---
title: "When the World Stayed Home in 2020, Its Fires Fell 12.3 Percent, Though Not Everywhere"
description: "A Chinese Academy of Sciences team used COVID lockdowns to measure how people both start fires and put them out"
author: "Jan Sokol"
published: 2026-11-15T12:07:34Z
modified: 2026-11-15T21:43:13Z
url: https://rews.cc/a/when-the-world-stayed-home-in-2020-its-fires-fell-12-3-perce-88baf5
language: en
tags: ["climate", "covid", "wildfires", "research", "environment", "science"]
publisher: "Rews (https://rews.cc)"
---

# When the World Stayed Home in 2020, Its Fires Fell 12.3 Percent, Though Not Everywhere

*A Chinese Academy of Sciences team used COVID lockdowns to measure how people both start fires and put them out*

By Jan Sokol · November 15, 2026 · https://rews.cc/a/when-the-world-stayed-home-in-2020-its-fires-fell-12-3-perce-88baf5

## In brief

- A Chinese Academy of Sciences team used 2020 COVID lockdowns as a natural experiment to measure human influence on fire
- After adjusting for weather, areas under lockdown had about 12.3% fewer active-fire detections than the same weeks in 2013–2019
- Forest fires fell about 27.7% and grassland fires 8.4%; without the weather adjustment the overall drop was 5.8%
- Fires rose in Honduras (70.7%), Colombia (31.4%) and India (25.4%), which researchers infer was due to weaker suppression
- The authors say the figure is a net effect, the study is associational, and fire policy should differ by region

In the first half of 2020 governments ordered their people indoors, and the people mostly went. Ministries published decrees. Telephones in pockets kept a record of where their owners stopped going. Above all of it the Suomi NPP satellite, operated by NASA and the National Oceanic and Atmospheric Administration, went on doing what it always does, which is to look down and count heat.

Those two records, the human one and the thermal one, are the material for a study led by Jianghao Wang, a professor at the Chinese Academy of Sciences, which [Ars Technica’s Yook Ji Hun reported on](https://arstechnica.com/science/2026/11/how-the-worlds-fires-changed-while-humanity-hit-pause/). The paper, “Unraveling the dual roles of human activity in shaping global fire regimes,” appeared in *Cell Reports Sustainability* on October 9, and according to the [Chinese Academy of Sciences’ release](https://www.eurekalert.org/news-releases/1147334) Wang led it with Ge Yong at the Institute of Geographic Sciences and Natural Resources Research. Its question is old. How much of the world’s burning is our own doing?

The question matters. Between 2019 and 2020 bushfires crossed southeastern Australia and destroyed thousands of homes. In 2025 fire spread through Los Angeles. Lightning, drought and wind all start fires or feed them. Of all the causes, though, only people can also choose to prevent one.

## A laboratory nobody wanted

Normally the human hand can’t be separated from the weather. Seasons, harvest calendars and dry spells shape both how fire behaves and how people behave, so the two move together. The lockdowns broke that link. “The early-2020 lockdowns produced an abrupt, large-scale drop in human mobility, and the timing differed across countries while the weather did not change in the same way,” Wang said. “This gave us a natural experiment.”

![A circular glass enclosure contains cabins linked by winding paths amid dense evergreen forest, with small wildfires burning along its edges.](https://rews.cc/img/d852cb4a9b5d13297d5cb9eafe7d0b95a9f3cc58.webp)

Disasters have become laboratories of this kind. A war has lately been read the same way, for [what it did to the world’s carbon curve](https://rews.cc/a/the-iran-war-bent-the-world-s-carbon-curve-down-and-the-bend-1816f7). Here the team dated each region’s restrictions using lockdown orders and phone location data. The satellite’s VIIRS sensor picks up hot spots at a resolution of 375 meters, so a single large fire can register as many detections. The sensor also cannot distinguish a wildfire from a deliberate burn, and so it says nothing about intent. The researchers compared detections during the 2020 restrictions with the same weeks in 2013 to 2019 and adjusted for differences in weather.

The result: about 12.3 percent fewer active-fire detections in areas under lockdown.

Without the weather adjustment the drop was only 5.8 percent, which was not statistically significant. Once weather was accounted for, the decline more than doubled. One possible reading is that 2020’s weather favored fire, and this concealed part of what people had stopped doing.

Forests showed the steepest fall, about 27.7 percent. Grasslands fell 8.4 percent. Shrublands, savannas and croplands barely changed. The researchers conclude that people are a major cause of fire in forests and grasslands. The academy’s release gives a 22.8 percent drop for the United States.

## The two hands

This is the difficult part. The person whose campfire gets away and the crew that puts the fire out both stayed home in the same spring. A stay-at-home order does not separate them. Wang is careful on this point:

![Two hands hover over a forest, one holding a smoking match near flames and the other dripping water from a green hose.](https://rews.cc/img/a3c2bed279a3cf74a33d8beed3ef64bbfbe1ecc7.webp)

> The 12.3 percent is a net effect, not the share of fires caused by humans; opposing changes in ignition and suppression mean it does not capture the full magnitude of human influence.

Fewer ignitions and weaker suppression partly cancel each other out. The real size of the human role is larger than the figure that survives that cancellation, and the study cannot say how much larger.

## Where the fires rose

In much of South America, southern Africa and northeast Asia, fire increased during the lockdowns. Detections rose 31.4 percent in Colombia, 25.4 percent in India and 70.7 percent in Honduras.

The team suspects that fire suppression broke down in those places, partly because earlier studies linked weaker fire management to more fires in Colombia and Madagascar. Wang does not claim more than that: “It is an inference, not a direct observation.” According to the academy’s release, the study “did not directly measure changes in firefighting or patrol activity.”

The strongest evidence comes from 0.5-degree grid cells. In countries with fewer fires, fire fell almost in a straight line as mobility fell. In countries with more fires it rose slowly and leveled off at about 27 percent on average. The researchers read the first pattern as fewer people starting fires and the second as fewer people putting them out. Wang allows that other factors, such as changes in local land use, could also be involved.

![A vintage-style illustration shows an empty fire lookout tower overlooking forested mountains with scattered wildfires and rising smoke beneath a coral sky.](https://rews.cc/img/c9729d68302b24e6363ae28256825d0b4baa5728.webp)

## The burns that were never lit

Fire is also a management tool. Land managers set controlled burns to clear the fuel that would feed a worse fire later. In 12 countries with records of prescribed burning, the area burned this way in 2020 was roughly half the 2013 to 2019 average. That comparison is imperfect, because the location and size of these burns change from year to year with land-management cycles. When the areas subject to prescribed burning were removed, the decline in those 12 countries shrank from 45.3 percent to 27.8 percent. A drop in detections, the release notes, does “not necessarily mean fewer harmful wildfires alone.”

## What the method cannot see

Phone data may miss rural areas where coverage is thin, and those are often the places that burn. The analysis covered only the first half of each year, so regions whose fire season comes later, such as the western United States, may be underrepresented. Wang describes the design as strong associational evidence but not a controlled experiment, and says the study does not evaluate any fire policy.

Even so, the results do not support a single policy for every country. Where fires fell, ignition seems to have been the main human effect, which argues for public education against accidental fires and for keeping useful prescribed burns. Where fires rose, the priority is keeping firefighting crews and patrols working through a crisis. [Wang told Earth.com](https://www.earth.com/climate/covid-19-lockdowns-changed-global-fire-activity-in-ways-researchers-didnt-expect/) that the team now plans to separate the different kinds of fire. It also wants to find out whether the fuel left unburned in that quiet spring of 2020 fed the fire seasons that came after.
