Losing the Dark

The short version

• The average resident’s exposure to artificial light at night rose about 6% between 2014 and 2024 (population-weighted radiance 14.5 → 15.3), and roughly two-thirds of tracts got brighter. Satellite sensors almost certainly understate the real trend (Kyba et al. 2023).
• The light is wildly uneven across space: the brightest tract (Uptown Charlotte) is about 850 times brighter than the darkest rural one, and the fastest brightening is in Uptown and South End redevelopment and the same fast-growing southern suburbs that are also losing tree canopy.
• Charlotte is not unusual. More than 99% of Americans live under light-polluted skies, and about 80% cannot see the Milky Way from home (Falchi et al. 2016). That has measurable costs for sleep and health (American Medical Association, Council on Science and Public Health 2016; Chepesiuk 2009) and for wildlife (Longcore and Rich 2004).
• The burden is not equally shared. People of color in the region experience about 60% more night-light than white residents, and Black residents are more than twice as likely to live in the brightest fifth of the region. Here the axis of inequity is race and urban density, not income, which mirrors the first national environmental-justice study of light pollution (Nadybal et al. 2020).

A pollutant hiding in plain sight

Most pollution announces itself: smog you can smell, water you would not drink. Light pollution is different. It is the one pollutant we deliberately manufacture and often admire. Yet artificial light at night (ALAN) has grown into a genuine environmental force. Ground-based measurements from tens of thousands of citizen scientists found the night sky brightening about 10% per year across North America from 2011 to 2022, fast enough to double in under a decade (Kyba et al. 2023). A child born where 250 stars are visible would see barely 100 by adulthood. Satellites record a gentler rise of roughly 2% per year, but that is largely because their sensors are blind to the blue-rich light of modern white LEDs, which scatters most in the atmosphere. The satellite view, including the data in this story, is a conservative one.

Why does it matter beyond stargazing?

• Human health. Light at night suppresses melatonin and disrupts the circadian clock that governs sleep, metabolism, and mood. Blue-rich LED light suppresses melatonin roughly twice as long as warmer light and can shift the body clock by hours (American Medical Association, Council on Science and Public Health 2016). The American Medical Association has formally warned about the health effects of nighttime lighting, and reviews link chronic circadian disruption to sleep loss and elevated risks for several diseases (Chepesiuk 2009; American Medical Association, Council on Science and Public Health 2016).
• Wildlife and ecology. “Ecological light pollution” reshapes the nights that half of all animal species depend on (Longcore and Rich 2004). Artificial light disorients migrating birds, drawing them into lit skylines and toward fatal collisions; it traps and exhausts nocturnal insects already in steep decline, with knock-on effects for pollination and food webs; it disrupts amphibians, bats, and firefly courtship.
• Energy and cost. Light that shines sideways or up is light doing no useful work. Poorly aimed outdoor lighting wastes energy and money and adds carbon for no benefit.
• A shared inheritance. A dark, star-filled sky is a cultural and scientific commons. Its loss is effectively total for city dwellers: more than one-third of humanity, and four in five North Americans, can no longer see the Milky Way (Falchi et al. 2016).

How to read the metric

The Explorer measures light pollution as upward radiance captured by the VIIRS satellite sensor, in nanowatts per square centimeter per steradian (nW/cm²/sr), averaged over each census tract (Elvidge et al. 2021). Higher means more artificial light escaping to the sky. It is a satellite measurement, not a survey, so it carries no sampling error, but two caveats matter. First, upward radiance is a proxy for the skyglow a person experiences on the ground, not a direct measure of it; the world atlas that translates satellite radiance into human sky brightness required a radiative-transfer model calibrated against tens of thousands of ground observations (Falchi et al. 2016). Second, as noted, VIIRS is relatively blind to blue light, so it understates the brightening from the LED transition (Kyba et al. 2023). Read the numbers here as a floor, and as a strong indicator of where light concentrates and how it is changing.

To anchor the scale: the region’s darkest rural tracts sit near 1 nW/cm²/sr (a reasonably dark sky where the Milky Way is faintly visible); the dense urban core exceeds 130 and peaks around 220 (an inner-city sky where only the brightest stars survive and the Milky Way is long gone).

1. How bright, and getting brighter

The region’s nights are brightening. Weighted by where people actually live, average radiance rose from 14.5 in 2014 to 15.3 in 2024, about 6%, after a dip in 2019 (Figure 1). The median tract climbed from 10.7 to 12.0, and 63% of all tracts were brighter in 2024 than a decade earlier.

The regional average conceals an enormous spatial range. Figure 2 maps radiance on a logarithmic scale, and the result looks like the satellite night-lights image it is: a blazing core over Uptown Charlotte and the dense Mecklenburg ring, fading through the suburbs to near-darkness at the rural edge. The brightest tract is roughly 850 times brighter than the darkest.

Where has it grown fastest? Figure 3 maps the decade’s change. The strongest gains cluster in the redeveloping center city and the growing southern suburbs; the faint dimming at the rural edges is small in absolute terms.

The brightest places are the Uptown wards (Table 1), where radiance runs from 130 to 220. More striking is where the light grew fastest. The same Uptown and Gateway tracts that top the brightness list also rose the most, by 50 to 80 units in a decade, as the center city’s building boom lit up. Just behind them sits Providence / Blakeney, one of the fast-growing southern suburbs that our companion story found losing tree canopy to new subdivisions. Development brings both more rooftops and more light.

By county (Table 2), Mecklenburg is in a class of its own at 26.6, more than twice any other county, followed by the suburban ring (Cabarrus, York). The rural counties remain comparatively dark. Almost every populated county brightened over the decade; the few that dimmed are small, rural, and lightly populated.

What is driving it: development

Why are the region’s nights brightening, and so unevenly? The pattern points to development: the same construction that adds rooftops, roads, and parking lots also adds light. Two development signatures stand out (Figure 4).

New housing. Tracts that built the most new homes over the decade saw the largest light-pollution increases. The top fifth by new-home construction gained a median of about +2 nW/cm²/sr, while the tracts that built the least were flat or slightly darker. Light tracks the building boom.

Tree-canopy loss. The same relationship appears in reverse for trees. Tracts that lost the most canopy gained the most light (a median of about +1.5), while tracts whose canopy held steady or grew saw essentially no change. Clearing land for development removes the tree cover that once darkened the night and installs the lighting that replaces it, the twin signatures of the same growth, and a direct link to the region’s documented canopy loss.

These are preliminary, descriptive findings from ongoing analysis for a forthcoming working paper, which models the drivers of light-pollution change more formally, accounting for baseline brightness, income, neighborhood composition, employment density, and spatial structure. That fuller analysis will be released as a companion academic paper and referenced here; the association above is the headline it is built on.

2. What the light costs

Put the region’s 3.1 million residents on the brightness ladder and the exposure is broad. About half of all residents (47%) live in a tract brighter than the regional median, and roughly 508,000 people live in the brightest fifth of tracts, where the median tract sits near 37, effectively an inner-suburban to urban sky in which the Milky Way is invisible. Only the darkest quintile, about 608,000 mostly rural and exurban residents, enjoys anything close to a natural night.

This is where the regional numbers meet the global literature. The Charlotte region is a normal American metro, and normal American metros sit almost entirely under light-polluted skies (Falchi et al. 2016). For the hundreds of thousands of residents in its bright core, that means the documented downsides are not abstract: disrupted sleep and circadian rhythm (American Medical Association, Council on Science and Public Health 2016; Chepesiuk 2009), and an urban ecology in which migrating birds, insects, and other nocturnal wildlife contend with a night that never fully arrives (Longcore and Rich 2004). It also means the region carries a slow, compounding loss most people never notice, because the dark that is missing cannot be seen.

There is a quiet paradox here. Light at night is a byproduct of exactly the growth the region celebrates: the cranes over South End, the subdivisions pushing south. The same places gaining people, buildings, and economic energy are the places losing their nights, and often their trees along with them.

3. An unequal night

Light pollution, it turns out, is not distributed the way you might guess. Nationally, the first environmental-justice study of ALAN found that Black, Hispanic, and Asian residents, and lower-income communities, live in significantly brighter areas than white and higher-income residents (Nadybal et al. 2020). The Charlotte region fits that pattern on race, sharply.

Weighting each group by where its members live (Figure 5), the average Black resident is exposed to radiance of 20.0, versus 12.5 for the average white resident, about 60% more. Hispanic (19.7) and Asian (19.6) residents are close behind, and residents below the poverty line (17.4) are also more exposed. The gap widens at the top of the scale: 26.5% of Black residents and 25.7% of Hispanic residents live in the brightest fifth of the region, compared with 10.6% of white residents (Figure 6).

But the axis of inequity here is race and urban density, not income in the simple sense. Radiance correlates strongly with the share of people of color in a tract (Spearman ρ = 0.54) and even more strongly with population density (ρ = 0.82), yet barely at all with median income (ρ = −0.03). Sorted into income thirds (Figure 7), exposure is highest in lower-income tracts but is not lowest in the middle: it rises again in the highest-income group, lifted by the bright, affluent core of Uptown and South End. The mechanism is spatial. Artificial light concentrates in the dense urban core and its commercial and industrial corridors, and in this region, as in most, people of color are disproportionately housed there. Density lights the sky; segregation decides who lives beneath it.

There is a greener dimension to the inequity, and it ties directly to the region’s development. Sort residents by the tree canopy of their neighborhood and the gap is stark: those in the least-green tracts are exposed to about 2.5 times more light at night than those in the leafiest ones (a population-weighted 23.0 versus 9.0 nW/cm²/sr). Nearly a million people live in low-canopy neighborhoods under the brightest nights, while a comparable million in high-canopy neighborhoods keep the darkest. Tree cover and darkness go together, and their absence compounds: the same neighborhoods clearing canopy for development are gaining the most light, so the loss of shade by day and the loss of dark by night fall on the same residents.

The equity finding is therefore both real and specific. It is not that poor neighborhoods are singled out for floodlights. It is that the region’s brightest ground, its urban core and industrial edges, is disproportionately home to Black, Hispanic, and Asian residents, so the health and ecological costs of the night sky’s loss fall first on them.

Bending the curve

Light pollution is unusual among environmental problems in that the remedy is cheap, fast, and reversible. Unlike carbon already in the atmosphere, skyglow disappears the instant a light is shielded, dimmed, or switched off. DarkSky International’s widely adopted Five Principles for Responsible Outdoor Lighting are the template (DarkSky International 2020):

1. Useful: light only where there is a clear need.
2. Targeted: direct light down, to the ground, with full shielding. Full shielding alone can cut skyglow by 50% to more than 90% (DarkSky International 2020).
3. Low level: no brighter than necessary.
4. Controlled: use timers, dimmers, and curfews so light is on only when useful.
5. Warm-colored: cap correlated color temperature at 3000 K or lower (2200–2700 K where possible) to limit the blue light that most affects health, wildlife, and skyglow, a threshold the AMA also endorses (American Medical Association, Council on Science and Public Health 2016).

Cities have shown it works. Flagstaff, Arizona, the world’s first International Dark Sky City, has held its skyglow nearly flat for decades through shielding standards, amber lighting, and caps on total light per acre, even as it grew (DarkSky International 2024). Tucson converted its street lighting to shielded, dimmable LEDs and measurably reduced its skyglow. Pittsburgh and Fort Collins have adopted dark-sky ordinances for public lighting. None of these are dark rural outposts; they are functioning cities that decided their nights were worth protecting (DarkSky International 2024).

For a fast-growing region like Charlotte, the priorities are clear: adopt a modern outdoor-lighting ordinance for new development and public projects (full shielding, 3000 K caps, curfew-dimming); specify warm, shielded, dimmable fixtures in the region’s ongoing LED streetlight conversions rather than the harsh blue-white default; and treat lighting as part of the same growth-management conversation that governs tree canopy and stormwater. The cheapest way to keep the nights is to build the next decade’s lights correctly the first time.

What it means

The Charlotte region’s nights are getting brighter, unevenly, and the trend is likely stronger than the satellites can see. The costs, to sleep and health, to wildlife, and to a shared night sky, are real if easy to ignore, and they land first on the region’s Black, Hispanic, and Asian residents, who are concentrated where the ground glows most. None of this is inevitable. Light pollution is the rare environmental harm that can be undone almost overnight, one shielded, warmer, well-aimed fixture at a time. The dark is not gone. It is only switched on.

Data and methods

Light pollution. Area-weighted mean upward radiance per census tract (nW/cm²/sr) from the Earth Observation Group’s VIIRS annual nighttime-lights composites (VNL V2) for 2014, 2019, and 2024, as compiled in the Carolinas Regional Explorer (Elvidge et al. 2021). VIIRS measures light escaping to the satellite, a proxy for, not a direct measure of, ground-level sky brightness, and it is relatively insensitive to blue-wavelength light, so it likely understates recent brightening (Kyba et al. 2023; Falchi et al. 2016). It is a satellite measurement and carries no sampling margin of error.

Population weighting. Regional and group averages are population-weighted: each tract’s radiance is weighted by the population (or group population) living in it, so the figures describe the light the average person experiences, not the average acre. Group populations are tract population multiplied by each group’s share (race/ethnicity, poverty, age) from the American Community Survey via the Explorer. This is the standard approach in distributive environmental-justice analysis (Nadybal et al. 2020).

Equity analysis. “People of color” is the non-white share of population. Group exposure is the population-weighted mean radiance for that group across all 752 tracts; “brightest quintile” tracts are the top 20% by 2024 radiance. Correlations are Spearman rank coefficients (descriptive, not causal). Income thirds are tertiles of tract median household income.

Reproducibility. All processing and figures were produced in R (analyze.R, figs.R) from the Explorer’s public data contract; the tract-level table and county/equity summaries are in this story’s data/ folder. Explore the underlying Light pollution (night lights) indicator, alongside income, race, and density, for any tract in the Carolinas Regional Explorer.

References