Skip to main content
NCBI home page
Scientific Reports logoLink to Scientific Reports
. 2024 Sep 28;14:22489. doi: 10.1038/s41598-024-73287-x

Accelerating growth of human coastal populations at the global and continent levels: 2000–2018

A G Cosby 1, V Lebakula 2,✉, C N Smith 3, D W Wanik 4, K Bergene 1,5, A N Rose 2, D Swanson 6,7, D E Bloom 8
PMCID: PMC11438952  PMID: 39341937

Abstract

Current human population growth along Earth’s coasts is on a collision path with anticipated consequences of increasing natural and anthropogenic induced coastal hazards. Using recently-available ambient, dasymetric data, we developed methods to estimate annual continental and global coastal populations from (2000–2018) measured horizontally from the shoreline inward. We found: (1) large concentrations of population in relatively small bands and regions along the coast (~ 2 billion within 50 km and ~ 1 billion within 10 km); (2) higher growth rates of coastal population than inland population (an addition of 463 million within 50 km and 233 million within 10 km); (3) strong influence of distance from the coast to predict population distribution; and (4) that macro population patterns and growth could be expressed and modeled as a power function at continental and global levels. Findings point to emerging macro population patterns along the coast as contributing to increasing anthropogenic effects on Earth systems and increasing human risks associated with sea-level rise, land subsidence, extreme weather, and public health. Reliable data tracking of the magnitude, spatial distribution and change of human populations in the coastal regions is essential for comprehensive coastal monitoring.

Subject terms: Population dynamics, Environmental impact

Introduction

The world’s population surpassed 8 billion in 2023, an increase of 1 billion in just over one decade. This benchmark signals the profound importance of the anthropogenic effects of an increasing human population on the global environment and resulting risks entailed for Earth and its human inhabitants. Among factors shaping anthropogenic impacts is the spatial distribution of population, especially concentrations in sensitive locations such as coastal regions. Here we applied a novel taxonomy and methodology with recently-available LandScan Global population distribution data to estimate and characterize high-resolution ambient coastal population through the early years of this century.

Natural aspects (e.g., physical, biological, and chemical) of coastal environments have substantial consequences for humans1,2. In turn, human population patterns have significant anthropogenic impacts on coastal ecosystems and ultimately Earth systems3–10. The Earth systems perspective emphasizes the integration and interconnection of major subsystems such as the atmosphere, biosphere, hydrosphere, and geosphere. As such, it is a holistic view of the Earth processes and interdependence between these quite different subsystems. The dramatic growth of human population and its associated impacts are seen as having potentially profound consequence on the stability of the Earth. Anthropogenic impacts of coastal populations are broad-based and include “urbanization, agriculture, nutrient runoff, engineering works, fisheries, oil and gas production, dredging, and various forms of pollution”11,12. Higher levels of economic activity characteristic of many coastal regions can intensify anthropogenic effects13–15. Population growth not only influences the level of hazards in coastal regions but also exposes more humans to increasing risk10,16–20. For example, the health risks from the array of waterborne and mosquito borne pathogens are tied to anthropogenic-induced consequences such as seawater rise and flooding and in turn health consequences for human population concentrations along coasts7,8,21. Hence, the intersection of human population growth with sensitive environments is highly problematic. From this perspective, reliable and current estimates of populations magnitude, spatial distribution, and change are fundamental components of Earth and coastal systems22.

Given significance of coastal population patterns for Earth’s systems, it is surprising that current macro-coastal data is limited. Earlier literature is replete with inconsistent estimates of coastal populations. Cohen and colleagues made this point in a letter to Science in 1997 and provided rigorous estimates of coastal populations at 100 km, 150 km, and 200 km from shoreline23. Using the Gridded Population of the World dataset (GPW2), Small and Cohen comprehensively analyzed human settlement patterns at global levels24,25. This pioneering research resulted in estimates of coastal population for 1994; approximately 37% (2.07 billion) of the globe’s population lived within 100 km of the coast and 44% (2.45 billion) within 150 km23,25,26. Leveraging digital global population data, they were the first to calculate coastal estimates in both distance from the shore and elevation. Their work provided the best knowledge of global coastal population magnitude; settlement patterns; and impacts of elevation, climatic conditions, and distances from shore. Small and Nicholls’ estimates were derived utilizing data from the early to mid-1990s3,23,24,27.

Our research builds on Cohen and colleagues work23,27. First, we estimate global and continental horizontal population patterns in near-coastal regions (50 km from shore) from 2000 through 2018 (See extensive Supplemental Tables 1-8). Second, we model distributions of near-coastal population “buildup” as a function of distance-from-shore. Third, we estimate rates of near-coastal population growth in 5 km bands from 2000 through 2018 for the globe and continents. Fourth, we develop and report a set of cross-sectional and temporal metrics to assist characterizing macro-coastal population patterns.

Our approach generated estimates from 2000 to 2018 utilizing Oak Ridge National Laboratory’s (ORNL) LandScan Global data28,29. LandScan Global exploits spatial data, machine learning, and imagery analysis technologies in a multivariable dasymetric modeling approach to spatially disaggregate subnational census or other enumeration information through several novel and dynamically adaptable algorithms27,29.

Results

Cohen et al. established the 1994 global coastal population as 2.07 billion (37% of global population) within 100 km and 2.45 billion (44%) within 150 km of shoreline23. These are considered the standard for coastal population. We updated these estimates for 2018: the coastal regions had grown to 2.86 billion (38.1% of global population) at 100 km and 3.34 billion (44.6%) at 150 km.

High-resolution population estimates provided from LandScan and our developed coastal regional/band shape files allowed us to focus on population patterns within near-coastal (≤ 50 km) and ultra-coastal (≤ 10 km) regions. We estimated that 1.72 billion people were within 50 km of shoreline in 2000. Over the next 18 years, coastal populations increased over 2 billion, accounting for more than 29% of global population (see Table 1). In context, in the 1930s the world’s population was 2.07 billion30. Current near-coastal population exceeds the world population that existed less than a century ago. Remarkably, this 50 km coastal region has 29% of the global population inhabiting 11.1% of landmass. Population buildup is more concentrated in ultra-coastal regions. In 2018, 1.09 billion were within a narrow 10 km band along the shore—a number approximating one half 1930’s global population. Ultra-coastal population encompasses 14.6% of the global population on 4% of the landmass. Previously, less granular estimates made at 100 km, 150 km, or greater do not capture near- and ultra-coastal population patterns. This is evident in patterns of 50 km coastal population buildup.

Table 1.

World coastal population estimates in millions: 2000–2018.

Year Percentage of world population Increase in population Percent increase in population
2000 2003 2006 2009 2012 2015 2018 2018 2000–2018 2000–2018
Total population 6086.6 6303.0 6529.3 6773.1 7014.6 7284.3 7502.3 100.0% 1,415.7 23.3%
Coastal region (50 km) 1742.8 1849.3 1901.9 1970.6 2061.0 2143.0 2206.1 29.4% 463.3 26.6%
Inland population 4343.8 4453.7 4627.4 4802.5 4953.5 5141.2 5296.2 70.6% 952.4 21.9%
Coastal region
 5 km 586.8 629.2 650.2 672.7 702.9 729.1 748.1 10.0% 161.3 27.5%
 10 km 857.8 917.1 943.6 976.1 1020.8 1062.1 1091.6 14.6% 233.8 27.3%
 15 km 1039.6 1109.3 1140.2 1181.7 1237.8 1289.8 1325.1 17.7% 285.5 27.5%
 20 km 1190.4 1267.0 1301.4 1349.2 1413.5 1474.1 1515.8 20.2% 325.4 27.3%
 25 km 1311.5 1393.5 1432.0 1485.2 1555.8 1622.0 1670.0 22.3% 358.5 27.3%
 30 km 1409.9 1496.8 1538.7 1595.8 1672.6 1743.0 1795.5 23.9% 385.6 27.4%
 35 km 1501.1 1592.6 1636.8 1697.8 1780.0 1853.6 1910.4 25.5% 409.3 27.3%
 40 km 1587.9 1683.9 1730.2 1794.3 1880.6 1957.0 2016.2 26.9% 428.3 27.0%
 45 km 1664.9 1766.0 1815.6 1881.8 1970.7 2050.3 2111.4 28.1% 446.4 26.8%
 50 km 1742.8 1849.3 1901.9 1970.6 2061.0 2143.0 2206.1 29.4% 463.3 26.6%
Coastal band
 0–5 km 586.8 629.2 650.2 672.7 702.9 729.1 748.1 10.0% 161.3 27.5%
  > 5–10 km 271.0 287.9 293.4 303.4 317.9 333.0 343.5 4.6% 72.5 26.8%
  > 10–15 km 181.8 192.2 196.6 205.6 217.1 227.6 233.5 3.1% 51.7 28.4%
  > 15–20 km 150.8 157.8 161.2 167.5 175.7 184.3 190.7 2.5% 40.0 26.5%
  > 20–25 km 121.1 126.5 130.6 136.0 142.4 147.9 154.2 2.1% 33.0 27.3%
  > 25–30 km 98.4 103.3 106.7 110.7 116.7 120.6 125.5 1.7% 27.1 27.6%
  > 30–35 km 91.2 95.9 98.2 102.0 107.4 111.0 115.0 1.5% 23.7 26.0%
  > 35–40 km 86.8 91.3 93.4 96.4 100.6 103.5 105.8 1.4% 19.0 21.9%
  > 40–45 km 77.0 82.0 85.4 87.6 90.1 93.3 95.1 1.3% 18.1 23.5%
  > 45–50 km 77.9 83.3 86.3 88.8 90.4 92.7 94.8 1.3% 16.9 21.7%
Open in a new tab

The coastal population curve

Previous studies refer to human population buildup along the coast; however, findings reporting the form and magnitude of coastal buildup are lacking. While Small and Nicholls24 report a concentration of coastal people, we expressly quantified this buildup at global and continental levels yearly. Using these data, we identified the buildup patterns as a power function of form y = axb, where y is population of coastal bands, x is distance from shoreline, and a and b are real numbers. Bands closest to shoreline had the largest population concentrations, and population systematically decreased in subsequent bands farther from the shoreline. We refer to this pattern as the Coastal Population, or CoPop, Curve (Fig. 1). We estimated 748.1 million people live within the first 5 km from shoreline; this constituted approximately 10% of the world’s population. The next 5 km band (i.e., > 5–10 km) contained 343.5 million people and accounted for an additional 4.6% of Earth’s population. Subsequent bands moving inward declined gradually to 1.3% of the Earth’s population in the > 45–50 km band. The CoPop Curve for 2018 was expressed as power function y = 7E + 08x−0.9 with a goodness of fit, where r2 = 0.99.

Fig. 1.

Fig. 1

Open in a new tab

Global coastal population curve (CoPop curve) expressed as population estimates in 5 km bands: 2018.

Exponent estimates for our global coastal population range from − 0.80 to − 0.90 with goodness of fit ranging from 0.98 to 0.99. Whereas, for continents exponent estimates varied from − 0.82 to − 1.03 with goodness of fit ranging from 0.84 to 0.99. Previous research by Small and others have utilized power functions to study population patterns including urban centers and networks in coastal regions, population as a function of elevation, anthropogenic impacts using nighttime lights as an index31–33. Their research resulted in power function exponents of approximately − 1. In contrast, our estimates for coastal band data are more variable and concentrated around − 0.90. This difference from previous research can be due to several factors that include the scale of data (coastal bands of continents and of the globe are at different scales than cities), foundational data sources (use of LandScan or Nighttime lights and other gridded population datasets), and possibly more disperse population patterns that occur in large scale coastal bands. Also, our estimation procedures are based on horizontal measurements rather than elevation from sea level which logically produces a different set of population estimates and possibly different exponent values. Additionally, it is interesting that the power exponential varied among continents and this pattern of variation tends to exist across the 19-year time series.

The shape of the CoPop Curve, with the largest population concentration near the shoreline and gradually smaller population bands moving inland, was a recurring pattern. We found similar power function fits for the coastal band data for each year of the global coastal population data from 2000 through 2018. Typically, the power function provided the best fit with strong r2 values. Furthermore, the CoPop Curves and power functions consistently modeled the coastal buildup patterns for continents throughout the time series (see Fig. 2). The consistency of CoPop Curves, across temporal and spatial scales (i.e., global and continental), indicates the existence of a strong, coastal pattern of human settlement that can be modeled at macro levels and should prove valuable in estimating anthropogenic impacts on Earth systems and risk to humans.

Fig. 2.

Fig. 2

Open in a new tab

Coastal population curve (CoPop curve) for major continents expressed as percentage distribution of near-coastal population for 5 km bands: 2018.

Global coastal population growth patterns

We found that while the total global population was growing substantially at the start of the twenty-first century, the coastal population was growing even faster, and that growth was most evident in the bands closest to the shoreline. Figure 3 provides estimates of the population increase of coastal and inland regions between 2000 and 2018. Coastal regions increased by 26.6%, resulting in an additional 463.33 million people living in the near-coastal region. This new coastal population is equivalent to 46 additional “megacities” (i.e., population of 10 million or more) suddenly appearing within the 50 km coastal region, with 15 of these megacities within the first 5 km of the shoreline. Regions nearest the shoreline were growing faster; 50% of the new near-coastal population (233.8 million) occurred within the ultra-coastal (< 10 km) region. In assessing the environmental impact of an increase of 463 million new coastal inhabitants, it is critical to appreciate the massive amount of new infrastructure needed to support such a large population increase in a limited geographic area.

Fig. 3.

Fig. 3

Open in a new tab

World coastal population: 2000–2018.

Global distribution of near- and ultra-coastal populations

We found both significant similarities and differences in coastal patterns among continents. In Table 2, coastal population estimates are provided for each continent with data for 2000 and 2018; also, Table 2 highlights several differences in key population metrics. The metrics for each continent include (1) near-coastal, or 50 km, population for the most recently available year (2018); (2) ultra-coastal, or 10 km, population for 2018; (3) near-coastal growth rate; (4) ultra-coastal growth rate; (5) number of new coastal residents added between 2000 and 2018; (6) ratio of coastal growth rates to inland growth rates; and (7) coastal prevalence, or percentage of total population that lives in near- or ultra-coastal regions. Detailed coastal estimates for major continents for this time period are provided in Supplemental Tables 1–8.

Table 2.

Coastal population profiles for the five most populated continents derived from landscan datasets: 2000–2018.

Africa Asia Europe North America South America
Near-coastal population within 50 km (2018) Africa had the second largest near-coastal population with 256.2 million coastal inhabitants. This number represents 11.6% of the global coastal population Asia had the largest near-coastal population (1,317.5 million) or 59.7% of the global coastal population Europe had the third largest coastal population with 232.3 million or 10.5% of the global coastal population North America had a relatively small coastal population with 206.9 million living within 50 km of the coast, or 9.4% of the globe’s coastal population South America has the smallest coastal population with 160.3 million living within 50 km and accounting for 7.3% of the global coastal population
Ultra-coastal population within 10 km (2018) Africa’s ultra-coastal population was substantial with 144.4 million located within the first 10 km Asia’s ultra-coastal population was the largest with 620.9 million people living within the first 10 km Europe’s ultra-coastal population also ranked third with 118.6 million North America’s ultra-coastal population was 101.4 million South America’s ultra-coastal population was 82.8 million
Near-coastal growth rate (50 km) in 2000–2018 Africa’s near-coastal growth rate was substantially higher than the other continents with an 18-year growth rate of 60.5% Asia’s near-coastal growth rate in 2000–2018 was a mid-ranged 25.6% Europe’s near-coastal region grew substantially more slowly than other continents with an 18-year growth rate of 9.3% North America’s near-coastal growth rate was relatively slow at 18.0% in 2000–2018; only Europe had a slower rate of growth South America’s growth rate in 2000–2018 was 30.5% for the near-coastal region
Ultra-coastal growth rate (10 km) in 2000–2018 The ultra-coastal growth rate increased even faster than the near-coastal rate with an 18-year 66.6% increase The ultra-coastal growth rate was a mid-ranged estimate at 25.1% Europe’s ultra-coastal region grew even slower at 7.3% North America’s ultra-coastal growth rate was a modest 17.4% South America had a rapidly growing ultra-coastal region with a 35.2% within 10 km between 2000 – 2018
Near-coastal population (50 km) added in 2000–2018 Africa added 96.6 million new near-coastal inhabitants Asia added 268.2 million new near-coastal inhabitants Europe added only 19.8 million new near-coastal inhabitants North America added 31.5 million new near-coastal inhabitants South America added 37.5 million new near-coastal inhabitants
Ultra-coastal population (10 km) added in 2000–2018 Africa added 57.7 million new ultra-coastal inhabitants Asia added 124.7 million new ultra-coastal inhabitants Europe added 8.1 million new ultra-coastal inhabitants North America added 15.0 million new ultra-coastal inhabitants South America added 21.5 million new ultra-coastal inhabitants
Ratio of coastal growth rate/inland growth rate (2018) Africa’s overall continent population increased rapidly in both coastal and inland areas. The coast increased at 1.1 times the inland population The coastal areas of Asia grew 1.4 times faster than the inland rates Europe’s inland population 18-year growth rate was slightly negative (approximately zero). Hence, the ratio is undefined North America was the only continent whose inland population grew faster than its coastal population producing a 0.9 ratio South America had the largest shift toward a coastal population with its near-coastal population growing 1.8 times faster than its inland population
Coastal prevalence (2018) Africa had the smallest percentage of coastal population with 20.1% in the near-coastal region and 11.3% in the ultra-coastal region Asia had a mid-range coastal prevalence for both the near-coastal (29.4%) and ultra-coastal (13.8%) regions Europe had a moderately high coastal prevalence with 32.9% of its total population living within 50 km and 16.8% living within 10 km North America had a relatively high coastal prevalence with 35.9% of its residents living within 50 km of the coast and 17.6% living within 10 km of the coast South America now has the largest percentage of its population living in the near-coastal region (38.0% within 50 km) and the ultra-coastal region (19.6% within 10 km)
CoPop curve equation (2018) y = 9E + 07x−1.03 y = 4E + 08x−0.875 y = 7E + 07x−0.914 y = 6E + 07x−0.841 y = 4E + 07x−0.851
r2 = 0.9676 r2 = 0.9947 r2 = 0.9736 r2 = 0.9665 r2 = 0.8677
Open in a new tab

Near-coastal population metrics for 2018

Although total coastal populations differ among continents, all continents had large, substantial coastal populations. Asia contains four of the top five largest populated countries: China, India, Indonesia, and Pakistan29. Asia also contains the largest near-coastal population (1.318 billion within the 50 km region in 2018), 59.7% of global near-coastal population. Remaining continents shared roughly 40% of the globe’s near-coastal population: Africa (11.6% or 256.2 million), Europe (10.5% or 232.4 million), North America (9.4% or 206.9 million), and South America (7.3% or 160.3 million) (see Table 2).

Near- and ultra-coastal growth metrics from 2000 to 2018

We found large differences in coastal growth rates among continents from 2000 to 2018. Africa had the most dynamic near-coastal settlement patterns with a growth rate of 60.5% and an even faster growing settlement pattern in the ultra-coastal region (66.6%; see Table 2). South America had a high growth rate; its near-coastal region grew at a rate of 30.5%, and its ultra-coastal at a rate of 35.2%. In contrast, Europe’s coastal patterns appeared more stable with growth of 9.3% for the near-coastal region and slower growth of 7.3% in the ultra-coastal region. Asia, with its large population, had the greatest cumulative increase. Africa had the most rapidly growing coastal population (see Fig. 4a and b). These growth patterns in near- and ultra-coastal regions depict a global acceleration of coastal populations with the greatest impact along the coasts of Asia and Africa.

Fig. 4.

Fig. 4

Open in a new tab

Cumulative continental ultra coastal (10 km) population growth in 2000–2018.

New coastal population metric

Earlier, we reported that 463.3 million new coastal inhabitants were added between 2000 and 2018 (see Table 1). The distribution among continents is a function of the coastal population base and growth rate. Asia, with its large coastal population and moderate growth rate of 25.6%, added 268.2 million near-coastal residents (Table 2). Of these, 124.7 million were in the ultra-coastal region. Africa, with the largest growth rate at 60.5%, added an additional 96.6 million near-coastal population, 57.7 million being in the ultra-coastal region. Combined new coastal populations in Asia and Africa accounted for 79% of new coastal population in the first 18 years of this century. Of the remaining 21% of the Earth’s new coastal population, South America accounted for 8%, North America 7%, Europe 4%, and Oceania (not included in Table 2) approximately 1%.

Ratio of coastal to inland growth metrics

Overall population growth in South America was strongly tied to coastal growth. Its near-coastal region grew 1.78 times faster than inland (Table 2). Asian growth patterns also heavily favored coastal regions; the near-coastal region grew 1.4 times faster than inland. Africa had a rapidly growing coast, and inland regions grew almost as fast. Hence, the ratio of Africa’s coastal to inland growth was 1.1. We were unable to calculate a ratio of coastal to inland growth for Europe because its inland growth was essentially zero and the ratio undefined. North America was the only continent in which inland regions grew faster than coastal regions, a ratio of 0.9 favoring inland growth.

Coastal prevalence metrics

The coastal prevalence metric is defined as the percentage of a continent’s population that resides either in the near-coastal or ultra-coastal regions. Using this metric, South America and North America were proportionately the most coastal continents (Table 2). South America had 38.0% of its population residing within 50 km of the coast and 19.6% of the population within 10 km of the coast. Africa and Asia had smaller proportions of their populations located in the coastal regions. Africa had 20.1% of its population living within 50 km and 11.3% within 10 km; Asia had 29.4% of its population living within 50 km and 13.8% within 10 km. Note that the continents that are adding the highest numbers of new coastal settlements (Africa and Asia) are also the continents with proportionately fewer people living in coastal regions.

Discussion

If teams of researchers visited random locations along coastal areas of the world, they would likely report on the great variety of human settlement patterns. Some would report there were no humans present, others there were small villages or farms, others sizable communities, and a few megacities. Such data could reasonably lead to the conclusion that coastal patterns of human occupancy is hardly systematic and difficult to predict. However, the macro perspective based on global gridded population estimates utilized in our research reveals a series of properties and regularities of coastal populations that would be difficult to detect in a series of site-specific studies. Among these are: (1) large concentrations of human population in relatively small geographic bands and regions along the coast; (2) higher rates of coastal population growth than inland regions; (3) a strong influence of distance from the coast to predict population distribution; and (4) that macro population patterns and growth could be expressed mathematically as a power function in the near-coastal region. With few exceptions, the foregoing coastal population regularities were found for the major continents over the 19-year study period (see Supplemental Table 9). Given that the continents have much different geographies, climates, histories, cultures, economies, and administrative systems, it is remarkable that they share these macro properties of coastal human population distributions.

High-resolution human coastal population estimates and population change patterns have important implications for many issues involving the coastal regions of the globe. Coastal regions are often areas of intense economic activity and infrastructure development13,14,34. Since the coastal population is growing at a faster rate than the inland population, we anticipate that the anthropogenic effects are more pronounced in the regions with higher population growth and higher economic activity. Hence, coastal regions and coastal growth would be associated with greater impacts on such anthropogenic associated effects such as climate change. While the general population is the driver of such effects as growing CO2 levels, sea level rise, and ocean acidification2,5,11, coastal population with its greater level of economic activity would proportionately (per capita) be more impactful than inland populations. We find that population growth patterns, and presumably the associated anthropogenic effects on the globe are becoming more concentrated in near-coastal regions.

Every metric we calculated supports the contention that coastal populations are rapidly expanding. Near- and ultra-coastal populations are large, significant segments of the entire human global population. Near- and ultra-coastal populations are becoming more concentrated in the geographic bands closer to the shoreline and are growing at faster rates than inland areas. Additionally, these new population data highlight the importance of the massive Asian coastal population and the rapidly growing African coastal population as dominant anthropogenic forces for Earth systems research.

Just as a rapidly growing coastal population impacts Earth systems, it also poses threats to the populations. Coastal regions are especially vulnerable to many consequences of climate change35 as well as existing coastal hazards. Sea-level rise, land subsidence, soil erosion, and exposure to extreme events (e.g., hurricanes, tsunamis, flooding, etc.) are well-documented risk factors for coastal ecosystems and the people who live there2,36–38. Simply put, the more humans who live in high-risk regions, the greater the risks they are exposed to. The human population growth patterns along the coast that we identified in our analysis coincide and overlap with growth in risk exposure in the near- and ultra-coastal regions. The magnitude of human presence in these coastal regions amplifies the level of risk.

The findings that a substantial portion of Earth’s human population is located near its coastlines and that this coastal population is growing at a rate exceeding inland population raises more questions than it resolves. For example, is there utility in conceptualizing the growth of coastal populations as a demographic trend similar to the growth of urban places? If so, how much of the coastal growth can be attributed to natural increase, from rural migration, from urban migration, and from urban/coastal overlap? Significantly, recent research documents that the low elevation coastal zone (LECZ) population is disproportionately more urban than rural and that urban LECZ population is growing faster than urban populations outside the zone39.

The finding that a power function describes both the pattern of coastal human presence and the growth rate of coastal population suggests that a power function could be a useful mathematical property for modeling Earth’s systems. For example, the function could be used as a coastal gravity model complementing the existing gravity model often used to estimate urban influence. The fundamental difference being that the coastal function indicates a line of influence along the coastline rather than a central point. To integrate our findings with current trajectories of coastal population research it would be useful to understand how coastal growth patterns overlap with the emergence and growth of megacities and the overlap between coastal growth and LECZs. Answers to these and a myriad of related questions about the nature of coastal growth patterns are needed to better understand the nature and extent of human risk resulting from accelerating coastal populations and the anthropogenic impacts coastal populations are having on Earth’s systems.

Methods

Our methodology generated populaiton estimates from 2000 to 2018 utilizing Oak Ridge National Laboratory’s (ORNL) LandScan Global data28,29. LandScan Global exploits spatial data, machine learning, and imagery analysis technologies in a multivariable dasymetric modeling approach to spatially disaggregate subnational census or other enumeration information through several novel and dynamically adaptable algorithms28,40. It uses subnational-level census counts for each country as a primary data source that has auxiliary data layers including “land cover, roads, slope, urban areas, village locations, and high-resolution imagery analysis”28,40. Modeling these factors produces a likelihood coefficient that then yields a spatial cell population estimate. LandScan is tailored to unique cultural and economic conditions with the resultant data representing an average, or ambient, population that integrates diurnal movements and collective travel habits into a single measure. This ensures that the population distribution represents the full scope of human activity space not simply residential accounts as are captured in censuses. LandScan is the only global gridded population dataset that provides high-resolution ambient and integer population counts rather than residential estimates. These features are considered useful for studying populations at risk40,41. Ambient population is especially relevant for coastal regions that often have nonresidential populations such as tourists, part-time residents, and workers from inland. LandScan estimates are provided in approximately 1 km × 1 km resolution and are considered the highest-resolution ambient data for global population estimates29,42. Manual ground-truthing is conducted periodically using high-resolution satellite imagery. As of May 2020, longitudinal LandScan data were made available to the scientific community at https://landscan.ornl.gov/, which made this study possible. LandScan extended worldwide access to the general population in April 2022. Utilizing mainland Southeast Asia as a testbed, LandScan was found to be superior in terms of spatial accuracy and estimated errors when compared with other world grid datasets such as Worldpop, Global Human Settlement Layer-Population, and the Gridded Population of the World43.

To estimate the change in horizontal coastal populations, we used an annual “best estimate” approach wherein we calculated coastal bands and regions via LandScan Global for the years 2000 to 2018. LandScan Global developers caution against comparing longitudinal patterns at the grid cell level because of potential estimation errors. However, we did not compute the percent change for each individual coastal cell (1square km) and then average these changes. Instead, to avoid cell-by-cell comparisons, we calculated the annual coastal population by aggregating (summing) the population of several coastal cells each year. Subsequently, we determined the percent change from year to year for the entire coastal region or band. Rather than computing percent change for each individual coastal cell.

We utilized a combination of Environmental Systems Research Institute (ESRI) and National Oceanic and Atmospheric Administration (NOAA) coastal boundaries to calculate horizontal population distributions occurring in the 50 km of land adjacent to the shoreline. In our view, distances of 100 km to 400 km that were widely utilized in studying human coastal populations in the past did not include the more pragmatic and colloquial contexts of coastlines and thus were unable to capture more nuanced population patterns. Next, we segmented the global population into “bands” and “regions” within 50 km of the coast. Bands are defined as discrete 5 km intervals, and regions are cumulative intervals. Starting at the shoreline and moving inland, population estimates were aggregated for each of the 5 km bands (i.e., 0–5 km, > 5–10 km, > 10–15 km, …, > 45–50 km). Similarly, coastal regions are defined in cumulative segments increasing by 5 km at each point, starting at the shoreline and moving inward (i.e., 0–5 km, 0–10 km, 0–15 km, …, 0–50 km).

Estimating human populations within any geographic area that includes coastlines can be challenging because of the roughness of the shoreline edge and the difficulty in determining the true length of the coast (i.e., the “coastal paradox”)18,44. Neither ESRI nor NOAA geospatial data perfectly align with the existing and constantly changing coastlines of the globe. NOAA data are more robust than ESRI’s when capturing irregularities of the shoreline and were used to derive global coastal estimates. However, NOAA data are not available at the continent level. Therefore, ESRI data were utilized for continent estimates. Additionally, ESRI coastal boundaries may not capture barrier islands and near-shore islands that include coastal populations. To estimate these complex near-shore population geographies, we extended a one-dimensional 15 km buffer into the ocean to capture coastal populations that were not included in the coastal bands. The buffer procedure had a minimal effect on estimates; approximately 500,000 individuals were added to the global estimate, which accounted for less than one-tenth of 1% of the 5 km band.

We used the curve fitting tool provided by Microsoft Excel to fit coastal band population as a function of distance from the shoreline in 5 km bands extending horizontally inland to 50 km. We are using a standard curve fitting procedure that does not consider more involved procedures to adjust for bias as in45. However, we have provided supplemental material on coastal band data that is available for other researchers to model using more sophisticated techniques for each continent and for the globe.

There are substantial limitations in estimating global and sub-global human population estimates especially dealing with the uncertainty of grid cell level and aggregate level estimates. These issues are many and complex. For example, time-wise comparisons of the world grid model estimations are challenged by (1) the frequency and accuracy of censuses both overtime and between countries, (2) the algorithms used by grid models and improvements made in algorithms overtime, and (3) a myriad of methodological and statistical issues. For example, Modifiable Areal Unit Problem (MAUP) and quality of data sources that drive smart interpretation are only two of many methodological challenges46. The foregoing list is only a partial inventory of the complexity in addressing uncertainty in this research area. Consequently, a comprehensive discussion on this topic is beyond the scope of this manuscript.

Supplementary Information

Supplementary Information 1. (630.2KB, docx)
Supplementary Information 2. (299.1KB, pdf)

Acknowledgements

The authors are most appreciative to Dr. Joel Cohen, Dr. Deborah Balk, and Dr. Neil Bennett for their insightful and valuable comments and suggestions during an earlier presentation of this research. We are also appreciative of Oak Ridge National Laboratory for making its annual LandScan Global data available to the research community. Any errors present in this study are the sole responsibility of the authors listed above. This manuscript has been co-authored by UT-Batelle, LLC, under contract DE-AC05-00OR22725 with the US Department of Energy (DOE). The US government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for US government purposes. DOE will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan). 

Author contributions

A. G. Cosby1: Project design, interpretation, drafting the manuscript, and project management. V. Lebakula2†: Research design, drafting the manuscript, analysis, visualization, software development, and data curation. C. N. Smith3: Drafting of the original version of the manuscript. D. W. Wanik4: Data curation, visualization, and review. K. Bergene5,1: Assisted in the drafting of the original version of the manuscript, reviewing, and editing. A. N. Rose2: Assisted in the drafting of the original version of the manuscript, validated the methods and reviewed the manuscript. D. Swanson6: manuscript review and interpretation. D. E. Bloom7: Interpretation, has substantially revised manuscript.

Funding

This research is not supported by a grant or contract. It is a collaborative effort by researchers from several universities and Oak Ridge National Laboratory.

Data availability

The data for this study are available free of charge from the LandScan Project at Oak Ridge National Laboratory. Interested researchers can download LandScan data by visiting https://landscan.ornl.gov/download.

Declarations

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-024-73287-x.

References

  • 1.Agardy, T. et al. Coastal systems. In Ecosystems and Human Well-Being: Current State and Trends (ed. Agardy, T.) (Island Press, 2005). [Google Scholar]
  • 2.Allison, E. H. & Bassett, H. R. Climate change in the oceans: Human impacts and responses. Science350(6262), 778–782 (2015). [DOI] [PubMed] [Google Scholar]
  • 3.Burke, L. et al. Pilot Analysis of Global Ecosystems: Coastal Ecosystems (World Resources Institute, 2001). [Google Scholar]
  • 4.Martínez, M. L. et al. The coasts of our world: Ecological, economic and social importance. Ecol. Econ.63(2–3), 254–272 (2007). [Google Scholar]
  • 5.Erlandson, J. M. & Rick, T. C. Archaeology, marine ecology, and human impacts on marine environments. In Human Impacts on Ancient Marine Ecosystems: A Global Perspective (eds Erlandson, J. M. & Rick, T. C.) (University of California Press, 2008). [Google Scholar]
  • 6.Creel, L. Ripple Effects: Population and Coastal Regions (Population Reference Bureau, 2003). [Google Scholar]
  • 7.Colwell, R. R. Global climate and infectious disease: The cholera paradigm. Science274(5295), 2025–2031 (1996). [DOI] [PubMed] [Google Scholar]
  • 8.Harvell, C. D. et al. Emerging marine diseases—Climate links and anthropogenic factors. Science285(5433), 1505–1510 (1999). [DOI] [PubMed] [Google Scholar]
  • 9.Balk, D. et al. Mapping urban settlements and the risks of climate change in Africa, Asia and South America. In Population Dynamics and Climate Change (eds Balk, D. et al. et al.) (UNFPA and IIED, 2009). [Google Scholar]
  • 10.Oliver-Smith, A. Sea Level Rise and the Vulnerability of Coastal Peoples: Responding to the Local Challenges of Global Climate Change in the 21st Century (UNU-EHS, 2009). [Google Scholar]
  • 11.Wright, L. D., Syvitski, J. P. M. & Nichols, C. R. Coastal systems in the Anthropocene. In Tomorrow’s Coasts: Complex and Impermanent (eds Wright, L. D. et al.) (Springer International Publishing, 2019). [Google Scholar]
  • 12.Crossett, K. M., Culliton, T. J., Wiley, P. C. & Goodspeed, T. R. Population trends along the coastal United States: 1980–2008 (NOAA, 2005). [Google Scholar]
  • 13.Gallup, J. L., Sachs, J. D. & Mellinger, A. D. Geography and economic development. Int. Reg. Sci. Rev.22(2), 179–232 (1999). [Google Scholar]
  • 14.Sachs, J. D., Mellinger, A. D. & Gallup, J. L. The geography of poverty and wealth. Sci. Am.284(3), 70–75 (2001). [DOI] [PubMed] [Google Scholar]
  • 15.Hausmann, R. Prisoners of geography. Foreign Policy122, 45–53 (2001). [Google Scholar]
  • 16.Neumann, B., Vafeidis, A. T., Zimmermann, J. & Nicholls, R. J. Future coastal population growth and exposure to sea-level rise and coastal flooding—A global assessment. PLoS One10(6), e0131375 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Cicin-Sain, B. et al. (eds) Trends and Future Challenges for US National Ocean and Coastal Policy: Proceedings (NOAA, 1999). [Google Scholar]
  • 18.Ache, B. W., Crossett, K. M., Pacheco, P. A., Adkins, J. E. & Wiley, P. C. The coast is complicated: A model to consistently describe the nation’s coastal population. Est. Coasts38(1), 151–155 (2015). [Google Scholar]
  • 19.Mimura, N. Sea-level rise caused by climate change and its implications for society. Proc. Jpn. Acad. Ser. B89(7), 281–301 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Small, C., Gornitz, V. & Cohen, J. E. Coastal hazards and the global distribution of human population. Environ. Geosci.7(1), 3–12 (2000). [Google Scholar]
  • 21.Ruiz, G. M. et al. Global spread of microorganisms by ships. Nature408(6808), 49–50 (2000). [DOI] [PubMed] [Google Scholar]
  • 22.McKee, J. J., Rose, A. N., Bright, E. A., Huynh, T. & Bhaduri, B. L. Locally adaptive, spatially explicit projection of US population for 2030 and 2050. Proc. Natl. Acad. Sci.112(5), 1344–1349 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Cohen, J. E., Small, C., Mellinger, A., Gallup, J. & Sachs, J. Estimates of coastal populations. Science278(5341), 1209–1213 (1997).9411741 [Google Scholar]
  • 24.Small, C. & Nicholls, R. J. A global analysis of human settlement in coastal zones. J. Coast. Res.19(3), 584–599 (2003). [Google Scholar]
  • 25.Small, C. & Cohen, J. Continental physiography, climate, and the global distribution of human population. Curr. Anthropol.45(2), 269–277 (2004). [Google Scholar]
  • 26.Nicholls, R. J. & Small, C. Improved estimates of coastal population and exposure to hazards released. Eos Trans. Am. Geophys. Union83(28), 301–305 (2002). [Google Scholar]
  • 27.Cohen, J. E. & Small, C. Hypsographic demography: The distribution of human population by altitude. Proc. Natl. Acad. Sci.95(24), 14009–14014 (1998). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Oak Ridge National Laboratory, “Documentation” (n.d.; https://landscan.ornl.gov/documentation).
  • 29.Hall, O., Stroh, E. & Paya, F. From census to grids: Comparing gridded population of the world with Swedish census records. Open Geogr. J.5, 1–5 (2012). [Google Scholar]
  • 30.United Nations, Department of Economic and Social Affairs. World population prospects 2019. https://population.un.org/wpp/ (Accessed 19 June 2020).
  • 31.Small, C., Sousa, D., Yetman, G., Elvidge, C. & MacManus, K. Decades of urban growth and development on the Asian megadeltas. Glob. Planet. Change165, 62–89 (2018). [Google Scholar]
  • 32.Small, C., Elvidge, C. D., Balk, D. & Montgomery, M. Spatial scaling of stable night lights. Remote Sens. Environ.115(2), 269–280 (2011). [Google Scholar]
  • 33.Small, C. & Elvidge, C. D. Night on earth: Mapping decadal changes of anthropogenic night light in Asia. Int. J. Appl. Earth Obs. Geoinform.22, 40–52 (2013). [Google Scholar]
  • 34.Rappaport, J. & Sachs, J. D. The US as a coastal nation. In Research Working Paper RWP 01-11 (eds Rappaport, J. & Sachs, J. D.) (Federal Reserve Bank of Kansas City, 2001). [Google Scholar]
  • 35.McGranahan, G., Balk, D. & Anderson, B. The rising tide: Assessing the risks of climate change and human settlements in low elevation coastal zones. Environ. Urban.19(1), 17–37 (2007). [Google Scholar]
  • 36.Syvitski, J. P. M., Vörösmarty, C., Kettner, A. & Green, P. Impact of humans on the flux of terrestrial sediment to the global coastal ocean. Science308(5720), 376–380 (2005). [DOI] [PubMed] [Google Scholar]
  • 37.Nagelkerken, I. & Connell, S. Global alteration of ocean ecosystem functioning due to increasing human CO2 emissions. Proc. Natl. Acad. Sci. U. S. A.112(43), 13272–13277 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.McCauley, D. J. et al. Marine defaunation: Animal loss in the global ocean. Science347(6219), 1255641 (2015). [DOI] [PubMed] [Google Scholar]
  • 39.MacManus, K., Balk, D., Engin, H., McGranahan, G. & Inman, R. Estimating population and urban areas at risk of coastal hazards, 1990–2015: How data choices matter. Earth Syst. Sci. Data13(12), 5747–5801 (2021). [Google Scholar]
  • 40.Dobson, J. E., Bright, E. A., Coleman, P. R., Durfee, R. C. & Worley, B. A. LandScan: A global population database for estimating populations at risk. Photogramm. Eng. Remote Sens.66(7), 849–857 (2000). [Google Scholar]
  • 41.Leyk, S. et al. The spatial allocation of population: A review of large-scale gridded population data products and their fitness for use. Earth Syst. Sci. Data11(3), 1385–1409 (2019). [Google Scholar]
  • 42.Bhaduri, B., Bright, E., Coleman, P. & Urban, M. L. LandScan USA: A high-resolution geospatial and temporal modeling approach for population distribution and dynamics. GeoJournal69(1–2), 103–117 (2007). [Google Scholar]
  • 43.Yin, X. et al. Which gridded population data product is better? Evidences from mainland southeast asia (MSEA). ISPRS Int. J. Geo-Inform.10(10), 681 (2021). [Google Scholar]
  • 44.Mandelbrot, B. How long is the coast of Britain? Statistical self-similarity and fractional dimension. Science156(3775), 636–638 (1967). [DOI] [PubMed] [Google Scholar]
  • 45.Clauset, A., Shalizi, C. R. & Newman, M. E. Power-law distributions in empirical data. SIAM Rev.51(4), 661–703 (2009). [Google Scholar]
  • 46.Calka, B. & Bielecka, E. Reliability analysis of LandScan gridded population data. The case study of Poland. ISPRS Int. J. Geo-Inform.8(5), 222 (2019). [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information 1. (630.2KB, docx)
Supplementary Information 2. (299.1KB, pdf)

Data Availability Statement

The data for this study are available free of charge from the LandScan Project at Oak Ridge National Laboratory. Interested researchers can download LandScan data by visiting https://landscan.ornl.gov/download.

Articles from Scientific Reports are provided here courtesy of Nature Publishing Group

RESOURCES