Journal of Global Change Data & Discovery2021.5(2):148-155

[PDF] [DATASET]

Citation:Su, Q. X., Li, Z. Q.Coastline datasets cover the Guangdong-Hong Kong- Macao, Tokyo and San Francisco Bays (1980–2020)[J]. Journal of Global Change Data & Discovery,2021.5(2):148-155 .DOI: 10.3974/geodp.2021.02.06 .

Coastline Datasets Cover the Guangdong-Hong Kong-Macao, Tokyo and San Francisco Bays (1980–2020)

Su, Q. X.  Li, Z. Q. *

Department of Ocean Technology, School of Electronic and Information Engineering, Guangdong Ocean

University, Zhanjiang 524088, China

 

Abstract: The coastline datasets of the Guangdong-Hong Kong-Macao, Tokyo, and San Francisco Bays were developed using Landsat remote sensing images from 1980 to 2020, and high-resolution Google Earth images. The coastlines were defined by the mean high-water line, and were divided into two categories, natural and artificial, and seven sub-categories. Furthermore, the intensity of coastline length change, type structure change, and utilization index were calculated. Subsequently, the total coastline lengths of the Guangdong-Hong Kong-Macao Bay Area, Tokyo Bay, and San Francisco Bay in 2020 were 2,243.17, 580.68, and 689.11 km, respectively, while the average annual change intensities of the coastline length from 1980 to 2020 were 0.22%, 0.37%, and 0.09%, respectively; furthermore, the proportion of artificial coastlines in 2020 were 57.62%, 95.87%, and 55.36%, respectively, while the coastline utilization indexes were 55.20%, 83.22%, and 57.70%, respectively. The datasets included data of the three Bay Areas for the following: (1) spatial distribution of coastlines and their types (.shp, .kmz), (2) coastline type structure (.xlsx), (3) coastline length change intensity (.xlsx), and (4) coastline utilization index (.xlsx).The datasets were archived in the .shp, .kmz and .xlsx format, and consisted of 192 data files with a size of 146 MB (compressed to 69 MB in three data files).

Keywords: bay area; coastline; intensity of coastline length change; coastline type; utilization index

DOI: https://doi.org/10.3974/geodp.2021.02.06

CSTR: https://cstr.escience.org.cn/CSTR:20146.14.2021.02.06

Dataset Availability Statement:

The dataset supporting this paper was published and is accessible through the Digital Journal of Global Change Data Repository at: https://doi.org/10.3974/geodb.2021.04.07.V1 or https://cstr.escience.org.cn/CSTR:20146.11.2021.04.07.V1;

https://doi.org/10.3974/geodb.2021.04.08.V1 or https://cstr.escience.org.cn/CSTR:20146.11.2021.04.08.V1;

https://doi.org/10.3974/geodb.2021.04.09.V1 or https://cstr.escience.org.cn/CSTR:20146.11.2021.04.09.V1.

1 Introduction

Due to global warming, the global mean sea level has been increasing continuously, thereby posing serious challenges to the survival and development of human civilization. Due to the high degree of economic and social development, the Bay Area has gained tremendous significance on the international socio-economic map. Its coastal zone represents the most active site, which witnesses intense human activity, however, it is severely affected by sea level rise, thus, indicating an evident impact of climate change and human activities on coastal resources[1,2]. Therefore, extracting information on the spatial distribution, change intensity, and utilization index of the Bay Area coastlines on a large temporal scale can assist in assessing the climate change impacts on the coastal zone, protecting and restoring regional coastlines, and providing a database for effective management and planning of the coastal environment.

This study aimed to extract the location and length of coastlines of the Guangdong-Hong Kong-Macao, Tokyo and San Francisco bays from Landsat remote sensing images and high-resolution Google Earth images for seven periods (1980, 1990, 2000, 2005, 2010, 2015, and 2020). Furthermore, the spatiotemporal distributions of the coastlines in the three Bay Areas were acquired after calculating the intensity of coastline length change, coastline structure change, and utilization index.

2 Metadata of the Dataset

The metadata for the three datasets, including Coastline types and their spatiotemporal variations in Guangdong, Hong Kong and Macao Bay Area (1979–2020)[3], Coastline types and their spatiotemporal variations in Tokyo Bay (1980–2020)[4] and Coastline types and their spatiotemporal variations in San Francisco Bay (1980–2020)[5] is summarized in Table 1.

3 Methods

Coastlines have multiple definitions and its selection significantly influences the research results. This study used mean high tide line to define coastlines[7]. The Landsat multi-spectral images (1980–2020) downloaded from the United States Geological Survey (http://glovis. usgs.gov/) were used as data sources (Table 2). Images with less than 5% cloud coverage, and an imaging time from October to March were selected. Moreover, images with a high spatial resolution from Google Earth were used to correct the extracted coastlines.

According to the types of basic functional plans of the national coasts, i.e., natural state of the coastlines and coastlines for man-made utilization, the coastlines in the selected study areas were divided into two categories: natural coastlines and artificial coastlines. Natural coastlines were divided into bedrock, gravel, muddy, and biological coastlines, whereas artificial coastlines were divided into farmland aquaculture, port wharf, and other artificial coastlines. Furthermore, the outer edge of the surface by the sea is usually selected as the coastline position[8–10].

Based on the analysis of different reflective spectral characteristics of the ground objects near the coastlines, modified normalized difference water index (MNDWI), threshold

Table 1  Metadata summary of the Coastline datasets of the Guangdong-Hong Kong-Macao, Tokyo and San Francisco Bays (19802020)

Items

Description

Dataset full name/short name

Coastline types and their spatiotemporal variations in Guangdong, Hong Kong and Macao Bay Area (1979–2020) /GHM_Coastline_1979-2020

Coastline types and their spatiotemporal variations in Tokyo Bay (1980–2020) /TK_Coastline_1980-2020

Coastline types and their spatiotemporal variations in San Francisco Bay (1980–2020) /SF_Coastline_1980-2020

Authors

Su, Q. X. AAD-2930-2021, Department of Ocean Technology, School of Electronic and Information Engineering, Guangdong Ocean University, qianxinsu77@163.com

Li, Z. Q. 0000-0001-9139-9579, Department of Ocean Technology, School of Electronic and Information Engineering, Guangdong Ocean University, qiangzl1974@163.com

Geographical region

Guangdong-Hong Kong-Macao Bay Area, Tokyo Bay Area, and San Francisco Bay Area

Year

1980, 1990, 2000, 2005, 2010, 2015, and 2020

Temporal resolution

Five and ten years

Spatial resolution

10 m

 

Data format

.xlsx, .shp, .kmz

Data size

69 MB

Data files

The dataset comprises the spatial distribution of coastlines in the three Bay Areas as follows: coastline type structure, intensity of coastline length change, and coastline utilization index

Foundations

National Natural Science Foundation of China (41676079) and Innovation Fund of Guangdong Ocean University (Q18307)

Data publisher

Global Change Research Data Publishing & Repository, http://www.geodoi.ac.cn

Address

No. 11A, Datun Road, Chaoyang District, Beijing 100101, China

Data sharing policy

 

Data from the Global Change Research Data Publishing & Repository includes metadata, datasets (in the Digital Journal of Global Change Data Repository), and publications (in the Journal of Global Change Data & Discovery). Data sharing policy includes: (1) Data are openly available and can be free downloaded via the Internet; (2) End users are encouraged to use Data subject to citation; (3) Users, who are by definition also value-added service providers, are welcome to redistribute Data subject to written permission from the GCdataPR Editorial Office and the issuance of a Data redistribution license; and (4) If Data are used to compile new datasets, the ??ten per cent principal?? should be followed such that Data records utilized should not surpass 10% of the new dataset contents, while sources should be clearly noted in suitable places in the new dataset[6]

Communication and searchable system

DOI, CSTR, Crossref, DCI, CSCD, CNKI, SciEngine, WDS/ISC, GEOSS

 

Table 2  List of Landsat series remote sensing image data used for data R&D

Location

Time (yyyy–mm–dd)

Location

Time (yyyy–mm–dd)

Location

Time (yyyy–mm–dd)

Sensor

Guangdong-Hong Kong-Macao Bay Area

1979–10–01

Tokyo Bay

1980–11–11

San Francisco Bay

1980–11–14

MSS (80 m)

1990–10–13

1990–12–07

1990–10–10

TM (30 m)

2000–01–27

2000–11–24

2000–12–16

2005–11–23

2005–03–19

2005–11–20

ETM (30 m)

2010–10–28

2010–01–20

2010–10–25

2015–10–18

2015–10–25

2015–12–26

2020–02–18

2020–02–09

2020–10–12

OLI (30 m)

 

segmentation method, mathematical morphological method, and Sobel operator method were comprehensively adopted to extract the coastline data for each period; additionally, the position and type of the required coastlines were modified based on visual interpretation[11].

3.1 Algorithm Principle

Coastline length change intensity is a measure of the difference in the coastline length over time, and is calculated as follows[12].

                                                                                         (1)

where, is the length change intensity of a particular coastline type from periods  to , and  and indicate the lengths of the coastline during  and .

The changes in the coastline type structure indicate a proportional relationship of various coastlines over a period of time, and is calculated as[12].

                                                                                                                 (2)

where,  indicates the proportion of the length of a particular coastline type to the total coastline length,  is the length of the coastline type, and n is the total number of coastline types. The evaluation method using the principal degree of coastline development and utilization of China as proposed by Xiao[13] was used to quantify the coastline structures in the study area. Subsequently, the coastlines of most areas were in the form of a single main body, or a binary or ternary structure. When the proportion of a particular coastline was more than 45%, it represented a single main structure. Furthermore, if the proportion of all the coastlines was less than 45%, and two or more coastlines exhibited a proportion of more than 20%, they represented a binary or ternary structure.

Based on the impact of various human activities on coastlines, the grading coefficient is determined by using the expert scoring method, and the coastline utilization index is calculated as follows[12] :

                                                                                                               (3)

where, S indicates the coastline utilization index of the area in a particular year and  is the grading coefficient for various coastlines based on the grading coefficient of the coastline types in the Pearl River Estuary region given by Liu et al.[12]. According to the characteristics and development utilization of the coastline type, the coastlines were divided into four levels, among which the natural coastlines were the least affected by humans; correspondingly, the grading coefficient was the lowest (0.25). Conversely, the port wharf coastline and other artificial coastlines were strongly affected by humans and could not be recovered; subsequently, their grading coefficients were 0.85 and 0.9, respectively. Among these, since the composition of other artificial coastlines was highly complex. However, although the farmland aquaculture coastline was artificial, the development utilization was relatively low; thus, the grading coefficient was 0.5.

3.2 Technical Route

The Landsat remote sensing images of the three Bay Areas were collected and pre-processed, which included geometric correction and registration, band synthesis, image mosaic, and cropping of study areas (Figure 1). Subsequently, the classification system and interpretation signs of the coastlines in the study area were established, and the spatial distribution of the Bay Area coastlines was extracted using MNDWI, and gray threshold, mathematical morphology, and Sobel operator methods. Furthermore, the accuracy of the coastline results was verified on Google Earth. Later, the intensity, type structure, and utilization index of the coastline length changes in the three Bay Areas were calculated using the respective formulae. Additionally, the degree of coastline changes in the three Bay Areas during 1980–2020 was quantitatively analyzed.

 

Figure 1  Technology roadmap of the dataset development

4 Data Results

4.1 Data Composition

The acquired datasets of the spatial distribution map of the three Bay Area coastlines mainly consisted of vector files (.shp and .kmz formats), while the results of the changes in the coastline length change intensity, type structure change, and utilization index calculated using the relevant formulae were provided as a table file (.xlsx). The vector files of the spatial distribution maps of the three Bay Area coastlines in 1980, 1990, 2000, 2005, 2010, 2015, and 2020 were derived from ArcGIS10.6. The data file acronyms are given in Table 3. Results from the .shp data calculations and the corresponding combination charts were conducted in Microsoft Excel. Further, the datasets were compressed to 69 MB.

Table 3  Acronym comparison table

Location

Acronym

Guangdong-Hong Kong-Macao Greater Bay Area

GHM

Tokyo Bay

TK

San Francisco Bay

SF

4.2 Data Results

The Landsat remote sensing images of the coastlines of the seven periods from 1980 to 2020 were interpretated to acquire the distribution of the coastline types in the three Bay Areas (Figure 2). Moreover, the length percentages of different types of coastlines at different times on the Bay Area Coast Zone scale were measured, and the coastline type structure was analyzed (Figure 3). Subsequently, significant inter-annual and spatial differences were observed in coastline types of the three Bay Areas. Furthermore, the proportion of gravel, bedrock, and muddy coastlines in the total coastline length in the three Bay Areas decreased continuously from 1980 to 2020, while the proportion of artificial coastlines (farmland aquaculture, port wharf, and other artificial) increased continuously. Additionally, the biological coastline fluctuations in the Guangdong-Hong Kong-Macao Bay Area increased. In 2020, the proportion of artificial coastlines in the three Bay Areas was over half, reaching 57.62%, 95.87%, and 55.36%, respectively.

The intensity of coastline changes in the seven periods in the three Bay Areas was calculated using Equation (1) (Figure 4). Subsequently, the intensity of coastline length change in the three Bay Areas fluctuated at each stage, with the largest change observed in the Tokyo Bay, which was followed by the Guangdong-Hong Kong-Macao Bay Area. The San Francisco Bay Area was the most stable and exhibited the least fluctuations. The average annual coastline length changes of the Guangdong-Hong Kong-Macao Bay Area, Tokyo Bay, and San Francisco Bay during 1980–2020 were 0.22%, 0.37%, and 0.09%, respectively.

 

 

Figure 2  Spatial distribution of different types of coastlines in the three Bay Areas from 1980 to 2020

 

Figure 3  Coastline types and their proportions in the three Bay Areas from 1980 to 2020

Figure 4  Intensity of coastline length changes in the three Bay Areas from 1980 to 2020

 

Figure 5  Coastline utilization index of the three Bay Areas from 1980 to 2020

Furthermore, Equation (2) was used to divide the main types of the three Bay Area coastlines (Figure 3) to reflect the changes in the coastline complexity in each Bay Area. Subsequently, the Guangdong-Hong Kong-Macao Bay Area exhibited the most complex composition and changes in the coastline types, which ranged from a binary structure comprising bedrock and farmland aquaculture coastlines in 1979 to a ternary structure comprising bedrock, farmland aqua-culture, and port wharf coastlines in 2005, and further to a binary structure comprising bedrock and port wharf coastlines in 2020. Contrastingly, the coastline type of the San Francisco Bay exhibited a muddy and port wharf coastline duality structure throughout the study period, while the coastline of the Tokyo Bay was a single main structure of the port wharf coastline.

To investigate the impact of various human activities on the coastlines, the coastline utilization index was calculated according to Equation (3) and the grading coefficient. Figure 5 indicates that the coastline utilization index of the Guangdong-Hong Kong-Macao Bay Area, Tokyo Bay Area, and San Francisco Bay Area increased from 1980 to 2020 by 7.09%, 1.81%, and 0.63%, respectively. Thus, the Guangdong-Hong Kong-Macao Bay Area changed the most, with a notable ??steep slope?? observed between 2005 and 2010. The order of the utilization indexes of the three Bay Areas was Tokyo Bay (83.22%) > San Francisco Bay (57.70%) > Guangdong-Hong Kong-Macao Bay Area (55.20%).

 

5 Conclusions

Using the Landsat remote sensing and Google Earth images, the coastlines of the three Bay Areas were extracted for 1980, 1990, 2000, 2005, 2010, 2015, and 2020 at a spatial resolution of 10 m. The variations in the length, type, and utilization of each Bay Area coastline were compared and analyzed. Consequently, significant inter-annual and spatial differences were observed between the three Bay Area coastline types from 1980 to 2020. Furthermore, the average length change intensities of the coastlines in the Guangdong-Hong Kong-Macao Bay Area, Tokyo Bay, and San Francisco Bay were 0.22%, 0.37%, and 0.09%, respectively. The composition and changes in the coastlines of the Guangdong-Hong Kong-Macao Bay Area were the most complex. The proportion of the farmland aquaculture coastline declined in 2020 and was converted to a binary structure of bedrock and port wharf coastlines. Conversely, the Tokyo Bay coastline remained as a single main structure composed of port wharf coastlines, while the San Francisco Bay coastline exhibited a binary structure comprising muddy and port wharf coastlines. In 2020, the proportion of artificial coastlines of the Guangdong-Hong Kong-Macao Bay Area, Tokyo Bay, and San Francisco Bay was over half, at 57.62%, 95.87%, and 55.36%, respectively, and the utilization indexes were 55.20%, 83.22%, and 57.70%, respectively. During 1980–2020, the Guangdong-Hong Kong-Macao Bay Area changed the most at 7.09%.

Existing coastline datasets are concentrated in islands and areas with intensive human activities, and the Bay Area, as a representative of global economy, includes detailed information regarding coastline changes. Thus, the Bay Area coastline dataset plays an important role in studying the coastline changes in the Pacific Coastal Bay Area and analyzing the changes of regional landscape patterns and the degree of coastline development and utilization. Therefore, the findings serve as a strong reference for rational exploitation of coastal resources in the Bay Area and promotion of its harmonious development with human civilization.

 

Author Contributions

Li, Z. Q. designed the overall plan and technical framework of this dataset. Su, Q. X. contributed to the data processing, verification and analysis, and wrote the data paper.

 

Conflicts of Interest

The authors declare no conflicts of interest.

References

[1]    Gao, S. Regional differences in coastal zones and classification of Marine economic potential: a land-sea perspective [J]. Democracy & Science, 2020(1): 29–31.

[2]    Zhang, J. J., Su, Z. F., Zuo, X. L., et al. Research on the spatial differentiation of coastal land development surrounding South China Sea [J]. Acta Geographica Sinica, 2015, 70(2): 319–332.

[3]    Su, Q. X., Li, Z. Q. Coastline types and their spatiotemporal variations in Guangdong-Hong Kong-Macao Bay Area (1979–2020) [J/DB/OL]. Digital Journal of Global Change Data Repository, 2021. https://doi.org/10.3974/geodb.2021.04.07.V1. https://cstr.escience.org.cn/CSTR:20146.11.2021.04.07.V1.

[4]    Su, Q. X., Li, Z. Q. Coastline types and their spatiotemporal variations in Tokyo Bay (1980–2020) [J/DB/OL]. Digital Journal of Global Change Data Repository, 2021. https://doi.org/10.3974/geodb.2021.04.08.V1. https://cstr.escience.org.cn/CSTR:20146.11.2021.04.08.V1.

[5]    Su, Q. X., Li, Z. Q. Coastline types and their spatiotemporal variations in San Francisco Bay (1980–2020) [J/DB/OL]. Digital Journal of Global Change Data Repository, 2021. https://doi.org/10.3974/geodb.2021.04.09.V1. https://cstr.escience.org.cn/CSTR:20146.11.2021.04.09.V1.

[6]    GCdataPR Editorial Office. GCdataPR data sharing policy [OL]. https://doi.org/10.3974/dp.policy.2014.05 (Updated 2017).

[7]    Hou, X. Y., Wu, T., Wang, Y. D., et al. Extraction and accuracy evaluation of multi-temporal coastlines of mainland China since 1940s [J]. Marine Sciences, 2014, 38(11): 66–73. DOI: 10.11759/hykx 20131217001.

[8]    Zhao, Y. L. Remote sensing survey and proposal for protection of the natural resources in Guangdong-Hong Kong-Macao, Greater Bay Area [J]. Remote Sensing for Land & Resources, 2018, 30(4): 139–147.

[9]    Liu, C., Shi, R. X., Zhang, Y. H., et al. Global multi-scale shoreline dataset of land and sea based on Google Earth remote sensing image (2015) [J/DB/OL]. Digital Journal of Global Change Data Repository, 2019. https://doi.org/10.3974/geodb.2019.04.13.V1. https://cstr.escience.org.cn/CSTR:20146.11.2019.04.13.V1.

[10]  Chen, B. Q., Xiao, X. M., Li, X. P., et al. Spatial distribution data of mangroves in China in 2015 [J/DB/OL]. Digital Journal of Global Change Data Repository, 2017. https://doi.org/10.3974/geodb. 2017.03.06.V1.

[11]  Gao, Z. Q., Liu, X. Y., Ning, J. C., et al. Analysis on changes in coastline and reclamation area and its causes based on 30-year satellite data in China [J]. Transactions of the Chinese Society of Agricultural Engineering, 2014, 30(12): 140–147.

[12]  Liu, X. L., Deng, R. R., Xu, J. H., et al. Spatiotemporal evolution characteristics of coastlines and driving force analysis of the Pearl River estuary in the past 40 years [J]. Journal of Geo-information Science, 2017, 19(10): 1336–1345.

[13]  Xiao, R. Analysis of change and driving force of the coastline of mainland in nearly 35 years [D]. Shanghai: East China Normal University, 2017.

Co-Sponsors
Superintend