Abrupt, climate-induced increase in wildfires in British Columbia since the mid-2000s

British Columbia (BC), Canada, has witnessed a dramatic escalation in wildfire activity in recent years, with four of the most severe fire seasons in the last century occurring within the past seven years (2017, 2018, 2021, and 2023). These events were characterized by extreme weather, exemplified by the 2021 heat dome that shattered temperature records across western Canada, fueling wildfires from central Canada to the Pacific Coast. The unprecedented heat, culminating in a record temperature of 49.6°C in Lytton, BC, led to widespread devastation and numerous evacuations. The 2023 wildfire season has already surpassed previous records in area burned, highlighting the urgent need to understand the underlying causes of this dramatic shift. While BC's fire activity in the 20th century was comparatively low or moderate, with infrequent very large fires (>50,000 ha), the interplay of climatic and anthropogenic factors has shaped wildfire trends. The early 20th century saw frequent, smaller fires due to droughts and human activities like logging, mining, and land clearing. However, since the 1950s, a cooler and wetter climate, coupled with increased fire suppression efforts, led to a steady decrease in wildfire activity until approximately 2000. Existing studies unanimously project significant increases in BC's future fire activity due to projected warming and drying trends. The recent surge in fire activity, therefore, is not entirely unexpected. What is surprising, however, is the early onset of this increase around the year 2000—decades earlier than anticipated—and the extreme magnitude of the fire seasons, with three of the past seven years experiencing over 1 Mha (1% of the land area) burned. This study investigates the contemporary trends in wildfire activity in BC, examining the interplay of climate change and bottom-up factors to determine if BC has entered a new fire epoch. The study analyzes century-long trends in fire activity against annual climate variables and explores how climatic and non-climatic factors have shaped current fire regimes.

Previous research has established a link between climate change and increased wildfire activity globally and in various regions of North America. Studies have highlighted the overriding role of temperature increases on the recent uptick in fire activity, particularly in western North America. However, other aspects of weather, including precipitation, relative humidity, and wind, also influence fuel moisture and flammability. Studies in the western US have convincingly demonstrated the sharp increase in fire activity is largely a result of prolonged annual moisture deficits. While the impact of anthropogenic climate change on fire weather has been established, the specific influence of these changes on fire activity in BC necessitates a region-specific investigation. The long-term impact of fire suppression on wildfire potential in North America is also increasingly being recognized. Decades of fire suppression, coupled with the suppression of Indigenous cultural burning practices, have contributed to increased forest density in some areas of BC, subsequently increasing the likelihood of large, high-intensity wildfires. The study acknowledges the complex interplay between climate change and pre-existing conditions like forest density, insect outbreaks, and land-use practices. These factors need to be considered to obtain a holistic understanding of BC's recent fire regime shift.

This study employed a comprehensive methodology to investigate the trends in wildfire activity in British Columbia and their relationship to climate change and other factors. The research integrated historical and projected climate data, wildfire perimeter data, fire suppression costs, and forest disturbance maps to paint a complete picture. **Climate and Fire Data:** Historical and projected climate data (1900–2100) were obtained for BC and three nested zones (Central, Coastal, and Northern), along with data from five selected US states and Canadian provinces. ClimateNA v7.3 was used for data interpolation at a 50-km resolution, incorporating three CMIP6 Shared Socio-economic Pathway (SSP) scenarios (ssp245, ssp370, ssp585) for future projections. To minimize collinearity, seasonal mean temperature, total precipitation, Hargreaves climatic moisture deficit (CMD) for spring and summer, and the annual number of frost-free days were analyzed. Spearman correlation tests were used to assess associations among variables. Mapped wildfire perimeters from the Canadian National Fire Database (1919–1984) and the National Burned Area Composite (1985–2021) provided fire data. Adjustments were made to account for inaccuracies in historical fire polygon data. Fires smaller than 20 ha were excluded due to inconsistent reporting. Annual area burned, number of fires, and the day of the year when cumulative area burned reached 1%, 2%, 5%, and 10% of the annual total were calculated. Fire statistics for areas outside BC were obtained from various sources including the Canadian National Fire Database, Fire and Resource Assessment Program, and Monitoring Trends in Burn Severity. Fire suppression costs (1919–2021) were collected from BC Ministry of Forests reports and adjusted for inflation. Evacuation data were drawn from the Canadian Wildland Fire Evacuation Database. **Trend Analysis:** Segmented linear regression was used to analyze trends in climate and fire variables from 1919–2021. A modified Mann–Kendall trend test was employed to assess the significance of trends. **Forest Disturbance Maps and Statistics:** Data on harvested cutblocks and mountain pine beetle infestations were obtained from the CanLAD dataset. Fire history data and cropland information were also incorporated. Spatial statistics of forest disturbances and land cover were summarized at the provincial and ecozone levels. The study also presented comprehensive data availability and code availability information for reproducibility.

The study revealed several key findings regarding the increase in wildfire activity in British Columbia: **Climate Change as a Primary Driver:** Analysis showed a strong correlation between increased wildfire activity and changes in climate variables. The wetting trend observed in BC until approximately 2005 reversed, with a drying trend observed since then. This drying trend is primarily attributed to rapid warming and increased evaporative demand, even though total precipitation levels remain relatively high. The study found that an increase in precipitation of at least 15% per degree of warming is needed to offset the increased biomass flammability; however, the observed precipitation increase was significantly lower than this threshold. This supports the significant role of climate change as a primary driver of the increased wildfire activity. **Spatiotemporal Variability:** The study highlighted spatiotemporal variability in the effects of climate change on wildfire activity across BC's three distinct zones (Central, Coastal, and Northern). While the Central zone experienced the most dramatic increases in fire activity, largely due to increased temperature, reduced summer precipitation, and higher summer climatic moisture deficits, the Coastal zone, despite having the highest temperature increase and increasing CMD, did not experience a corresponding increase in fire activity. This is attributed to inherently less flammable conditions due to less lightning, high forest canopies less prone to crown fire, and historical heavy industrialization practices. **Influence of Bottom-Up Factors:** The study emphasizes the important role of bottom-up factors (non-climatic biophysical and anthropogenic factors) which influence wildfire activity. These factors include: * **Legacy of past disturbances:** Previous wildfires, insect outbreaks (particularly the mountain pine beetle epidemic), and land-use practices have significantly altered fuel loads and created conditions more conducive to fire spread. * **Fire suppression policies:** Decades of fire suppression practices, coupled with a decline in Indigenous cultural burning, have resulted in a fire deficit in some areas, contributing to increased forest density and the likelihood of severe wildfires. * **Human ignitions:** Human activities, although a smaller percentage compared to lightning strikes continue to be an important factor of wildfire occurrences. **Statistical Correlations:** The study found significant correlations between wildfire activity (area burned and number of fires larger than 20 ha) and several climate variables (CMD, temperature, precipitation, and number of frost-free days). These correlations further support the link between climate change and increased wildfire activity. **Future Projections:** CMIP6 projections suggest that the potential for wildfire will continue to increase throughout the 21st century, under various climate scenarios. This indicates that the current increase in wildfire activity is likely to persist and potentially intensify.

The findings of this study strongly support the hypothesis that climate change has played a significant role in the abrupt and dramatic increase in wildfire activity observed in British Columbia since the mid-2000s. The observed increase in fire activity coincides with a sharp reversal in the historical wetting trend, resulting from increased temperatures and evaporative demand despite relatively high precipitation levels. The study highlights the need to move beyond simply attributing the increased fire activity solely to temperature, acknowledging the complex interactions between climate variables and bottom-up factors such as fuel loads, land-use history, and fire management policies. The study's findings are particularly relevant in the context of future climate projections, which indicate a continued increase in wildfire potential. Understanding the interplay between climate and non-climatic drivers is crucial for developing effective mitigation and adaptation strategies. This understanding needs to factor in the diverse vegetation types and fire regimes across British Columbia's varied ecosystems. The study's findings call for a shift towards integrated fire management strategies that incorporate climate change projections and consider the legacy effects of past disturbances and human land use practices. The need to integrate Indigenous knowledge and practices into fire management is emphasized to develop more holistic and sustainable solutions. Overall, the results underscore the urgency for proactive wildfire management and climate change adaptation strategies to protect both human communities and ecosystems.

This study demonstrates a clear link between climate change and the abrupt increase in wildfire activity in British Columbia since the mid-2000s. The findings highlight the importance of considering both climate-driven top-down controls and bottom-up factors such as fuel loads and land management practices when assessing wildfire risk. Future research should focus on improving the accuracy of wildfire projections by refining the modeling of complex interactions between climate, fuel dynamics, and human interventions. Further investigation into the specific influence of altered large-scale weather patterns due to Arctic warming and their effects on fire weather in BC is also warranted. The results emphasize the need for innovative and place-based adaptation strategies to mitigate the escalating risks posed by wildfires in BC and similar regions.

While the study provides strong evidence of a link between climate change and increased wildfire activity, some limitations exist. The analysis relies on existing climate and fire data, which may have uncertainties or limitations, especially in the early part of the 20th century due to sparser data collection. The study focuses primarily on large fires, potentially overlooking trends in smaller fires. While the study acknowledges the role of various bottom-up factors, quantifying their precise effects on wildfire activity remains challenging due to their complex and often interconnected nature. Future studies should strive to incorporate additional data sources and improve the modeling of these interactions to enhance understanding of BC's wildfire dynamics.