Analysis of Population Spatial Distribution in Counties of Guangdong

Province Based on Spatial Autocorrelation

Yang Yu

1,2 a

, Boyuan Liu

and Wei Feng

1,2,* c

China Institute of Geo-Environment Monitoring, Beijing 100081, China

Key Laboratory of Mine Ecological Effects and Systematic Restoration, Ministry of Natural Resources,

Beijing 100081, China

No.1 Institute of Geology and Mineral Resource Exploration of Shandong Province, Jinan 250100, China

Keywords: Population Spatial Distribution, Spatial Autocorrelation, Population Density, County-Level Regions of

Guangdong Province.

Abstract: A comprehensive examination of the characteristics and patterns of population spatial distribution is

imperative for achieving a balanced dispersal of population, fair allocation of resources, and harmonised

socio-economic development. This paper employs spatial autocorrelation analysis, complemented by

population density and range methodologies, to scrutinise the county-level permanent population figures of

Guangdong Province for the years 2010, 2015, and 2020. The focus lies in exploring the spatial distribution

features of the population across Guangdong Province. The findings indicate that the population in

Guangdong Province predominantly congregates along the coastal regions, exhibiting a discernible gradient

in population density from north to south, accompanied by notable disparities in spatial distribution. A

conspicuous clustering tendency is observed in the population spatial distribution, characterised chiefly by

high-high and low-low agglomerations, which tend to be concentrated and contiguous, displaying an

escalating clustering intensity. High-high agglomerations are predominantly concentrated in the developed

zones of the Pearl River Delta, whereas low-low agglomerations are primarily situated in the northern and

eastern regions of Guangdong.

1 INTRODUCTION

Population distribution refers to the aggregation and

dispersion of the population within specific

geographical spaces at a given time. It reflects the

spatial manifestation of population activities, which

is influenced by various factors such as modes of

social production, economic development levels, and

natural conditions, resulting in different distribution

patterns across regions. Currently, with regional

economic development being uneven, there is a clear

spatial concentration and distribution difference in

population, where economically developed central

cities and urban clusters become major agglomeration

areas, while regions with relatively lagging

economies face severe population loss and economic

development dilemmas. An in-depth study of the

https://orcid.org/0000-0001-6584-4050

https://orcid.org/0009-0008-5078-9958

https://orcid.org/0009-0009-1917-8769

characteristics and laws of population spatial

distribution is of great significance for achieving

reasonable population distribution, balanced

allocation of resources, and coordinated socio-

economic development (Ju, 2022; Yu, 2022; Zhang, Z. Q.,

2022; Yin, 2022; Li, B., 2022; Jin, 2022).

Guangdong Province holds a prominent position

on China's economic landscape, serving as a pioneer

in the nation's economic advancement, with its

overall economy and populace ranking at the

forefront nationwide. In 2020, Guangdong Province,

representing 8.73% of the country's population,

contributed 10.90% to China's GDP. In 2017, the

Guangdong-Hong Kong-Macao Greater Bay Area

was integrated into the national development strategy,

aiming to establish a world-class urban cluster and

bolster global competitiveness. The Pearl River Delta

Yu, Y., Liu, B., Feng and W.

Analysis of Population Spatial Distribution in Counties of Guangdong Province Based on Spatial Autocorrelation.

DOI: 10.5220/0013572900004671

In Proceedings of the 7th International Conference on Environmental Science and Civil Engineering (ICESCE 2024), pages 24-30

ISBN: 978-989-758-764-1; ISSN: 3051-701X

in Guangdong, as a major component of the

Guangdong-Hong Kong-Macao Greater Bay Area,

plays a pivotal role. The execution of this strategy is

anticipated to attract more external talents, further

stimulating significant population concentration in

the region. Thus, the distinctive economic backdrop

and geographical location determine the

distinctiveness of population spatial distribution in

Guangdong Province. A survey of existing literature

underscores substantial progress in the exploration of

population issues in Guangdong, encompassing

focused investigations on migrant populations, labor

demographics, and aging trends. While some scholars

have employed population density and gravity

models to scrutinize the temporal and spatial

distribution patterns of the population across

Guangdong Province as a whole, few have delved

into the inherent connections within the province's

population spatial distribution (Shi, 2022; Zhou, 2022;

Bao, 2022; Gao, 2022; Wang, 2021; Dang, 2022; Liao,

2022; Ye, 2019; Hou, 2023).

Building upon the aforementioned points, this

paper will employ the permanent population data

from 2010, 2015, and 2020, utilising counties as the

focal research units. It will apply the spatial

autocorrelation method to investigate the prevailing

characteristics of population spatial distribution in

Guangdong Province and unveil the inherent

correlations within population distribution

phenomena. This endeavour aims to furnish insights

for governmental policymakers to craft strategies for

enhancing population distribution, resource

allocation, and fostering coordinated economic

growth (Figure 1).

Figure 1: Study Area Map.

2 STUDY AREA OVERVIEW

Guangdong Province is located at the southernmost

tip of mainland China, between east longitude 109°45′

and 117°20′ and north latitude 20°09′ and 25°31′,

adjacent to Fujian, Jiangxi, Hunan, and Guangxi, with

the South China Sea to the south, and Hong Kong and

Macao on either side of the Pearl River estuary. The

province falls within the East Asian monsoon climate

zone, maintaining an average annual temperature

above 20℃, with an average annual precipitation

between 1333 and 2254mm. Guangdong features a

complex and varied topography, including mountains,

hills, and plains, generally showing a high north to

low south trend, with the north dominated by

mountains and high hills, while the south is mainly

plains. In 2020, the total permanent population of the

province was 126.0125 million, with Dongguan City

having the highest population distribution of 10.4666

million people, and the least populated county being

Nanao District of Shantou City, with 64,400 people.

3 DATA SOURCES

This paper centres on the enduring populace figures

of the county-level administrative regions in

Guangdong Province for the years 2010, 2015, and

2020. The population data originates from the

"Seventh National Population Census Bulletin" and

"Sixth National Population Census Bulletin" of

Guangdong Province and its cities, along with the

2016 Statistical Yearbook. To enable longitudinal

scrutiny, the population statistics for all years are

harmonised with the county-level administrative

divisions as of 2010. These administrative regions

encompass counties, county-level cities, municipal

districts within prefecture-level cities (e.g.,

Shaoguan's municipal districts), and prefecture-level

cities devoid of subordinate counties (e.g., Dongguan

City, Zhongshan City). GDP and additional economic

metrics are extrapolated from the 2021 China

Statistical Yearbook and the Guangdong Provincial

Statistical Yearbook. Regional area information is

acquired from the National Administrative Division

Information Query Platform (Xie, 2023; Ma, 2023;

Guan, 2023; Kang, 2023; Ye, 2023; Li, 2023; Zhang,

Y., 2022; Xu, 2023; Li, X. R., 2022).

4 RESEARCH METHODS

4.1 Population Density

Population density is defined as the ratio of the

number of people in a region to its area. It is one of

the important indicators reflecting the form of

Analysis of Population Spatial Distribution in Counties of Guangdong Province Based on Spatial Autocorrelation

population distribution and differences between

regions.

(1)

In the formula, d

represents the population

density of the i

county，x

represents the population

number of the i

county；s

represents the area of the

county region.

4.2 Range Method

The range method is a simple and intuitive statistical

analysis method used to describe the range of

variation or difference in a dataset. It can be used to

quantify the degree of difference in population

distribution within a region.

R =

max

min

(2)

In the formula, R is the range of population

density within the counties of Guangdong Province,

max

is the maximum value of county population

density, and x

min

is the minimum value of county

population density.

4.3 Spatial Autocorrelation

4.3.1 Global Autocorrelation

Global autocorrelation analysis can determine

whether there is significant spatial clustering or

dispersion of population distribution within the study

area, measured by the Global Moran's I index. The

Global Moran's I index values range between [-1,1].

When the index value is greater than 0, the regional

population distribution shows a positive spatial

correlation, indicating a certain trend of clustering in

space; when the index value equals 0 or is close to 0,

the distribution of the population across different

areas is random, or there is no spatial autocorrelation;

when the index value is less than 0, the regional

population distribution shows a negative spatial

correlation, indicating a tendency to disperse in space.

The formula for calculating the Global Moran's I

index is as follows:

𝐼



𝑛

∑∑

𝑊





𝑋



−𝑋





𝑋



−𝑋















∑∑

𝑊













∑

𝑋



−𝑋











(3)

In the formula, I

is the Global Moran's I index, n

is the total number of regions, x

，x

are the population

numbers of regions i and j respectively, represents any

element of the spatial weight matrix, defining the

adjacency relationship between spatial objects. In this

study, the adjacency rule is used to construct the

weight matrix, where w

=1 if regions i and j are

adjacent, and w

=0 if not.

4.3.2 Local Spatial Autocorrelation

Global spatial autocorrelation can only indicate

whether there is spatial association in the distribution

of population on a global scale but cannot explain the

characteristics of local spatial association. Local

autocorrelation allows for the analysis of spatial

correlations between local areas within the study

region, measured by the Local Moran's I index. The

Local Moran's I index is calculated using the

following formula:

𝐼



𝑛



𝑋



−𝑋





∑

𝑋



−𝑋













𝑊



𝑋



−𝑋









(4)

In the formula, I

represents the Local Moran's

Index, with the meanings of the other variables

consistent with Equation (3).

5 RESULTS ANALYSIS

5.1 Characteristics of Population

Density Distribution

The range of population density in Guangdong

Province for the years 2010, 2015, and 2020 were

33,964, 33,947, and 30,470 persons/km

, respectively,

indicating significant differences in the distribution of

population density. Over time, the disparity in

population density has slightly decreased, but the

differences remain pronounced.

Utilizing ArcGIS software for spatial

visualization of population density in Guangdong

Province over these three years allowed for an

intuitive display of the distribution of population. The

distribution maps of population density across three

periods (Figure 2) reveal significant spatial variations

in population density within the province. Regions

with higher population numbers are concentrated

along the coastal areas, with population density

decreasing towards the northern regions. There is a

clear gradation in population density from north to

south across the province. High-density areas (more

than 1,000 persons/km

) are primarily located in the

Pearl River Delta, western Guangdong, and parts of

eastern Guangdong. The extent of these high-density

areas, such as the Pearl River Delta, has gradually

expanded over time; low-density areas (less than 200

persons/km

) are steadily distributed in the northern

parts of Guangdong (Table 1).

ICESCE 2024 - The International Conference on Environmental Science and Civil Engineering

Figure 2: Map of Population Density Distribution in Guangdong Province Range.

Table 1: Population Density Statistics by Year

(persons/km

Year Max Min Range

2010 34,038 74 33,964

2015 34,024 77 33,947

2020 30,548 78 30,470

5.2 Global Spatial Autocorrelation

Analysis

Global spatial autocorrelation analysis reflects the

overall characteristics of population distribution

across the entire spatial region. Utilizing GeoDa

software and a weight matrix based on adjacency

rules, the global Moran's I for the population

distribution of Guangdong Province's counties in

2010, 2015, and 2020 was calculated, with the results

presented in Table 2. For all three periods, Moran's I

and Z-value were positive, and P-values were less

than 0.05. At a significance level of α=0.05, the

spatial distribution of population in Guangdong

Province shows a positive correlation, indicating

significant clustering characteristics of population

spatial distribution. Observing the trend, the global

Moran's I indicates an increasing trend, suggesting

that the clustering of population spatial distribution in

Guangdong Province is becoming more pronounced.

Table 2: Global Spatial Autocorrelation Parameters of

Population Distribution.

2010 2015 2020

Moran'I 0.1302 0.1523 0.2691

Z-value 2.8243 3.2369 5.332

P-value 0.009 0.006 0.001

5.3 Local Spatial Autocorrelation

Analysis

Local spatial autocorrelation analysis facilitates a

thorough exploration of the spatial relationships

between local geographic entities and their

neighbouring regions. Utilising GeoDa software,

local autocorrelation analysis was conducted on

county population counts in Guangdong Province for

the years 2010, 2015, and 2020, generating local

Moran’s I scatter plots and LISA (Local Indicators of

Spatial Association) cluster maps for each year. The

Moran’s I scatter plots are segmented into four

quadrants, representing distinct types of local spatial

associations: High-High clusters, High-Low outliers,

Low-Low clusters, and Low-High outliers. High-

High and Low-Low clusters denote positive spatial

correlation, while High-Low and Low-High outliers

indicate negative spatial correlation. As depicted by

the Moran’s I scatter plots across the three periods,

the majority of data points fall within the first and

third quadrants, indicating that the population spatial

distribution in Guangdong Province predominantly

exhibits High-High and Low-Low clustering

tendencies. The local Moran's I values for 2010, 2015,

and 2020 stood at 0.1302, 0.1523, and 0.2691,

correspondingly, all positive and demonstrating an

upward trajectory. This indicates a strengthening

positive spatial correlation among neighbouring

county units over time, signifying an augmentation in

clustering.

The LISA (Local Indicators of Spatial Association)

cluster map visually represents the local spatial

relationships depicted by the local Moran's I scatter

plot, showcasing the distribution and location of

spatial clustering types. From Figure 3 and Figure 4, it

is observed that High-High clusters are predominantly

located in the Pearl River Delta region and parts of

eastern

Guangdong, exhibiting a contiguous

Analysis of Population Spatial Distribution in Counties of Guangdong Province Based on Spatial Autocorrelation

Figure 3: Local Moran’s I Scatter Plot.

Figure 4: LISA Cluster Map of Population Distribution in Guangdong Province.

distribution. High-High clusters in the Pearl River

Delta are influenced by the radiating effect of urban

agglomerations, expanding outward and increasing in

clustering intensity. However, the clustering trend in

parts of eastern Guangdong weakens, with the spatial

distribution correlation becoming insignificant by

2020. Low-Low clusters mainly occur in northern and

eastern Guangdong, also showing a contiguous

distribution characteristic. The number of Low-Low

clusters exhibits a fluctuating growth trend.

Influenced by natural geographical conditions and

slow economic development, counties within this

clustering type have a stable but sparse population

distribution. Low-High clusters are found surrounding

population core areas, affected by the "population

siphon" effect from nearby economically developed

regions. These areas have relatively fewer inhabitants.

As the surrounding economically developed areas

continue to grow, their economic influence and

radiating effects strengthen, gradually attracting

population inflow. This dynamic causes some areas

initially classified as Low-High clusters to gradually

evolve towards High-High clustering. High-Low

clusters, observed in Yingde city from 2010 to 2015,

became less significant by 2020.

6 CONCLUSION

This study, based on spatial autocorrelation analysis

combined with population density, has analyzed the

spatial distribution characteristics of Guangdong

Province's population over the past decade, yielding

the following conclusions:

Analysis of population density across three

periods in Guangdong Province reveals significant

spatial distribution differences. High population

density areas are concentrated along the coastal

regions, with density decreasing progressively

towards the northern regions, indicating a clear

gradation in population distribution levels.

The global Moran's I indices for population

distribution in Guangdong Province across three

periods exhibit positive values. Statistical tests

confirm a significant positive spatial correlation in

population distribution, indicative of spatial

clustering characteristics with an escalating

clustering intensity over time.

Distinct local clustering traits are discernible in

population distribution across the three periods, with

High-High and Low-Low clusters predominating.

These clusters demonstrate a pattern of concentrated

ICESCE 2024 - The International Conference on Environmental Science and Civil Engineering

and contiguous distribution, with clustering intensity

strengthening over time. High-High clusters are

chiefly situated in the developed areas of the Pearl

River Delta, while Low-Low clusters are primarily

observed in the northern and eastern regions of

Guangdong.

The spatial distribution of population to some

extent mirrors the level of economic development and

resource allocation within a region. The noticeable

disparities in population distribution across

Guangdong underscore the imbalance in regional

economic development and resource allocation levels.

To address this, Guangdong should bolster policy

support for relatively underdeveloped areas in terms

of population and economy, harnessing the radiating

effect of the Pearl River Delta urban agglomeration to

propel swift development in other regions. However,

in the pursuit of economic advancement, it is

imperative to comprehensively consider

enhancements in population quantity, quality, and

productivity levels to foster a balanced population

distribution. Such an approach will ensure

coordinated development among population

dynamics, natural environment, and socio-economic

facets.

ACKNOWLEDGMENTS

Correspondence should be addressed FENG Wei;

ecorestoration2023@163.com.

REFERENCES

Ju, X. 2022. Advances in remote sensing monitoring of key

elements of cultivated land ecological environment.

Natural Resources Informatization, (03): 59-65.

DOI:10.3969/j.issn.1674-3695.2022.03.010.

Yu, Y. 2022. Applications of 3 D laser scanning technology

in the extraction of vegetation parameters. Journal of

Henan Polytechnic University (Natural Science),

41(04): 51-57. DOI:10.16186/j.cnki.1673-97

87.2020090105.

Zhang, Z. Q. 2022. The evolution pattern and influence of

human activities of landslide driving factors in Wulong

section of the Three Gorges Reservoir area. The

Chinese Journal of Geological Hazard and Control,

33(03): 39-50. DOI:10.16031/j.cnki.issn.1003

-8035.2022.03-05.

Yin, Y. 2022. A Quantitative Study on Cultivated Land

Compensation Based on Ecological Value

Accounting—Taking Shenyang City As an Example.

Natural Resource Economics of China, 35(11): 18-24.

DOI:10.19676/j.cnki.1672-6995.000750.

Li, B. 2022. The Coordination and temporal-spacial

evolution pattern between population urbanization and

land urbanization——A case study on 18 cities in

Sichuan. Natural Resources Information, (09): 29-36.

Jin, S. Z. 2022. Prediction of Carbon Emission in

Chongqing Based on Support Vector Regression.

Journal of Chongqing University of Science and

Technology (Natural Sciences Edition), 24(01): 110-114.

DOI:10.19406/j.cnki.cqkjxyxbzkb.2022.01.018.

Shi, S. H. 2022. Heavy Metal Migration and Soil Pollution

Assessment in an Intensive Mining Area in the

Southwest China. Metal Mine, (02): 194-200.

DOI:10.19614/j.cnki.jsks.202202026.

Zhou, X. 2022. Analysis on spatio-temporal measure and

evolution characteristics of information entropy of land

use structure in the Yangtze River Delta urban

agglomeration. Natural Resources Informatization, (01):

44-51. DOI: 10.3969/j.issn.1674-3695.2022.01.008.

Bao, J. J. 2022. Temporal and spatial evolution of

coordinated development between new urbanization

and intensive use of cultivated land in Gansu. Natural

Resources Information, (08): 42-50. DOI:10.3969/j.is

sn.1674-3709.2022.08.007.

Gao, M. M. 2022. Spatial and temporal variations in

vegetation index and its impact factors in the West

Liaohe Plain in Inner Mongolia. Hydrogeology &

Engineering Geology, 49(01): 175-182. DOI:10.16030

/j.cnki.issn.1000-3665.202104046.

Wang, N. 2021. Mine Environment Investigation and

Research Based on Remote Sensing Technology:A

Case Study of the Jidong Iron Mine. Metal Mine, (10):

192-198. DOI:10.19614/j.cnki.jsks.202110026.

Dang, Y. G. 2022. Discussion on the Top-level Design and

Optimization of the System of Cultivated Land

Requisition-Compensation Balance. Natural Resource

Economics of China, 35(06): 43-48, 88. DOI:

10.19676/j.cnki.1672-6995.000669.

Liao, Z. M. 2022. Habitat Quality Assessment Based on

InVEST Model in Dianchi Basin. Ecology and

Environmental Monitoring of Three Gorges, 7(02): 42-

52. DOI:10.19478/j.cnki.2096-2347.2022.02.05.

Ye, S. S. 2019. Ecological Environmental Cost Accounting

of Mining Area Based on the Green Mine: A Case from

a Mining Area in the North China Plain. Metal Mine, (04):

168-174. DOI:10.19614/j.cnki.jsks.201904031.

Hou, J. W. 2023. Discourse on scientific advancements in

mining ecological restoration. Mining Safety &

Environmental Protection, 50(06): 1-6,15. DOI:

10.19835/j.issn.1008-4495.2023.06.001.

Xie, Y. 2023. Analysis of land use change and influencing

factors based on GIS and RS:A case of Hefei. Natural

Resources Informatization, (04): 18-23. DOI:

10.3969/j.issn.1674-3695.2023.04.003.

Ma, J. X. 2023. Spatio-temporal change characteristics of

water conservation function in the Zhang-Cheng

district based on the InVEST model. Hydrogeology &

Engineering Geology, 50(03): 54-64. DOI:

10.16030/j.cnki.issn.1000-3665.202208084.

Guan, D. J. 2023. Study on the Spatial-temporal Coupling

Relationship between New Urbanization and Logistics

Analysis of Population Spatial Distribution in Counties of Guangdong Province Based on Spatial Autocorrelation

Industry: Based on the Panel Data of Anhui Province.

Journal of Cangzhou Normal University, 39(01): 51-55.

DOI:10.13834/j.cnki.czsfxyxb.2023.01.012.

Kang, S. Y. 2023. Analysis of land use change and driving

factors in Chengdu from 2010 to 2020. Natural

Resources Information, (03): 43-50.

DOI:10.3969/j.issn.1674-3709.2023.03.006.

Ye, S. S. 2023. Comprehensive Management of Ecological

Protection and Restoration Funds. Finance and

Accounting for International Commerce, (20): 54-58.

DOI:10.3969/j.issn.1673-8594.2023.20.009.

Li, X. L. 2023. Theoretical analysis and engineering

practice of dynamic pre-reclamation in coal mining

subsidence area. Mining Safety & Environmental

Protection, 50(01): 86-91. DOI:10.19835/j.issn.1008-

4495.2023.01.015.

Zhang, Y. 2022. Selection of pioneer plants for repairing

limestone high and steep slopes in North China. The

Chinese Journal of Geological Hazard and Control, 33(05):

109-118. DOI:10.16031/j.cnki.issn.1003-8035.202110012.

Xu, L. 2023. Carbon Storage Change in Xishuangbanna

Based on PLUS and InVEST Model. Ecology and

Environmental Monitoring of Three Gorges, 8(02): 75-

87. DOI:10.19478/j.cnki.2096-2347.2023.02.10.

Li, X. R. 2022. Study on the development situation and

countermeasures of non-grain utilization of cultivated

land in China. Natural Resources Information, (08): 29-

35. DOI:10.3969/j.issn.1674-3709.2022.08.005.

ICESCE 2024 - The International Conference on Environmental Science and Civil Engineering