By: Ariel Levi Simons, Ph.D. Candidate, USC

Published June 30, 2020

Humans, and some of their predecessors, have been pushing back against the night for hundreds of millennia. Up until the late 19th century, this was confined to some form of fire. Once we figured out how to generate electricity, though, things really began to take off in terms of lighting up the night sky. Originally, the only group of people who voiced any complaints about this lighting trend were astronomers. As cities grew and illumination technology improved in both brightness and efficiency, astronomers were forced to retreat farther and farther away from any human habitation in order to see anything besides the Moon.

How is all of this light affecting coastal ecosystems in and around human settlements? Are species migrating away from humans to cope? Are some species attracted to the light?

It turns out the astronomers weren’t the only ones picking up and moving away from the glow of human activity. In recent decades, ecologists, especially those working in coastal habitats, have begun to piece together a picture of the role of light in dictating how species communicate, avoid being eaten, and find their way. When humans started to flood large parts of the planet with artificial light, we were accidentally running a planetary-scale experiment. How is all of this light affecting coastal ecosystems in and around human settlements? Are species migrating away from humans to cope? Are some species attracted to the light? These are questions I’ve recently been working on with colleagues at the University of Southern California as we’ve studied light pollution along the coasts of southern California.

Why not just use satellite photos to do all of this? While we’re using nighttime satellite images of southern California, this type of data only represents the light that gets transmitted straight up through the sky from the surface. This leads to a number of limitations for using satellite-based measurements of nighttime lights as a proxy for actual exposure to artificial light at night (ALAN). First, the spatial resolution of the most common type of satellite-based measurements for ALAN is about 750 m. This means that all of the light coming out of a square measuring 4 city blocks per side gets averaged together. Second, we only get the light that reaches from the ground all the way up to the satellite. With this measurement, we don’t necessarily capture skyglow from distant sources along the horizon, or even the total exposure from streetlamps with shades (Figure 1). As a result of these limitations, we suspected that those popular photos of Earth at night were giving an incomplete picture of the lighting environment as experienced by organisms living along the beach.

However, getting all of those photos of the sky gets to be a labor-intensive process, even with an army of graduate students. So, we made the problem a bit more manageable by imaging the night sky with a hemispherical lens at 515 locations along the coast from southern Orange to northern Ventura counties. At each location, we set up our camera in a standard way, took multiple exposures of the night sky, and later extracted image data to calculate our total exposure to nighttime light from the full sky hemisphere (Figure 2). We then compared these ground-based measurements to those derived from satellite-based measurements of brightness at corresponding locations, in order to build a model to predict what the night sky should look like along the coast using just satellite-based measurements. Our goal was to build this model so that other people interested in light pollution, whether from an ecological or urban planning perspective, would be able to quickly predict exposure from readily available satellite-based sources.

From June of 2018 through February of 2019, I, along with our team at USC, friends, family, and some awesome high school students at the LA Zoo Science Magnet, drove and took night sky photos at every beach along the southern California coast. What followed was a riveting tale of programming and manuscript revisions leading to our first paper: "High correlation but high scale-dependent variance between satellite measured night lights and terrestrial exposure"

Here is a summary of key findings in our first paper:

1. Nighttime satellite photos give us pictures which average out all of the upwards glow within pixels approximately four blocks on a side. In measuring exposure to light pollution on the ground we found, in some locations, that shadows and direct glare can cause locations which are only 100 m apart to be 100 times brighter or darker than each other.

2. Satellite-based measurements can be used to make a model which is a reliable proxy for total exposure to nighttime artificial light. However, these measurements need to be combined with a model which includes light scattered by air and reflected off of surfaces in the environment.

With these findings we are now working on our next step, which is to apply our new light pollution model to directly address the ecological question of how light pollution shapes the distribution of species in coastal environments. In particular, we focused on two iconic species of the California coast: California grunions and federally threatened western snowy plovers (Figure 3). While preliminary, our findings show that ALAN does in fact play a significant role in determining where these two species are found. Grunions tending to spawn in places brighter than about ½ a full moon, and plovers avoid areas with only half that brightness.