 50-75a cycles, and has a strong negative correlation with PDO for 50-70a cycle. It also has a strong negative correlation with SST of the northern Indian Ocean from the winter of the previous year to the autumn of the current year, and SST of the equatorial Pacific Ocean in the summer and autumn of the current year, which means warmer (colder) SST tended to link with less (more) rainy season precipitation.
Rural and township areas are multi-factor territorial units. Due to the complex and diverse location conditions and the obvious differences in resource endowments, the rural and township development has multiple development paths. This paper constructs a classification index system for rural and township development based on the ‘pressure-state-response’ analytical framework from the perspective of resource and environmental carrying capacity. Taking townships as the analytical unit, the paper adopts a combination of one-way advantageous indicators and composite indicators to determine their relative advantages. A comprehensive program of rural and township development classification at the provincial level has been established, taking into account of national and provincial functional zoning, coordination of function-oriented development of townships and articulation of provincial territorial and spatial development strategies. Jiangsu Province was selected as the case area to classify the types of rural and township development. The results show that there are obvious regional differences in the classifications of rural and township development in northern, central and southern Jiangsu Province, and this was used as the basis for proposing solutions to improve the resource and environmental carrying capacity of different types of township development.
The coastline classification is the basis of conducting marine natural resources investigation and management and implementing coastline protection and utilization. Sorting out different classification systems, this article constructs a new classification system for strictly controlling natural coastline in the new era with the principles of easy operation and good practicability. The new classification system includes three one-class types, seven second-class types and 12 third-class types. Especially, the classification of natural coastline is further refined. For example, the natural coastline can be divided into three second-class types, and each second-class type is subdivided into four corresponding third-class types. The establishment of the new classification system is a positive practice for Management Measures for Coastline Protection and Utilization, and likewise is beneficial for attentive management and renovation-restoration of coastline.ass type is subdivided into four corresponding third-class types. The establishment of the new classificationsystem is a positive practice for Management Measures for Coastline Protection and Utilization, and likewiseis beneficial for attentive management and renovation-restoration of coastline.
Based on the random forest model, the GF-6 image and ALOS DEM data were used as the basic information source, combined with 11 factors such as elevation, topographic relief and topographic moisture, to spatially zone the landslide sensitivity of Liulin County, Shanxi Province. The evaluation of model accuracy shows that the random forest model has an accuracy of 0.75 and the support vector machine model has an accuracy of 0.7, indicating that the random forest is more suitable for the landslide sensitivity evaluation in Liulin County. The results of the importance analysis of indicators show that: elevation, slope, distance from roads and distance from rivers are the main factors affecting the development of landslides in Liulin County. The results of the sensitivity spatial zoning indicate that the highly sensitive areas account for about 28% of the total area of Liulin County and are mainly distributed within the north and south boundaries of the Sanchuan River Basin and adjacent areas, with Jiajiaguan Township having the most widely distributed area. In terms of time cost, training difficulty, stability and accuracy, the random forest model is more suitable for such non-linear computational problems as landslide sensitivity evaluation.
Since the beginning of the 21st century, the Poyang Lake has undergone significant hydrological change. However, so far, the complex driving mechanism behind is still not clear. Based on the in-depth analysis of the characteristics and influencing factors of the open hydrological system of Poyang Lake, two combined neural network models were constructed to quantitatively distinguish the impacts, spatial-temporal differences and development trends of compounding influencing factors to this hydrological change, such as lake bottom topography change, the operation of the Three Gorges Dam (TGD), and climate change and other human activities over the Yangtze River basin. The driving mechanism of the influencing factors were then explored. Results indicate that with reference to the period 1980-1999, the average contributions of lake bottom topography change, the operation of the TGD, climate change and other human activities over the Yangtze River basin were 50%, 18% and 32% respectively to lake decline during 2003-2014. The response of the lake water level to the three driving factors shows obvious spatiotemporal differences due to different influencing mechanisms. The decline of lake water level in winter and spring is mainly caused by the change of lake basin topography, while the decline of water level in summer and autumn is mainly due to the comprehensive influence of climate change and other human activities in the Yangtze River Basin. The influence of lake bottom topography change on the decline of lake water level is the most prominent near Duchang, and the effect is still increasing. The decline of lake water level caused by the TGD is the largest at Hukou and gradually weakens to the south of the lake, while this is becoming more and more stable. The role of climate change and other human activities in the Yangtze River Basin deserves special attention. The inter-annual fluctuation of this effect is large. In some years (such as 2006, 2011), it can be the dominant driving factor of lake water level decline. However, the long-term trend of this effect is not obvious.
