The degree to which forest fires are caused by fossil fuel-driven climate change

I happened to run across something that contradicts Bob Zybach’s repeated assertions that, “these fires have been clearly predicted by me and others because of USFS management policies and Wilderness designations and have zero to do with warming climate or drier fuels.”  The Union of Concerned Scientists calculated the effect of warming climate and drier fuels on burned area, and the result from their peer-reviewed analysis is not “zero.”

Climate change is causing hotter, drier conditions that are also fueling these increasingly large and severe wildfires. In particular, vapor pressure deficit (VPD), a measure of atmospheric “thirst,” has emerged as a key way of tracking how climate change is amplifying wildfires because of its role in regulating water dynamics in ecosystems and, together with rising temperatures, contributing to increasing dryness (Box 1).

UCS used a combination of data and modeling to determine how much the carbon emissions associated with 88 major carbon producers (hereafter, the “big 88”) have historically contributed to increases in VPD and burned forest area across the western United States and southwestern Canada (see Methodology).

Across western North America, the area burned by forest fires increases exponentially as VPD increases, which means that relatively small changes in VPD result in large changes in burned forest area. The observed rise in VPD has enabled a steep increase in the forest area that has burned across the region since the mid-1980s. Since 1986,1 a cumulative 53.0 million acres of forest area has burned across western North America as VPD has risen. Without emissions tied to the big 88, the rise in VPD would have been much smaller, and 33.3 million acres (IQR 27.7 million–38.5 million) would have burned (Figure 4). That means that 37 percent (IQR 26–47 percent) of the cumulative burned forest area from 1986 to 2021 is attributable to emissions from the big 88. This represents nearly 19.8 million acres of burned forest area, or an area roughly the size of Maine.

You can criticize UCS for being agenda-driven (and we’ve talked about the limitations of “burned area” as a metric), but I’d challenge Bob or others to provide a similarly peer-reviewed research paper that attributes fire effects to his chosen causes.


13 thoughts on “The degree to which forest fires are caused by fossil fuel-driven climate change”

  1. I don’t buy it; correlate the beginning and intensification of WFU in federal policy, and the “change” in firefighting response. Up until the late 1990’s, the FS (anyway) put most emphasis on putting fires out while they were small, and doing so quite aggressively. These factors must have some way being factored into the graph. Otherwise, it is a nice picture – of something…..

    • There are three primary reasons for the horrific destruction due to unplanned wildfires over the last 15 years, especially. And it’s a tie for the top spot. See page 15 of “A Call to Action.” I don’t spend a lot of time quibbling about the top three. But we better address these three, for sure. The lack of care [i.e., maintenance; management] for our forests [forest are more than just trees] is clearly a major culprit. We all know this. Right now, just for the Forest Service, the annual budget to effectively address this issue is about $3 billion short. Yes, $3 billion. Yet we seem okay with spending +$1 million/hour fighting fires with absolutely no end in sight. Annual damages from wildfires are approaching $1 trillion. Let’s do the math. What would +$3 billion for the next 5-7 years contribute to reducing suppression costs and damages? Clearing more than $3 billion beginning in years 6-7; it’s not even a close call. I have found myself not debating the “top three” reasons any longer. I just wish we could band together and be an influencing force to effectively address this national emergency. Properly coalesced, we certainly have the position power to do exactly that.

      Very respectfully,

    • I agree that, in addition to the effect of temperatures, it’s not hard to make the case that there should be a correlation between the number of standing trees and burned area. But it’s harder to make the case for how much more logging we should be doing than we have been, especially in light of the adverse effects (and doubtful benefits) of removing larger trees. That’s the research I’d like to see, but it really plays out in site-specific projects that have to make that case on their own.

      • What do you mean by logging exactly? I see some possibilities
        cutting trees and leaving them.. or burning in piles
        cutting trees and removing them, whether or not they are used for personal or commercial purposes
        I also don’t understand what you mean by removing “larger” trees.. larger than the trees they are next to? Commercial sized?

        • “Larger” = ecologically rare and valuable. Also, most people seem to agree that larger trees are more fire resistant and removing them is therefore not likely to make a forest more fire resistant. Both are reasons to not cut them down. The instances where overstory trees pose a fire threat should be viewed as the exception rather than the rule, and a project should have to make that case based on site-specific information. (Instead of simply labeling every timber sale as “fuels reduction.”)

          • They do make site specific cases..that’s what silvicultural and fuels prescriptions are??
            Is a mature white fir next to a resilient P pine “ecologically rare and valuable)? Is a large white fir ever “fire resistant”?

      • It would be nice if the correlation was just the number of standing trees and burn areas. The fuel mixture (fine up to large), structure, topography, and weather all play a part in the the burn characteristics. An examination of the Oregon Rum Creek Fire fire pattern and behavior makes this very clear. Some of the areas that had extreme stand replacing burn behaviors were very heavily managed/treated. While right next to it was very thick high density stands that burned at low intensity or not at all. Other areas that have had no modern harvest/treatment showed a matrix of burn from no/low to medium to high burn severity. What is obvious is the time frame when the fire took off after an agressive initial attack was due to weather/wind and and all fuels seemed to burn. Once the weather calmed down then fuels and topography seem to be the drivers. And the suppression effort I’m sure played a part. There was a lot of aerial support (retardant) on some of the areas with mixed results (limited access for ground crews to follow up after drops in the northwest section).

        So I would be surprised if there would be a simple x number of trees vs burn area equation.

  2. So I don’t think it’s 0, but I know the UCS doesn’t know, in fact no one knows for sure because all the changes are too difficult to separate and in fact, there aren’t even data for many of the variables.

    Also I looked at this and I didn’t see exactly how it was peer-reviewed.

    When I know that they don’t know… and yet they claim they are Very Sure!!! I wonder what’s behind it.. Do they really not know how things work (scary) or don’t they care (possibly scarier?)

    • “The authors thank three anonymous scholars who reviewed either this report or the associated publication in Environmental Research Letters.”

  3. Annual wildfire area burned in coming decades is likely to be highly geographically heterogeneous, reflecting interacting regional and seasonal climate drivers of fire occurrence and spread.

    Weather conditions at local and regional scales during the fire season are dominant factors influencing fuel flammability and annual area burned [17–20]. Warmer temperatures in concert with drought decrease fuel moisture, increase lightning ignitions, and lengthen the period over which fires occur [21], [22]. At regional scales, the occurrence of severe fire weather is regulated strongly by seasonal climatology as reflected in synoptic-scale atmospheric circulation [23–26]. As a consequence, the potential impacts of seasonal weather on fire behavior and extent are spatially heterogeneous, for several reasons. Growing season temperature influences water balance deficits (PET-AET)–an important variable controlling annual area burned by regulating biomass and fuel moisture–variably across climates and ecosystems [22], [26–28]. … In addition to fire-season weather, antecedent temperature and precipitation over prior months and years regulate fire frequency and extent to varying degrees across ecosystem types. These effects are especially strong where winter and spring moisture are limiting to growth and conditioning of fine fuels prior to fire season [26], [33], [34]. Ecosystems will thus generate differential responses in vegetation growth in response to warming and changes in precipitation, resulting in significant differences in fuel types and loads under altered climate [7], [35]. For example, although warming may increase area burned under extreme warming/drying conditions in more productive, less fire-prone ecosystems due to increased fuel desiccation, it may produce only modest changes or even decreased area burned in less productive, fuel limited ecosystems by restricting fuel production [3], [6], [15].

    Seasonal climate variation exerts a primary control on the length, location, and intensity of fire seasons in western North America and worldwide. The amount of area burned by wildfires represents a complex integration of productivity, ignition patterns, landscape configuration (i.e., fuel connectivity), synoptic and local weather, and seasonal climatic conditions that condition fuels and influence fire spread and the length of fire season, along with the important role of anthropogenic ignitions [18], [20], [32].

    Wildfire is a global Earth system process that both integrates and influences many other interactions between ecosystem and the climate system. Fire mediates other ecosystem responses to changing climate, for example by modulating forest density and composition, and thus providing a mechanism by which ecosystems adapt to changing climate conditions. Uncertainties in key elements of climate projections could be compounded by nonlinear responses of fire to climate variability. Fires may also act as triggers for abrupt and irreversible change to novel configurations under Direct and indirect controls on annual wildfire area burned future climate [44], [83]. As climate change progresses, the projected changes in the area affected annually by fire may be an important multiplier of these effects in coming decades.

    Kitzberger T, Falk DA, Westerling AL, Swetnam TW (2017) Direct and indirect climate controls predict heterogeneous early-mid 21st century wildfire burned area across western and boreal North America. PLoS ONE 12(12): e0188486.

    Using stochastic machine learning (SML) framework, SMLFire1.0, to model observed fire frequencies and sizes in 12 km × 12 km grid cells across the WUS [western United States]. … The modeled WUS fire frequency matches observations at both monthly (r=0.94) and annual (r=0.85) timescales, as do the monthly (r=0.90) and annual (r=0.88) area burned. Moreover, the modeled annual time series of both fire variables exhibit strong correlations (r≥0.6) with observations in 16 out of 18 ecoregions. Our ML model captures the interannual variability and the distinct multidecade increases in annual area burned for both forested and non-forested ecoregions. … [W]e find that fire-month vapor pressure deficit (VPD) is the dominant driver of fire frequencies and sizes across the WUS, followed by 1000 h [3-8” diameter] dead fuel moisture (FM1000), total monthly precipitation (Prec), mean daily maximum temperature (Tmax), and fraction of grassland cover in a grid cell.

    Buch, J., Williams, A. P., Juang, C. S., Hansen, W. D., and Gentine, P.: 2023. SMLFire1.0: a stochastic machine learning (SML) model for wildfire activity in the western United States, Geosci. Model Dev., 16, 3407–3433,

    • Second, I’d add that in contrast to “Weather conditions at local and regional scales during the fire season are dominant factors influencing fuel flammability and annual area burned [17–20].” I’d say that weather, the efficacy of suppression, and suppression strategies and tactics are dominant factors influencing annual area burned.
      It’s kind of like… predicting river floods assuming no levees.


Leave a Comment