New Paper: A Compendium of Photopigments and Visual Spectral Response Curves

One of the ways to reduce the adverse impacts of light at night is to adjust the spectrum of the lights, but let me say at the outset of talking about this new paper that it is more important to reduce the intensity, control the direction, and reduce the duration of lights first, and then pick the most suitable spectrum. As some other research in the pipeline will show, the color doesn’t matter if the lights are dim enough.

I conceived of this new paper, published in Basic and Applied Ecology in a special issue edited by Eva Knop and Davide Dominoni, years ago, when working on 2018 paper (https://onlinelibrary.wiley.com/doi/abs/10.1002/jez.2184) that formalized a way to predict impacts of lights on species with different spectral sensitivities. For that paper, we had something like a dozen spectral response curves. I knew there were more buried in the literature but we ran with what was available in the interest of moving the conversation along. Then, when the California Department of Transportation approached me in 2019 about doing a study on the effects of LEDs on terrestrial wildlife, I knew finding more curves to better describe the “average” visual response of wildlife groups would be part of the project. Four years later and it is finally published, the product of many a pandemic evening scouring the old literature for photopigment information and visual response curves. Here’s the formal abstract:

The presence and proportions of photopigments, which are responsible for the visual and physiological effects of light, vary between taxonomic groups. This leads to differing wavelength sensitivities ranging from ultraviolet (UV; <400 nm) to infrared (IR; >780 nm) and complicates the balancing of spectra used for outdoor lighting to maximize human visual performance while mitigating light pollution effects on wildlife. I developed a database of spectral response information for terrestrial wildlife to create generalized spectral response curves by taxonomic phylum, class, and order. Existing data on species visual sensitivity were collected from previously published research that used behavioral responses, electroretinograms (ERGs), and reflectance within the eye. Resulting summaries of photopigment peak sensitivities (n=968) and sensitivity curves (n=177) allow for general observations. Overall, longer wavelengths provide the highest possibility for supporting human visual performance at night while reducing intrusive overlap with the vision of other species, because many taxonomic groups are sensitive to light in the blue and into the ultraviolet. Comparison of average response curves at the class level and the spectral power distribution of lamps suggests that spectral tuning might reduce the apparency of the lowest correlated color temperature (CCT) lamps to insects, spiders, and non-human mammals the most, with substantial but smaller reductions for reptiles, birds, and amphibians. Spectral tuning, most simply by reducing CCT, should be considered an additional benefit to be used in concert with other mitigation measures such as dimming, shielding, and part-night lighting.

The figures tell the story (click to enlarge).

I was frankly sort of surprised that a paper like this one did not already exist — I’m not trained as a visual ecologist and I’d figured it would have already been done. But it turns out that was not the case and I’ve simply unlocked a bunch of information that was there in old scanned images but not in a usable digital format. I’ve posted the entire datasets, including the oil droplets associated with photopigments for those interested, on GitHub and would happily take additional contributions from people (aquatic species need to be done) and update the resource for public use.

The big take-homes are the variability between taxonomic groups in the average visual sensitivity, as seen in this figure.

Notwithstanding the variation in these visual responses, when I used our (Longcore et al. 2018) rapid assessment technique to predict the effect of various lighting sources on the different classes, it shows quite convincingly that if you want to reduce the visual apparency of the light while still being visible to humans, you should avoid the blue end of the spectrum and use more light at the longer wavelengths. That’s illustrated in this figure showing an index of predicted effects for each of the classes and their average:

I hope this paper helps move along the discussion of spectrum in light pollution mitigation and can be used by visual ecologists as they investigate animal perception. For example, I didn’t use any phylogenetic information to interpret this dataset and extrapolate from it to other species, because I haven’t learned how. But if anyone wants to do it, I’ve made the data available and I’d be happy to collaborate.

I have to send out thanks to the three anonymous reviewers, who each had very complimentary things to say about the paper (“a remarkable review containing a wealth of information relevant for visual ecology and outdoor lighting design,” “I expect that this paper will inspire a great deal of future research and that the author’s database (which they promise to update as new data come out) will become a vital resource for people in our field,” “an ambitious undertaking [that will] likely provide a useful resource for the field”) that were followed by extensive, detailed, constructive comments that I know improved the final product. This time, Reviewer #2 was top notch, as were #1 and #3, so thank you to them and to the editors for recruiting them.

The paper is open access and the data are open access, so go have a look if it interests you.

Thank you to Simon Bisrat at California Department of Transportation and the technical advisory board for recommending and funding this research as part of the “Effects of LED Lighting on Terrestrial Wildlife” project.

Longcore, T. (2023). A compendium of photopigment peak sensitivities and visual spectral response curves of terrestrial wildlife to guide design of outdoor nighttime lighting. Basic and Applied Ecology 73:40-50. doi: 10.1016/j.baae.2023.09.002