Historical Geographic Data of the Silk Road Zone II Versions EN1 Vol 3 (3) 2018
Land use-based human activity intensity along the Yangtze River Economic Belt, China (1970s – 2015)
: 2018 - 05 - 21
: 2018 - 05 - 30
: 2018 - 09 - 21
1066 16 0
Abstract & Keywords
Abstract: Data of human activity intensity (HAI) can be used to assess the effects of human activities on biodiversity loss. Based on China's land use/cover datasets and a land use-based index measuring human disturbance on ecosystem, we developed the dataset of HAI along the Yangtze River Economic Belt (YREB) for seven time periods, namely, late 1970s, late 1980s, 1995, 2000, 2005, 2010, and 2015. The dataset has a resolution of 1 km. The correlation coefficients between HAI and county-level population density, and between HAI and county-level gross domestic production (GDP) are 0.615 and 0.709 respectively, indicating that the dataset is of good rationality. A comparison with the global human footprint dataset indicates that our dataset has an accuracy at least 26% higher. The dataset presented here can advance our understanding of the human-nature relationship, and provide references for policymaking with regard to ecological conservation and economic development in the YREB.
Keywords: Yangtze River Economic Belt; land use; human activity intensity; long time series; raster format
Dataset Profile
English titleLand use-based human activity intensity along the Yangtze River Economic Belt, China (1970s – 2015)
Corresponding authorZhang Xuezhen (xzzhang@igsnrr.ac.cn)
Data authorsLi Shicheng, Zhang Xuezhen
Time rangeFrom late 1970s to 2015
Geographical scopeYangtze River Economic Belt in China (including Shanghai, Jiangsu, Zhejiang, Anhui, Jiangxi, Hubei, Hunan, Chongqing, Sichuan, Yunnan, and Guizhou)
Spatial resolution1000 mData volume106.82 MB (4.02 MB when compressed)
Data format*.GRID (32 bit float)
Data service system<>;
Sources of fundingNational Key Research and Development Program of China on Global Change (No. 2017YFA0603304), Fundamental Research Funds for the Central Universities, China University of Geosciences (Wuhan) (No. CUGL170823), National Natural Science Foundation of China (No. 41790424, 41561020), and the CAS Key Research Program of Frontier Sciences (No. QYZDB-SSW-DQC005).
Database compositionThe dataset consists of seven subsets on human activity intensity. They are: 1970s.zip, 1980s.zip, 1995.zip, 2000.zip, 2005.zip, 2010.zip, and 2015.zip.
1.   Introduction
Human activities have significantly changed the earth's surface, causing severe damage to its ecological environment.1 Mapping the spatial distribution of human activity intensity (HAI) and its changing trends can deepen our understanding on the human-nature relationship and contribute to sustainable development.2 HAI data can be applied in multiple fields, particularly to assess the impacts of human activities on biodiversity loss and ecosystem service degradation,3 or the effectiveness of ecological protection in nature reserves.4,5
A large number of studies have been performed to map HAI from regional to global scales, through which corresponding datasets are developed, including the global human footprint datasets,6,7 the human footprint datasets for the northern Appalachian/Acadian ecoregion,8 the comprehensive ecosystem anthropogenic disturbance datasets for China,9 and the human activity intensity datasets for the Tibetan Plateau.3,4 However, most studies for HAI mapping are static, or just cover a short time period, and long-term studies are rare. Additionally, most studies focus on administrative division-based mapping10 which cannot reflect finer details of the spatial variation of human activities and hence have limited applicability.
Accounting for more than 40% of China's total population and gross domestic product (GDP),11 the Yangtze River Economic Belt (YREB) sees prominent contradiction between human activities and ecological protection.12,13 Over the past few decades, the region’s GDP increased rapidly at the cost of resource consumption and environment pollution. For environmental restoration, the Chinese government came up with policies to promote the region’s sustainable development.14 For scientific implementation of large-scale protection policies, it is necessary to systematically evaluate the status and changing trends of the region’s human activity at a longer time period.
In view of this, this study aims to evaluate HAI in the YREB from late 1970s to 2015. We developed HAI datasets with a resolution of 1 km, which provide scientific data support for the implementation of protection policies for YREB and even the Yangtze River Basin.
2.   Data collection and processing
2.1   Data sources
Liu et al.15,16 reconstructed China's Land-Use/cover Datasets (CLUDs) at 1km scale with six classes of land use/cover, including cropland, woodland, grassland, water body, unused land, and built-up land for late 1970s, late 1980s, 1995, 2000, 2005, 2010, and 2015, primarily based on Landsat TM/ETM+ satellite remote sensing data. A field survey found that it had a classification accuracy of over 90%.15,16 Our study uses CLUDs of the seven time periods to map HAI of the YREB, the outcomes of which can be downloaded from the Resource and Environment Data Cloud Platform (http://www.resdc.cn/).
2.2   Methods
Various forms of human activities have an impact on the ecosystem, and the intensity of such activities is difficult to assess quantitatively. The change of land cover caused by human land use activities is the most direct manifestation of human disturbance to the ecosystem17 and the primary threat to biodiversity.1 Therefore, the intensity of human activity on the land surface can be defined as the extent to which human land use activities develop, utilize and transform the cover of the land surface, which can be characterized by land use/cover.10 In addition, land use-based HAI mapping can be used to describe the human-nature relationship and denote the disturbance of human activity on ecosystem and biodiversity. Sanderson et al.6 used a combination of indicators to map human activity intensity, including population density, land transformation, accessibility, and electrical power infrastructure. However, population density, accessibility, and electrical power infrastructure are all highly correlated with, and can be sufficiently characterized by, land use. In view of this, we use land use to characterize HAI, and use the comprehensive ecosystem anthropogenic disturbance index (Table 1)9 to develop the dataset of HAI for the YREB from the late 1970s to 2015.
Table 1   Land use types and the human influence scores assigned to each type9
Land use typesUnused land, permanent ice/snowWoodland, grassland/pasture, waterbody (excluding permanent ice/snow)Paddy land, dry landBuilt-up land, rural settlements, lands used for factories, quarries, mining, oil-field slattern, and lands for special uses such as transportation and airport
Human influence scores00.250.501
3.   Sample description
The HAI data for late 1970s, late 1980s, 1990, 2000, 2005, 2010, and 2015 are included in this dataset. These data are stored in 7 folders with a total amount of 106.82 MB, and 4.02 MB when compressed in a single file named "Land use-based human activity intensity along the Yangtze River Economic Belt, China (1970s – 2015)". Each folder stores the GRID raster data of a single period, with a 32-bit data structure. The coordinate system used is Krasovsky_1940_Albers. Figure 1 shows sample data for the land use-based human activity intensity along the YREB of China in the year 2015.

Fig.1   Spatial distribution of land use-based human activity intensity along the Yangtze River Economic Belt of China in 2015 (Censoring No. GS(2018)4935)
4.   Quality control and assessment
HAI is a comprehensive concept, whose rationality cannot yet be verified by direct validation against ground observation data. So indirect validation is used for verification purposes in this study. Reflecting the spatial distribution of population which is the subject of human activity, population density plays a leading role in the formation and evolution of the spatial pattern of HAI. As GDP is also an important indicator of HAI, we perform correlation analyses between population density and HAI,10 and between GDP intensity and HAI as a way to verify the rationality of HAI. The 1 km HAI for the year 2010 is spatially aggregated to county level, and the correlation coefficients at county scale are 0.615 and 0.709, respectively. Additionally, the county level HAI is fitted with county level population density and GDP intensity respectively. It is found that the HAI is exponentially related to population density and GDP density (Figure 2). Both the correlation coefficients and the exponential curves indicate that our HAI dataset is reliable.
In addition, HAI in the YREB has a spatial distribution pattern generally consistent with population density, which reinforces the reliability of our HAI dataset.

Fig.2   Correlation analysis and exponential fitting between human activity intensity, population density and gross domestic product density (GDPD)
5.   Value and significance
Using four human pressures on the land, including population density, land transformation, accessibility, and electrical power infrastructure, Sanderson et al.6 mapped human influence intensity (human footprint) on the entire land surface. Their study has been updated for 1993 – 2009 using the latest global satellite images and ground surveys, suggesting that human beings are stewards of nature.7 We compared our HAI dataset with the global datasets for the YREB to identify the significance of our datasets. Because of different human pressures considered, their absolute value was not comparable, in which circumstance the values were normalized and we only considered differences in the spatial pattern (Figure 3).

Fig.3   Comparison of the human activity intensity between this study and the global datasets for the YREB (Censoring No. GS(2018)4935)
The source data used in this study, CLUDs, had a classification accuracy of 94.3% for first level land use classes and 91.2% for second level land use classes.15,16 The UMD dataset used by the global HAI datasets had an average accuracy of only 65%,7 and the 10 km pasture data18 used by the global datasets also had great uncertainty. An analysis of their cropland data showed an obviously overestimated distribution of cropland in southern Sichuan Basin and central-western Guizhou, which led to a subsequent overestimation of HAI in these regions. In addition, the global dataset used multiple sources of remote sensing products to map HAI, including UMD dataset for the cropland data, a fusion of MODIS and GLC2000 for the pasture data calibrated by the inventory data. Each satellite-based dataset might have also introduced uncertainty during the assimilation process. Besides, the population data used in the global datasets, which had a resolution of 2.5' (equivalent to about 5 km in the equatorial region), was also of poor precision and was only applicable to certain plains and basins in China.19 The above analysis showed that the accuracy of this dataset was at least 26% higher than the global datasets in terms of land use.
This dataset also involves some uncertainties that need to be noted by data users. First, this dataset represents a conservative estimate of human activity intensity. Human activities bring various disturbances and damages to the environment, such as pollution (waste water, waste gas, and solid waste), noise,20 artificial lighting.21 Despite being the dominant human pressure, land use still gives a conservative estimate of human activity intensity. Second, this dataset is applicable to macro-level decision-making related to human activities along the YREB. The YREB has a large area of about 2.05 million km2, and its natural geographical environment varies significantly across its vast width. The land use classification system of CLUDs and the HAI assignment principle can reflect the overall changing trend of HAI along the YREB for the past 40 years, rather than the trends at local to regional scales, and are thus only suitable for macro-level decision-making.
The HAI dataset has a wide range of applications.3,7 For example, it can be used to assess the effects of human activities on animal behavior,22 or the impact on ecosystem service degradation and biodiversity loss.3 It can also be used to assess the effectiveness of ecological engineering,23 or measure the efficiency of human production activities.7, 11
We are grateful to the Resource and Environment Data Cloud Platform (http://www.resdc.cn/) for providing the CLUDs.
Foley JA, Defries R, Asner GP et al. Global consequences of land use. Science 309(2005): 570-574.
Johnson CN, Balmford A, Brook BW et al. Biodiversity losses and conservation responses in the Anthropocene. Science 356(2017): 270-274.
Li SC, Zhang YL, Wang ZF et al. Mapping human influence intensity in the Tibetan Plateau for conservation of ecological service functions. Ecosystem Services 30(2018): 276-286.
Li SC, Wu JS, Gong J et al. Human footprint in Tibet: Assessing the spatial layout and effectiveness of nature reserves. Science of the Total Environment 621(2018): 18-29.
Tapia-Armijos MF, Homeier J & Draper Munt D. Spatio-temporal analysis of the human footprint in South Ecuador: Influence of human pressure on ecosystems and effectiveness of protected areas. Applied Geography 78(2017): 22-32.
Sanderson EW, Jaiteh M, Levy MA et al. The human footprint and the last of the wild. Bioscience 52(2002): 891-904.
Venter O, Sanderson EW, Magrach A et al. Sixteen years of change in the global terrestrial human footprint and implications for biodiversity conservation. Nature Communications 7(2016). DOI: 10.1038/ncomms12558
Woolmer G, Trombulak SC, Ray JC et al. Rescaling the Human Footprint: A tool for conservation planning at an ecoregional scale. Landscape and Urban Planning 87(2008): 42-53.
Zhao GS, Liu JY, Kuang WH et al. Disturbance impacts of land use change on biodiversity conservation priority areas across China: 1990-2010. Journal of Geographical Sciences 25(2015): 515-529.
Xu Y, Xu XR & Tang Q. Human activity intensity of land surface: Concept, methods and application in China. Journal of Geographical Sciences 26(2016): 1349-1361.
Jin G, Deng XZ, Zhao XD et al. Spatiotemporal patterns in urbanization efficiency within the Yangtze River Economic Belt between 2005 and 2014. Journal of Geographical Sciences 28(2018): 1113-1126.
Xu XB, Yang GS, Tan Y et al. Ecosystem services trade-offs and determinants in China's Yangtze River Economic Belt from 2000 to 2015. Science of the Total Environment 634(2018): 1601-1614.
Song Y & Hou XY. A dataset of land cover classification for 25 port cities and their surrounding areas along the Belt and Road (2015). China Scientific Data 2(2017). DOI: 10.11922/csdata.170.2017.0131
Xu XB, Yang GS & Tan Y. Identifying ecological red lines in China’s Yangtze River Economic Belt: A regional approach. Ecological Indicators 96(2019): 635-646.
Liu JY, Kuang WH, Zhang ZX et al. Spatiotemporal characteristics, patterns, and causes of land-use changes in China since the late 1980s. Journal of Geographical Sciences 24(2014): 195-210.
Liu JY, Liu ML, Tian HQ et al. Spatial and temporal patterns of China's cropland during 1990-2000: An analysis based on Landsat TM data. Remote Sensing of Environment 98(2005): 442-456.
Zhu Y, Liu Y & Zhao L. Grid data on land use and land cover of the Loess Plateau region (2015). China Scientific Data 2(2017). DOI: 10.11922/csdata.170.2017.0137
Ramankutty N, Evan AT, Monfreda C et al. Farming the planet: 1. Geographic distribution of global agricultural lands in the year 2000. Global Biogeochemical Cycles 22(2008): GB1003.
Bai ZQ, Wang JL, Wang MM et al. Accuracy assessment of multi-source gridded population distribution datasets in China. Sustainability 10(2018): 1363.
Buxton RT, Mckenna MF, Mennitt D et al. Noise pollution is pervasive in US protected areas. Science 356(2017): 531-533.
Koen EL, Minnaar C, Roever CL et al. Emerging threat of the 21st century lightscape to global biodiversity. Global Change Biology 24(2018): 2315-2324.
Tucker MA, Bohning-Gaese K, Fagan WF et al. Moving in the Anthropocene: Global reductions in terrestrial mammalian movements. Science 359(2018): 466-469.
Xu WH, Xiao Y, Zhang JJ et al. Strengthening protected areas for biodiversity and ecosystem services in China. Proceedings of the National Academy of Sciences of the United States of America 114(2017): 1601-1606.
Data citation
1. Li S & Zhang X. Land use-based human activity intensity along the Yangtze River Economic Belt, China (1970s – 2015). Science Data Bank. DOI: 10.11922/sciencedb.631 (2018).
Article and author information
How to cite this article
Li S & Zhang X. Land use-based human activity intensity along the Yangtze River Economic Belt, China (1970s – 2015). China Scientific Data 3(2018). DOI: 10.11922/csdata.2018.0025.zh
Li Shicheng
data analysis, visualization, and initial draft preparation.
PhD, Assistant Professor; research area: historical land use reconstruction and human activity mapping.
Zhang Xuezhen
study design, manuscript review and editing.
PhD, Associate Professor; research area: climate and environmental effects of land use/cover change.
National Key Research and Development Program of China on Global Change (No. 2017YFA0603304), Fundamental Research Funds for the Central Universities, China University of Geosciences (Wuhan) (No. CUGL170823), National Natural Science Foundation of China (No. 41790424, 41561020), and the CAS Key Research Program of Frontier Sciences (No. QYZDB-SSW-DQC005).
Publication records
Published: Sept. 21, 2018 ( VersionsEN1
Released: May 29, 2018 ( VersionsZH2
Published: Sept. 21, 2018 ( VersionsZH3
Updated: Sept. 21, 2018 ( VersionsZH4