Research article
Light pollution in USA and Europe: The good, the bad and the ugly

https://doi.org/10.1016/j.jenvman.2019.06.128Get rights and content

Highlights

  • New maps of USA and Europe show light pollution from different perspectives.

  • Night light flux per capita varies greatly between administrative units in USA and in Europe.

  • Night light flux per dollar varies greatly between administrative units in USA and in Europe.

  • Germany results to be the best overall in light pollution global ranking.

  • USA on average pollutes three times more than Europe in night light flux per capita.

Abstract

Light pollution is a worldwide problem that has a range of adverse effects on human health and natural ecosystems. Using data from the New World Atlas of Artificial Night Sky Brightness, VIIRS-recorded radiance and Gross Domestic Product (GDP) data, we compared light pollution levels, and the light flux to the population size and GDP at the State and County levels in the USA and at Regional (NUTS2) and Province (NUTS3) levels in Europe. We found 6800-fold differences between the most and least polluted regions in Europe, 120-fold differences in their light flux per capita, and 267-fold differences in flux per GDP unit. Yet, we found even greater differences between US counties: 200,000-fold differences in sky pollution, 16,000-fold differences in light flux per capita, and 40,000-fold differences in light flux per GDP unit. These findings may inform policy-makers, helping to reduce energy waste and adverse environmental, cultural and health consequences associated with light pollution.

Introduction

Light pollution (LP), resulting from the alteration of natural night light levels by artificial light sources is one of the most evident pollutant in the Anthropocene (Cinzano et al., 2000), is continuously increasing in magnitude (Cinzano, 2000; Garstang, 2004; Bennie et al., 2014), notwithstanding, or, perhaps, due to the raising efficiency in producing light (Kyba et al., 2017). LP is a major environmental and health problem, known to be associated with depression, insomnia and other health disorders in humans (Pauley, 2004; Haim and Portnov, 2013; Hatori et al., 2017) and potential changes in foraging, navigating and reproductive behaviour in wildlife species (Rich and Longcore, 2005). The widespread introduction of high intensity white light-emitting diodes (LEDs), praised by many for their high efficiency, does, in fact, only exacerbates the problem due to light emissions with “bluer” and more polluting light spectra compared to more yellow light emitted by previous lighting technologies, such as incandescent and low pressure sodium lights (American Medical Association, 2016; Aubé et al., 2013). As a result, more short wavelength (commonly called blue) light, is introduced into the night environment.

The technical parameters of light sources and actions required to lower artificial light at night (ALAN) pollution are well known (Falchi et al., 2011), and some of them are already implemented in regional and national laws in several countries, including Italy, Slovenia, Chile, Spain, France and Croatia. These actions include: aiming the lights only downwards, instead of wasting light by directing it above the horizontal plane; orienting street lights towards the target (e.g. on the road or pathway, not towards private properties or windows), and turning lights on at the correct timing, using smart and adaptive lighting technologies. Other regulatory measures to reduce LP include regulating, on sound scientific basis, the absolute minimum lighting levels necessary to perform the action (e.g., driving or walking on a sidewalk) and using light sources emitting less impacting, blue poor spectra, while avoiding high intensity blue emission sources, such as e.g., white LEDs. The use of these strategies, suggested by light pollution experts, can reduce, by an order of magnitude or more, LP in heavily polluted areas.

This work presents the amount of LP produced by different geographic units, such as States and counties in the USA and NUTS2 and NUTS3 (Nomenclature des Unités Territoriales Statistiques, the French for Nomenclature of Territorial Units for Statistics) regions of the EU. It lists all the administrative units in Europe and USA from the best to the worst examples (in light pollution of the sky, in light emissions per capita and light emissions per unit of income). This ‘catalogue’ will be useful to the scientific community and the policy-makers as a basis to explore the multiple causes of the more and less virtuous administrative units, helping to find better solutions to the global problem of light pollution.

For the analysis we use data on artificial night sky brightness from the New World Atlas of Artificial Night Sky Brightness (Falchi et al., 2016a), data on the flux emitted by light sources obtained from the VIIRS satellite images, the population densities and per capita income data for the above regional subdivisions, obtained from the Eurostat and the US Census Bureau. We use these data to calculate the following five measures for each administrative subdivision:

  • a)

    the percent of the area of an administrative subdivision with a given level of artificial night sky brightness (subdivided into 6 classes of ALAN pollution);

  • b)

    the percent of population living under a given artificial night sky brightness (subdivided into 6 classes of ALAN pollution);

  • c)

    the average artificial night sky brightness of the considered territory;

  • d)

    the artificial light flux per capita (FpC);

  • e)

    the artificial light flux per GDP unit (FpD), with GDP measured in US$ using purchasing power parity.

The FpC and FpD allowed to analyse LP in a new perspective, showing often that the most polluted areas, such as the metropolis are those that pollute less per inhabitants or per unit of income. These data also allowed to compare the different polluting power per capita and per income of the various cities and administrative units, finding that a similar light pollution can derive from cities with a very different number of inhabitants. It is worth mentioning that the flux data is different from the radiance of the night sky given by the New World Atlas of Artificial Night Sky Brightness. The main difference for the purposes of this study is that the flux gives how much light is produced (and escapes to space) in each pixel area, while the second gives the artificial night sky brightness looking at the zenith from the centre of each pixel. The two things are related, but not in a trivial way. As an example, the flux coming from the Upper Bay in New York is essentially zero (no light sources are on the water), while the night sky observed from the centre of the bay is extremely light polluted, due to the lights coming from the surrounding sources. In fact, the World Atlas radiance data for each pixel was computed taking into account the lights coming from a circle of 200 km radius.

Combined with the abovementioned LP-reduction strategies, this knowledge can help to developed targeted policies aimed to achieve a substantial decrease in LP, thus de facto solving the problem and arriving to a sustainable use of artificial light at night. If we are unable to solve this problem, for which the countermeasures are well known, easy to implement, and can bring energy and direct and indirect monetary savings, along with benefits to biodiversity, health and culture, then our ability of solving more complex environmental problems, such as e.g., global warming, will remain in doubt.

Section snippets

Results

The analysis reveals extremely high discrepancies in LP (intended as artificial night sky brightness), in FpC and in FpD between areas in both the USA and Europe, with discrepancies, both absolute and relative, varying by magnitudes. On a smaller scale, the FpC and FpD was already investigated in 1997 for Italy (Falchi, 1998).

Artificial night sky brightness

The three provinces with the best sky in Europe are all in Scotland. Of the best 25 NUTS3 (that is regions with the darkest skies), 5 belong to Austria, 4 to the UK, 3 are in Latvia, 2 are in Lithuania and Bulgaria each, 1 in Denmark, Estonia, Greece, Iceland, Norway, Romania, Spain and Sweden. In USA, the three darkest skies counties are in Alaska (which hosts 12 counties in the first 25 places). Other US states with the least light polluted counties in the top 25 are: Montana (3), Oregon (3),

Materials and methods

For the present analysis we used several datasets, as detailed below:

The light flux (from radiance detected by Suomi NPP satellite), artificial night sky brightness at zenith (form the New World Atlas), population density (Landscan), per capita income (OECD's online regional database and States Census's American Community Survey), vector files of the European NUTS administrative subdivisions (//ec.europa.eu/eurostat/web/gisco/geodata/reference-data/administrative-units-statistical-units/nuts

Conclusions

In the present analysis we found that there are great differences in the studied parameters between Europe and USA, with, e.g. USA having almost three times the Flux per Capita compared to Europe. We also found differences between countries inside EU (e.g. Portugal with four times the FpC of Germany) and USA (e.g. South Dakota with five times the FpC compared to New York). Greater differences are found between smaller administrative units, in part due to differences in population densities,

Funding

No special funds were used for the research that carried to this publication.

Author contributions

F.F. and R.F. conceived the research, F.F. wrote most of the manuscript, N.R. and B.P. performed the statistics of population and territory using the World Atlas sky brightness data, F.F. and R.F. performed the QGIS analysis and produced the maps, graphs and tables, T.G. performed the economic data search and analysis and wrote the pertaining parts. K.B. and C.D.E. provided the radiance data. All authors read and approved the manuscript.

Competing interests

The authors declare to have no competing interests. Notwithstanding this, F.F. retains correct to say that he's president of CieloBuio, an Italian association for the protection of the night sky.

References (40)

  • P. Cinzano

    The growth of light pollution in North-Eastern Italy from 1960 to 1995

    Memorie della Società Astronomica Italiana-J. Ital. Astron. Soc.

    (2000)
  • P. Cinzano et al.

    The artificial night sky brightness mapped from DMSP satellite Operational Linescan System measurements

    Mon. Not. R. Astron. Soc.

    (2000)
  • Da Silva et al.

    Artificial night lighting rather than traffic noise affects the daily timing of dawn and dusk singing in common European songbirds

    Behav. Ecol.

    (2014)
  • Gerhard Eisenbeis et al.

    Light pollution and the impact of artificial night lighting on insects

    Ecol. cities towns

    (2009)
  • C.D. Elvidge et al.

    A fifteen year record of global natural gas flaring derived from satellite data

    Energies

    (2009)
  • F. Falchi

    Luminanza artificiale del cielo notturno in Italia

    (1998)
  • F. Falchi et al.

    The new world atlas of artificial night sky brightness

    Sci. Adv.

    (2016)
  • F. Falchi et al.

    Supplement to the New World Atlas of Artificial Night Sky Brightness

    (2016)
  • Terrel Gallaway

    On light pollution, passive pleasures, and the instrumental value of beauty

    J. Econ. Issues

    (2010)
  • Terrel A. Gallaway et al.

    Blinded by the light: economic analysis of severe light pollution

    J. Econ.

    (2013)
  • Cited by (88)

    View all citing articles on Scopus
    View full text