Luginbuhl, Walker, and Wainscoat reply: We appreciate the concern of Mark Rea, John Bullough, and Jennifer Brons for aspects of light pollution other than sky glow, and we support increased awareness and reduction of all forms of light pollution and energy waste. Our article was explicitly about the impacts on astronomical observatories and science, though to a large extent the careless lighting practices that increase sky glow over observatories also cause light trespass, glare, and energy waste. Nevertheless, many techniques used to protect observatories, in particular the full shielding of fixtures to control direct uplight, have wide applicability for national parks, coastlines, and anywhere citizens want to preserve or restore their ability to see stars in the night sky.
Of course, distance and lighting amounts are critical factors. If direct uplight were a tiny fraction of total light output, approaches other than improving shielding might be more productive for reducing sky glow. But the observational evidence does not support the low uplight fraction Rea and coauthors suggest. Using measurements of how sky glow varies with distance and inventories of light fixtures, other researchers find uplight percentages of 8-15%. 1–3 So direct upward emission, whatever the distance or amount, dominates sky glow. Eliminating it in communities with typical shielding and near-ground blocking would reduce sky glow by 35-75% for observation distances of 50-200 km.
The writers’ suggestion that vegetation could be intentionally used to decrease light pollution impacts may have some merit, though vegetation is not something on which astronomers or lighting designers have much influence. In any case, the modeling shows that the impact of direct upward emissions remains disproportionate even in the presence of substantial blocking by structures or vegetation.
Whatever uses the outdoor site-lighting performance metric may have for evaluating other aspects of light pollution, it is not a good metric for evaluating sky glow. It contains no information about the direction light is propagating away from a lighting installation, nor does it distinguish between upward- and downward-directed light. As our work and that of others 1 , 2 , 4 demonstrate, direction is critical in considering most aspects of light pollution.