To the Editor:
We read with interest the investigation of Hurley et al.1 into the relation between outdoor light at night (OLAN) and breast cancer risk. Stevens2 proposed a causal relation between artificial light exposure and breast cancer risk, and one possible consequence of this hypothesis is increased risk where outdoor lighting penetrates sleeping environments. Hurley et al. used data from the Defense Meteorological Satellite Program (DMSP) that they described as “the best available satellite imagery data to estimate OLAN.” In fact, more precise data are available and should be used to improve exposure assessment in future studies.
Astronaut photographs of Earth provide color images of individual cities at up to 10-m resolution (street level). Calibrated images of the entire Earth are produced by the Visible Infrared Imaging Radiometer Suite Day-Night Band (DNB) at 750-m resolution, corresponding to the neighborhood level. In contrast, the 2.7-km resolution data used by Hurley et al. are far less precise (Figure).
Researchers examining relations between OLAN and health effects should no longer use DMSP data, even for retrospective studies. Because street lighting typically changes on a 15- to 30-year time scale, the DNB data from 2012 provide a better indicator of past OLAN exposure than DMSP at most urban locations, for example, among study participants who have remained at the same residence for several years. Higher resolution data will also reduce the apparent correlation between degree of urbanization and light: if OLAN is a true cause of breast cancer (rather than a correlate of another urban parameter), then estimated risk in studies using DNB data should be higher than those using DMSP. Of course, measuring each participant’s light exposure would be ideal compared with using remotely sensed light data, but this is possible only in cross-sectional and prospective studies.
An interdisciplinary collaboration with researchers in remote sensing would benefit future epidemiologic studies. For example, the light -emitting diodes replacing traditional street lamps in many cities radiate a large fraction of light in the spectral range 440–500 nm, a critical range for human physiologic response, but unfortunately one to which neither DNB nor DMSP is sensitive.4 Calibration and analysis of astronaut photographs could potentially allow for the determination of associations between specific wavelengths and health risk. Better exposure assessment will lead to more precise evidence about the potential relation between ambient light at night and health effects.
We thank Helga Kuechly for producing the Figure. Image and data processing by NOAA’s National Geophysical Data Center. DMSP data collected by the US Air Force Weather Agency. Astronaut photograph courtesy of the Earth Science and Remote Sensing Unit, NASA Johnson Space Center.
Christopher C. M. Kyba
Deutsches GeoForschungsZentrum GFZ
Leibniz Institute of Freshwater Ecology and Inland Fisheries
Kristan J. Aronson
1. Hurley S, Goldberg D, Nelson D, et al. Light at night and breast cancer risk among California teachers. Epidemiology. 2014;25:697–706
2. Stevens RG.. Electric power use and breast cancer: a hypothesis. Am J Epidemiol. 1987;125:556–561
3. Sanchez de Miguel A, Gómez Castaño J, Zamorano J, et al. Atlas of astronaut photos of earth at night. Astron Geophys. 2014;55:4–36
4. Kyba CCM, Garz S, Kuechly H, et al. High-resolution imagery of earth at night: new sources, opportunities and challenges. Remote Sens. 2015;7:1–23