Perspective The Dark Side of Light: A Transdisciplinary Research Agenda for Light Pollution Policy

Although the invention and widespread use of artificial light is clearly one of the most important human technological advances, the transformation of nightscapes is increasingly recognized as having adverse effects. Night lighting may have serious physiological consequences for humans, ecological and evolutionary implications for animal and plant populations, and may reshape entire ecosystems. However, knowledge on the adverse effects of light pollution is vague. In response to climate change and energy shortages, many countries, regions, and communities are developing new lighting programs and concepts with a strong focus on energy efficiency and greenhouse gas emissions. Given the dramatic increase in artificial light at night (0 - 20% per year, depending on geographic region), we see an urgent need for light pollution policies that go beyond energy efficiency to include human well-being, the structure and functioning of ecosystems, and inter-related socioeconomic consequences. Such a policy shift will require a sound transdisciplinary understanding of the significance of the night, and its loss, for humans and the natural systems upon which we depend. Knowledge is also urgently needed on suitable lighting technologies and concepts which are ecologically, socially, and economically sustainable. Unless managing darkness becomes an integral part of future conservation and lighting policies, modern society may run into a global self-experiment with unpredictable outcomes.


INTRODUCTION
In 2009, the UN's Year of Astronomy drew worldwide attention to an area affected by a long neglected environmental stressor: the increasing illumination of our nightscapes.The Year of Astronomy coincided with the 400th anniversary of Galileo's first observations with a telescope in Padua.However, to look at today's firmament Galileo would have to escape to remote areas for his research.This is because the Earth has become brighter at night.The rapid proliferation of electric lights has drastically reordered nightscapes across the globe, in terms of both light intensity and light spectrum (Cinzano et al. 2001, Elvidge et al. 2007).Although artificial lighting has clearly enhanced the quality of human life (Jakle 2001, Doll et al. 2006), the benefits are accompanied by hidden costs.Astronomers were the first to recognize that sky glow hampers the detection of faint celestial objects, obliging them to conduct their observations from darker areas or from orbit (Riegel 1973, Smith 2009).It is only very recently that the multiple negative effects of artificial lighting on ecology, human health, and social well-being have gained broader recognition (Jakle 2001, Rich and Longcore 2006, Navara and Nelson 2007).http://www.ecologyandsociety.org/vol15/iss4/art13/Light pollution is now a widely accepted term for adverse effects of artificial light on nature and humans (Longcore andRich 2004, Navara andNelson 2007).Nearly all living organisms, including human beings, have evolved under a natural rhythm of day and night.Interestingly, around 30% of all vertebrates and more than 60% of all invertebrates world-wide are nocturnal (Hölker et al. 2010).As lighting becomes brighter and extends farther into rural areas and offshore in marine systems (see Appendix 1), the distinction between day and night becomes blurred.Our understanding of the adverse effects of light pollution is vague and based mostly on purely observational case studies.Nonetheless, there is clear evidence that artificial lighting can alter physiology, including hormonal balance, as well as behavior, orientation, organism fitness, food web interactions, and biotope connectivity (Rich andLongcore 2006, Navara andNelson 2007).The artificial disturbance of the natural day/night cycle may, as a result, have serious psycho-physiological and even medical consequences for humans, along with ecological and evolutionary implications for animals, plants, and even entire terrestrial, freshwater, and marine ecosystems (Rich andLongcore 2006, Navara andNelson 2007).Light pollution is most probably an important but underestimated driver behind the erosion of provisioning, e.g., loss of light-sensitive species and genotypes; regulating, e.g., decline of nocturnal pollinators such as moths and bats; and cultural ecosystem services, e.g., loss of aesthetic values such as the visibility of the Milky Way (Rich and Longcore 2006, Carpenter et al. 2009, Smith 2009).The principal effects become most apparent at the interfaces between the physiological, ecological, and socioeconomic realms (Fig. 1).The problem is escalating worldwide as artificial lighting is rapidly increasing by around 6% per year (range: 0-20%; Table 1).

THE CURRENT FOCUS ON ENERGY EFFICIENCY
Artificial lighting consumes 19% of total global electricity, accounting for greenhouse gas emissions of 1900 Mt of CO 2 per year (OECD/ IEA 2006).It is no surprise that current artificial lighting policies focus primarily on energy efficiency and greenhouse gas emissions (e.g., OECD/IEA 2010), although safety, astronomical, and other considerations appear sporadically (see Appendix 2).The International Energy Agency has calculated that the systematic use of 'least life-cycle cost' lighting solutions (see Appendix 3) from 2008 onward would reduce the electricity consumption attributable to lighting until 2020 by 1311 TWh and 763 Mt of CO 2 emissions per year compared to projections on the basis of current policies (OECD/ IEA 2006).
Recently, the European Ecodesign Directive established a framework to phase out the incandescent lamp and other particularly energyintensive lighting products, e.g., high-pressure mercury lamps (The European Parliament and the Council of the European Union 2009).This step could reduce CO 2 emissions in the EU by approximately 42 Mt per year, corresponding roughly to a 10% reduction of the greenhouse gas emissions the EU promised to achieve under Kyoto (Denneman 2009, Managenergy 2010).In the United States, President Obama has proposed a scheme for more energy-efficient lamps and lighting equipment as part of his climate change policy.This would result in savings of approximately 20 Mt CO 2 annually (The White House 2009).Similar activities are reported inter alia for China, Australia, and New Zealand (OECD/ IEA 2006IEA , 2010)).
Within such policy frameworks, many countries, regions, and communities are developing new lighting programs and concepts.For example, the EU has launched a number of programs, e.g., GreenLight www.eu-greenlight.org,E-Street www .e-streetlight.com, to adopt efficient lighting systems and to initiate a permanent market transition.Although most of these programs and concepts are driven by energy efficiency motives alone, there remain causes for concern.For example, technological innovations that help improve the efficiency of energy appliances and systems often lead to greater energy use because of direct 'rebound' effects (Herring andRoy 2007, Charles 2009).New technologies and reduced costs could generate steep increases in the overall use of lighting and may stimulate innovative additional uses for lighting (Herring andRoy 2007, Fouquet andPearson 2006).Lighting efficiency has doubled over the past 50 years in the UK; however, per capita electricity consumption for lighting increased fourfold over the same period (Fouquet and Pearson 2006).Due to the development and use of new lighting technologies, e.g., compact fluorescent http://www.ecologyandsociety.org/vol15/iss4/art13/Table 1.Pristine sites with natural sky background brightness can be identified by measuring the sky background (astronomical unit mag/arcsec² in the visual range corresponds to a negative logarithmic scale of sky luminance) at modern astronomical observatories.Older observatories were built at dark sites close to cities. Today, these peri-urban sites are characterized by bright skies.Some observatories in the southwest of the USA were able to preserve relatively dark skies because of lighting ordinances (low pressure sodium lamps, full cut-off luminaires) in the nearby cities, resulting in a decrease of sky brightness in Flagstaff, Arizona since 1982 (Lockwood et al. 1990).The increase of the sky background depends on different model assumptions for several observatories (Garstang 1989).Indirect values can be derived from the increase of power consumption and lamp efficiency for street lighting.Note: IAO India = Indian Astronomical Observatory, NAO Japan = National Astronomical Observatory of Japan lamp (CFL), light-emitting diode (LED), organic light-emitting diode (OLED), we can expect a dramatic drop in the cost of lighting services, a desirable end in itself, but with possibly higher energy consumption and wider loss of dark nightscapes as a consequence.Technological innovations should, therefore, not only save consumers money, but also consider human health, ecological, and socioeconomic aspects.

BEYOND ENERGY EFFICIENCY
Whereas air, noise, or water pollution have been high priority policy issues for decades, light pollution remains scientifically, culturally, and institutionally in the dark.Given the dramatic increase in artificial light in recent years, we see an urgent need for research on the physiological, human health, ecological, and socioeconomic significance of the loss of the night that addresses how illumination can be improved both technically and institutionally yet having fewer adverse effects.
Managing darkness has to be an integral part of future conservation planning and illumination concepts.If not, our modern society may run into a global self-experiment with unpredictable outcomes (Fig. 1).
Any attempts to reduce light pollution run up against positive connotations of lighting which are deeply ingrained in modern societies.Culturally, light is a symbol of enlightenment, modernity, urbanity, and security (Jakle 2001).Policy initiatives against light pollution therefore need to take into consideration the many advantages of artificial lighting, real and perceived, for economic production, social lifestyles, and security while at the same time addressing its negative side effects.For this, a sound understanding of the historical, socioeconomic, and cultural reasons for the emergence and dissemination of lighting systems is needed.We then need to ask how far recent changes in attitudes, in particular relating to the environment and human health, are creating openings for a shift in policy and practice.Part of this process involves identifying and building up a coalition of interest around the light pollution issue, incorporating such diverse stakeholder groups as ecologists, astronomers, and health professionals, but also electricity utilities, lamp manufacturers, property owners, local businesses, city planners, or those concerned about nighttime security.
Thus, the research needed is transdisciplinary, i.e. it should cut across boundaries between scientific disciplines and between science, policy, and practice and should address facts, practices, and values (Wiesmann et al. 2008).The following natural, social, and engineering science questions are central to this research agenda: q What characteristics of light disrupt human health and ecological communities?1997,2000,2003).Promising options are, for example, lamps that direct their light more accurately toward where it is needed, lamps that emit light with a spectral distribution causing minimal harm, timers and sensors to turn lights on only when needed, and the consideration for lightsensitive areas, especially the periphery of residential areas, forests, parks, and shores of water bodies.The comprehensive and transdisciplinary research advocated here will result in more advanced regulations and guidelines at, in particular, the national level and the development of intelligent, i.e., adaptive and context-dependent, lighting concepts for local communities.These will help countries, regions, and cities to maximize the social and economic benefits of artificial light at night, while minimizing its negative and unintended ecological and health impacts.On this basis, future generations will be able to experience nightscapes comparable to those which Galileo knew without having to travel to the Australian Outback or the Chilean Andes.

Fig. 1 .
Fig. 1.Hypothetical impacts of exposure to artificial light at night.The main impacts, both with respect to light intensity and color spectra, emerge at the interfaces between the different physiological, ecological, and socioeconomic aspects.