Spatial distribution of urban artificial light at night and the ecological impacts of a streetlight retrofit

Abstract

Artificial light at night (ALAN) is a major environmental pollutant, but one that has only relatively recently received attention in ecological research. ALAN is globally prevalent and has a wide range of impacts at multiple scales in diverse ecosystems. Artificial lightscapes are currently undergoing a shift in urban areas, as old streetlights are increasingly being replaced by new energy-efficient lights that have different intensities and spectra. Therefore, there is a need to understand the impacts of ALAN in an urban landscape, and how these might change as a result of the altered lightscapes. In this thesis I use Auckland as a case study to investigate the levels of light pollution in and around the city, and how the impact of ALAN changes as a result of a streetlight retrofit. Using continuous and annual measurements of night sky brightness (sky glow), I discovered that light pollution in Auckland was high enough at some sites to mask the lunar cycle, and that areas of ecological importance and natural significance within and around the city were put at risk. Proximity to the city centre and industrial and commercial land-use zones were the most significant predictors of elevated sky glow, suggesting that regulation of ALAN in these areas and the restriction of urban sprawl near ecologically valuable areas should be a management priority. To investigate the impact of a streetlight retrofit on urban wildlife I utilised the city-wide retrofit in Auckland, where high-pressure sodium (HPS) streetlights are being replaced with white light-emitting diodes (LED). An 18-month experimental field study at residential properties across Auckland revealed that the retrofit caused mixed, slightly contradictory responses in urban wildlife. Dawn song initiation was slightly advanced for one bird species (the introduced common myna (Acridotheres tristis)) and delayed dependent on month for another (the endemic tūī (Prosthemadera novaeseelandiae), while bird community composition was significantly different between control and retrofit sites, but only in backyards. Additionally, total and introduced bird species richness and the relative abundance of three bird species were higher under LED streetlights, as was detection of ground-insect presence. No effect of the retrofit was found for frequency of nocturnal owl vocalisations, relative total abundance of trapped invertebrates or detection of small pest mammal presence. Finally, I investigated the effects of white LED ALAN on pollination of a native diurnally and nocturnally pollinated plant (mānuka, Leptospermum scoparium) by comparing invertebrate visitation rate and plant reproductive success between experimentally lit and unlit sites. Invertebrate visitation rates, seed weights and success rates of nocturnally pollinated seed capsules were higher in lit sites, but so too was capsule herbivory. Overall, this thesis emphasises the threat that ALAN poses to areas of ecological significance and its ability to act at multiple scales, affecting individuals, communities and ecosystem services. Additionally, it suggests that white LED streetlight retrofits may not have large impacts on urban wildlife, and highlights the difficulties involved in extrapolating small-scale studies to large-scale urban contexts. Given the ubiquity of ALAN and its potential and known impacts, reduction of light pollution should be a higher management priority than it is at the moment and a continued topic of focus for research. Further investigation into the effects of sky glow, the community-level impacts of ALAN, and the complexities of predicting impacts from streetlight retrofits is particularly needed.