Invited research article
Decades of urban growth and development on the Asian megadeltas

https://doi.org/10.1016/j.gloplacha.2018.03.005Get rights and content

Highlights

  • Population, night light and digital elevation models are coregistered to quantify development of nine Asian megadeltas.

  • Over 174 million people live within 7 m of sea level on the Asian megadeltas.

  • Rapidly growing spatial networks of urban development maintain unity power law scaling since 1992.

  • The Huanghe Delta + North China Plain now contains the largest spatial network of urban development in Asia.

Abstract

The current and ongoing expansion of urban areas worldwide represents the largest mass migration in human history. It is well known that the world's coastal zones are associated with large and growing concentrations of population, urban development and economic activity. Among coastal environments, deltas have long been recognized for both benefits and hazards. This is particularly true on the Asian megadeltas, where the majority of the world's deltaic populations reside. Current trends in urban migration, combined with demographic momentum suggest that the already large populations on the Asian megadeltas will continue to grow. In this study, we combine recently released gridded population density (circa 2010) with a newly developed night light change product (1992 to 2012) and a digital elevation model to quantify the spatial distribution of population and development on the nine Asian megadeltas. Bivariate distributions of population as functions of elevation and coastal proximity quantify potential exposure of deltaic populations to flood and coastal hazards. Comparison of these distributions for the Asian megadeltas show very different patterns of habitation with peak population elevations ranging from 2 to 11 m above sea level over a wide range of coastal proximities. Over all nine megadeltas, over 174 million people reside below a peak population elevation of 7 m. Changes in the spatial extent of anthropogenic night light from 1992 to 2012 show widely varying extents and changes of lighted urban development. All of the deltas except the Indus show the greatest increases in night light brightness occurring at elevations <10 m. At global and continental scales, growth of settlements of all sizes takes the form of evolving spatial networks of development. Spatial networks of lighted urban development in Asia show power law scaling properties consistent with other continents, but much higher rates of growth. The three largest networks of development in China all occur on deltas and adjacent lowlands, and are growing faster than the rest of the urban network in China. Since 2000, the Huanghe Delta + North China Plain urban network has surpassed the Japanese urban network in size and may soon connect with the Changjiang Delta + Yangtze River urban network to form the largest conurbation in Asia.

Introduction

The current and ongoing expansion of urban areas worldwide represents the largest mass migration in human history. The percentage of global population considered to be urban has risen from 30% in 1950 to 54% in 2014 and is projected to reach 66% by 2050 (United Nations, 2015). Although the population of Asia is currently only 40% urban, it already accounts for 53% of the world's urban population. By 2050, India and China are projected to have urban populations of 404 and 292 million (respectively). However, almost half of the global urban population lives in cities of <500,000 while <1/8 lives in megacities of >10 million inhabitants.

It is well known that the world's coastal zones are associated with large and growing concentrations of population, urban development and economic activity. The global population living within 100 km of a shoreline at elevations <100 m has been estimated at 1.2 × 109 people (circa 1990) with an average population density nearly three times higher than global average density (Small and Nicholls, 2003). Within this coastal zone, the highest mean population densities occur over a range of coastal proximity reflecting large populations in low-lying river basins and deltas. Consistent with the findings of the United Nations (2015)), this comparison of satellite observed night light and census-derived population indicates that most of this coastal population occurs in relatively densely populated rural areas and small to medium size cities rather than large cities.

Deltas have long been recognized for both benefits and hazards. The benefits of fertile soils and fluvial to coastal transportation corridors are balanced by a variety of coastal and flood hazards (Nicholls et al., 2007). Over 325 million people live on deltas worldwide (Syvitski and Saito, 2007) with >250 million (circa 2000) on nine Asian deltas alone (Woodroffe et al., 2006). These so-called Asian megadeltas are of particular interest because they are among the most densely populated and rapidly urbanizing environments on Earth. In addition to the well known coastal and fluvial hazards posed by the deltaic environment, human activities are also known to impact the dynamics of deltas through the disruption of water flow (Vörösmarty and Sahagian, 2000) and sediment delivery (Syvitski and Saito, 2007). Interruption of sediment delivery by upstream capture (Syvitski et al., 2005) and flood control measures, compounded by natural and human-induced subsidence (Syvitski et al., 2009) results in a rise in relative sea level generally in excess of the rate of eustatic sea level rise, contributing to increased levels of inundation and coastal erosion (Ericson et al., 2006). This feedback between increasing levels of development and increasing vulnerability depends critically on the spatial extent, location and type of development within the deltaic environment. This is particularly true on the Asian megadeltas where current rate of urban development has been increasing over the past 20 years.

In this study, we combine recently released gridded population density (circa 2010) with a newly developed night light change product (1992 to 2013) and a digital elevation model to quantify the spatial distribution of population and development within the nine Asian megadeltas. While the earlier study of (Small and Nicholls, 2003) quantified the global distribution of population and development in coastal zones at continental and regional scales, the population and elevation products available at that time did not have sufficient spatial resolution or vertical accuracy (respectively) to quantify the relationship at spatial scales appropriate for most deltas. The gridded population data used in this study vary in spatial resolution from country to country, but generally provide sufficient resolution (10–50 km) to distinguish the higher densities associated with large cities and their location with respect to shorelines and sea level. We quantify the distribution of population with respect to coastal proximity and elevation to allow comparison among deltas. We use satellite-derived maps of night light brightness as a proxy for urban development coinciding with outdoor lighted infrastructure. While not all forms of development coincide with outdoor lighting, and some outdoor lighting is not related to urban development (e.g. resource extraction, fishing, fires, etc), the land areas under consideration in this study do not include significant areas of non-urban lighting. The night light change product provides much higher spatial resolution (0.5–2.5 km) and clearly resolves the location and spatial extent of urban development since 1992. We quantify this change in lighted development on the megadeltas in the context of rapid spatiotemporal evolution of networks of urban development throughout southeastern Asia over the past 25 years. We take the work of (Woodroffe et al., 2006) as a starting point for an intercomparison of the Asian megadeltas in the context of population distribution and evolution of lighted development and attempt to interpret this evolution in the context of the ongoing rapid urban growth occurring throughout Asia.

Section snippets

Data

The spatial extent of each delta is based on the spatial extent of the maximum transgression at highstand 6000 years ago as defined by Woodroffe et al. (2006). Each delta study area was chosen to completely encompass the corresponding delta as defined by Woodroffe et al. (2006). The boundaries of the delta are also apparent on the bivariate distributions as departures from the overall elevation gradient extending inland. As noted by Woodroffe et al. (2006), “SRTM analysis of those deltas for

Results

The results of the analysis of individual deltas are presented in Fig. A1, Fig. A2, Fig. A3, Fig. A4, Fig. A5, Fig. A6, Fig. A7, Fig. A8, Fig. A9 as a set of maps and population distributions for each delta with population distributions and elevation maps in the a) figure and VIIRS night light + SRTM shaded relief with dOLS+VIIRS change composite in the b) figure. Brief summaries of principal observations are given below.

Discussion

The spatial resolution of the population density and night light brightness data used in this analysis allow for a more spatially explicit analysis of the relationship between the morphology of each delta and the distribution of its human inhabitants and development. In comparison to earlier global analyses, the intersection of the anthropogenic and geodynamic characteristics of individual deltas allow each to be considered in the context of its specific combination of morphology and

Conclusions

The analysis of Asian megadeltas described here combines recently released gridded population density (circa 2010) with a newly developed night light change product (1992 to 2012) and a digital elevation model to quantify the spatial distribution of population and development on the nine Asian megadeltas. Bivariate distributions of population as functions of elevation and coastal proximity quantify potential exposure of deltaic populations to flood and coastal hazards. While these data do not

Acknowledgements

CS gratefully acknowledges the support of NASA (grant NNX12AM20G) and the Office of Naval Research (grant N00014-11-1-0683). Work done by D. Sousa was conducted with Government support under FA9550-11-C-0028 and awarded by the Department of Defense, Air Force Office of Scientific Research, National Defense Science and Engineering Graduate (NDSEG) Fellowship, 32 CFR 168a. D. Sousa thanks E. Sousa for blazing the trail. The authors thank the editor and two anonymous reviewers for helpful comments

References (34)

  • C.C. Carabajal et al.

    SRTM C-band and ICESat laser altimetry elevation comparisons as a function of tree cover and relief

    Photogramm. Eng. Remote Sens.

    (2006)
  • A. Clauset

    Power-law distributions in empirical data

    SIAM Rev.

    (2009)
  • C.D. Elvidge

    What is so great about nighttime VIIRS data for the detection and characterization of combustion sources?

    Asia Pac. Adv. Netw.

    (2013)
  • C.D. Elvidge

    VIIRS night-time lights

    Int. J. Remote Sens.

    (2017)
  • D. Gesch

    Validation of the Aster Global Digital Elevation Model Version 2 over the Conterminous United States. 2012 XXII ISPRS Congress

    (2012)
  • M. Hofton

    Validation of SRTM elevations over vegetated and non-vegetated terrain using medium footprint Lidar

    Photogramm. Eng. Remote Sens.

    (2006)
  • S.M. Hvidegaard

    ASTER GDEM validation using LiDAR data over coastal regions of Greenland

    Remote Sens. Lett.

    (2012)
  • Cited by (38)

    • Development of a three-source remote sensing model for estimation of urban evapotranspiration

      2022, Advances in Water Resources
      Citation Excerpt :

      Urban areas with predominantly impervious surfaces and intensive anthropogenic activities, and hence cause significant changes in the regional hydrologic cycle, turbulent heat fluxes, and microclimate (Owen et al., 1998; Kondoh and Nishiyama, 2000; Zhang et al., 2009; Small et al., 2018; Small, 2019; DiGiovanni et al., 2018). Owing to the heterogeneity associated with the urban land cover, urban ET poses significant spatial variability (Small et al., 2018; Wang et al., 2016; Qiu et al., 2017). Further, urban soil evaporation, vegetation transpiration and impervious surface evaporation cool the land surface and mitigate the urban heat island effect (Fletcher et al., 2013; Haider et al., 1997; Prathap et al., 2014; Matthias et al., 2014; Hecht et al., 2020).

    • Anisotropic characteristic of artificial light at night – Systematic investigation with VIIRS DNB multi-temporal observations

      2019, Remote Sensing of Environment
      Citation Excerpt :

      The Defense Meteorological Satellite Program/Operational Linescan System (DMSP/OLS) was originally designed to detect nighttime clouds, but scientists have found that DMSP/OLS has the ability to detect artificial light at night (Croft, 1978). Nighttime light images can be used in a variety of applications: (1) nighttime light dynamics are able to reflect socioeconomic development such as GDP growth (Elvidge et al., 2009, 2012; Henderson et al., 2011; Klomp, 2016; Li et al., 2013a); (2) spatiotemporal change of the nighttime light is a good proxy for urbanization analysis (Liu et al., 2012; Small et al., 2011; Small and Elvidge, 2013; Small et al., 2018; Zhou et al., 2014); (3) the reduction of nighttime light can be used as evidence for evaluating humanitarian disasters (Li and Li, 2014); (4) nighttime light affects the health of human beings and animals, so that the images are useful for investigating light pollution (Bauer et al., 2013; Katz and Levin, 2016; Kloog et al., 2008; Kuechly et al., 2012; Marcantonio et al., 2015); (5) the gas flaring recorded by satellites can be used to evaluate carbon emission and oil production (Elvidge et al., 2009); (6) the satellite-recorded artificial light on the sea can be used to monitor fishery (Choi et al., 2008; Kiyofuji and Saitoh, 2004; Waluda et al., 2008). The Visible Infrared Imaging Radiometer Suite (VIIRS), carried by Suomi National Polar-Orbiting Partnership (Suomi NPP) satellite launched in October 2011, provides a new source of nighttime light imagery.

    • A new ranking of the world's largest cities—Do administrative units obscure morphological realities?

      2019, Remote Sensing of Environment
      Citation Excerpt :

      While many EO-based studies approached the global urban extent (e.g. Schneider et al., 2009; Elvidge et al., 2007; Pesaresi et al., 2013; Esch et al., 2012), only few studies, mostly relying on night light emission and/or gridded population data have approached the delimitation of city extents on a global scale. In these studies the power law scaling of rank-size distributions is documented (Decker et al., 2007; Small et al., 2011; Small and Sousa, 2016) and consistency is confirmed over time (Small et al., 2018). Small et al. (2011) found spatial networks of urban development are vastly larger than the administratively-defined cities they contain.

    View all citing articles on Scopus
    View full text