All updates

NASA Black Marble & satellite-derived indicators ·

Asia night lights, October 2026: haze dims Malaysia 24%, India dips on later Diwali

Satellite night-light readings for Malaysia fell 24% in October 2026 from September, consistent with haze from Indonesian fires. India fell 5% from a year earlier because Diwali moved into November.

A data update from FactIQ, written from NASA Black Marble & satellite-derived indicators figures. What is FactIQ?

Key figures

Malaysia mean night-light radiance, October 2026
1.25down from 1.65 in September (−24%) and 1.41 a year earlier2022-11-012026-10-01Nighttime lights — Malaysia, mean monthly radiance
Kuala Lumpur mean night-light radiance, October 2026
62.2down from 83.3 in September2022-09-012026-10-01Nighttime lights — Kuala Lumpur, Malaysia, mean monthly radiance
India mean night-light radiance, October 2026
1.49down from 1.57 in October 2025 (−5%)2022-11-012026-10-01Nighttime lights — India, mean monthly radiance
China mean night-light radiance, October 2026
0.637down 8.7% from 0.698; weakest October since 20222022-11-012026-10-01Nighttime lights — China, mean monthly radiance

Haze from Indonesian fires dimmed Malaysia's lights in October

Radiance measures how much light satellites see at night. It is reported in nanowatts per square centimetre per steradian (nW/cm²/sr), and analysts use it as a rough guide to economic activity. In the Black Marble night-lights data, Malaysia's mean radiance fell to 1.25 in October 2026. That was down from 1.65 in September (−24%) and from 1.41 a year earlier. The drop is consistent with smoke from Indonesian fires blocking the satellite's view.

  • Kuala Lumpur: 62.2, down from 83.3. This is its lowest recent reading.
  • Selangor: 11.3, down from 13.8.
  • Johor: 3.66, down from 4.53.
  • Riau Islands in Indonesia, near Singapore: 1.57, down from 2.16.

The glow from Indonesia's fires faded, but the smoke over Malaysia remained

Indonesia fell to 0.54, from 0.93 in September, when fires pushed the reading up. The provinces hit by fire are back near normal. Central Kalimantan read 0.11, against 0.63 in September. West Kalimantan read 0.15, against 0.52, and Riau 0.32, against 0.89. Jakarta was flat at 47.8.

India's drop from a year earlier is mostly about when Diwali fell

India read 1.49, down from 1.57 in October 2025 (−5%). Diwali fell on 20 October last year but on 8 November this year. Delhi fell to 35.9 from 43.9 (−18%), and Uttar Pradesh fell to 2.47 from 2.83. Maharashtra (1.74 vs 1.69) and Gujarat (1.95 vs 1.85) were higher than a year earlier. November readings are expected to rebound against last year.

China's weakest October since 2022 came from its north and west

China read 0.637, down 8.7% from 0.698. Coastal areas were flat: Shanghai 22.2 vs 22.4, Guangdong 3.86 vs 3.91, Jiangsu 6.02 vs 6.05. The declines were in Inner Mongolia (0.117 vs 0.152), Xinjiang (0.157 vs 0.203), Hebei (1.43 vs 1.69) and Shandong (2.63 vs 2.91). The national figure is an average over land area. Large, thinly lit regions therefore move it a lot, so it should not be read directly as a sign of weaker activity. Elsewhere, Vietnam rose to 1.73 (+13% from a year earlier) and Korea rose to 5.46 (+4%). Japan was flat at 1.44. Myanmar (0.079) and Sri Lanka (0.62) fell from their highs in the monsoon season.

Source: NASA Black Marble & satellite-derived indicators. See all SAT data in FactIQ.

About FactIQ and these updates

  • What FactIQ is. FactIQ is a database of official statistics from government agencies, central banks and international organisations in the US, China, India, Europe and elsewhere. AI assistants such as Claude and ChatGPT connect to it to look up the exact figures and cite the agency that published them. It also holds what companies report in their filings and say on their earnings calls.
  • How an update is made. When NASA Black Marble & satellite-derived indicators publishes new figures, FactIQ loads them and an AI model writes a short summary of what changed. Before it is published, code checks that every number in the summary matches FactIQ's record of the release, and a second review checks that nothing is overstated. Values that look like errors in the source data are left out.
  • Where the numbers come from. Each key figure links to the full series, with its source, units and history.

More about FactIQBrowse the dataUse it in Claude or ChatGPT