Implications of climate change on fire and thinning prescriptions Jessica Halofsky1, 2 Morris Johnson2 1 University of Washington, School of Forest Resources 2 U. S. Forest Service, Pacific Wildland Fire Sciences Laboratory, Seattle, WA

Climate controls ecosystem processes The hydrologic cycle Plant establishment, growth, and distribution Disturbance Drought Fire Flooding Insect outbreaks Many ecosystem processes are either directly or indirectly tied to climate, and so climatic change is likely to affect them in the future. Processes influenced by climate include the hydrologic cycle, plant species establishment, vegetation productivity and growth, and disturbance, including insects, and fire.

Fire Behavior Triangle Topography Weather I’m sure most of you are familiar with the fire behavior triangle, which shows that fire behavior is controlled mainly by topography, weather and fuels. Topography – slope steepness, aspect, etc. Weather – wind, temp, relative humidity, precip Fuels – fine or heavy, arrangement or continuity, composition, and fuel moisture. Fuels

Fire Behavior Triangle Topography Weather Increasing temperatures with climate change will likely influence both the weather and fuel component of this triangle. Increasing summer temperatures, even if coupled with increases in summer precipitation will likely lead to increased summer drought and lower levels of relative humidity, and lower levels of fuel moisture, which can all lead to more extreme fire behavior. Fuels

Area burned – Western U.S., 1916 - 2007 Many of you have probably seen this figure as well, which shows the annual area burned in the Western U.S. from 1916-2007.

Area burned – Western U.S., 1916 - 2007 One explanation for the trends in this figure is related to fuels and fire suppression. You can see the pattern of relatively high area burned until after World War 2, then we got really good at suppressing fires, but then fuels built up in systems that historically experienced frequent, low severity fires, which lead to increased area burned starting in the 80’s. Fire Suppression Fire Exclusion Fuel Accumulation

Area burned – Western U.S., 1916 - 2007 But another pattern that stands out is the relationship between annual area burned and a El Nino-like climate cycle called the Pacific Decadal Oscillation, or the PDO. Like El Nino, the PDO has a warm phase and a cool phase, but the PDO has 20-30 year cycles instead of 3-7 year cycles. Fire Suppression Fire Exclusion Fuel Accumulation

Area burned – Western U.S., 1916 - 2007 You can see that area burned was greater during warm phases of the PDO, but area burned decreased during the cool phase of the PDO during mid-century. I should note that the positive relationship between warm PDO and fire only exists for the Pacific Northwest. There is actually a negative relationship between fire and warm PDO in the southwest. Still, during the 20th century, wildfire area burned has been strongly associated with low precipitation, drought, and high summer temperatures. Fire Suppression Fire Exclusion Fuel Accumulation Lots of Fire Much Less Fire Lots of Fire

Cascade Mixed ecoregion Fire area burned and PDO Cascade Mixed ecoregion These are reconstructed ecoprovince area burned data derived from Oregon and Washington state-level data for 1916-2003. We do not have good area burned data prior to 1916. Mean area burned is, however, greater for the 1925-46 warm phase (~51,000 acres / yr) and 1977-1998 warm phase (18,000 acres/yr) than for the cool phase (1947-76) period (~4500 acres). The dotted red line indicates the increase in mean value if data up to 2004 is included with the later warm phase. It is probably important to note the log scale on the y axis here; clearly more area was burning on average earlier in the century. The PDO phases also are contemporary with anecdotal timelines of fire suppression and exclusion Figure courtesy of J. Littell

Years with fire area > 200,000 acres Warm-Phase PDO Cool-Phase PDO Idaho 15 7 Oregon 14 5 Washington 11 2 TOTAL 40 (74%) 14 (26%)

Climate Change and Fire Warmer and drier spring conditions = early snowmelt lower summer soil and fuel moisture longer fire seasons increased fire frequency and extent Fire intensity and severity may also increase Trends in wildfire and climate in the western United States from 1974-2004…show that both the frequency of large wildfires and fire season length increased substantially after 1985,and that these changes were closely linked with advances in the timing of spring snowmelt, and increases in  spring and summer air temperatures.” Intensity of fires may also increase in some areas if higher temperatures interact with fuel characteristics to increase fire intensity.

Trends in Area Burned •Average Area Burned 1970-1979 3,000,000 ac

Recent trends also show increased: Length of fire season Fire intensity/severity Number of fires Invasive species (cheatgrass) Time needed to suppress average wildfire Cost of suppression Number of structures lost Strain on fire management resources

How much will area burned increase with climate change? Analysis of wildfire data since 1916 for the 11 contiguous Western states shows that for a 4°F increase that annual area burned will be 2-3 times higher. McKenzie et al. (2004), Conservation Biology 18:890-902 McKenzie and colleagues looked at how fire extent was related to climate in the past, and then projected those relationships into the future with climate predictions from global climate models. McKenzie et al. (2004)

Wildfire area burned in Oregon with 2°C warming Area burned for Oregon. Data for 1 year are represented by a circle whose position indicates the summer temperature (x-axis) and precipitation (y-axis) anomalies and whose area is proportional to area burned. Largest circles tend to appear in the lower right of each panel (warm, dry summers). Contour lines indicate the mean area burned from multiple regression, as a ratio to the value at the origin; contours are at 0.1, 0.2, 0.5, 1, 2, 5, and so forth. The arrow shows the direction of climatic change McKenzie et al. (2004) McKenzie et al. (2004), Conservation Biology 18:890-902

Wildfire area burned in Washington with 2°C warming 3x more fire with 2 degree celsius warming McKenzie et al. (2004) McKenzie et al. (2004), Conservation Biology 18:890-902

Projected changes in area burned in the PNW These are similar results from work by Jeremy Littell and others that show the projected changes in area burned for the PNW based on statistical models describing the relationship between climate and area burned in the past. We have area burned on a log scale for the historical period, the 2020s, 2040s, and 2080s with results from two global climate models. Littell et al. 2010

Projected changes in area burned in the PNW 2.0 M Ac 1.1 M Ac 0.8 M Ac 0.5 M Ac This shows the average number of acres burned for each time period. Future fire projected from the best statistical model suggests a doubling or tripling of area burned by the 2080s The median regional area burned, averaged over both GCMs, is projected to increase from about 0.5 million acres to 0.8 million acres in the 2020s, 1.1 million acres in the 2040s, and 2.0 million acres in the 2080s. These models suggest that summer precipitation and temperature play a large role in the area burned by fire Littell et al. 2010

Projections of future area burned in WA ecosections The text numbers below each set of box-and-whiskers plots indicate the average of A1B and B1 future area burned estimates for the ecosections in acres. Patterns differ by area.

Fire Regimes Vary by Environment Warm In addition to area burned, the frequency and intensity of fires will likely change with climate change, and this will vary by fire regime. This shows a generalized classification of fire regimes based on temperature growth index and moisture stress. In general, we have the forest with the highest moisture levels in the high severity fire regime, the forests with the greatest moisture stress index and highest temperatures in the low severity fire regimes, and the ones that are intermediate between those two extremes in the mixed severity fire regime. Cold Wet Dry Agee 1993

Gradients of Fire Regime Controls Mixed Severity High Severity Weather Driven Mixed Severity And because of the variation in moisture stress among forests in those fire regimes, there are differences in the drivers of fire in the different systems. Drier forest types, as well as many grasslands and shrublands, often have drought and weather conditions that allow for fires, but because of the moisture limitations, fuels are limiting. Forests in high severity fire regimes usually have lots of biomass and fuels to carry a fire, but those fuels aren’t often dry enough to carry a fire. So fires only happen when weather conditions allow for fuel drying. In mixed severity fire regimes, climate and fuels interact in a complex manner to control the frequency and severity of fires. However, recently we’ve seen that weather can be a dominant controller of mixed severity fires under extreme conditions and almost completely override fuel conditions in controlling burn severity. Increased fuel levels because of fire suppression could be a factor in those cases. Low Severity Fuel Driven Adapted from J. Agee

Relative influence of climate and fuels on fire regimes in common western US ecosystems This figure illustrates the same concept, except we’ve added some vegetation types here. PNW-IM = Pacific Northwest and intermountain region of the West SW = American Southwest

Relative influence of climate and fuels on fire regimes in common western US ecosystems High elevation forests usually require major drought and ignition for fire to occur. These forests are not fuel-limited. In lower elevation and drier forest types, drought is frequent, but fuel production is limiting. The forests between these two extremes are primarily considered mixed severity. In mixed severity fire regimes, climate and fuels interact in a complex manner to control the frequency and severity of fires. PNW-IM = Pacific Northwest and intermountain region of the West SW = American Southwest

Relative influence of climate and fuels on fire regimes in common western US ecosystems Climate Change With climate change, we might expect these high and mixed-severity systems to move in this direction to become less climate limited and more fuel limited. PNW-IM = Pacific Northwest and intermountain region of the West SW = American Southwest

Some energy (climate) limited forests may become water (fuel) limited forests This shows the current area of energy-limited forests in WA, which have fire regimes that are mainly climate-limited. With climate change, some of these energy-limited forests may shift to water-limited forests, which means that fire will likely be less limited by climate and more limited by fuels. The Okanogan highlands and the foothills of the north-eastern Cascades contain most of the area that climate projections indicated will transition from energy- to water-limited forest by the 2080s. Littell et al. 2010

Variation in Fire Severity within a General Fire Regime Low High Moderate Proportion We may also expect some changes in fire severity with climate change. This is a conceptual figure showing the proportion of low, moderate and high severity burn area in low, mixed and high severity fire regimes. Low Mixed High Severity Adapted from Agee 1993

Initially, with more frequent extreme burning conditions? High Low Proportion Moderate Extreme climate and weather conditions can override the influence of stand structure and fuels on fire behavior. We might expect much more high severity fire initially. Low Mixed High Severity Adapted from Agee 1993

With eventual drought- and fire-induced reductions in fuel in drier forest types? Low High Proportion Moderate Eventually, more frequent, high-severity fires will likely lead to fuel reductions, leading to the mixed severity fire regimes looking more like the low severity fire regimes. And it’s likely that some areas that were historically high severity fire regimes could start to look more like mixed. However, with increased drought and reduction in fuels from high severity fire, there may be more low severity patches after that. Low Mixed High Severity Adapted from Agee 1993

Fire Regimes and Landscape Patterns Low-Severity Fire Regime Mixed-Severity Fire Regime High-Severity Fire Regime And these changes would have implications for landscape structure. Initially, with more high severity fires, forest in mixed severity fire regimes could start looking more like this. However, as the system equilibrates with the new climate, and fuels are reduced by high severity fires and drought, forest in mixed severity fire regimes may start looking more like this. Agee 1998

Percent change in biomass consumed by fire A1B B1 MIROC3_MEDRES HADCM3 CSIRO_MK3 This shows output from the MC1 model for the Pacific Northwest, which is a process-based model that looks at changes in vegetation over time. MC1 also has a fire submodel, MCFIRE, so this model shows interactions between vegetation change and fire. You can see that many areas in the Pacific Northwest, particularly with the Hadley model, which shows a drier Pacific Northwest, the Percent biomass consumed by fire is expected to increase in many areas. You can also see areas where there will likely be reductions in biomass burned, even with a warmer and drier scenario like Hadley. These areas are east of the cascades and in the Okanogan highlands. This is likely because fuel levels will decrease due to water stress in these areas, so even though climate will be amenable to fire, fuels will become limited and the biomass consumed will decrease. percent Percent change in biomass consumed by fire 2051-2100 vs. 1951-2000 R. Neilson et al., USFS and OSU MAPSS Team, Corvallis, OR

Fire interacts with other disturbances and vegetation/fuel conditions But, fire is not the only disturbance in forests. Insect outbreaks may become more frequent and widespread because warmer temperatures may accelerate insect life cycles and allow them to expand their distribution to areas where they haven’t been previously found. In addition to effects of increased temperatures on insect life cycles, increased temperatures will also increase drought stress of some forest tree species, thus making some forests more susceptible to insect infestation. In addition, insect infestations can interact with fire. Recently burned forests may be more susceptible to insect damage. In turn, dead and weakened trees that have been infested with insects increase fire risk.

The Disease Spiral Manion 1991 This is the disease spiral which shows factors that eventually lead to tree death. Manion 1991

Disturbance are drivers of major ecosystem changes. Warmer temperatures and drought will lead to increased fire frequency and more area burned, which will facilitate further habitat changes and species responses that are already happening with climate change. McKenzie et al. 2009

Adaptation strategies for natural resource management? So what should be done about these probable changes in fire and other disturbances with climate change? We don’t know exactly what to do – there’s so much uncertainty in the future. There will likely be a lot of good outcomes and a lot of bad outcomes, so you’re damned if you do and damned if you don’t. But being proactive will likely lead to more good outcomes than bad.

Adaptation strategy #1 Increase landscape diversity Thin forest stands to create lower density, and diverse stand structures and species assemblages that reduce fire hazard and increase resilience to wildfire. I’m going through 7 general adaptation strategies from recent literature and resource managers. These are ideas or suggestions. Not all recommendations listed are appropriate to all situations; methods must be carefully evaluated to fit the case in hand. Then Morris is going to go into more detail on potential changes to thinning prescriptions. The more diversity of landscape structures, ages, species, etc. we have, the better. Idea of not putting all your eggs in one basket.

Adaptation strategy #2 Increase resilience at large spatial scales Implement thinning and surface fuel treatments across large portions of landscapes where wildfires may occur Orient the location of treatments to modify fire severity and fire spread Focus the spatial scale of treatments on units of hundreds to thousands of acres

Adaptation strategy #3 Maintain biological diversity Modify genetic guidelines Experiment with mixed species, mixed genotypes Assist colonization, establish neo-native species Identify species, populations, and communities that are sensitive to increased disturbance We have spent decades developing seed zone maps, but these might not work anymore. We might want to plant stock from a little further south or the next zone or two down the mountain. Again, hedging your bets with mixed species to prevent stand failure. For example, if we think Monterey pine will do really well up here 50 years from now, should we go get some and plant them. See what you have now and know where the vulnerabilities lie.

Adaptation strategy #4 Plan for post-disturbance management Treat fire and other ecological disturbance as normal, periodic occurrences Incorporate fire management options directly in general planning process Do a better job of planning for post-disturbance management, so when it does happen you know exactly what you’re going to do. Use disturbance events as opportunities. Incorporate long-term experiments. Increase capacity to restore forest lands after large disturbances Maintain a tree seed inventory with high quality seed for a range of species

Adaptation strategy #5 Implement early detection / rapid response Eliminate or control exotic species Monitor post-disturbance conditions, reduce fire-enhancing species (e.g., cheatgrass) This has been applied a lot to exotic species – jump on it as soon as you can to prevent it from getting worse.

Adaptation strategy #6 Collaborate with a variety of partners Develop mutual plans for fire and fuels management with adjacent landowners to ensure consistency and effectiveness across large landscapes

Adaptation strategy #7 Promote education and awareness about climate change Facilitate discussion among management staff regarding the effects of a warmer climate on fire and interactions among multiple resources Educate local residents about how a warmer climate will increase fire frequency and how fuel reduction can protect property