PolyU research finds frequent Arctic wildfires could cut snow cover by 18 days, impacting global climate and ecology

The correlation between Arctic wildfires and abnormal snow cover under global warming is of growing concern. A comprehensive quantitative assessment by researchers at The Hong Kong Polytechnic University (PolyU) has shown that increasingly frequent seasonal wildland fires across the Arctic in recent years have delayed snow cover formation by at least five days and could lead to a future 18-day reduction of snow cover duration, with implications for global ecosystems. Against the backdrop of the United Nation’s “Decade of Action for Cryospheric Sciences”, this study not only underscores the urgency of addressing climate change, but also provides critical scientific evidence to inform global climate adaptation strategies.

Snow cover in the Arctic plays a key role in the global climate system. It reflects solar radiation back into space thus keeping the surface cool, while its meltwater is an important source of freshwater. Snow is therefore central to the planet’s energy balance, hydrological cycles and weather patterns. Anomalies such as delayed snow formation or earlier melt can intensify warming, affect water supplies, and reduce forest ecosystem productivity and carbon sequestration beyond the Arctic, ultimately disrupting global ecosystems and biodiversity.

Led by Prof. Shuo WANG, Associate Professor of the PolyU Department of Land Surveying and Geo-Informatics, a core member of the Research Institute for Land and Space, and a member of the State Key Laboratory of Climate Resilience for Coastal Cities, the study is conducted in collaboration with international researchers from the University of California, Irvine, and Columbia University. The findings have been published in the international journal Nature Climate Change.

Prof. Wang elaborated, “Global warming has intensified Arctic wildland fires, making such fires increasingly frequent, larger in scale and in some cases more intense. In 2023, Canada experienced record-breaking fires, with over 45 million acres burned - nearly 10 times the average annual burned area over the past 40 years. This research aims to quantify the links among wildfires, snow formation and snow cover duration, thereby advancing our understanding of land-atmosphere interactions under climate change.”

The research team compiled long-term satellite remote sensing data of the burned area together with the start day and end day of snow cover in the Arctic from 1982 to 2018. They integrated these data with an artificial intelligence model built on the state-of-the-art XGBoost machine learning algorithms, incorporating a range of climate factors before, during and after fires (such as albedo, surface temperature and air temperature), as well as fire location, to evaluate the influence of these variables on snow cover.

The satellite data indicated that as burned area in the Arctic increased, the duration of snow cover decreased. Between 2001 and 2018, the average snow cover lasted 205 days, 10 days shorter than that from 1982 to 2000. The team further utilised the CMIP6 climate model projections to simulate future changes in Arctic wildfires and snow under different emission scenarios. They discovered that, under the high-emission scenario SSP5-8.5, the annual burned area of the Arctic could expand by 2.6 times by year 2100, while snow duration may shrink to about 130 days — approximately 18 days shorter than the historical average from 1950 to 2014.

The study also found that major wildland fires significantly delay the formation of snow cover. Through regional impact analysis, the team determined that in the first year following a major wildfire, the snow start date is postponed by more than five days compared with the three-year average prior to the fire; moreover, the larger the burned area, the longer the delay.

The research team identified the underlying physical mechanism as the deposition and persistence of black carbon on the ground after fires, which reduces surface albedo and enhances the absorption of solar radiation. This additional energy increases both land surface temperature and near-surface air temperature, thereby suppressing effective snow accumulation and ultimately postponing snow formation.

 “Wildland fires alter surface properties in the Arctic and subsequently shorten the duration of regional snow cover,” Prof. Wang added. “The reduction of snow cover further disrupts surface energy balance, prolongs land exposure, and leads to warmer, drier surfaces, which create favourable conditions for an earlier start and broader spread of fires. Such a feedback loop underscores the vulnerability of Arctic ecosystems to cascading climate impacts.”

The research team envisions these findings will not only provide solid evidence for predicting the future hydrological cycle and climate dynamics of the Arctic, but also offer scientific guidance for assessing ecosystem resilience and formulating effective climate adaptation strategies to help mitigate the chain effect of climate change.

***END***