Elsevier

Journal of Cleaner Production

Volume 276, 10 December 2020, 123244
Journal of Cleaner Production

A new perspective to map the supply and demand of artificial night light based on Loujia1-01 and urban big data

https://doi.org/10.1016/j.jclepro.2020.123244Get rights and content

Highlights

  • The supply and demand status of ANL in open areas is mapped.

  • The relationship between ANL and PD has been quantified.

  • The significant HSLD status of ANL calls for high attention.

  • Unbalanced status of ANL is highly related to housing price.

  • Light regulation is necessary to alleviate the disruptive effects.

Abstract

The notable increase in artificial night light (ANL) induced by the rapid urbanization process has been widely studied, but a deep understanding of the supply and demand status of ANL is still lacking. This paper attempts to map the supply and demand of ANL from the human perspective by using advanced Loujia1-01 nighttime imagery and social media derived population density (PD) data, which provides a new tool for light regulation in urban management. The bivariate clustering based k-means algorithm and template matching technique are integrated to delineate mismatch regions at the block scale to further analyze the underlying reason for unbalanced status. The results showed that the high supply but low demand (HSLD) ANL status was the leading component in the mismatch regions, occupying more than 650,000 ha and mainly occurring in the city center. The HSLD proportion was considerable in terms of public services (44%), commercial (40%), industrial (39%), transportation (56%), and green space areas (53%). Moreover, the HSLD area notably increased 946 ha over time from 18:00 to 22:00. The measurements for validation obtained by field investigation showed highly linear relationship with ANL (R2 = 0.75) and PD (R2 = 0.62), and the mapping results were consistent with the actual conditions. This study reveals the highly unbalanced ANL status, and appeals to planners for the establishment of optimal lighting regulations to alleviate disruptive effects.

Introduction

In the last decade, artificial night light (ANL) has rapidly increased both in intensity and density, accompanied by the rapid development of urbanization (Chang et al., 2020; Kyba et al., 2017). The multiple sources of ANL include street lighting, lighting from buildings and advertising, vehicles, etc. (Gaston et al., 2015; Katz and Levin, 2016). ANL has some clear benefits for humans, including illumination, recreation in festivals, extension of human activities into the night, and promotion of production activities (Boyce, 2019; Gaston et al., 2014b). However, several concerns have been raised about its negative influence, especially for ANL in open areas. Selecting the U.S. as an example, approximately 120 TW-hours of energy is consumed by outdoor lighting in an average year, mostly to illuminate streets and parking lots, and at least 30% of outdoor lighting is wasted, which releases 21 million tons of carbon dioxide per year (International Dark-Sky Association, 2020). Therefore, the absence of optimal planning of the ANL supply in open areas based on the human demand can cause serious problems. If the ANL supply greatly exceeds the need, this can lead to considerable electrical energy consumption as well as profound and lasting impacts on ecosystems (Katz and Levin, 2016; Levin et al., 2020). For example, previous studies have demonstrated that ANL changes the natural light cycle of wildlife organisms (Gaston et al., 2014a), and disrupts circadian rhythms (Ciach and Fröhlich, 2019; Davies et al., 2017). Conversely, if the ANL supply is lower than the demand, insufficient light might reduce the human sense of security at night, especially for elderly people and children, thus making nocturnal activities inconvenient (Perkins et al., 2015). Hence, there is an urgent need to study how to satisfy human activity demands and establish effective lighting regulations simultaneously to alleviate the disruptive effects of ANL (Gaston, 2013).

China, the largest developing country, is now suffering from an imbalanced supply and demand of ANL due to uncontrolled urban sprawl and the corresponding illumination facilities in most cities (Han et al., 2014; Yao et al., 2020). Effective regulation measures have not been established due to lack of ANL supply and demand patterns. However, few current studies have focused on the assessment of the rationalization of the ANL supply based on the human demand at the block scale. Therefore, mapping the supply and demand of ANL is imperative and meaningful for planners and researchers to better understand the current status, and thus help to make optimal urban management decisions.

Most studies have confirmed that the satellites-recorded ANL has a great potential in modelling demographic and socioeconomic variables, such as the gross domestic product (GDP) (Deville et al., 2014), population density (PD) (Ma, 2018), electric power consumption (Hu and Huang, 2019), and CO2 emission from the city level to the global level (Shi et al., 2019). Estimating ecological light pollution by ANL data is currently a hot research topic (Davies and Smyth, 2018). Several studies have adopted ANL data to represent the human footprint to map the human impact on Earth ecosystems (Halpern et al., 2015; Venter et al., 2016), and most of these studies have focused on a large geographic scope. Other studies have estimated the impact of ANL on specific species, such as birds (Horton et al., 2019), sea turtles (Hu et al., 2018) and bats (Azam et al., 2016). However, few studies have evaluated the ANL status from the human perspective, and most employed ANL data in previous studies are too coarse to meet the requirements of light regulation in urban areas. Specifically, the Defense Meteorological Satellite Program/-Operational Linescan System (DMSP/OLS) and Visible Infrared Imaging Radiation Suite (VIIRS) have provided the longest publicly available time series of ANL data (Fehrer and Krarti, 2018; Yu et al., 2018; Zhang et al., 2016). But the coarse spatial resolution of DMSP/OLS (2.7 km) and VIIRS (740 m) data limits their capacity to accurately depict the spatial pattern of the ANL supply within the urban environment (Levin et al., 2014; Zheng et al., 2018c, Zhang et al., 2018b, Zhang et al., 2018a). However, the recently launched Luojia1-01 satellite provides a new generation of ANL imagery with a higher spatial resolution (130 m) (Li et al., 2018; Zhang et al., 2018a; Zhang et al., 2019). Compared to DMSP/OLS and VIIRS data, the improved spatial resolution allows detailed analysis of various illumination mechanisms among different functional zones at the block scale, which are naturally separated by road networks (Jiang et al., 2018; Zhang et al., 2019).

Moreover, the increasingly available urban big data maps are providing massive spatiotemporal geo-information nowadays, thereby helping us understand the urban structure and population dynamics in a deep and timely manner (Cai et al., 2017). An important source of big data are the location requests derived from social media platforms, such as Facebook, Twitter, Weibo, and WeChat, which have been confirmed to be a good surrogate of PD (Cai et al., 2017; Jiang et al., 2016; Ma, 2018). The social media derived PD has irreplaceable advantages, as it attains a finer spatio-temporal resolution than that of the traditional PD obtained from statistical census data, thus making it a powerful tool for the detailed description of human activities (Steiger et al., 2015). Additionally, in contrast to censuses, which are usually conducted based on administrative divisions without considering individual movements, social media data dynamically capture individual information marked by GPS locations and time stamps (Jiang et al., 2016). Therefore, the combination of Loujia1-01 nighttime imagery and social media derived PD provides a good opportunity to estimate the supply and demand of ANL from the human perspective, which offers a reference for effective lighting regulation.

This paper aims to map the supply and demand of ANL from the human perspective, and thus provides a new tool for planners and researchers to deeply understand the relationship between the ANL and PD for further making optimal decisions in urban management. To achieve this objective, this study has to (1) assess the overall spatial pattern between the ANL and PD; (2) delineate mismatch and match regions at the block scale; (3) validate the mapping results by field investigation; and (4) analyze the underlying mechanism of the delineation results to formulate light regulation recommendations.

This paper is organized as follows. Section 2 describes study area and the data sources applied in this study. Section 3 introduces the proposed framework for mapping the supply and demand status of ANL. Results and discussion are presented in section 4 Results, 5 Discussion, respectively. And section 6 sets out main conclusions.

Section snippets

Study area

Our study area, Hangzhou, located in Southeast China, is the capital city of Zhejiang Province (Fig. 1). Over recent decades, Hangzhou has experienced rapid urbanization, and the per capita GDP in 2017 was 135,113 CNY (equivalent to 19,083 USD), thereby making it one of the most developed cities in China (Du and Huang, 2017). Along with the high speed of development, electricity consumption also increased rapidly from 521,925 104 kW h in 2010–7380,288 104 kW h in 2017 according to the Hangzhou

Methods

The step-by-step procedures in Fig. 3 were implemented to study the supply and demand status of ANL from the human perspective. First, raw ANL and PD datasets were generated as layer stacking raster data after calibration and resampling, respectively (section 3.1). Second, the block was adopted to integrate all the geo-information. Blocks generation was achieved by the morphologic operations of dilation and thinning of the road networks (section 3.2). Third, the study applied bivariate

Overall spatial patterns of bivariate clustering

After data preprocessing, the ANL and PD maps overlapped with the same spatial resolution and geo-locations (Fig. 8 a-c), which was necessary for block generation and bivariate clustering. As shown in Fig. 8 d-e, the comparative relationships between ANL and PD are distinct and highly related to the development level of the districts. Cluster C1 was clearly located in the city center, indicating that both ANL and PD were high due to the intense socioeconomic activities. Cluster C2 was

Negative impacts of the current ANL status

According to the delineation results of mismatch regions, the areas belonging to the HSLD status were far more abundant than those belonging to the LSHD status, which provides new evidence that the energy waste problem is substantial by the current unbalanced light situation. Based on the results in Table 2, industrial land has the second largest area within the HSLD regions, and is worthy of further discussion. (1) The industry in Hangzhou has gone through an important suburbanization process

Conclusion

This study proposed an integral framework to map the supply and demand status of the ANL in open areas from the human perspective. The improved k-means algorithm was applied for bivariate spatial clustering, and the NCC index was calculated via the template matching technique to quantify the mismatch degrees. As a result, four types of delineations were defined for further analysis of the control factors, and the negative effects in terms of energy consumption and ecological protection were

CRediT authorship contribution statement

Yang Ye: Conceptualization, Methodology, Writing - original draft, Software, Formal analysis. Xingyu Xue: Investigation, Writing - original draft, Visualization, Validation. Lingyan Huang: Writing - review & editing, Resources. Muye Gan: Resources, Funding acquisition. Ke Wang: Investigation, Data curation, Supervision, Project administration, Funding acquisition. Jinsong Deng: Supervision, Project administration, Writing - review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (grant numbers. 41701171); the Natural Science Foundation of Zhejiang Province (grant numbers. LY18G030006) and the Ministry of Science and Technology of the People’s Republic of China (grant numbers. 2016YFC0503404).

References (58)

  • Y. Katz et al.

    Quantifying urban light pollution - a comparison between field measurements and EROS-B imagery

    Remote Sens. Environ.

    (2016)
  • N. Levin

    The impact of seasonal changes on observed nighttime brightness from 2014 to 2015 monthly VIIRS DNB composites

    Remote Sens. Environ.

    (2017)
  • N. Levin et al.

    A new source for high spatial resolution night time images - the EROS-B commercial satellite

    Remote Sens. Environ.

    (2014)
  • C.Z.W.X. Li

    Potentiality of using luojia1-01 night-time light imagert to estimate urban community housing price-A case study in Wuhan,China

    Sensors

    (2019)
  • K. Shi et al.

    Spatiotemporal variations of CO2emissions and their impact factors in China: a comparative analysis between the provincial and prefectural levels

    Appl. Energy

    (2019)
  • E. Steiger et al.

    Twitter as an indicator for whereabouts of people? Correlating Twitter with UK census data

    Comput. Environ. Urban Syst.

    (2015)
  • K.Y. Wu et al.

    Land use dynamics, built-up land expansion patterns, and driving forces analysis of the fast-growing Hangzhou metropolitan area, eastern China (1978-2008)

    Appl. Geogr.

    (2012)
  • X. Xue et al.

    Mapping the fine-scale spatial pattern of artificial light pollution at night in urban environments from the perspective of bird habitats

    Sci. Total Environ.

    (2020)
  • L. Zhang et al.

    Suburban industrial land development in transitional China: spatial restructuring and determinants

    Cities

    (2018)
  • Q. Zheng et al.

    A new source of multi-spectral high spatial resolution night-time light imagery—JL1-3B

    Remote Sens. Environ.

    (2018)
  • C. Azam et al.

    Disentangling the relative effect of light pollution, impervious surfaces and intensive agriculture on bat activity with a national-scale monitoring program

    Landsc. Ecol.

    (2016)
  • J. Bennie et al.

    Mapping artificial lightscapes for ecological studies

    Methods Ecol. Evol.

    (2014)
  • P.R. Boyce et al.

    Road lighting and energy saving

    Light. Res. Technol.

    (2009)
  • M. Ciach et al.

    Ungulates in the city : light pollution and open habitats predict the probability of roe deer occurring in an urban environment

    Urban Ecosyst.

    (2019)
  • T.W. Davies et al.

    Why artificial light at night should be a focus for global change research in the 21st century

    Global Change Biol.

    (2018)
  • T.W. Davies et al.

    Multiple night-time light-emitting diode lighting strategies impact grassland invertebrate assemblages

    Global Change Biol.

    (2017)
  • P. Deville et al.

    Using luminosity data as a proxy for economic statistics

    Proc. Natl. Acad. Sci.

    (2014)
  • K.J. Gaston

    Sustainability: a green light for efficiency

    Nature

    (2013)
  • K.J. Gaston et al.

    Reducing the ecological consequences of night-time light pollution: options and developments

    J. Appl. Ecol.

    (2012)
  • Cited by (12)

    • A feasible framework to downscale NPP-VIIRS nighttime light imagery using multi-source spatial variables and geographically weighted regression

      2021, International Journal of Applied Earth Observation and Geoinformation
      Citation Excerpt :

      (2) Loujia1-01 imagery was applied to validate the downscaled results. Compared to the NPP-VIIRS NTL data, the imagery from Loujia1-01 has a higher spatial resolution and larger spectral range (Table 1), which further enables finer-scale applications (Wang et al., 2020; Ye et al., 2020). However, the impact of cloud coverage has not been eliminated in Loujia1-01 imagery.

    • Fine-scale mapping of urban ecosystem service demand in a metropolitan context: A population-income-environmental perspective

      2021, Science of the Total Environment
      Citation Excerpt :

      First, spatial location and aggregation extent of the beneficiaries should be clear. Apart from population density, alternative indicators such as building density and light intensity, can also be applied to characterize beneficiaries' distribution (Koprowska et al., 2020; Ye et al., 2020). The second perspective is that urban ES demands are profoundly affected by socio-economic factors, such as income, wealth, education, and ethnicity (King et al., 2014; Rounsevell et al., 2010; Villamagna et al., 2013, 2014; Wilkerson et al., 2018).

    • Light Pollution Assessment Decision Model Based on Entropy Weight Topsis and Decision Tree

      2023, International Conference on Integrated Intelligence and Communication Systems, ICIICS 2023
    View all citing articles on Scopus
    View full text