Abstract
Disturbances are the chief cause of ecosystem change across much of the West. The important disturbances on the mountain landscape are wildfire, insect outbreaks, windstorms, and avalanches and landslides. I define disturbance a “relatively discrete event in space and time”, and discuss how different disturbances may change in a warming West. Much has been made of predicted increases in wildfire extent and severity, but strong negative feedbacks, such as diminishing available vegetation (fuel), may offset the projected increases. Warmer and drier is better for most insects, and some of our most widespread, e.g., bark beetles, could be big winners in that outbreaks may be more extensive. Projecting changes in avalanches is uncertain because snow dynamics are complex and local changes in layer formation could increase or decrease stability. The principal driver of landslides is precipitation, so in regions in which precipitation is projected to increase, landslides may also. Air pollution, in the form of haze and visibility reduction at regional scales and toxic effects of smoke at local scales, is associated with both wildfire and prescribed fire, so changes in air quality will track changes in wildfire.
Some say the world will end in fire,
Some say in ice.
From what I’ve tasted of desire
I hold with those who favor fire.
—Robert Frost
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Early in Earth’s history, there were of course volcanic eruptions and other fiery events. I am referring to the origins of wildfire as we know it, fueled by plant biomass and supported by enough oxygen in the atmosphere to sustain combustion, between about 16–30%, with the current level being 21%. Below 16% fire will not propagate, and levels above 30% are possible only with enough moisture in the air to extinguish any fire that might start even if it had plenty of oxygen.
- 2.
For example, some deserts, polar regions, and rainforests, but not all. Just a little fuel, and a dry spell, can be enough.
- 3.
We wrote a whole book about this, also published by Springer. D. McKenzie, C. Miller, and D.A. Falk. 2010. The Landscape Ecology of Fire. Springer, Dordrecht, The Netherlands.
- 4.
The first phrase in quotes is from one (or more) standard definitions in the literature. A classic treatise is an edited volume, S.T.A. Pickett and P.S. White. 1985. The Ecology of Natural Disturbance and Patch Dynamics, Academic Press. The second phrase is deliberately non-technical, and my addition.
- 5.
But later in this chapter and in Chapter 9 we will get to how these events interact with climate-change forcing.
- 6.
As in the spread of infectious disease. A sneeze (for example) provides the force, and dense enough populations the medium.
- 7.
For example, some will claim that the atmosphere is the medium for hurricanes, tornadoes, and other windstorms. Even further down that path, one could argue that the Earth’s gravitational field is the medium for land movements.
- 8.
Fire “scientists” (who study the physical elements of fire) and fire ecologists (who study the biological and ecological elements) distinguish between intensity, basically the total heat produced by the fire, with its associated metrics (see Chapter 5), and severity, the damage caused, such as tree mortality, nutrient depletion, or erosion.
- 9.
These are the elements of the classic “Fire Triangle”, which can be seen in many forms, both verbal and graphic, with an internet search on the term.
- 10.
In general, more intense fires are more severe, per unit area, for obvious reasons: flames, heat, and smoke kill organisms. This will not hold perfectly when organisms vary in their vulnerability. For example, large trees with thick bark can survive more intense fire than can seedlings, saplings, of mature trees with thin bark. See the discussion of adaptations below.
- 11.
Such as power plants. Obviously industrial accidents can start fires that can be intense and cause a lot of change in ecosystems as well as killing humans and damaging structures.
- 12.
And that is one of those regional-scale changes that we have a hard time modeling, so far. See Chapter 3.
- 13.
Discussion of these would take us far afield, but the so-called wildland-urban interface (WUI), land areas in which human structures, usually homes, and open-space vegetation are interspersed, sometimes in complicated spatial patterns, presents a huge conundrum around regulating development, landscaping within developments, and liabilities, all in the context of what seem to be unfailingly lucrative financial situations.
- 14.
For example, in the American Southwest, “warm and dry enough” begins in April or May, and “dry enough” ends with the arrival of the monsoon (see Chapter 4), even though temperatures remain high.
- 15.
For example, Florida, USA, where most wildfire happens in summer or winter, and the seasonally dry tropics, where the fire season is the dry season.
- 16.
Yes, it might be energy-limited, but for West-wide averages, water limitation is more pervasive, and more limiting, something that will only become more unbalanced as the climate warms (see Chapter 5). This logic therefore holds at the broad scales we are considering right here.
- 17.
For the enterprising reader, we wrote two papers about this. D. McKenzie and J.S. Littell. 2017. Climate change and the eco-hydrology of fire: will area burned increase in a warming western USA? Ecological Applications 27:26–36. J.S. Littell, D. McKenzie, et al. 2018. Climate change and future wildfire in the western USA: an ecological approach to non-stationarity. Earth’s Future 6. https://doi.org/10.1029/2018EF000878.
- 18.
It is easy to see why this is so. Imagine a vertical flame on a slope. Every part of the flame except what is next to the ground is closer to the land surface on the uphill side than on the downhill side, making it easier for any flammable substance to ignite on that uphill side.
- 19.
For example, as most of us know, the Northern Spotted Owl was declared endangered as a result of massive clearcut logging in the Olympic Mountains and the west side of the Cascade Range, decimating its habitat. It turns out that population pressures (we will see more about this in Chapter 7) drove a considerable percentage of the survivors across the Cascade crest into marginal, but still livable, habitat on the eastern slope, which (typical of eastern slopes) has more frequent fire than the western side. It turned out that a large proportion of these dispersal sites had been fire refugia in historical times, but in an already warming climate they are now quite susceptible to wildfires severe enough to render them useless as habitat for the owl.
- 20.
For example, it takes a surprisingly small percentage of carbon monoxide (CO), present in high concentrations in smoke, to replace enough oxygen to kill any organism. The “Immediately Dangerous to Life or Health Concentrations” (IDLH) for humans is said (by the CDC) to be 1200 ppm, but this is a general criterion and will vary among both people and circumstances (e.g. length of exposure, other ambient concentrations).
- 21.
Worst of these are the smallest, particulate matter less than 2.5 μm (a millionth of a meter) in diameter, called “PM2.5”. There is good medical evidence that PM2.5 from fire is more toxic to humans than from other sources, even of the same size.
- 22.
I won’t mention these much until Chapter 9, but prescribed fires are fires that are set intentionally by land managers to achieve some objective. Common ones are (1) to reduce fuels on a site so as to limit or prevent high-intensity wildfires, and (2) to eliminate or control some species of vegetation to promote others or increase the diversity of species overall.
- 23.
During the summers of 2017 and 2018, living in Seattle, I was asked by the media if climate change meant that all the summers from then on would be smoky. No, they won’t all (and as I write (2019) Seattle is having its coolest and wettest July in years); weather will always be variable, and the smoke was a “perfect storm” of the coincidence of anomalous weather and wildfires inland. That said, if wildfire increases in British Columbia, Canada, in a warming climate, and regional circulations change, both of which are very possible, we could see more smoky days in a average summer, and more frequent summers with smoke invasions.
- 24.
This network of sites is called IMPROVE (Interagency Monitoring of PROtected Visual Environments). Their website has stunning visuals as well as data. http://vista.cira.colostate.edu/Improve/.
- 25.
Recall, though, that we are now “off the (Holocene) chart”. Global average temperatures are higher than they have been in the Holocene.
- 26.
Technically, both fire history and climate simulations are what are known as stochastic processes. Each realization (instance that they happen) will be different because they don’t start in exactly the same place. Fire history is “real”, and climate simulations are “fake,” but they both have this property. This is easier to see with a simulation; we start with initial values of parameters and these change when we run the simulation again. With fire history, in order to understand it properly, we proceed with our analysis AS IF it were only one instance, instead of its being the only instance that happened. If we have a long enough record, different chunks of time can be the “repeats,” and we can infer, for example, the “average” influence of climate on wildfire over 100 years.
- 27.
To do these statistics right is not for the timid, and it also requires care in avoiding an over-reach of inference. For the interested, a moderately deep dive into the methods is P.E. Higuera, et al. 2010. Peak detection in sediment-charcoal records: impacts of alternative data analysis methods on fire-history interpretations. International Journal of Wildland Fire 19:996–1014.
- 28.
This vicinity can be defined only imprecisely, because charcoal fossils can be deposited some distance from the combustion that produced them. Wind, topography, and fire intensity all contribute to the variability in distance traveled. This is OK. In rare cases we might like to reconstruct the perimeters of specific fires, but normally these data are used for broader inferences.
- 29.
Can you guess why? Which level of fire severity—low, high, or mixed—would be easiest to detect in the charcoal record? Hint: the northern boreal forests are another region with numerous sediment-charcoal records.
- 30.
Stumps, logs, or snags (dead trees still standing) whose tissues are preserved enough for tree rings to be evident.
- 31.
For example, giant sequoias 3000+ years old in the Sierra Nevada, and bristlecone pines 4000+ years old in the White Mountains, a Basin Range (see Chapter 2).
- 32.
I am making this sound straightforward, for our larger purpose, but this process can be quite delicate. Rings in some trees can be missing, but not in others nearby, so that with many trees, if one is not careful the number of years with a fire scar can be over-counted. Recall again that “absence of evidence is not evidence of absence.” Trees that experience a fire will not always be scarred. In particular, the first scar is the hardest to make.
- 33.
And some researchers have managed to place fire scars precisely enough within the annual growth cycle so as to estimate the season in which a tree was scarred. This is impressive, but we won’t need the details for our discussion here.
- 34.
This good estimate can be tricky to verify. Because trees establish at different times after a stand-replacing fire, the end result will be a distribution of ages for each patch. A reasonable choice for the presumed year of the fire is the assumed establishment date of the oldest tree (“fire was no later than this year”), but there are arguments pro and con other choices.
- 35.
For a toy example, consider a fairly large landscape, probably sampled coarsely by remote sensing (images from aircraft or even satellites) to define the patches, and then by hand to get the tree ring counts. Suppose, for simplicity, that all patches were the same size, and that the following patch ages were estimated: 500, 475, 450, 425, 400, 250, 100, 75, 50 years since fire. It would be reasonable to assume that there was more fire in the early and late parts of the record, and less in the middle. Knowing the climatology (e.g., the Little Ice Age was in the middle of the period), we have the basis for inferences about the effects of climate on wildfire in this region.
- 36.
Some good overviews of this work are J.E. Halofsky, et al. 2011. Mixed-severity fire regimes: lessons and hypotheses from the Klamath-Siskiyou Ecoregion. Free online from Ecosphere 2(4):art40. doi: 10.1890/ESlO-00184.1. P.Z. Fulé, et al. 2003. Mixed-severity fire regime in a high-elevation forest of Grand Canyon, Arizona, USA. Landscape Ecology 18:465–486.
- 37.
At least not in any clear way that can be incorporated in the analysis. For example, would there have been more controlled burns, or fewer, in a hot dry summer conducive to large lightning-caused fires? We don’t know.
- 38.
For example, probably the most detailed and comprehensive fire-scar record we have in the West is from eastern Washington State. In every one of seven watersheds sampled, trees have “stopped” recording fire by 1900.
- 39.
But as in other disciplines, some of these variables are fake. A classic, and a nice example of “the tail wagging the dog,” is the fire season, and its length. Papers have been published in prestigious journals, and cited profusely, that make claims like “Wildfires are increasing around the West because the fire season is getting longer”. But this is circular logic. The fire season is defined or bounded by the fires that occur, it does not cause them, any more than being in the rainy season causes rain.
- 40.
Right here we are focused on the “winners”. In Chapter 7 there will be more about some “losers”.
- 41.
Of course there are two ways to do this. What has little to do with warming climate directly is the success of invasive species, both plants and animals, such as cheatgrass across much of the West and buffelgrass in the Southwest, which were given a running start by being transported here when they would have not dispersed so far on their own. The second way, dispersing from adjacent habitat by taking advantage of a new more favorable climate, is the one I will discuss.
- 42.
For example, the mosquito whose bite carries dengue fever, Aedes aegypti, could have about six times as long a breeding season in middle-to-high elevations in the West by the end of the century, in a “business as usual” scenario for climate change. For details see S.J. Ryan et al. 2019. Global expansion and redistribution of Aedes-borne virus transmission risk with climate change. PLOS Neglected Tropical Diseases (a free online journal) 13(3), https://doi.org/10.1371/journal.pntd.0007213.
- 43.
And we won’t deal with these, for various reasons, including their sheer number and variety and a lack of confidence, for most, that we know what will happen with a warming climate. For some good information on pathogens, though, see https://www.fs.fed.us/research/invasive-species/plant-pathogens/, which covers both invasive and native pathogens.
- 44.
We don’t have any particular reason to believe that has changed over time, but read on.
- 45.
How much of each? The alert reader may see the opportunity for a detection and attribution lesson or analysis here, which could provide valuable data for a land manager. Our focus being on climate, we have something different to emphasize, but if you want to drill down, think limiting factors.
- 46.
This is a fair amount of technical detail already, and there is more in the full story. There are relatively stable “states” of the MPB’s cycle; it has been biennial in the past, but temperatures have warmed enough in large areas of its range that it has switched to annual, meaning that a new cohort of beetles is produced twice as fast as before. These are the sheer numbers that overwhelm the host trees. For the technical details unabridged, see J.A. Powell and J.A. Logan. 2005. Insect seasonality: circle map analysis of temperature-driven life cycles. Theoretical Population Biology 67:161–179.
- 47.
Remember that bark beetles attack pines. With broadleaf trees, this example is less clear.
- 48.
For much more on this topic, see J.A. Hicke et al. 2012. Effects of bark beetle-caused tree mortality on wildfire. Forest Ecology and Management 271:81–90.
- 49.
With respect to the classic elements of disturbance: frequency, severity, and (particularly) extent.
- 50.
For example, the western spruce budworm is a defoliator that has reached outbreak numbers in the Southwest nine times in the last 300 years. Unlike bark beetles, however, its strongest climatic driver seems to be precipitation; in the Southwest wetter springs are associated with increased activity. For the classic study, with our invaluable tree rings, see T.W. Swetnam and A.M. Lynch. 1993. Multicentury regional-scale patterns of western spruce budworm outbreaks. Ecological Monographs 63:399–424.
- 51.
Full disclosure: here is one forest type for which I mostly agree with the “universalists” who believe that wildfire will increase everywhere.
- 52.
This is not universal logic for bark beetles; it applies only to live-cambium feeders. Other species colonize wood that is already dead, and so the logic changes.
- 53.
In other words, changes in one don’t affect changes in the other. For example, there is no particular evidence that it takes both of them to kill a tree; either one will suffice.
- 54.
A clear example of this, though somewhat technical, is Kaspari et al. 2015. Accelerated glacier melt on Snow Dome, Mount Olympus, Washington, USA, due to deposition of black carbon and mineral dust from wildfire. Journal of Geophysical Research: Atmospheres 120:2793–2807.
- 55.
Earthquakes create some of the biggest mass movements, of course. Even if we someday find a way to predict earthquakes, projecting their changes in a warming climate is likely to remain beyond our reach.
- 56.
OK yes, mass movements change the specifics of local topography, but not its general characteristics, e.g., steep and rugged vs. gentle or flat.
- 57.
On average, and all other things being equal, which they never are exactly. For example, if a slide-prone slope gains tree cover over time from growth accelerated by warming, its inherent stability will increase, offsetting its increased annual likelihood of failing because of more frequent severe rainfall. For a technical exposition of the expectations under warming climate, see S.L. Gariano and F. Guzzetti. 2016. Landslides in a changing climate. Earth-Science Reviews 162:227–252.
- 58.
For example, backcountry snowsports enthusiasts are familiar with the complexity of the layering of sequential deposits of snow in avalanche-prone terrain. Depending on their depth, moisture content, and the pressure from layers above them, “weak” layers (i.e., prone to slide) can persist under what may seem to be stable snow.
- 59.
In this example, a different species that is not vulnerable to the insect and therefore has not “spent” resources on the same defenses.
- 60.
These are the “actors” at center stage in the climate-change story. Other plants and microorganisms are certainly important in forests and other vegetation types, and we will look at the animal kingdom in Chapter 7. The erudite reader must forgive me for keeping the “casting” at a manageable scale.
- 61.
A key point here, which I come to below, is that adaptation to an environmental factor requires evolutionary time. In this example, the fire regime must be stable enough temporally in its severity for evolution to take its course. If fire regimes change quickly in a warming climate, many “bets” are off.
- 62.
For those who like classifications with evocative names, a researcher named J.S. Rowe wrote a chapter in 1981 in a now out-of-print book that named five classes: invaders, evaders, avoiders, resisters, and endurers. I would call all of these except resisters evasive strategies. For more details, track down R.W. Wein and D.A. Maclean. 1981. The Role of Fire in Northern Circumpolar Ecosystems. Or see James K Agee’s classic Fire Ecology of Pacific Northwest Forests. 1993. Island Press. pp. 135–136, although the entire book is worth a careful read.
- 63.
We can see a subtlety in the evolutionary game by looking at how severe are fire regimes where lodgepole pine is dominant or common, and asking what percentage of the cones at each site are truly serotinous. Not surprisingly, the percentage of serotiny increases with fire severity, although nowhere does it reach 100%.
- 64.
See just below where we revisit “Forests on the brink.”
- 65.
Forecasting changes in fire frequency is not straightforward, despite claims in the media and elsewhere that a warming climate will cause more frequent fires. Consider the negative feedback to fire frequency from more area burned and more severe fire (both also being predicted for the future), if fire renders a site much less flammable, even if for only a few years. More on this later in Chapter 6, in the section on detection and attribution.
- 66.
Fire does kill whitebark pines, but their seeds are often cached by birds (see Chapter 7) some distance from their source, thereby varying the locations of seedlings.
- 67.
There are subtleties here that I am not bringing in, to keep the narrative clear. Late-successional species’ adaptations to the climate that was in place when they were seedlings would not necessarily extend to regenerating successfully in the open; even then they need the moderating environment under a canopy. If heat or drought were the “worst” parts of that open environment, then the situation is worse in a warming climate. If cold and snow were the limiting factors, then they might regenerate better in a new (warmer) climate.
- 68.
In area, compared to their proportional area in the mountains as a whole. I make no value judgments here (but see Chapter 9).
- 69.
In general, over all life stages. In some arid mountain environments, it is actually desiccation that kills seedlings. This is a combination of water limitation and energy limitation, the latter in that cold prevents plants from accessing available water, either if it is frozen or their tissues cannot transport it.
- 70.
This discussion draws heavily on a recent paper (free online): C.A. Cansler, et al. 2018. Fire enhances the complexity of forest structure in alpine treeline ecotones. Ecosphere (2):e02091. https://doi.org/10.1002/ecs2.2091. This was an interesting case where the authors started out expecting one of the two “forces” to win out. Thomas Huxley said that “many a beautiful theory was killed by an ugly fact”. But the facts, like these, can be more beautiful themselves, and still kill a theory.
- 71.
That visibility is a reason that this is a good example, but it is not the only one. The same interactions—warming climate, disturbance, succession—will play out in interesting and complex ways on many landscapes, such as closed-canopy forests, in which the changes may be harder to see from a vista point.
- 72.
OK I am intentionally “passing the buck” of attribution. How do we know that these events are “caused” by warming climate? More on that in Chapter 8. For now, it is simply a criterion.
- 73.
In some cases this is an obvious evaluation, in some not. For example, replacement of lodgepole pine by sagebrush would qualify, but what about replacement of pinyon pine by juniper, the big dieback events in the Southwest, or other low-stature pines by scrubby oaks, in California?
- 74.
Even though not all sea-level rise is from melting snow and ice—thermal expansion causes a bit of it—the attribution is still quite clear. As ice on Greenland and Antarctica melts from global warming, sea level rises.
- 75.
The main technical problem is that the fires thought to be “climate-change” fires are rare. Both detection and attribution, of anything, are much easier when there are large samples. The statistics involved are what is called “extreme-value” theory, which compensates for the rarity of events with certain compromises that lead to more uncertainty than ordinary statistics. The main practical problem with fires is (curiously) the reverse of rarity: There are so many different circumstances and types of weather and vegetation that it is difficult to isolate the important factors in each case.
- 76.
More on these in Chapter 8. Their rarity is one difficulty, but not the only one.
- 77.
Full disclosure: Unlike with other model types I have discussed, I have commissioned, written, reviewed, and used landscape disturbance models. I hope that the reader will see quickly that this does not make me blind to their limitations, even though I have just labeled them as the sine qua non.
- 78.
Meaning that they are different in each simulation. See Note 34, Chapter 3.
- 79.
Fire is the only example I will use for this first step. The empirical models for fire are the best developed for any disturbance, in the West. There are also models of insect outbreaks, but many of the issues are the same, so I won’t cover them separately. For a good discussion of bark beetles and climate change, see B.J. Bentz, et al. 2010. Climate change and bark beetles of the Western United States and Canada: direct and indirect effects. BioScience 60:602–613.
- 80.
There are also models of “fire probability”. This turns out to be a convoluted statistical transformation of area burned, but with enough numerical noise that the models are weaker than area-burned predictions.
- 81.
Meaning that a process is being interpolated or extrapolated outside its range of validity. For example, if we ran a GCM at 1-km grid spacing (instead of 1°), there would be no way we could represent circulation meaningfully. Specifically, our outcomes would have false precision. In the case of fire, a recent paper modeled the association between climate and area burned at a very fine scale (~1°). The result was much obvious false precision, to be expected when climate and fire regimes were represented uniquely at that scale.
- 82.
Yes, we did that. C.L. Raymond and D. McKenzie. 2012. Carbon dynamics of forests in Washington, U.S: projections for the twenty-first century based on climate-driven changes in fire regime. Ecological Applications 22:1589–1611. The outcome was similar to that from the eco-hydrological models of Chapter 4. We asked what the difference was in carbon storage between present and future in Washington State.
- 83.
Why approximate physics? The true physics is understood at the scale of centimeters or even finer. This quickly becomes impractical on a computer when trying to cover even one 30 × 30 m area, taking longer by orders of magnitude to simulate combustion than it takes the same process to happen in real time.
- 84.
For a roadmap, as yet unrealized, see R.E. Keane, et al. 2015. Representing climate, disturbance, and vegetation interactions in landscape models. Ecological Modelling 309:33–47.
- 85.
For a review of these in landscape modeling, see E.A. Newman, et al. 2019. Scaling and complexity in landscape ecology. Frontiers in Ecology and Evolution doi:10.3389/fevo.2019.00293. (free online)
- 86.
Specifically, strong (chinook) easterly winds, low humidity, and hot temperatures. Predicting these in association with global climate, in the Cascade Range and elsewhere such as Southern California, is difficult.
- 87.
But not overly so in Yellowstone. It may just be a ticking time bomb, or have escaped the levels of damage further south (in Colorado) and north (in British Columbia, Canada) by having an active fire regime (thanks to mostly flat terrain and less fire exclusion than in other areas dominated by lodgepole pine) that keeps the proportion of the most vulnerable age class (80 years and older) down.
- 88.
In the Western Mountains, populations of oaks are rarely called forests, rather woodlands, if large trees such as the oaks of the Pacific Coast Ranges, or shrublands, if the smaller oaks inland. This is partly subjective, and can be misleading in areas like carbon accounting, especially because the large-oak woodlands in the Coast Ranges have a lot of biomass.
- 89.
Remember that we need a critical mass of resources, whether food (for defoliators) or tissue for cambium feeders and their galleries, to enable and sustain an outbreak.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this chapter
Cite this chapter
McKenzie, D. (2020). Ecological Disturbance. In: Mountains in the Greenhouse. Springer, Cham. https://doi.org/10.1007/978-3-030-42432-9_6
Download citation
DOI: https://doi.org/10.1007/978-3-030-42432-9_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-42431-2
Online ISBN: 978-3-030-42432-9
eBook Packages: Biomedical and Life SciencesBiomedical and Life Sciences (R0)