Geography of China in the Big Data Era
Lao, X. H.
Institute of Geographic Sciences and Natural Resources
Research, Chinese Academy of Sciences, Beijing 100101, China
Abstract: This article discusses the
state of geography in China in the big data era along with the importance of
geographic spatiotemporal big data in geography research. Remote sensing
satellite data are the largest source of geographic big data. Since 2012, the National
Remote Sensing Centre of China has used Chinese and international remote
sensing satellite data to analyze the global ecological environment, issued a
series of annual reports that propose solutions for global environmental
problems, and helped to expand geographical research in China from the regional
to the global scale. In 2018, the State Council of China proposed the
Scientific Data Management Measures, which contribute to the effective
management of geographic big data and promote data sharing. In the same year,
the Big Geographical Data Working Committee of the Geographical Society of
China was established. This committee has played an active role in addressing
national needs and fostering talent in geography and geographical big data. The
application of big data in contemporary geographical research will strengthen
the ability to solve problems and make the results of research more applicable
in practice. Geographic big data also allow various theories, methods, and
models in geography to be tested and refined. Thus, geographic big data are expected
to play an important role in solving resource-related and environmental
problems resulting from social progress and economic development.
Keywords: geographic big data; geography of China; remote sensing; global
observation; ecological environment
1 Introduction
The processes of informatization, digitization, and
automation have generated huge and diverse data resources on different scales
throughout the world. Big data thus present both significant challenges and
opportunities[1]. As of 2003, 5 terabytes (TB) of data have been
created, and the amount of generated data has exploded in the past 10 years.
Globally, the amount of data generated is expected to reach nearly 40 zettabytes (ZB) (1 ZB = 1012 TB)
by 2020[2]. This trend in data growth has exceeded most previous
predictions and is profoundly affecting scientific research.
Nature and Science
have published special journal issues to discuss the opportunities and
challenges related to big data research[3–4]. In March 2012, the
United States government announced the launch of the Big Data Research and
Development Program[5]. In May 2012, the United Nations issued a
white paper entitled “Big data development, opportunities and challenges.”[1].
In December 2011, China’s capital markets released their first big data-themed
report entitled “The Big Data Era is Coming.”[6]. In August 2015,
the State Council of China issued a notice entitled “Promoting the Action for
the Development of Big Data.”[7]. On December 8, 2017, Chairman Xi
Jinping stressed the implementation of the national big data strategy and
accelerated the digitization of China during the Second Collective Study
Conference[8]. In March 2018, the State Council of China proposed
the Scientific Data Management Measures (GBF [2018]17) [9]. In May
2018, Chairman Xi Jinping proposed the full implementation of the national big
data strategy in the congratulatory letter of the China International Big Data
Expo 2018[10].
Big data are proliferating in all walks of life, and
the big data era has arrived. The size of a country’s data and the country’s
ability to use the data have become important indicators of the country’s
national strength. The possession and control of the data are also expected to
become critical both for nations and enterprises. The big data era has spread
to all areas, including geography and its various subfields.
2
Characteristics of Geographic Big Data
Geography mainly concerns the spatial structures and
temporal processes of natural and human elements along with the internal
mechanisms and external representations of the interactions among them.
Geographic science cannot be separated from the study of spatial structures,
temporal processes, surface characteristics, and extreme environments. From a
data perspective, scientific studies are critical to the spatial, temporal, and
spectral resolutions of the data along with data availability. More detailed
and accurate data will result in a higher level of cognition and a better
understanding of reality. Therefore, geography has welcomed big data from their
inception.
Geographic data are typical big data with the
following characteristics [11–13]: (1) Large volume. The size of
remote sensing data, wireless sensor network data, volunteered geographic
information (VGI) data, and other geographic data are on the TB to petabyte
(PB) scale or even the exabyte (EB) scale; (2) Fast updates. An integrated
network of air, space, and ground sensors means that network data are generated
quickly, and VGI data can be generated and transmitted at any time and place; (3)
Multimodal. Geographic big data include structured remote sensing data, ground
observation network data, and semi-structured or unstructured VGI data; (4)
Uncertain accuracy. A large amount of contaminated data exists, and heavy data
cleaning must be conducted before using the data; (5) High value. Geographic
big data can be widely used in agricultural services, urban management,
resource surveys, environmental protection, disaster prevention, etc.
Eighty percent of the world’s information is related
to geographic location[14]. For example, data related to resource
(e.g., land, minerals, and the environment) management, urban planning,
transportation, water conservation, agriculture, forestry, environmental
protection, emergency decision making, and so on can be generated via spatial
visualization, spatial inquiry, thematic mapping, spatial analysis, and other
spatial methods.
Geographic spatiotemporal data are an important
component of big data. These data can be generated by various means, including
remote sensing satellites (e.g., land, atmosphere, ocean, and sea ice series
satellites). The number of remote sensing satellites worldwide exceeds 1,000[15];
archived data from remote sensing satellites have reached the EB level, making
these data the largest source of geographic big data. Some regular observation
data such as meteorological and hydrological data also exist in large
quantities. There are currently thousands of observation networks around the
world, including land-based meteorological networks, hydrological networks,
sea-based buoy networks, and deep-sea exploration networks. VGI-based big data
include mobile global position system, tagging, image, and other structured
data on professional platforms along with unstructured data containin g spatiotemporal
information produced by social networks. The number of global Internet users
has exceeded four billion[16], which has resulted in a sharp
increase in VGI-based big data. Other big data related to earth systems, the
atmosphere, pavement, sea ice, and so on are also available. The CMIP6 project estimated that the quantity of
geoscience-related big data is approximately 20–40 PB[17].
The inclusion of big data in the field of geography
has produced numerous advances, including the following: (1) The combination of
traditional research with in-depth data mining allows complex associations to
be resolved; (2) Geographic big data are extensible, and the spatiotemporal
information embedded in big data can be applied in spatial analyses through big
data mining; (3) New techniques and methods that have arisen from the emergence
of big data can be widely applied in geography research; (4) Transmission
networking, data clouds, computing cloudshave become ubiquitous; (5) The fourth scientific research paradigm (i.e.,
data-driven scientific discovery) is widely applied.
As an example, the most recent exploration survey of land and resources
(arable land, permanent basic farmland, and woodland) in China employed several
data-related methods that were not used in previous surveys[18–20].
For instance, the latest survey considered professional and social big data
related to land use, basic geography, and management operations along with big
data obtained by remote sensing satellites and unmanned aerial vehicles. It
also employed methods such as supercomputing, deep learning, and cloud
computing. Another example is traditional sampling analysis; it is difficult to
study the spatial characteristics of urban populations and analyze migration
without the help of big data and the associated techniques. This work can be
carried out rapidly based on data available from mobile phone signals.
Geographic information system (GIS) data are essential
for the application of big data in geography. GIS is a computer system for
collecting, storing, managing, analyzing, displaying, and applying geographic
information. GIS is a general technology for analyzing and processing massive
geographic data in the real world (resources and environment) [21].
GIS has undergone several development stages, including the initial development
of GIS, which occurred in the 1970s and focused on the development and
management of map data, the statistics of spatial data, and map production. In
the 1980s, GIS entered a stage of consolidation, which included the development
of mixed-data models, comprehensive spatial analysis, and the development of
professional software modules and environmental resource applications. These
two eras are collectively referred to as the development stage of GIS[22–23],
which is also known as the mainframe era. The 1990s were characterized by the
development of geographic information science and are referred to as the PC
era; in this stage, large-scale databases and networks were developed to
support, geographic ontology, system modeling, and web GIS. In the 21st
century, during which the Internet era began, GIS entered the public service
stage. This stage was characterized by the development of location service
applications, computing services, grid GIS, and virtual environment
applications. In the past 10 years, GIS has entered the social services stage
of development, also known as the development stage of the geographic
information world; this stage is characterized by automatic processing, massive
storage, efficient computing, and knowledge-based services[24].
Remote sensing satellite data account for the largest proportion of
geographic big data and have allowed Chinese geographical research to expand
from a national scale to a global one. In previous scientific studies,
developed countries mainly used primary observation data to study global
problems. Now, relying on remote-sensing satellites at home and abroad, Chinese
geoscientists have been able to do research on transnational and global
problems. China has developed one of the most complete and expansive networks
of remote sensing satellites in the world, rivaling that of the United States[25].
Chinese scientists are now able to use the
basic data from foreign remote sensing satellites along with their own
algorithms to produce data products and analytics with international influence.
Chinese remote sensing satellites can also be used to obtain data from all over
the world. For example, starting in 2012, the National Remote Sensing Centre of
China has organized efforts to analyze the global ecological environment using
Chinese and foreign remote sensing satellite data, resulting in a series of
annual reports[26] covering topics such as ecology, vegetation, land
use, agriculture, wetlands, and urbanization. These reports have had a
significant impact on the activities of international scientific organizations
such as the Earth Observation Organization.
3 Geographical Big Data Enhance China’s Ability to
Solve Practical Problems
|
Figure 1 Big data "injection"
geography promotes the further development of academic theories and methods
|
With societal development and increasing economic
growth, problems related to resources and the environment are becoming more
prominent and complex. Traditional simplified models often cannot solve these
complex practical problems, and solutions to these problems need to be
developed and refined based on large sample sizes. In addition, many geographical
theories, methods, and models require accurate boundary constraints and field
data as inputs. The history of geography shows that obtaining reliable,
multifactorial analysis results is related to the availability of high-quality
input data. The incorporation of big data into contemporary geography will
improve the ability to solve real problems[27] and ensure that the
prediction results of models are consistent with reality (Figure 1).
The advances associated with utilizing geographic big
data are attributed to several factors, including the following. First, big
data allow the scope of geography research to range from local sampling to
global coverage (e.g., using remote sensing satellite
data to estimate global vegetation biomass or crop yield). This type of
research has been widely applied in the areas of tourism, natural resources,
agriculture, and so on. Second, by incorporating big data, geography research
can be extended from understanding the current conditions to generating
high-precision historical reconstructions and highly reliable predictions of
future scenarios. This type of research has been widely applied to study
regional development, global change, and earth processes based on numerical
models. Third, big data mining can be applied to high-resolution spatiotemporal
data fitting and the analysis of spatiotemporal associations, leading to an
understanding of complex relationships involving multiple factors. Finally, the
large sample sets made possible by big data can be used to test and improve
geographical theories, models, and methods that were formulated based on
relatively small samples in the past. In this way, big data can improve the
reliability of scientific research and help researchers develop solutions that
are more appropriate to real-world problems.
4 Considerations Related to Geographic Big Data
Application Management
The meaningful application of geographic big data
cannot be separated from orderly data management, the protection of
intellectual property, data security, and quality control. In August 2018, the
Big Geographical Data Working Committee of the Geographical Society of China
was established[28]. The demand-oriented approach proposed by the
working committee is aimed at addressing the most important bottlenecks in the
development of geographic big data in China, which include:
(1) Data management and data sharing
Guided by the State Council of China’s Scientific Data
Management Measures, it is expected to take five years of effort to provide a
basic solution to the bottleneck in geographical data sharing. The national big
data strategy to address this bottleneck involves setting up a platform for
data publishing and sharing, constructing a national scientific data center,
and encouraging society members to protect intellectual property rights.
(2) Intellectual property protection and scientific
evaluation
The establishment of standards, norms, and methods for
data authentication and intellectual property rights will promote the data
sharing and scientific achievements, pilot and demonstration work in a period
of one to two years, and popularization throughout the country in the next five
years.
(3) Deficiency in global-scale research in China
The Geographical Society of
China will hold a global research and development conference and will become
actively involved in any relevant national science and technology activities
(e.g., the remote sensing monitoring of the global ecological environment),
which will help enable the publication of representative global datasets within
five years. This objective is in line with the national development strategy
and the United Nations sustainable development goals.
(4) Deficiency in academic papers based on scientific
data
The Big Geographical Data Working Committee and the
Academic Editorial Committee will work closely together to promote the
publication of academic papers and original scientific data. The proportion of
academic papers and data published in academic journals sponsored by the
Geographical Society of China (or co-sponsored) is currrently less than 1%.
However, this is expected to increase to more than 30% over the next five years
as a result of the Committee’s efforts.
(5) Construction of a data computing environment
The Geographical Society of
China will help realize the value of geographic big data by (a) selecting and
promoting practical examples of valuable computing environments for geographic
big data and (b) promoting the use of big data in scientific discovery and
sustainable development.
(6) Scientific application of geographic big data to
sustainable development
Making big data play a positive role in the
sustainable development of society is one of the important tasks of the Big
Geographical Data Working Committee. The Geographical Society of China intends
to explore how geographical big data can be applied to promote national and
local sustainable development and ensure national and local ecological
security.
(7) Fostering talent in geographic big data
The Big Geographical Data Working Committee will
continue to promote the Capacity Building in 100 Universities Program on Global
Change Research Data Publishing & Sharing along with the development of
related textbooks and university courses with the goals of (a) introducing big
data in 100 colleges and universities by 2025 and (b) making courses in
geographical big data available in more than 10 universities.
(8) Data security and scientific ethics
Data security and scientific ethics are key issues
that the Geographical Society of China must emphasize. The Big Geographical
Data Working Committee will prioritize data security and scientific ethics in
geographic big data and work to create associated standards and guidelines.
5 Conclusion
With the advent of geographical big data, the
Geographical Society of China welcomes new opportunities for development. Geographical
theories, methods, and application practices at the regional and global scales
will be continuously tested and refined using the large sample sets provided by
big data. With the development of global-scale geographical big data, particularly
the rapid proliferation of earth observation satellite data in China, important
research results are being produced. Thus, geographic big data will play an
important role in social progress and economic development, both globally and
within China, and help solve pressing challenges related to resources and the
environment.
Acknowledgement
This article is based on the report of the annual
conference of the Big Data Working Committee of the Geographical Society of
China on September 21, 2019. Ma, J. H. and Deng, X. M. contributed to the
revision and improvement of this article.
References
[1]
UN Global Pulse. Big data for
development: challenges & opportunities [R/OL]. [2012-10-02]. http:
//www.unglobalpulse.org/projects/BigDataforDevelopment.
[2]
https://www.idc.com/.
[3]
https://www.nature.com/.
[4]
https://www.sciencemag.org/.
[5]
The U.S. Government Released
“Big Data Research and Development Initiative”.
[6]
https://blog.sina.com.cn/zhaogd.
[7] State Council.
Platform for Action to Promote the Development of Big Data (Guofaban [2015]
NO.50) [Z]. 2015-8-31.
[8]
http://www.xinhuanet.com/politics/leaders/2017-12/09/c_1122084706.htm.
[9] State Council.
Measures for the Administration of Scientific Data (Guofaban [2018] NO.50) [Z].
2018-03-17.
[10]
Xi, J. P. Congratulatory Letter
from China International Big Data Industry Expo 2018[R]. People’s Daily,
2018-05-27 (01).
[11]
Guo, H. D. Big Data, Big
Science, Big Discovery—a summary of the International Symposium on Big Data and
Scientific Discovery [J]. Proceedings of
the Chinese Academy of Sciences, 2014, 299(4): 500–506.
[12]
Guo, H. D. Earth Big Data
Science Engineering [J]. Proceedings of
the Chinese Academy of Sciences, 2018, 33(8): 818–824.
[13]
Han, P., Zhang, J. W., Wang, Y.
W. Application of Geographic Information and Location Big Data in Map Compilation
[J]. Gansu Science and Technology,
2016, 32(1): 334–336.
[14]
Cao,
S. B. Overview of the remote sensing satellite market in 2017 (Part 2) [J]. China Aerospace, 2018(6): 73–78.
[15]
https://new.qq.com/omn/20190128/20190128B00CCS.html.
[16]
http://www.xinhuanet.com/expo/zt/sjdlxxdh/index.htm.
2018.
[17]
https://esgf.llnl.gov/search/cmip6.
[18]
Zheng, J. Y. Analysis of
Technical Exploration Based on the Third National Land Survey [J]. China Resources Comprehensive Utilization,
2019, 37(9): 64–66.
[19]
Zhu, W., Wang, J. C., Yuan, R.
C. Talking about the Application of Intelligent Mobile Terminals in the Three
Tunes of the Territory [J]. Science &
Technology and Information, 2017(9): 155.
[20]
Department of Natural Resources
Investigation and Monitoring, Guangxi Natural Resources Department. Where is
the Three Tunes and New? —— A picture to understand the new changes in the
third national land survey [J]. Southern
Land Resources, 2019(6): 18–19.
[21]
Chen, S. P. Introduction to
Geographic Information System [M]. Beijing: Science Press, 2003.
[22]
Zhou, C. H. New era of
Geographic Information System: Grid Geographic Information System [J]. Geographic Information World, 2007(4):
17.
[23]
Zhang, H. Y., Wang, Q. M.,
Zhou, C. H., et al. "Digital Earth" and Geographic Information
Science [J]. Geoinformatics, 2001(4):
1–4.
[24]
Zhang, S. L. The improvement of
geographic information service capabilities effectively promotes social
progress [J]. Surveying and Mapping
Technology & Equipment, 2015, 17(1): 53–54.
[25]
http://www.broadcast.hc360.com.
[26]
http://www.nrscc.gov.cn/.
[27]
Wang, P. Big data plugged in
machine learning wings to provide new fuel for 5G [J]. Communications World, 2019(11): 47–48.
[28]
Geographical Society of China.
The Big Geographical Data Working Committee of Geographical Society of China
(GSC_BigData) established [R]. Journal of
Global Change Data & Discovery,
2018, 2(3): 354-356. DOI: 10.3974/geodp.2018.03.18.