Abstract
Forests cover about half of the American West, but a much greater proportion of the Western Mountains. Most of them are dominated by conifers rather than deciduous trees. We start with some basic forest ecology, and then explore the vulnerability of the forests of western mountains to a warming climate. Some forests and woodlands may be converted to shrublands, whereas others will be resilient enough, at least temporarily, that they will endure, albeit with changing species density and composition. I discuss potential dispersal and migration of forest species in the face of climate change, and note the considerable barriers to these in the West, from both mountain topography and disconnected natural areas. I introduce the idea of source vs. sink with respect to carbon sequestration, and discuss the tradeoffs between CO2 fertilization and water stress, and the feedbacks (positive if a source, negative if a sink) to climate warming.
The woods are lovely, dark, and deep
But I have promises to keep
And miles to go before I sleep
And miles to go before I sleep.
—Robert Frost
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
So how much exactly? There is no one correct answer, because there is no clear dividing line between forest and non-forest. The terms “closed forest”, “open forest”, “woodland”, and “savanna” are used world-wide to refer broadly to varying levels of tree cover, but these have their own gradations, and even the coarse-scale thresholds are not agreed upon universally. In keeping with this uncertainty, I will use the terms but not define specific thresholds between them.
- 2.
Once again, no precise numbers exist. A further complication is that not everyone will agree as to whether such things as “rolling hills”, mesas, bluffs, etc. should be included in mountain acreage.
- 3.
Depending on how rocky and steep those summits are, of course. See Chapter 2.
- 4.
The Earth’s boreal (referring here to high latitudes, not boreal vs. austral for the Northern vs. Southern Hemispheres) forests, in Alaska, Canada, Scandinavia, and Siberia, are primarily coniferous. There seem to be physiological limits to “how cold you can go” for broadleaf trees. A scholarly review of this is in C. Körner et al. 2016. Where, why and how? Explaining the low-temperature range limits of temperate tree species. Journal of Ecology 104:1076–1088.
- 5.
More technically, if the most favorable seasons for biomass accumulation via photosynthesis are when only conifers have foliage, then they have a strong competitive advantage. Mild winters, particularly in the Cascade Range in which conifers dominate almost totally, and hot dry summers across most of the West, during which growth partly or wholly shuts down due to water scarcity, both favor a life form that can add biomass year-round.
- 6.
Scientific (in pseudo-Latin) names comprise the genus (singular of genera) and species, along with a “subspecies” if it exists. The main taxonomic hierarchy, for those who don’t recall high-school biology, is Kingdom, Phylum, Class, Order, Family, Genus, Species (whose mnemonic is King Philip Cried Oh For Goodness Sake, or its less printable varieties). Unfortunately for simplicity, the most important broad distinction between tree life forms (a term I will use loosely throughout within trees and among trees, shrubs, grasses, and herbs) is gymnosperm vs. angiosperm, most simply cone-bearing vs. flowering. This is a division “between” Kingdom (plants) and Phylum, into six phyla of gymnosperms, but eight “groups” (and 84 classes) of angiosperms. For our purposes, “conifer” versus “broadleaf” covers the essentials.
- 7.
Sometimes we are lucky enough to identify them down to the level of species, but often it is a coarser level of the taxonomic hierarchy. The whole field of molecularly based fossil identification (from fossil DNA and other components) is outside our scope of discussion here.
- 8.
The aphorism “Absence of evidence is not evidence of absence” is relevant here. We can say, more or less for sure, which species or life forms were present; we cannot say which were not.
- 9.
“Of course some do go both ways”, as the Scarecrow said. We saw in Chapter 3 that tree-ring analysis can be used to infer climate, and here I am saying that it can be used to infer things about trees. With the caveat that correlation is not (always) causation, we prefer to start any analysis by being explicit about which inferences we want to make, and why.
- 10.
The translation across scales can be problematic, as can other extrapolations outside the domain that an initial inference was made. For example, if a bit more CO2 is a good thing, would a lot more be better? We will look at some of these limitations explicitly in Chapter 8, and I will be careful here to avoid the “if a little is good, more is always better” error.
- 11.
But there is virtually no “pure” anything in ecological systems. For example, a principal cause of seedling death at high elevations is desiccation, i.e., water limitation.
- 12.
Though far fewer than in the East. Large areas of the Western Mountains are dominated by one or a few conifer species. Broadleaf species other than quaking aspen are usually found at the lower elevations of forests (e.g., oaks) or in recently disturbed (e.g., red alder, bigleaf maple) or riparian (along rivers or streams) areas (e.g., cottonwoods, alder, ash).
- 13.
For example, in most stands of conifers, individual trees are connected belowground by ectomycorrhiza (literally, fungi [“myco”] on the surface [“ecto” as opposed to “endo”] of roots [“rhiza”]. These symbiotic fungi enable individual trees to share resources through a connected root system. There are also claims, of varying credibility, of more profound communication among organisms through ectomycorrhiza. I will leave the interested reader to explore this.
- 14.
In the same sense, as we learn in high-school physics, as for gravitational potential energy or electrical potential (voltage). When this energy is catalyzed and released, it becomes useful energy, for growth (biological), power (electrical), or colliding bodies (gravitational).
- 15.
Because of this, the paleo-climate reconstructions that I discussed in Chapter 3 require careful selection of trees for sampling tree rings. Trees experiencing minimal competition, often at the extremes of their ranges, are considered optimal.
- 16.
Recall our discussion of leaf-area index (LAI) in Chapter 2 (Note 12). The term “closed-canopy” is used loosely, and not always consistently, to indicate LAI > 1.
- 17.
For example, the gymnosperms (our conifers dominant in the West) probably originated in the late Carboniferous Period, around 320 million years ago. Angiosperms are probably more recent, but there is some disagreement about their origin, putting it somewhere between 250 and 200 million years ago.
- 18.
For example, a photosynthesis light-response curve is a graph with incoming light on the X-axis and a plant’s photosynthetic rate on the Y-axis. These graphs usually start up steeply and level off, under “light saturation”. A curve that starts the most steeply, but levels off quickly, is probably a shade-tolerant species. It doesn’t take much light for it to assimilate at its maximum rate.
- 19.
But probably well inside the projected extremes. More on this topic in Chapter 8.
- 20.
For example, the dominant trees in many forests across the West established in the Little Ice Age (LIA), roughly 1300–1850, depending on whom you ask, and with particularly cold periods around 1650, 1770, and 1850 (but which probably hit Europe harder than the American West). If ours were an equilibrium climate that had lasted a thousand years, instead of considerably warmer than the LIA, our “undisturbed” wilderness forests might look quite different.
- 21.
I will use this term as a rough equivalent of “adolescent”. Numerous more precise definitions, for various purposes, can be found in forestry.
- 22.
Remembering, of course, that evolutionary time is itself many orders of magnitude different for some organisms than for others. For example, evolutionary time scales for bacteria are shorter than ecological time scales for the long-lived trees that we are discussing. This scale comparison is limited to organisms with similar lifespans, specifically with respect to reproductive “turnover”.
- 23.
But this is partly a consequence of the unprecedented speed of the current warming. Most climate changes in Earth’s past, even the fairly rapid PETM (see Chapter 3), were more in sync with the evolutionary time scales of long-lived organisms.
- 24.
We will see this term later, in Chapter 8, as an element of complex ecological systems that can be difficult to predict even with detailed knowledge of fine-scale processes. That can be a challenge, because the emergent properties of systems are often the ones that we want to predict.
- 25.
This process is called nitrogen fixing, and it will suffice to know that no one expects this to change much in a warming climate, at least not until extreme changes occur that are outside our time frame here.
- 26.
For conifers, “usable” generally means ammonium ions (NH4+); for broadleaf trees nitrates (NOx−). Nitrate pollution, whether or not associated with climate change, can certainly affect ecosystems, but is unlikely to affect this early stage of succession.
- 27.
An experienced traveler in the West might object that there are many stands of pure aspen in the Rocky Mountains and eastern slopes of the Cascade Range, and mixed stands of broadleaf trees in riparian areas. True. Species composition in the riparian may be stable, in that frequent disturbance (e.g., flooding) brings back the same species that dominate. The case of aspen is unique, and it is not yet understood (meaning that there is considerable disagreement) how stable aspen stands are in space; they may be a continually moving mosaic.
- 28.
Partly because it is more or less linear, but also because it is the paradigm that was long taught in forestry schools, in which timber management is a prime objective. The key element was that the most valuable timber species, for example Sitka spruce and Douglas-fir in the Pacific Northwest, dominated in middle succession, so “late-successional” became synonymous with “over-mature” or simply “decadent”.
- 29.
A (really, almost prehistoric) classic paper laying out this paradigm, for one particular example of it, is P.J. Catellino et al. 1979. Predicting the multiple pathways of plant succession. Environmental Management 3:41–50. A more recent exposition is D.C. Donato et al. 2012. Multiple successional pathways and precocity in forest development: can some forests be born complex? Journal of Vegetation Science 23:576–584.
- 30.
For those of us who had elementary schooling in logic, and still have at least a dim memory of it, this is an example of the converse of a statement or theorem not necessarily being true even if the statement itself is true. Elementary, yes, but it can become complicated quickly, as can everything in ecology. A good way to avoid many pitfalls is to remember, as I said above (Note 8), that “absence of evidence is not evidence of absence”.
- 31.
Here we should distinguish among dispersal, migration, and “translocation”, as they are used technically. Dispersal is a permanent move, either by seeds (plants) or juvenile or adult animals away from the place of their birth. Migration is a seasonal move, followed by a move “back” within the year (the unfortunate phrase “assisted migration”, though widely used, really refers to assisted dispersal or translocation). Translocation, or assisted colonization, or “managed relocation” refer generally to human-assisted moves.
- 32.
A few conifer species can “move” by clonal expansion, as can many broadleaf species. This will certainly affect local species composition, as anyone who has watched the expansion of a clone of quaking aspen can report, but it is clearly much slower than seed or juvenile-animal dispersal.
- 33.
For example, as we will see in Chapter 7, the Clark’s nutcracker caches seeds of whitebark and limber pine, then often “forgets” them, to the advantage of hungry grizzly bears in autumn.
- 34.
In theory, with the right timing, a migrating bird could transport a seed for hundreds of miles, depending on the speed of both its flight and its digestion. For example, Arctic terns have been known to migrate tens of thousands of miles in a year, but birds that cache seeds (i.e., can hold on to them) do so within their local home ranges.
- 35.
Whether or not the climate is changing. That doesn’t really affect dispersal distances directly in any way that we can measure, but as I discuss next, it affects whether dispersal is successful.
- 36.
And of course they will never be equal. By definition, elevation changes quickly and continuously in mountains. Average lapse rates (relatively consistent changes in temperature with elevation) can be estimated, but they can vary widely, and with elevation-dependent warming (see Chapter 3) they may lessen. Topographic influences on precipitation also vary widely, meaning that the interplay of energy and water limitations is harder to estimate than if we could rely on broad averages.
- 37.
As far as I know, no one has tried to do these computations rigorously, for an obvious reason: barriers are virtually everywhere. This has not stopped people from creating climate velocity “grids”, for example, 1-km resolution grids for North America based on output from GCMs. For example, see https://adaptwest.databasin.org/pages/adaptwest-velocitywna.
- 38.
Easiest for birds, more difficult for un-winged animals that have to walk across, and for plants, which have to ride the wind.
- 39.
For a toy example, suppose we have a series of ridges about a half-mile (800 m) apart. There is a 90% probability that a seed will disperse 800 m with average winds before the environment becomes too hostile. Consider the simple probability that a species will disperse successfully across 5 ridges (with our numbers 0.95 = 59%). What if our 90% were off by 5%? 0.855 = 44%. More likely than not easily becomes less likely. This doesn’t mean that we shouldn’t try, only that we should be careful with our claims.
- 40.
One could argue that this is a bit too simple. It could be that there were enough dispersal “events” of a certain distance to enable this movement, but that seedlings were always out-competed by other species. An implicit assumption of climate-velocity models is that the “landing zones” for dispersing species will give them the same competitive advantage in the future as they have now in their current habitat.
- 41.
In a situation like this it is best to “blur” annual estimates of climatic variables, to avoid false precision.
- 42.
Why a lower bound? The species was at two places at time 1 and time 2, so it could move at least fast enough to span the distance. But remembering that “absence of evidence is not evidence of absence”, it might have dispersed farther, but we haven’t observed that.
- 43.
This is another instance of the problem of non-stationarity, to which I have referred before and will again.
- 44.
We wrote a paper about this. L. Iverson and D. McKenzie. 2013. Tree-species range shifts in a changing climate—detecting, modeling, assisting. Landscape Ecology 28:879–889.
- 45.
We all know that carbon comes in other forms besides CO2, but CO2 is the principal “vehicle” by which carbon enters and leaves ecosystems. I use it and “carbon” somewhat interchangeably.
- 46.
For example, a useful metric is the Net Ecosystem Exchange (NEE), the gain or loss per unit time of CO2 from an ecosystem. This can be measured in “bulk” by various sensors, thereby characterizing a forest as a net source or sink. On regional to global scales, it can be used to inform Earth-system models, which incorporate exchanges between the atmosphere and the land surface (and the oceans) to project response to climate change.
- 47.
Even though these latter are a tiny fraction of the total forest biomass, their biology can be important for the health and resilience of forests. See Note 13.
- 48.
A reminder that weight is actually a measure of force, not the English-unit analog of the metric units of mass. A person of a certain mass will have that same mass on the moon, but only about a sixth of the weight; weight = mass × the local acceleration due to gravity, in the right units, whether English or metric. For those who remember their elementary physics, F = ma.
- 49.
More of this below in “Feedbacks to climate change” and in Chapter 8.
- 50.
The evidence for this is from small-scale controlled experiments. The complex task of extrapolating these experiments to the larger-scale dynamics of forests is still a challenge.
- 51.
Which of course they are not, but bear with me for a bit. We will see that the energy-water gradient forms a robust baseline for variation introduced by both environmental pressures and human activities.
- 52.
I have shied away from drawing such a map, because it would be fake while tempting viewers, including me, to view it as real. This is something maps are known to do. We are nowhere close to having such a map, but here are some things that might happen in a warming climate. (1) Cold high-elevation forests that were deep green might become less so, (2) arid low-elevation forests that were brown (of all shades) would become more so, (3) forests near the center of the gradient, on the green side, could turn brown, (4) forests shifting toward the center would shift toward the “sink” end of the associated source-sink gradient, whereas those shifting away from the center would shift toward “source”. Do you see why #4 is true, from our discussion of limiting factors? And knowing what the forcings are, do you also see why practically all movement would be away from the green end and toward the brown end? Another thing that would happen is that the pieces of the patchwork would change shape and size.
- 53.
And I won’t do this explicitly, with numbers. Accounting methods vary, as do the numbers from year to year, and the stationarity of the calculations, whose parameters may change.
- 54.
For a not too technical summary of forest feedbacks to climate change, see G. Bonan. 2008. Forests and climate change: forcings, feedbacks, and the climate benefits of forests. Science 320:1444–1449.
- 55.
For perspective, the albedo of snow is about 0.8 (out of 0.0, the lowest possible, and 1.0, the highest). Deciduous forests are between 0.15 and 0.18; conifer forests between 0.08 and 0.15, about the same as worn asphalt; open ocean (without sea ice, which is 0.5–0.7) is 0.06. Green grass is about 0.25. So forest vs. non-forest matters, but snow vs. non-snow matters a lot more.
- 56.
More to the point, there is a positive feedback between water vapor and CO2 that amplifies global warming, because a warmer atmosphere can hold more water vapor.
- 57.
For the particulars, see A.L.S. Swann et al. 2010. Changes in Arctic vegetation amplify high-latitude warming through the greenhouse effect. Proceedings of the National Academy of Sciences, USA. 107:1295–1300.
- 58.
Where will these things happen? If you recall our discussion of leaf-area index (LAI: Chapter 2, Note 12), with LAI above 1 the albedo is not likely to change much, whereas moving from 0.3 to 0.8, for example, will clearly reduce it. In general, this feedback will be more relevant in arid mountain forests with less than 100% canopy cover. We have a lot of these in the Western Mountains.
- 59.
This discussion is based on A.L.S. Swann et al. 2018. Continental-scale consequences of tree die-offs in North America: identifying where forest loss matters most. Environmental Research Letters 13: (055014—paper identifier).
- 60.
The informed reader will notice that I am ignoring another contributor to sea-level rising: warming ocean temperatures. For purposes of comparison, I am also ignoring the loss of glaciers and perennial snow outside of Greenland and Antarctica, which contribute a bit to sea levels.
- 61.
And much more evident without accurate instruments. Whereas sea levels change in increments of millimeters, forests change in increments of squared kilometers (six orders of magnitude difference).
- 62.
A fairly recent review of observations, with some hypotheses about causes and uncertainties, is C.D. Allen et al., 2010. A global overview of drought and heat-induced tree mortality reveals emerging climate change risks for forests. Forest Ecology and Management 259:660–684.
- 63.
Recall from Chapter 3 that the 1° increase is our best evidence of a warming climate per se. Other observations, such heat waves, are considered to be the effects of the global rise, via feedbacks and changes in circulation.
- 64.
For a detailed discussion of such thresholds, see N. McDowell et al., 2008. Mechanisms of plant survival and mortality during drought: why do some plants survive while others succumb to drought? New Phytologist 178:719–739.
- 65.
A personal observation: the first timber management models began to proliferate about when the sweep of timber extraction across the US had more or less reached the Pacific Ocean. Interest in models rises in a world with limits.
- 66.
Recall though from our earlier discussions of models that this distinction is not so clear as it may appear to (or be wished for by) some. For example, “biologically based” here is the analog of “physically based” in climate models, but often the parameters in the mathematical representations of relationships are obtained empirically, probably through statistical estimation.
- 67.
Or we might even ask “What are the chances of forest dieback?” With the current imperfect state of forest models, we would be hard pressed to get that right.
- 68.
More precisely at 30 arc s, about 0.924 km2 at the Equator, but changing moving poleward. See D. Bachelet et al., 2018. Translating MC2 DGVM results into ecosystem services for climate change mitigation and adaptation. Climate. Free online at https://www.mdpi.com/2225-1154/6/1/1/htm.
- 69.
For empirical models these are the values of the driving variables, since there is no evolution from one state to the next, by definition.
- 70.
Simplifying again. We solve for parameters, but also for the uncertainty associated with them. Even if those parameters are “right”, the range of possibilities may be too large to give a meaningful answer.
- 71.
And as we proceed through chapters of the book, I trust that the reader will bear with me as I add specific inferences incrementally to these sections based on the topics in each chapter.
- 72.
Geographically speaking. Of course, as we have seen, things also change with elevation, particularly on the drier aspects.
- 73.
Recall that nutrients limited by competition are an energy source.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this chapter
Cite this chapter
McKenzie, D. (2020). Trees, Forests, and Carbon. In: Mountains in the Greenhouse. Springer, Cham. https://doi.org/10.1007/978-3-030-42432-9_5
Download citation
DOI: https://doi.org/10.1007/978-3-030-42432-9_5
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)