A firefighter sprays water onto burning brush beside a road.
Firefighters work to contain the Creek Fire in Sierra National Forest in California on 10 September 2020. Credit: Pacific Southwest Forest Service, U.S. Department of Agriculture

Droughts, heat waves, and wildfires are among the costliest and most life-threatening disasters in the United States and worldwide. Wildfires in the western United States burned nearly 3.56 million hectares (8.8 million acres) in 2020, or about 75% more area than expected in an average year. With 37 people killed, tens of thousands more displaced, and millions having experienced impaired air quality, the total cost, including from health issues and indirect impacts such as disruptions of supply chains nationwide, is expected to be hundreds of billions of dollars [Wang et al., 2021]. This record wildfire season, like many seasons in the past 20 years (Figure 1), occurred concurrently with a once-in-a-millennium drought and record heat across much of the southwest United States (part of the Northern Hemisphere’s warmest summer on record).

Plot showing annual area burned from 1984 to 2018 in the United States in total, in the western United States, and in Alaska and California combined
Fig. 1. Annual area burned from 1984 to 2018 in the United States in total, in the western United States, and in Alaska and California combined based on data from the federal Monitoring Trends in Burn Severity database.

How do droughts, wildfires, and heat waves interact? How do they shape each other’s likelihoods, magnitudes, and impacts? With millions of lives and billions of dollars at stake, answering such questions is vital in the United States and elsewhere around the world amid an uncertain climate future.

How will vast burned landscapes shape atmospheric conditions favorable to droughts and extreme heat in coming years?

Concurrent droughts and heat waves can substantially increase fire risk and the scale of burned areas (although the degree of their impacts varies considerably with different fire regimes and histories). For example, tens of millions of trees died during the 2012–2016 California drought, creating a massive fuel load for wildfires [Goulden and Bales, 2019]. During the 2020 fire season in California, unusually strong back-to-back summer heat waves, punctuated by dry spells, made conditions ideal for the August Complex and Creek Fires (Figure 2), which together burned more than 566,000 hectares (1.4 million acres), an area roughly the size of Delaware. Information about drought, heat waves, and the evaporative demand of the atmosphere is thus crucial for fire research and early-warning systems.

Less clear is how vast burned landscapes will shape atmospheric conditions favorable to droughts and extreme heat in coming years. With La Niña and warm North Pacific sea surface temperature anomalies lasting into early 2021, drought-favorable conditions are likely to persist over much of the western United States, independent of carryover effects from this past fire season. Will drought or extreme heat be established more quickly and intensely in newly burned regions or cause fire-favorable weather elsewhere? Global warming increases the co-occurrence of droughts and heat waves, along with concomitant wildfire risks. Scientists must characterize and quantify the potentially multiplicative impacts of these phenomena within and across warm seasons.

Challenges to Progress

We identify two principal challenges in advancing the science of and responses to drought–heat wave–wildfire events. First, feedbacks involved in shaping compound drought–heat wave–wildfire events are poorly understood, in part because there have been too few observed examples to identify robustly their shared origins and to model their co-occurrences statistically. Furthermore, many of the processes underpinning these events are also not represented well, if at all, in the physical models used to disentangle forcings and feedbacks. For example, models must better account for effects of fire-generated aerosols on subsequent rainfall; how reduced canopy densities of burned landscapes expose previously shaded winter snowpack to more solar radiation, leading to earlier melting in spring and reduced summer runoff [Gleason et al., 2019]; and how fire-scarred vegetation and surface soils reduce evapotranspiration and water infiltration and increase runoff.

Plot showing daily averaged vapor pressure deficit of surface air in Northern California between 1 May and 1 October 2020
Fig. 2. Observed anomalous daily averaged vapor pressure deficit (VPD = esea), vapor pressure (ea), and saturation vapor pressure (es) in hectopascals (hPa) of surface air in Northern California between 1 May and 1 October 2020 obtained from gridMET data and smoothed as a 5-day running mean. “VPD (Analogue)” represents averaged VPD values and their 25% to 75% uncertainty range (blue shading) for similar atmospheric circulation conditions in the same region from 1 May to 1 October of 1979–2019. The observed VPD for 2020 is mostly higher than we have ever seen in the recent past, mainly because of unprecedented warm surface temperatures during the 2020 fire season. Prior to the August Complex Fire in mid-August, dry anomalies (valleys in ea) from 1 July to 1 August followed by a heat wave (spike in es) in mid-August created ideal climate conditions for wildfire fueled by dead trees. Prior to and during the Creek Fire in early September, another spike in VPD due to a heat wave in early September followed by a dry spell in early mid-September again created conditions for sustained wildfires. This plot illustrates how drought and heat waves work in concert to exacerbate wildfire risk in a changing climate.

Identifying the drivers of compound effects requires impact data, such as data describing drought impacts on the area burned by wildfires or the combined economic, environmental, and human health costs of drought, heat waves, and wildfire [Zscheischler et al., 2020]. Until recently, however, wildfire was not widely considered to be an impact factor or feedback mechanism by the drought and heat wave research community.

The second challenge is that there are numerous barriers among disciplines and between researchers and decisionmakers in the study and risk management of drought–heat wave–wildfire events. Different disciplines view problems from their own angles. For example, drought researchers focus on variations in moisture supply (e.g., rainfall) as the primary cause of drought, whereas wildfire experts focus on the moisture demand of the atmosphere as indicated by vapor pressure deficits (Figure 2).

Addressing gaps between research disciplines requires collaboration among scientists with widely varying expertise.

Addressing gaps between research disciplines requires collaboration among scientists with widely varying expertise, from climate physics, hydrology, and ecology to fire behavior and management science, air and water quality, and public health and infrastructure engineering. Adding the decisionmaking dimension, unfortunately, reinforces existing barriers. For example, scientists build their reputation on the thoroughness of their work, especially on understanding the limitations and uncertainty in their results. As such, they are reluctant to advocate for imperfect solutions. In contrast, risk managers are accustomed to working with imperfect (but timely) information to save lives and manage the impacts of hazards. Reconciling these differences requires culture change in both fields.

Current Collaborations

In the past decade, coordinated efforts among scientists have significantly advanced drought research and early-warning capabilities. NOAA’s Drought Task Force (DTF), for example, is composed of U.S. drought experts selected through a competitive proposal process run by NOAA’s Modeling, Analysis, Predictions, and Projections Program in collaboration with the National Integrated Drought Information System (NIDIS). Over its 10-year history, the DTF has led community research efforts to study drought causes, predictability, modeling, and monitoring that enhance early-warning capabilities in support of the NIDIS mission.

DTF and NIDIS have drawn top researchers from different disciplines to work together to address science questions central to drought early warning and management, especially those identified by stakeholders. For example, during the 2011 Texas, 2012 Great Plains, and 2012–2016 California droughts, the DTF addressed the increasingly important role of heat in shaping these extreme events through timely reports about their dynamic causes, such as the role of sea surface temperature variability and anthropogenic forcing in producing them [Seager et al., 2014; Hoerling et al., 2015]. Separately, recent wildfire research has shown that warming, in addition to other factors like reduced precipitation and expansions of urban-wildland interfaces, is a leading cause of increasing wildfire in the western United States [Williams et al., 2019] and worldwide.

The wildfire and climate research communities have begun collaborating to create wildfire early-warning systems.

The wildfire and climate research communities have already begun collaborating to create wildfire early-warning systems. For example, the U.S. Forest Service Wildland Fire Assessment System (WFAS), which is supported by the National Interagency Coordination Center (NICC), uses real-time drought and temperature information provided by NOAA to assess wildfire risk.

In 2015, the NIDIS Drought and Wildland Fire Nexus (NDAWN) initiative was established to improve drought information products and communication to drought and fire management communities. NDAWN partners with other drought and fire research and early-warning communities, including NICC and WFAS, to break down barriers to effective interdisciplinary collaboration.

Meanwhile, the DTF supports projects that investigate the impacts of wildfire on clouds, precipitation, snowmelt, and streamflow over the western United States through both observational analysis and regional model simulations. And a growing body of research is aimed at developing a framework to characterize and quantify compound effects of droughts, heat waves, and wildfires [e.g., Zscheischler et al., 2020]. These, together with other existing efforts, will inform Earth system models with improved representations of ecosystem-fire-climate interactions, which, in turn, will ultimately enable researchers to dissect the underlying mechanisms of coupling among droughts, heat waves, and wildfires.

Opportunities for Improvement

How can we leverage current capabilities to make tangible progress? We need research support targeted toward early-career researchers who are working to increase the understanding of relationships among drought, heat, and wildfire and to develop key long-term collaborations with drought and wildfire managers. This support, along with revisions to the researcher evaluation metrics used by academic and research institutions to better account for and encourage high-risk research investments made by early-career scientists, is needed to sustain research programs that effectively address these societally impactful problems.

The combined impacts of droughts, heat waves, and wildfires are substantial and are outpacing mitigation and adaptation efforts.

We should also raise awareness of existing scientific efforts, identify stakeholder needs, and foster and communicate shared and open science. For example, NIDIS and NDAWN can translate the latest research results into actionable information for stakeholders and provide feedback from stakeholders to researchers. Interdisciplinary science socialization of this kind, particularly when motivated by research imperatives identified by NIDIS and other stakeholders, can lead to insights that otherwise are missed through narrower disciplinary research alone.

The combined impacts of droughts, heat waves, and wildfires are substantial and are outpacing mitigation and adaptation efforts. Consequently, there is broad shared interest among federal, regional, and state agencies and among private sector companies involved in weather, climate, wildfire, air and water quality information to create an effective predictive capability. Coordinated efforts among research and application programs that leverage pooled resources are important not only for sustaining ongoing research that is addressing questions about drought, heat waves, and wildfire but also for seeding and expanding future interdisciplinary collaborations to ensure we have the science necessary to manage risks from compound extremes.


We thank Daniel Barrie, the program manager of the NOAA DTF, for his support of this article; NIDIS for support of and collaboration with the NOAA DTF IV; Timothy Brown and Tamara Wall from NDAWN and Cenlin He, the principal investigator of a DTF project on drought-wildfire interaction, for invaluable feedback; and Yizhou Zhuang for creating Figure 2.


Gleason, K. E., et al. (2019), Four-fold increase in solar forcing on snow in western U.S. burned forests since 1999, Nat. Commun., 10, 2026, https://doi.org/10.1038/s41467-019-09935-y.

Goulden, M. L., and R. C. Bales (2019), California forest die-off linked to multi-year deep soil drying in 2012–2015 drought, Nat. Geosci., 12, 632–637, https://doi.org/10.1038/s41561-019-0388-5.

Hoerling, M., et al. (2013), An interpretation of the origins of the 2012 central Great Plains drought, Model., Anal., Predict., and Projections Program, Silver Spring, Md., cpo.noaa.gov/Meet-the-Divisions/Earth-System-Science-and-Modeling/MAPP/MAPP-Task-Forces/Drought-Task-Force-III/Drought-Task-Force-II/An-Interpretation-of-the-Origins-of-the-2012-Central-Great-Plains-Drought.

Hoerling, M., et al. (2015), Temperature and drought: A science assessment by a subgroup of the Drought Task Force, Model., Anal., Predict., and Projections Program, Silver Spring, Md., cpo.noaa.gov/Meet-the-Divisions/Earth-System-Science-and-Modeling/MAPP/MAPP-Task-Forces/Drought-Task-Force-III/Projects/Temperature-and-Drought-Assessment.

Seager, R., et al. (2014), Causes and Predictability of the 2011 to 2014 California Drought, Model., Anal., Predict., and Projections Program, Silver Spring, Md., cpo.noaa.gov/Meet-the-Divisions/Earth-System-Science-and-Modeling/MAPP/MAPP-Task-Forces/Drought-Task-Force-III/Drought-Task-Force-I/Causes-and-Predictability-of-the-2011-2014-California-Drought.

Wang, D., et al. (2021), Economic footprint of California wildfires in 2018, Nat Sustain., 4, 252–260, https://doi.org/10.1038/s41893-020-00646-7.

Williams, A. P., et al. (2019), Observed impacts of anthropogenic climate change on wildfire in California, Earth’s Future, 7, 892–910, https://doi.org/10.1029/2019EF001210.

Zscheischler, J., et al. (2020), A typology of compound weather and climate events, Nat. Rev. Earth Environ., 1(7), 333–347, https://doi.org/10.1038/s43017-020-0060-z.

Author Information

Rong Fu (rfu@atmos.ucla.edu), University of California, Los Angeles; Andrew Hoell, Physical Sciences Laboratory, NOAA, Boulder, Colo.; Justin Mankin, Dartmouth College, Hanover, N.H.; Amanda Sheffield, National Integrated Drought Information System, NOAA, Boulder, Colo.; and Isla Simpson, National Center for Atmospheric Research, Boulder, Colo.


Fu, R., A. Hoell, J. Mankin, A. Sheffield, and I. Simpson (2021), Tackling challenges of a drier, hotter, more fire-prone future, Eos, 102, https://doi.org/10.1029/2021EO156650. Published on 01 April 2021.

Text © 2021. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.