Where Italy shines at night
Mean night-time radiance of every Italian municipality, from NASA’s Black Marble annual product. Drag the slider or press play to watch 2012–2024; click a municipality for its series.
No municipality selected. Click the map or use the search box.
Growth and decline by inner-area class
Population-weighted mean radiance by SNAI 2020 class, indexed to 2012 = 100. Click a class to highlight its municipalities on the map.
SNAI classes (Agenzia per la Coesione, 2020): A poles, B inter-municipal poles, C belt, D intermediate, E peripheral, F ultra-peripheral, by travel time to essential services.
Does night light measure inequality?
Working paper, in preparationNight-light dispersion inside a municipality has been proposed as a proxy for income inequality. Replicating and extending that idea on all Italian municipalities gives a more precise answer: light reads the structure of a territory and, inside large cities, relative affluence — not municipal income inequality.
Median within-city correlation between a zone’s night light and its real-estate prices (OMI zones, 428 municipalities): brighter zones are pricier in 85% of cities, and the gradient survives a control for distance to the centre (+0.54).
Share of municipalities where brighter census sections are significantly less deprived: 81% of cities with 100,000–250,000 residents, 11% of municipalities below 5,000. The intra-urban light signal is a big-city phenomenon.
Share of the municipal light-Gini explained by the geometry of the built-up area versus by income inequality. The light-Gini is a settlement-structure index, not an inequality index.
Conditional versus marginal R² when the published Spain method (PLOS ONE, 2025) is replicated on Italy: the apparent fit comes from differences between municipalities, not from the light–income link.
Results from the author’s ongoing project Light inequality (Italian municipalities and census sections, VIIRS/Black Marble 2013–2024, MEF income data, OMI real-estate zones). Preprint link will be added here when available.
Data & method
Source
NASA Black Marble annual product VNP46A4 (Suomi NPP VIIRS, 15 arc-second grid, ~500 m), years 2012–2024, all-angle snow-free composite. Reference: Román et al. (2018), Remote Sensing of Environment, 210, 113–143.
Aggregation
Mean radiance (nW/cm²/sr) of the pixels falling inside each municipality, on the ISTAT administrative boundaries of 1 January 2021 (7,903 municipalities). Inner-area classes from the SNAI 2020 classification. Aggregate curves are population-weighted (2021 population).
What light measures, and what it does not
- Artificial light at night: mostly public lighting, roads and commercial areas. It tracks settlement structure and, between places, economic activity.
- It is not economic activity itself. Street-lighting conversions to LED (whose blue-rich spectrum the VIIRS sensor under-detects) lower measured radiance without any decline in activity: part of the decrease in large cities after 2015 is this.
- Bright cores saturate less in Black Marble than in older DMSP data, but small differences in dim rural areas remain noisy.
- Annual composites remove most cloud and snow effects; moonlight and stray light are corrected by NASA.
Processing
Data processed by the author from the NASA products (zonal statistics on the municipal boundaries). The full pipeline, part of a larger municipal geospatial database, is available on request. The map shows only the annual municipal means; no other variables are published here.
Citation
Licences
NASA Black Marble: public domain. ISTAT boundaries: CC BY 4.0. Basemap: © OpenFreeMap, © OpenMapTiles, data © OpenStreetMap contributors. Map built with MapLibre GL JS. App code: MIT, github.com/agapitosantangelo/nightlights.