Just What Are Temperate and Boreal Rainforests?

Abstract

When most people think of rainforests, they think of lush, tropical “jungles”teeming with poison arrow frogs (Dendrobates spp.), toucans (e.g., Ramphastos sulfuratus), mountain gorillas (Gorilla gorilla beringei), and jaguars (Panthera spp.).Tropical rainforests are indeed special places, as they account for over half the terrestrial species on Earth (Meyers
et al. 2000) while representing just 12 percent of the world’s forest cover (Ritter 2008). Their temperate and boreal counterparts
are another story, though, one yet to receive the kind of global recognition rightfully merited by tropical rainforests. Their
story is told here,beginning with historical and recent accounts to define and map the temperate and boreal rainforests of
the world.

Full text

C H A P T E R 1
P
Just What Are Temperate and
Boreal Rainforests?
Dominick A. DellaSala, Paul Alaback,Toby Spribille,
Henrik vonWehrden, and Richard S. Nauman
When most people think of rainforests, they think of lush, tropical “jungles”
teeming with poison arrow frogs (Dendrobates spp.), toucans (e.g., Ramphastos
sulfuratus), mountain gorillas (Gorilla gorilla beringei), and jaguars (Panthera spp.).
Tropical rainforests are indeed special places, as they account for over half the
terrestrial species on Earth (Meyers et al. 2000) while representing just 12 per-
cent of the world’s forest cover (Ritter 2008).Their temperate and boreal coun-
terparts are another story,though, one yet to receive the kind of global recog-
nition rightfully merited by tropical rainforests. Their story is told here,
beginning with historical and recent accounts to define and map the temperate
and boreal rainforests of the world.
Any discussion of rainforests must begin with what we mean by this term
and how we map rainforests. Definitions and mapping standards are the mortar
with which scientists visually construct biome delineations such as temperate
and boreal rainforests. Consequently, the modeling techniques used in this
chapter frame the entire book,as each of the regional chapters is built from the
approaches set herein. In cases where it is necessary to deviate from globally
based models and maps,explanations are given by regional authors of the book.
Nevertheless,we now build on earlier approaches and definitions of temperate
and boreal rainforests by providing a standardized modeling approach and a
consistent methodology for mapping these rainforests.While it was our original
intent that readers of this book would use our approach as the up-to-date stan-
dard for defining and delineating temperate and boreal rainforests,we note that
DOI 10.5822/978-1-61091-008-8_1, © Island Press 2011
1
,D.A. DellaSala (ed.), Temperate and Boreal Rainforests of the World: Ecology and Conservation

2t e m pe ra te an d b or e a l ra i n f o re s t s of th e w o r l d
this is a work-in-progress requiring further refinement and real-world verifica-
tion as new data sets become available.Similarly, in Chapter 10,we present stan-
dardized mapping techniques aimed at determining just how much of this rain-
forest biome is in strict protection, a necessary step for developing a unifying
vision for rainforests globally and for calling on decision makers to protect
these rainforests as we do in Chapter 11. Because the process used in this open-
ing chapter is central to the entire book,we put more emphasis here compared
to the regional chapters that follow.
SCIENTIFIC HISTORY OF TEMPERATE AND
BOREAL RAINFORESTS
Throughout this book we refer to either temperate or boreal rainforests that
differ mainly with respect to latitude, climate, and plant associations. For de-
scriptive purposes we separate these rainforest types in this chapter but refer to
them jointly throughout much of the book.
Temperate Rainforests
Temperate rainforests have been recognized in some fashion by ecologists for
nearly a century (Köppen 1918; Holderidge et al. 1971;Whittaker 1975; Jar-
mon and Brown 1983; Veblen 1985; Read and Hill 1985; Omernick 1987;
Moore 1990; Hickey 1990; Alaback 1991; Kirk and Franklin 1992; Kellogg
1992, 1995; Gallant 1996; Lawford et al. 1996; Schoonmaker et al. 1997; Moen
1999). Most researchers classify them as distinct biomes based on broad differ-
ences in dominant vegetation and/or climate, or as inclusions within larger
ecoregions (large areas distinguished by their dominant vegetation, climate, and
land form).Yet a simple internet search for “temperate rainforest” yields incon-
sistencies in mapping locations due to gross differences in definitions and map-
ping techniques.
An earlier term, “high-latitude rainforest,” was proposed by researchers to
describe the pan-American portion of the biome (Lawford et al. 1996), since
this is the most simple and unambiguous way to define temperate as contrasted
with tropical (low-latitude) rainforests, but “high-latitude rainforests” has in-
creasingly been replaced by “temperate rainforests,” which generally have
milder climates than boreal rainforests,due primarily to comparatively low lat-
itudes. A number of temperate rainforest subtypes are described later in this
chapter in order to distinguish rainforests from one another, and this terminol-
ogy is used throughout this book.

Boreal Rainforests
The border between boreal and temperate has traditionally been defined as the
zone where conifer forests give way to deciduous forests, or, in drier regions,
grasslands, roughly equated by Köppen (1918) with the –3°C January isotherm
in the south (Tuhkanen 1984).The delineation of boreal versus temperate is
blurred in montane regions,where temperate coniferous forest transitions seam-
lessly to boreal conifer forest.The important thing to note here is that boreal is a
latitudinal zone and should not be conflated with terms such as continental; bio-
geographers are unanimous in recognizing some high-precipitation oceanic re-
gions as part of the boreal zone.Tuhkanen (1984) compared a wide variety of
different approaches to delineating the northern and southern limits of the bo-
real zone,and in the integrated classification he proposed that several of the rain-
forest regions treated here as“temperate” would be considered part of the boreal
zone.Nonetheless, throughout this book,we use the term boreal to describe the
cold northern rainforests of what in other studies have been more generally
termed subpolar. As we will see later, these include the Pacific Coast of North
America north of ~55°N latitude (chapter 2), the northern half of the inland
rainforest of Northwestern North America (chapter 3),much of the wet forests
of Eastern Canada (chapter 4), portions of Norway (chapter 6), and Inland
Southern Siberia (chapter 9). Because there is no boreal zone in the Southern
Hemisphere,relatively colder areas in this hemisphere are considered subpolar.
In reality, many temperate rainforests straddle the abiotic (nonliving chem-
ical and physical factors) boundaries between temperate and boreal,both latitu-
dinally and altitudinally, and more so for oceanic boreal systems. Thus, these
rainforests serve as a phytogeographical bridge, facilitating the exchange of
mesic (moist) floral elements among neighboring systems and as corridors of
latitude- and slope- related south-to-north, north-to-south and slope-up,
slope-down migrations of wildlife during periods of climate change. How
much of the forests included in this book is boreal versus temperate depends on
which classification system chosen.The fact that highly similar forest-species
assemblages can be found on both sides of artificially drawn lines is a topic best
reconciled to biogeography debates.
RAINFOREST DEFINITIONS
Where and how to draw the line between temperate and boreal rainforests has
changed over time as more and better data have become available regarding
these unique rainforests and the conditions that have created them. Several
JustWhat AreTemperate and Boreal Rainforests? 3

geographers who developed classifications for the world’s climate included a cat-
egory for temperate rainforest based,for instance, on some combination of cool
temperatures and high rainfall,or cool temperatures and a small annual range of
temperatures (see below).Whittaker (1975) in his classic ecology text Communi-
ties and Ecosystems also identified a temperate rainforest type. Most of these early
efforts separated the Southern Hemisphere forests into a broadleaf evergreen
forest type,further complicating a comprehensive global definition.These classi-
fications vary widely in how they portray the distribution of temperate rain-
forests, and especially what types of temperate rainforests occur on Earth.
The prevailing definition of temperate rainforest began with work in the
1980s,when the environmental group Ecotrust and its collaborators proposed a
more precise definition so that more accurate global maps and conservation
strategies could be developed (Alaback 1991, 1996; Kellogg 1992, 1995). The
first iteration of this work included a definition for these rainforests consisting
of: (1) annual precipitation exceeding 1,200 millimeters with 10 percent or
more occurring during summer months; (2) mean July temperature of 16°C or
less; (3) cool dormant seasons; and (4) infrequent fire that is an unimportant
evolutionary factor (Alaback 1991). Soon it became apparent that this defini-
tion was too restrictive, and more important, it did not accurately characterize
availability of moisture,since there was no direct link between evaporation and
the required minimum amount of rainfall. The most biophysically precise
method of doing this would be to calculate potential evapotranspiration, which
corrects for latitude—with increasing latitude, less precipitation is required to
maintain the same humidity levels (Stephenson 1990). Potential evapotranspi-
ration was also later shown to precisely predict the distribution of at least one
common rainforest tree in northwestern North America, western hemlock
(Tsuga heterophylla), even including its distribution in interior rainforests of
northwestern North America (Gavin and Hu 2006). In the absence of detailed
models and global spatial coverages, a more inclusive definition was proffered
byAlaback (1996). In this case, temperate rainforests meeting the original crite-
ria for annual rainfall were divided into four subtypes (or zones,including bo-
real), analogous to subtypes of tropical forests, based on seasonality of precipita-
tion and annual temperatures:
•Subpolar—summer rainfall is above 20 percent of the annual total, sum-
mers are cool, and snow is persistent in winter, with mean annual tem-
perature below 4°C.
•Perhumid—summer rainfall is above 10 percent of the annual total, sum-
mers are cool, and typically transient snow is present in winter, with
4t e m pe ra te a n d b or e a l ra i n f o re s t s o f th e w o r l d

mean annual temperature of 7°C.“Cool-temperate”also has been used in
this context.
•Seasonal—summer droughts and fires can periodically occur, summer
rainfall is less than 10 percent of the annual total,with mean annual tem-
perature of 10°C.
•Warm-temperate—summer precipitation is less than 5 percent of the an-
nual total, winter snow is rare, drought can occur during any season, and
mean annual temperature is 12°C or above (Alaback 1996;Veblen and Al-
aback 1996;Alaback and Pojar 1997).
The threshold values of temperature and precipitation for each of the forest
subtypes was determined by examining climatic conditions in areas along the
west coast of North and South America that possessed key ecological charac-
teristics associated with rainforests.This has been the prevailing set of defini-
tional parameters for describing rainforest regions used throughout the chap-
ters of this book.
A NEW GLOBAL RAINFOREST MODEL
Building on concepts from Alaback (1991), we developed a strongly organ-
ism/ecosystem–driven model for temperate and boreal rainforests that has
identified a very small amount of land surface of the earth within the same
biome and sharing climatic characteristics and associated ecological processes
that rightfully and generally can be called temperate and boreal rainforest.The
processes described herein build on earlier work of rainforest ecologists by pro-
viding a broad suite of climatic criteria and a standardized approach to mapping
rainforests globally.
In this chapter, we use computer modeling to develop defensible criteria
for identifying temperate and boreal rainforests and to locate forests not widely
recognized as rainforest but meeting our criteria. Further,we create a computer
model with high-resolution climate data and compare it to maps created by re-
gional experts.
Rainforest Distribution Model
This book’s chapter authors, from a wide range of rainforest regions, provided
locations of sites they considered typical of temperate or boreal rainforest in
their area. Based on this input, we used climate data for 117 localities from
six regions for the initial modeling step: the Pacific Coast of North America
JustWhat AreTemperate and Boreal Rainforests? 5

(n= 55, mostly coastal); Chile and Argentina (n= 9); New Zealand (n= 10);
Tasmania (n= 6); Norway (n= 15); and Japan (n= 22).These regions were se-
lected because we had localities from collaborators, and because there was little
dispute that the locations represent rainforests (especially the Pacific Coast of
North America, Chile, and New Zealand). Baseline predictors were extrapo-
lated from a global climate data set (Hijmans et al. 2005); redundancy in the
model variables was reduced based on a principal-components analysis of the
complete data set.The final model was constructed using a MaxEnt modeling
approach (Phillips et al. 2006), consisting only of predictors that improved the
model.This yielded 11 discrete climate-related parameters.We used the Max-
Ent model since it is known to be more conservative compared to other
presence-only models, which tend to overestimate occurrence of a particular
variable of interest (in this case,temperate and boreal rainforest).
The model was evaluated with a bootstrapping method (Burnham and
Anderson 2002), resulting in strong support of the predictive ability of the
model (AUC = 0.90; values less than 0.5 indicate no predictive capabilities;see
Phillips et al.2006). Based on 100 repeated runs,we quantified the heterogene-
ity of the ground-truth climate data set, thus ensuring a demarcation of core
zones with a high probability of rainforest occurrence in comparison to areas
with a lower probability (for mapping simplicity, only high-probability areas
were depicted).
The rainforest distribution model generated four additional regions with
climate suitable for temperate and boreal rainforests: the Inland Northwest of
North America (figure 1-1,middle-right portion of panel a—inland British Co-
lumbia), Eastern Canada (figure 1-1, panel b), Great Britain and Ireland (figure
1-1,western corner of panel d), and portions of the Alps (figure 1-1,lower mid-
dle of panel d).Notably,two of these regions have not been widely recognized as
rainforest by scientists,including the wettest parts of Eastern Canada, which ap-
peared in some form in all map iterations,and some valleys of the eastern Alps,in
particular the Salzburg Alps and mountain ranges of western Slovenia. Interest-
ingly, these regions support rainforest lichen assemblages remarkably similar to
those of the Pacific Northwest of NorthAmerica or coastal Norway.
Two lower-latitude regions often considered rainforest by some (e.g., Kel-
logg 1992), such as the Colchic (Georgia) and Hyrcanic (Iran) forests of the
Western Eurasian Caucasus, and the forests of the southern cape of South
Africa, were shown to be in a class of their own compared to the more defini-
tive rainforests of the Pacific Coast of North America and Valdivia. Including
these warmer and drier outliers in the model calibration invariably resulted in
overestimating the global extent of these rainforests by also including South
6t e m pe ra te a n d b or e a l ra i n f o re s t s o f th e w o r l d

Figure 1-1. Temperate and boreal rainforests of the world based on the global rainforest
distribution model, including: (a) Pacific Coast and Inland Northwestern North Amer-
ica; (b) Eastern Canada; (c) Chile and Argentina; (d) Europe; (e) Japan and Korea; and
(f ) Australasia.

American páramo, high-elevationAfrican equatorial fog forests,and nearly half
of the Alps. Retention of the eastern Black Sea region (Colchic), in particular,
resulted in model inclusion of large areas of eastern North America, parts of
which are indeed climatically similar, but did not agree with our initial criteria
on several counts.We settled on a conservative definition of temperate and bo-
real rainforest based generally on the climate data (see table 1-1; figures 1-2,
1-3) presented for nine regions (some were combined from the set above) as
follows:
• Annual (minimum, maximum) temperatures from ~4 to 12°C.
• Annual (minimum, maximum) precipitation from 846 to 5,600
millimeters.
• Snowy winters in high latitudes.
• Significant precipitation (that is,up to 25 percent of annual precipitation)
during the driest quarter.
• Low annual temperature fluctuation (based on low annual temperature
variability).
• Temperature of warmest quarter (summer) from 7 to 23°C.
This is the first time a spatially explicit global data set was made available for
the world’s temperate and boreal rainforests that was based on a suite of climate
variables obtained from a global data set (available in raster—or grid—GIS for-
mat), improvements in computer processing capacity,and statistical models.The
model therefore represents an initial cut at producing a global rainforest map,
requiring further refinements through the use of regional climate data sets, re-
gional rainforest classifications,and regional maps. Notably,while the minimum
precipitation and maximum temperature values reported seem extreme in
comparison to earlier rainforest definitions, rainforest communities persist in
these regions due to compensatory factors as discussed below and in the re-
gional chapters of this book.This is why regional ground-truth of the model
and further study of rainforest classifications are essential.
CLIMATIC PATTERNS OF TEMPERATE AND
BOREAL RAINFORESTS
Based on the rainforest distribution model,rainforests were clustered along pre-
cipitation and temperature gradients that distinguished them from one another
and other forest types.
8t e m pe ra te a n d b or e a l ra i n f o re s t s o f th e w o r l d

Table 1-1. Abiotic conditions (climate,altitude) of temperate and boreal rainforests used in the rainforest distribution model,based on a
global climate data set.a
Inland Pacific Coast Japan Chile
Great Northwest of Eastern of North and and
Model Var iablebNorway Britain North America Canada Alps America Korea Argentina Australasia
Mean Diurnal Range Min 3.9 5.4 9.2 7.0 6.3 4.3 5.8 4.2 6.1
(mean of monthly; Max 7.3 7.6 12.1 11.9 10.4 11.9 11.1 12.1 11.4
max temp–min Mean 6.1 6.6 10.8 8.9 8.9 7.6 8.6 7.2 8.9
temp; °C)
Isothermality (diurnal Min 24.1 30.4 27.6 22.3 27.8 24.9 19.2 42.8 44.1
temperature/annual Max 31.2 42.2 33.2 32.3 34.6 63.6 31.9 60.1 52.2
range of temperature, Mean 27.9 36.4 30.2 25.9 31.8 33.4 25.1 48.3 47.6
°C)
MeanTemperature of Min –0.9 5.5 –0.1 –11.1 –4.8 –7.0 –14.8 1.2 –1.0
Driest Quarter (°C) Max 9.3 13.6 15.2 15.0 15.1 17.9 10.5 16.2 15.2
Mean 5.4 9.3 3.2 1.7 0.4 9.9 –2.6 9.9 10.0
MeanTemperature of Min 10.3 10.3 10.8 10.4 10.9 9.9 10.6 6.8 6.8
Warmest Quarter Max 14.8 15.2 17.9 18.7 19.5 17.9 22.8 16.3 17.8
(°C) Mean 11.5 12.9 13.5 13.6 15.4 12.5 17.9 11.9 13.2
Precipitation of Driest Min 124 169 106 162 188 6 97 74 143
Quarter (mm) Max 447 352 198 399 317 619 487 1378 1169
Mean 262 248 152 248 245 252 233 499 458

Table 1-1. Continued
Inland Pacific Coast Japan Chile
Great Northwest of Eastern of North and and
Model Var iablebNorway Britain North America Canada Alps America Korea Argentina Australasia
Precipitation of Cold- Min 210 244 187 209 209 125 97 230 241
est Quarter (mm) Max 859 697 458 439 388 1473 506 1403 1803
Mean 428 431 267 309 269 642 263 801 613
Altitude (m) Min 25 74 356 86 284 0 69 0 4
Max 754 842 1487 1319 1548 1263 2612 1066 2334
Mean 327 314 973 267 826 364 811 278 580
aSee Hijmans et al. 2005.
bAnnual mean temperature,temperature annual range, annual precipitation, and precipitation seasonality (coefficient of variation) were included in the model
but not in this table as the data are summarized in figures 1-2 and 1-3.

Figure 1-2. Annual precipitation (a) and variation in rainfall (b) of definitive temperate and
boreal rainforests,based on a global climate data set (Hijmans et al. 2005) and the rainforest
distribution model.
Figure 1-3.Annual temperature (a) and annual range of temperature (b) of definitive
temperate and boreal rainforests, based on a global climate data set (Hijmans et al.
2005) and the rainforest distribution model.

Precipitation Gradient
A broad range of annual rainfall amounts occurs in the“classic” temperate rain-
forests of the Pacific Coast of North America, Chile and Argentina, and Aus-
tralasia (see figure 1-2a).As in tropical rainforests,seasonality of precipitation is
a key element of rainforest climate that can influence rates of decomposition,
the roles of fire, drought, epiphytes, and species composition (Alaback 1996;
Losos and Leigh 2004). Just looking at the coefficient of variation of monthly
precipitation shows the greatest range in the rainforest regions with the greatest
latitudinal ranges (e.g., the Pacific Coast of North America, Chile and Ar-
gentina) and also the greatest seasonality, but a less clear pattern in the seasonal-
ity of precipitation in smaller regions (see figure 1-2b).More work is needed to
clarify how seasonality of precipitation helps effect such differences among
rainforest regions.
Temperature Gradient
Based on the global climate data set, Norway had the coolest annual tempera-
ture and Inland Northwestern North America (based on southern locales) the
warmest (see figure 1-3a). Notably, climate data sets derived from a global ref-
erence (Hijmans et al.2005) may differ from data sets presented in the regional
chapters due, for instance, to topographical influences on local climate and the
location and density of weather stations.
The comparatively wide range of annual temperatures on the Pacific Coast
of North America and in the Valdivian temperate rainforest reflects both its
broad latitudinal distribution and a large range in climates from boreal and sub-
polar to nearly subtropical. Similarly, the Japanese archipelago spans many cli-
mate types (alpine to subtropical, and continental to oceanic), with rainforests
distributed zonally.
The annual range of temperature provides a good measure of seasonality of
a given region (see figure 1-3b).The regions with the greatest influence from
interior climates, such as Inland Northwestern North America, Eastern Can-
ada, Japan, and Korea, all clearly show this influence.The more oceanic cli-
mates, such as Norway, the British Islands, and the Southern Hemisphere rain-
forests, by contrast show a much smaller range of monthly temperatures.This
also helps explain why some of the forests in these regions can develop rainfor-
est characteristics with less rainfall than in comparable continental regions.
In sum, rainforests can be grouped both by differences in annual tempera-
ture and annual precipitation,with the Inland Northwest of NorthAmerica the
warmest, driest rainforest globally, Norway the coolest (with moderate precipi-
12t e m pe ra te a n d b or e a l ra i n f o re s t s o f th e w o r l d

tation), and Chile, Argentina, and Australasia the wettest, with relatively cool-
to-moderate temperatures.
OUTLIERS AND OTHER CONSIDERATIONS
The rainforest distribution model did not predict some areas as rainforest
which, upon further inspection, showed signs of rainforest conditions or com-
munities.We chose to include some of these as “rainforests at the margins” (or
outliers), based on input from regional scientists specializing in the specific re-
gions (see chapter 9). For instance, in some places rainforest communities can
persist at precipitation levels lower than the range used in the model as long as
there is enough moisture at critical times of the year (e.g., warm summer
months) to support moisture-loving species such as lichens and mosses, either
directly through some rainfall or indirectly through compensatory mechanisms
(e.g.,low evapotranspiration rates,high humidity, cool summer nighttime tem-
peratures, and fog).Evidence for this exists for the Knysna-Tsitsikamma forests
of South Africa and the Colchic and Hyrcanic forests of the Western Eurasian
Caucasus, where persistent fog and high humidity compensate for low summer
precipitation and/or hot summers (chapter 9). Such conditions prove suitable
for oceanic lichens and humidity-dependent vegetation.The Ussuri taiga of the
Russian Far East and the Sayani Mountains of Inland Southern Siberia were
too dry for inclusion in the model but have relatively low temperatures and
high humidity (chapter 9). Low evaporative losses apparently compensate for
drier conditions, allowing humidity-dependent forests to flourish.
The rainforest distribution model also did not identify rainforest in some
areas previously suspected to be rainforest.For instance,while Taiwanese mon-
tane forests receive sufficient rainfall and cool-enough temperatures zonally (at
high elevations) to be considered “temperate rainforest” by some (see Wiki-
pedia1; also see Farjon 2005), the lack of a well-defined cool dormant season
makes them more ecologically equivalent to cloud or subtropical forests (Al-
aback 1991), and thus we did not give further consideration to these forests.
Iceland’s scant boreal forests,though recognized as rainforest by Kellogg (1992),
were not included in our rainforest model because the mean annual tempera-
ture is below even the minimum used to define rainforests.Icelandic forests also
lack the structural complexity associated with temperate and boreal rainforests,
JustWhat AreTemperate and Boreal Rainforests? 13
1www.en.wikipedia.org/wiki/Temperate_rain_forest

such as well-defined canopy layers and gap-phase disturbance dynamics,as trees
usually are not long-lived or productive due to severe weather. There are no
naturally occurring boreonemoral tree species (see chapter 6) such as elms (Ul-
mus sp.) and oaks (Quercus sp.), and there are few of the rainforest lichens com-
mon to Norway’s rainforests (e.g., Biatora toensbergii,Fuscopannaria ahlneri, and
Lobaria hallii).
Although the Appalachian mixed-mesophytic forest of the southeastern
United States has been recognized as temperate rainforests by some (see
Netencyclo.com2;Shanks 1954; see chapter 4), it was not predicted by the rain-
forest distribution model, presumably because the region has relatively high
year-round temperatures and dry summers. However, because there was evi-
dence of rainforest conditions at high elevations (moist pockets of spruce-fir
within the larger ecoregion), we briefly mentioned them as a southerly exten-
sion of Appalachian boreal rainforests from Eastern Canada that require further
study (see chapter 4).In sum,we hope the techniques used here will inspire ad-
ditional research into these areas in order to further refine our approach.
INTRODUCING TEMPERATE AND BOREAL RAINFORESTS
In the following sections of this book,we discuss seven definitive regions (some
regions from above were combined) identified by the model and three outlier
regions that collectively make up the global network of temperate and boreal
rainforests.3We generally organized regions north to south (Western Hemi-
sphere) and west to east (Eastern Hemisphere), as presented sequentially as the
book’s regional chapters.
Definitive Regions
• Pacific Coast of North America (chapter 2)
• Inland Northwestern North America (chapter 3)
• Eastern Canada (chapter 4)
• Chile and Argentina (chapter 5)
• Europe: Norway, Ireland, Great Britain, portions of the Alps, the Bo-
hemian region, and the Balkans (chapter 6)
• Japan (chapter 7—note that Korea was included in the Russian Far East
and Inland Southern Siberia profile,based on author expertise)
• Australasia: Australia,Tasmania, and New Zealand (chapter 8)
14t e m pe ra te a n d b or e a l ra i n f o re s t s o f th e w o r l d
2www.netencyclo.com/en/Temperate_rain_forest
3Maps available at www.databasin.org

Outliers (chapter 9)
• Western Eurasia Caucasus (Colchic and Hyrcanic forests)
• Russian Far East and Inland Southern Siberia
• South Africa (Knysna-Tsitsikamma forests)
REGIONAL VS. RAINFOREST DISTRIBUTION MAPS
While the global model was useful in predicting general locations of temperate
and boreal rainforests, we often found differences in global projections versus
regional delineations made by local experts (see table 1-2).Thus, comparing
predicted distributions with regional maps was necessary to ensure that an
agreed-upon set of maps was used in the regional chapters.Digital maps for this
step were obtained for the Pacific Coast of North America (Kellogg 1995; see
figure 4), Inland Northwestern North America (Craighead and Cross 2007),
Eastern Canada (described below), Chile and Argentina (provided by Patricio
Pliscoff—see below), Australasia (Kirkpatrick and Dickerson 1984), Japan
(Miyawaki et al. 1980–89), and Norway (described below). Here, we describe
the differences in mapping delineations and reasons for including regional
maps, where we had them,in the chapters of the book that follow.
Pacific Coast of North America
Differences in mapping estimates between the global model and regional map-
ping (Kellogg 1995) were fairly minor (see table 1-2; figure 1-4).The rainforest
distribution model yielded a rainforest estimate that was ~9 percent higher than
regional mapping (see table 1-2). We present the map by Kellogg (1995) in
Chapter 2 because it allowed us to base conservation priorities on regionally
specific zones (finer scale) that were not apparent from the coarser rainforest
distribution model.
Inland Northwestern North America
The model predicted rainforest to occur on nearly 2.2 million hectares, but
only for eastern British Columbia (see figure 1-5;table 1-2). In comparison, us-
ing the distribution of western red cedar (Thuja plicata) and western hemlock
(Tsuga heterophylla) (i.e., Interior Cedar Hemlock forests) yielded over 3 times
the amount of rainforest at 7.3 million hectares (Craighead and Cross 2007;see
chapter 3), with nearly equal amounts in British Columbia and the United
States.While the rainforest distribution model and the vegetation-based map
showed strong agreement in British Columbia, the Interior Cedar Hemlock
JustWhat AreTemperate and Boreal Rainforests? 15

Table 1-2. Global (rainforest distribution model, Kellogg 1992) and regional (based on
digital maps from published sources) estimates for temperate and boreal rainforests.
Rainforest Regionally
distribution based
model estimatesaKellogg (1992)
Region (ha) (%) (ha) (ha) (%)
Pacific Coast of
North Americab27,274,225 35.0025,097,930 20,726,700 50.30
Inland Northwestern
North America
British Columbia 2,179,733 2.8 3,879,730
United States 0 0.0 3,366,874
Total Inland Northwestern
North America 2,179,733 2.8 7,246,604
Eastern Canada 5,969,641 7.7 6,085,063
Valdivia
Chile 12,211,573 15.709,752,451 11,675,100 28.40
Argentina 348,371 0.4 2,211,888 323,300 0.79
TotalValdivia 12,559,944 16.1011,964,339 11,998,400 29.10
European Relicts
Iceland 195,200 0.47
Norway 4,887,739 6.3 3,747,090 1,459,000 3.5
Great Britain 5,064,759 6.5 1,149,300 2.8
Ireland/Republic of
Ireland 1,578,545 2.0 157,300 0.38
Northeast Alps and Swiss
Prealps 745,915 1.0
Bohemia 220,199 0.3
Southeastern Alps and
Northwest Balkans 577,425 0.7
Total European relicts 13,074,582 16.802,960,800 7.2
Japan and Korea 8,295,241 10.602,404,404
Australasia
Australia 55,989 0.07 1,652,933
New Zealand 5,458,170 7.0 4,969,590 4,040,400 9.8
Tasmania 3,132,684 4.0 692,300 551,700 1.3
Total Australasia 8,646,843 11.107,314,823 4,592,100 11.20
Total Rainforest 78,000,209 1.95c41,177,500 1.1
Outliersd
South Africa (Knysna-
Tsitsikamma) 235,483 1.2

map extends this rainforest type southward for roughly 430 kilometers into
northeastern Washington, northern Idaho,and northwestern Montana (see fig-
ure 1-5a). Based on local knowledge, we choose the map of Interior Cedar
Hemlock forests for Chapter 3.
Eastern Canada
For this region, we overlaid the Thornthwaite (1948) index for perhumid re-
gions (100+ moisture index) onto digital layers of vegetation obtained from
coniferous and mixed forest types (source: Canadian Vegetation and Land
Cover data set, www.nrcan.gc.ca). This shapefile is based on satellite data
obtained in 1995 by the Advance Very High Resolution Radiometer
JustWhat AreTemperate and Boreal Rainforests? 17
Table 1-2. Continued
Rainforest Regionally
distribution based
model estimatesaKellogg (1992)
Region (ha) (%) (ha) (ha) (%)
Western Eurasia
Hyrcanic 1,960,000 10.30
Colchice3,000,000 15.80899,500 2.2
Total Wester n Eurasia 4,960,000 26.10
Russia/Siberia
Russian Far East 6,800,000 35.80
Inland Southern Siberia 7,000,000 36.90
Total Russia/Siberia 13,800,000 72.60
Total Outliers 18,995,483 0.47c
Combined temperate
and boreal rainforest
total 96,995,692 2.42c
aRegional estimates were provided for comparisons to the rainforest distribution model but, due to
differences in mapping methodologies, did not include percentages except in the case of Kellogg
(1992), which was based on more consistent mapping methodologies.
bDifferences in rainforest estimates between the two Kellogg references (1992, 1995) are presumed
due to refinements in mapping techniques,mainly the addition of the western Cascades in Washing-
ton and Oregon,which were not included in the original maps.
cPercentages were derived from global forest cover (all forest types) estimated at 4 billion hectares
based on FAO (2005) estimates that define forests as >10% tree cover. Plantations are included in es-
timates.
dOutlier estimates,provided by regional authors,were derived from different mapping methodologies
not directly comparable to rainforest distribution estimates or other regional estimates.
eKellogg (1992) lists this region as Eastern Black Sea (Turkey, Georgia).

Figure 1-4. Temperate and boreal rainforests of the Pacific Coast of North America based
on (a) regional mapping (Ecotrust 1995) and (b) the rainforest distribution model.
Figure 1-5. Temperate and boreal rainforests of Inland Northwestern North America
based on (a) regional mapping (Craighead and Cross 2007) and (b) the rainforest distribu-
tion model.

(AVHRR) on board the NOAA-14 (National Oceanic and Atmospheric Ad-
ministration) satellite.We assumed these forest types were most likely to include
important lichen assemblages and rainforest structure that matched perhumid
climatic conditions in the region.
Both the rainforest distribution model and regional map (Thornthwaite
1948) yielded nearly identical area estimates (see table 1-2).However,predicted
locations of rainforests from the rainforest distribution model vs. regional map-
ping differed appreciably (see figure 1-6).Thus, we used the regional map in
JustWhat AreTemperate and Boreal Rainforests? 19
Figure 1-6. Perhumid boreal and hemiboreal rainforests of Eastern Canada based (a) on re-
gional mapping (modified from Thornthwaite 1948) and (b) the rainforest distribution
model.
a
b

20t e m pe ra te a n d b or e a l ra i n f o re s t s o f th e w o r l d
Figure 1-7. Valdivian temperate rainforests of Chile and Argentina based on (a) regional
mapping (digitized from national vegetation surveys) and (b) the rainforest distribution
model.
Chapter 4 because it was thought to have higher predictability and greater con-
cordance with forests supporting rainforest lichen assemblages based on local
knowledge.
Chile and Argentina
The primary map source for Chile was the national vegetation survey.This was
originally produced using aerial photography at a scale of 1:50,000 and with
varied level of verification on the ground. Later updates to this information
were produced using Landsat imagery and essentially serve to track loss of for-
est cover. As a representation of forest cover, the national vegetation survey is
widely used in Chile, is embraced the official source by Chile’s Native Forest
Law of 2008, and is fairly reliable. For Argentina no such forest survey exists;
thus we used the same criteria and methods from Chile’s national survey and a
series of aerial photos to produce a forest-cover map at 1:500,000 scale without
ground verification.
The rainforest distribution model and regional map yielded similar area es-
timates forValdivia (see table 1-2; figure 1-7). However, there were significant
differences in rainforest locations, with the rainforest distribution model ex-
tending farther south into the Magellanic (subpolar) rainforests, considered a
separate ecoregion by Chilean scientists (see chapter 5), but missing important

rainforest locations in the north and in Argentina. Notably, because the Magel-
lanic forests can be considered rainforest by the standards set forth herein, re-
gional authors included some mention of them in Chapter 5. We used the
regional map in Chapter 5 because it is widely accepted in regional conserva-
tion planning.
Europe
Norway was the only regional map available for comparisons to the rainforest
distribution model in Europe.The regional map in this case was based solely on
floristic data, namely distribution of epiphytic lichens housed at the Norwegian
Lichen Database.4Notably, the core area of boreal rainforest in Norway (and
Europe) is rather well outlined by the distribution of just two lichens—Rinod-
ina disjuncta and Pyrrhospora (Lecidea) subcinnabarina—also known from the Pa-
cific Coast of North America (seeTønsberg 1992, 1993; Sheard 1995).The dis-
tribution of three other lichens demark the northern and southern limits, with
Lobaria hallii delimiting boreal forests with occurrences in ravines and by water-
falls, and Leptogium burgessii and Pyrenula occidentalis the southern boreonemoral
(temperate) rainforests.
The rainforest distribution map of Norway estimated about 1.1 million
hectares (~30 percent) more rainforest than the estimate generated by regional
authors (see table 1-2; figure 1-8).In this case,the rainforest distribution model
may have correctly predicted conditions suitable for rainforests but local differ-
ences in soils, wind exposure, or human disturbance may preclude rainforest
development.Therefore,the Norway regional map was used because it was pre-
pared with regional forest inventories based on known rainforest lichen assem-
blages (see chapter 6).
Japan
About 5.9 million hectares (over 3 times) more rainforest was estimated by the
rainforest distribution model compared to a digitized map of Japan’s rainforest
zones (see table 1-2);figure 1-9), which were based on finer-scale mapping and
therefore used in Chapter 7.
Australasia
About 1.3 million hectares (18 percent, table 1-2) more rainforest was pre-
dicted by the rainforest distribution model compared to regional mapping (see
figure 1-10). Differences were greatest for Tasmania, where the rainforest
JustWhat AreTemperate and Boreal Rainforests? 21
4www.nhm.uio.no/botanisk/lav/index.html

Figure 1-9. Temperate rainforests of Japan based on (a) regional mapping (Miyawaki et al.
1980–1989) and (b) the rainforest distribution model.
Figure 1-8. Boreal and boreonemoral rainforests of Norway based on (a) regional mapping
(derived from lichen distribution maps) and (b) the rainforest distribution model.

distribution model estimated about 2.4 million hectares (over 4 times) more
rainforest than the regional map. Conversely, the rainforest distribution map es-
timated about 1.6 million hectares less rainforest along the Australian coastline
(New South Wales). Notably, about 151,173 hectares and 830,769 hectares of
the regionally based totals (Kirkpatrick and Dickerson 1984) were classified as
clear felled or forests patchily distributed, respectively, at the time. So the over-
estimate of rainforest by the model may have been partially compensated by the
JustWhat AreTemperate and Boreal Rainforests? 23
Figure 1-10. Temperate rainforests of Australasia based on (a) regional mapping (Kirk-
patrick and Dickerson 1984) and (b) the rainforest distribution model.

mapping of cleared forests by regional experts. Because the regional maps in-
cluded more of the Australian coastline where rainforests are known to occur,
they were used in Chapter 8.
In sum, the rainforest distribution model was useful in establishing an ob-
jective upper range of potential rainforest, was the only standardized data set
available for comparisons among regions, and provided a reliable global rainfor-
est total. However, the model had a tendency to overestimate rainforest extent
in most, but not all, regions when compared to site-specific mapping and re-
gional expertise.The rainforest distribution model was potentially confounded
by human disturbance and local site conditions. Rainforest estimates derived
from regional maps,however, also have limitations, as they cannot be compared
among regions due to differences in mapping techniques, data sources, and
mapping scales.Thus, in making relative comparisons among regions and pre-
dicting new localities, the global rainforest distribution model performs quite
well;however, for regional specificity we relied on regional maps, as they had a
higher degree of reliability at that scale. Follow-up mapping assessments and
modeling is recommended in both cases—regionally and globally—to improve
rainforest estimates and mapping techniques.
TEMPERATE AND BOREAL RAINFOREST TOTALS
Based on the rainforest distribution model,the Pacific Coast of North America
(British Columbia and the United States combined) by far contains the most
expansive temperate and boreal rainforests globally, representing over one-third
of the world’s totals (see table 1-2). Our estimate for this region is notably less
than prior estimates (50 percent). Differences are due largely to rainforest areas
added in the rainforest distribution model and different mapping techniques,
which obviously affected regional totals. Nonetheless, in decreasing order, rain-
forest extent was then highest for European rainforest relicts (disjunctly distrib-
uted); Chile and Argentina; Australasia; Japan; Eastern Canada; and Inland
Northwestern North America. However, these percentages do not indicate in-
tactness of rainforests within a given region. For instance, some of the last re-
maining large blocks of temperate rainforests in the world occur in Valdivia,
Tasmania, and New Zealand (see chapters 5 and 8), in comparison to highly
fragmented European relicts (see chapter 6); and some of the most intact old-
growth rainforests occur in the British Columbia and Alaska (see chapter 2).
However,regional totals are not affected by conservation status.
24t e m pe ra te a n d b or e a l ra i n f o re s t s o f th e w o r l d

In addition to definitive regions, outliers added nearly 19 million hectares
to the global temperate and boreal rainforest total (roughly 0.5 percent), with
the Russian Far East and Inland Southern Siberia by far containing the largest
(73 percent) expanse and South Africa the smallest (~1 percent, table 1-2).
In sum,our estimate for global temperate and boreal rainforest extent (2.42
percent) was more than twice that of previous estimates (1.1 percent; Kellogg
1992), due largely to additional regions estimated by the rainforest distribution
model and differences in mapping techniques.However,some regions (Iceland)
previously considered rainforest (Kellogg 1992) were not included here as they
do not appear to support rainforest communities. Nonetheless, despite these
differences there was considerable overlap in regional estimates, with the net
result that temperate and boreal rainforests still represent just a fraction of the
global forest cover.
RAINFOREST DIFFERENCES IN THE NORTHERN AND
SOUTHERN HEMISPHERES
In this section, we examine major differences in gross rainforest characteristics
that can be readily grouped by differences in biogeography between hemi-
spheres where these rainforests are found.
Northern Hemisphere
Temperate and boreal rainforests in the Northern Hemisphere are remarkably
similar in species composition, at least at the genus level.The largest of these
rainforests in terms of areal extent are dominated by conifers (e.g., Pacific
Coastal and Inland Northwest North America, parts of Japan, Norway),usually
broadly distributed but closely related species of the pine family, including
hemlock, true firs (Abies spp.), Douglas-fir (Pseudotsuga menzeisii), spruce (Picea
spp.) or pine (Pinus spp.), and species of Cupressaceae, especially red cedars
(Thuja spp.).Other, smaller regions are dominated especially by beeches (Fagus
spp.; found in Japan and central European fragments) or beech-spruce mix-
tures (found, for example, in Norway). In general, temperate and boreal rain-
forests of the Northern Hemisphere have a dense understory of largely decidu-
ous woody shrubs, a variety of widely distributed (often circumboreal) herba-
ceous plants and a thick mat of bryophytes (mosses and liverworts), lichens, and
many fern species. The broad commonalities among these rainforests make
sense from a biogeographical standpoint, since the floras of the Northern
JustWhat AreTemperate and Boreal Rainforests? 25

Hemisphere are believed to have been derived in large part from commonTer-
tiary ancestors 60–80 million years ago (see Axelrod 1976).
Southern Hemisphere
In Southern Hemisphere rainforests (southern Chile,Argentina, New Zealand,
Tasmania and nearby areas),most trees are broad-leaved evergreens,which form
a patchy canopy with many layers beneath the dominant overstory, including a
broad diversity of both evergreen and deciduous trees and shrubs.The trees are
tall and dense,with small tough leaves (Veblen et al. 1996).
Southern vs. Northern Hemisphere
Southern Hemisphere trees are unlike most of the familiar broad-leaved trees
in the North. The “southern beech” or Nothofagus trees, for example, are not
closely related to beeches of the Northern Hemisphere.They are in their own
family (Nothofagaceae) and originated in ancient Gondwana before it split
into what have become the small areas of temperate rainforest scattered across
the Southern Hemisphere (Veblen et al. 1996). This explains why there are
many species of trees that are shared at least at the genus level among rain-
forests in New Zealand, South America, and Australia (Ezcurra et al. 2008).
Another big surprise is in the pine family.While pines, spruces, firs, and related
species dominate high-latitude forests of the Northern Hemisphere, this entire
family is absent in the Southern Hemisphere (Lusk 2008). The principal tree
families shared are the most ancient ones, such as the cedars and cypress spe-
cies (family Cupressaceae), that were well developed before the continents
split apart.
While the Northern Hemisphere is dominated by conifers in the pine fam-
ily (Pinaceae), trees in temperate rainforests of the Southern Hemisphere be-
long to a wide assortment of mostly small, specialized families. Among these,
the myrtle family (Myrtaceae) is often the most diverse. Some other, more-
modern families are also shared between the Northern and Southern Hemi-
spheres, such as the heath and heather family (Ericaceae). In this case, these
plants are particularly well adapted to cool, moist conditions, either alpine or
subalpine, and have apparently been able to disperse along the Rockies and
Sierra Madre in North America down the Andes all the way to Tierra del
Fuego.The crowberry (Empetrum nigra), for example, has black berries in rain-
forests of the Northern Hemisphere, but red berries in the Southern Hemi-
sphere (E. rubrum), and otherwise looks very similar between hemispheres.The
occurrence of these two families may be,in part, attributable to dispersal of the
seeds by migratory birds moving between hemispheres, a prospect that also has
26t e m pe ra te a n d b or e a l ra i n f o re s t s o f th e w o r l d

been proposed for some lichens.A striking exception to the pattern of diver-
gence is the case of an increasing number of possibly relictual lichen lineages
being discovered to be shared between the Pacific Coast of North America and
Tasmanian and/or Valdivian rainforests (Spribille et al. 2010). However, the
overwhelming pattern is one of disparity, with contrasting assemblages recur-
ring with bryophytes, most nonmigratory birds, mammals, fishes, and insects.
Why are these forests so taxonomically different between hemispheres? Let’s
explore some of the leading hypotheses.
Continental Drift and Isolation
This is generally considered the key factor explaining hemispheric differences.
While the continents in the Northern Hemisphere were well connected many
times in the past, including as recently as a few tens of thousands of years ago
during glacial cycles, in the Southern Hemisphere many of the land masses that
now have temperate rainforests have been isolated from each other since the
lateTertiary period (over 60 million years ago—see Lawford et al.1996;Veblen
et al. 1996;Arroyo et al.2000).This has lead to adaptive radiation events in spe-
cies with ancient lineages, resulting in many unique forms (endemics).
Geography
Most of the Southern Hemisphere is dominated by ocean, and at the high lat-
itudes land masses are highly fragmented and have been since the upper
Tertiary some 2 million years ago, when the rainforest zone became progres-
sively isolated by xeric climates to the east and north triggered by the uplift of
the Andes (Arroyo et al. 1996).Thus, most temperate rainforests have milder
winter climates with rainfall evenly distributed over the growing season.This
unique climate leads to a more subdued role for wildfire and to a more lim-
ited adaptation to extreme cold. Even subalpine species from the Southern
Hemisphere are generally not hardy enough to survive in continental rain-
forests of the Northern Hemisphere (Lawford et al. 1996; Veblen and Kitz-
berger 2002).
Endemism
The vast majority of species in temperate and subpolar rainforests of the
Southern Hemisphere are unique to each continent (South America, Africa,
and Australasia), and sometimes to a specific area due to their relictual taxo-
nomic status and long periods of isolation (Lawford et al. 1996; Smith-
Ramirez 2004; Hinojosa et al. 2006; also see chapters 5 and 8). By contrast, in
the Northern Hemisphere fewer species are limited to specific habitats or
JustWhat AreTemperate and Boreal Rainforests? 27

areas, although island biogeographical effects in northern coastal latitudes have
triggered speciation events at the subspecies level (see chapter 2).
Species Mutualisms
Many species in the Southern Hemisphere evolved from tropical affinities (e.g.,
Valdivia—see chapter 5),including complex interactions between plants,herbi-
vores, pollinators, and seed-dispersing species. Further, most trees in Southern
rainforests produce edible fruits and have co-evolved with seed-dispersing ani-
mal species (Armesto et al. 1996). In contrast, most rainforest trees of the
Northern Hemisphere are conifers with less direct and specific co-evolution
with pollinators and seed dispersers (e.g.,Willson et al. 1990).
TEMPERATE AND BOREAL RAINFORESTS VS. TROPICAL
MOIST RAINFORESTS
Tropical rainforests, as their name implies, are bracketed by the tropics of Can-
cer and Capricorn (see figure 1-11; table 1-3).They cover about 6 times more
area than temperate and boreal rainforests (~2 percent versus 12 percent of the
world’s forests). Tropical rainforests are generally drenched in warm, moist
climates with little seasonal temperature variation within 1 kilometer of sea
level.On the other hand, temperate and boreal rainforests are generally but not
28t e m pe ra te a n d b or e a l ra i n f o re s t s o f th e w o r l d
Figure 1-11. Tropical moist (Olson and Dinerstein 1998) and temperate and boreal rain-
forests of the world.

Table 1-3. General features distinguishing tropical moist rainforests from temperate and
boreal rainforests.
Feature Tropical Moist RainforestaTemperate and Boreal Rainforest
Distribution up to 23° latitude from the equator: ~30–69° latitude, disjunct, mainly
large belts across South America, coastal: Pacific Northwest,Alaska,
Central America, Southeast Asia, British Columbia, Chile, Argentina,
and Africa Tasmania, New Zealand, Australia,
Japan, Europe
Extent ~12% of present global forest cover, ~2% of present global forest cover, re-
reduced by over half of estimated duced by ~half of estimated his-
historic levels toric levels
Deforestation 1-2% annualb,especially high in South forest cover generally increasing, but
(2000–2005) America and Africa, mostly con- old growth replaced by tree
verted to agriculture plantations
Annual Mean 23–27° Celsius ~4–12° Celsius
Temperature
Seasonality unifor m temperature with wide varia- varied temperatures, snow in winter,
tion in rainfall patterns (up to a greater precipitation in fall and
3-month dry season) winter with summer rains over
14% of annual precipitation
Moisture over 1,700 mm, high humidity, high 846–2658 mm, high humidity, low
evapotranspiration evapotranspiration
Canopy diversity multilayered,rich epiphytes (orchids, generally multilayered,rich epiphytes
bromeliads), and abundant lianas (lichens, mosses), lianas less
developed
Forest height 20–50 m 10–70 m
Soils thin litter layer, infertile and severely rich humus, highly productive and
leached except in volcanic and ri- rich in invertebrates,large amount
parian areas; large nutrient pools in of coarse, woody debris
trees
Biomass moderate (100–250 metric tons/ha), low (Europe) to exceptional (red-
highest in dipterocarps (Southeast woods, Pacific Nor thwest, Tasma-
Asia) nia,Valdivia) (100–1867 metric
tons/ha)
Productivity high-exceptional exceptional (mar ine, freshwater,
terrestrial)
Nutrient cycling rapid decomposition rates slow decomposition rates
Pollination exceptional low in conifers
Plant and animal exceptional, over half of terrestrial low (Europe) to moderate (Japan,Val-
richness species on Earth, generally 5–10 divia),but high for mosses and
times that of temperate forests lichens
Endemism exceptional, many species unique low (Europe), moderate (California),
high (Chile and Argentina)
Tree richness exceptional (50–200 species/ha) low to moderate (1–20 species/ha)
aSynthesized fromTerborg (1992); Richards (1996);Kricher (1997); Myers et al. (2000);and Losos and
Leigh (2004).
bDeforestation rates based on total forest cover lost on a continental scale (FAO 2005). Individual
countries with rainforest, however,may have higher or lower rates of deforestation or show afforesta-
tion due to tree planting.

exclusively found along coastlines at middle to upper latitudes, and can extend
to nearly timberline (exceptions include Inland Northwest of North America,
the Alps, and Inland Southern Siberia).
Climatically, temperate and boreal rainforests have a more distinctive sea-
sonality (especially wider temperature swings), and greater range of precipita-
tion types including snow and sleet, than tropical counterparts (see table 1-3).
High temperatures in the tropics lead to high evaporation rates and the devel-
opment of daily clouds above the forest, so that they can recycle 70 percent or
more of their annual rainfall.Temperate rainforests, on the other hand, are cool
and wet,with slower rates of decomposition and low evaporation rates.To bet-
ter understand the differences between these rainforest types, we turn to some
key concepts in forest ecology.
Ecologists today generally recognize that forest ecosystems are comprised
of three main “ingredients”: composition—the mix of species in a forest; struc-
ture—the vertical and horizontal dimensions and spatial patterns of a forest; and
function—the workings of a forest expressed through nutrient cycling, food-
web and disturbance dynamics, forest succession, pollination, and many other
processes (Perry et al. 2008).The regions identified as temperate and boreal
rainforest in this book have a suite of underlying characteristics along these
lines that can be used to further distinguish them from each other as well as
from their tropical counterparts.
Structure
Both temperate and tropical rainforests (boreal less so) have complex forest
canopies composed of many canopy layers, creating dense and continuous veg-
etation cover that provides for rich fauna from the ground up. In both forest
types,canopy gaps and emergent crowns of dominant trees create complex spa-
tial patterns in the lower strata.A key difference in rainforest canopies is that
temperate rainforests are dominated by conifers (except in the Southern Hemi-
sphere, where they are dominated by broadleaf evergreens, and in Japan and
Europe,where they can be deciduous),while tropical rainforests are dominated
by broad-leaved trees enveloped by numerous lianas (Valdivia, New Zealand,
Hyrcanic,and South African temperate rainforests also have lianas).Both rain-
forest types often have a high degree of standing dead trees (snags) and fallen
logs that provide structure and habitat for scores of plant and animal species
(Baker et al. 2007; Perry et al. 2008).
Function
Biomass in temperate rainforests is exceptional on a global scale, exceeding that
of tropical rainforests (Smithwick et al. 2002; Losos and Leigh 2004; Keith et al.
30t e m pe ra te a n d b or e a l ra i n f o re s t s o f th e w o r l d

2009; see table 1-3). For instance, one study of a young temperate rainforest in
Oregon showed that it could fix as much carbon per year as some mature trop-
ical rainforests (e.g., 36 metric tons of organic matter per hectare annually—
Fujimori 1971). Another study found primary forests in Australia capable of
storing up to 1,867 metric tons per hectare, the world’s highest known total
biomass carbon density (Keith et al. 2009). However, while tropical forests are
not exceptionally carbon-dense systems, they still play the dominant role for
forest contributions to global carbon cycles due to their high rates of produc-
tivity, decomposition, long growing seasons, and the large land area they still
occupy.
Evergreen needles (or leaves) are a common characteristic of the vast ma-
jority of tree species that grow in temperate and boreal rainforest climates.They
allow rainforest plants to photosynthesize throughout the year in most coastal
temperate areas, helping to explain the high productivity of these rainforests
(Waring and Franklin 1979).The mild climate of these rainforest regions may
explain why most of the tallest trees in the world grow there. Examples from
around the world include towering Eucalyptus forests in southeastern Australia,
massive coastal redwoods and alerce in California and Chile, respectively, and
ancient coastal Douglas-fir (Pseudotsuga menziesii var. menziesii) of northern
California and the Pacific Northwest. Finally, a continuously mild, wet climate,
combined with minimal genetic losses during Pleistocene glaciations, may have
played a role in maintaining the rich genetic diversity of conifer species in the
Pacific Northwest but led to losses in other regions (Waring and Franklin 1979;
Premoli et al. 2000).
Coastal rainforests also are productive places for marine life, with strong
linkages between marine and terrestrial ecosystems (Simenstad et al. 1997).
Well-known examples include the marbled murrelet (Brachyramphus marmora-
tus) of the Pacific Coast of North America, a coastal seabird that summers at sea
but breeds and nests in the tops of old-growth trees; and historical links be-
tween Pacific sea-run salmon (Oncorhynchus sp.) and terrestrial predators such as
bears (Ursus spp.) and wolves (Canis lupus), which, in the Great Bear Rainforest
of British Columbia, prey upon salmon and help fertilize coastal riparian forests
through their droppings (see chapter 2).
Composition
Compared to the tropics,in Northern Hemisphere rainforests plant and animal
species richness is generally low, and endemism low to moderate, with some
noted exceptions (see table 1-3), including island systems (e.g., Cook et al.
2001). However, lichens appear to be much more diversified at high latitudes
than in the tropics (witness ~750 species for a single southeastAlaskan rainforest
JustWhat AreTemperate and Boreal Rainforests? 31

fjord compared to ~550 species in all ofThailand; Spribille et al. 2010). Even if
many more lichens are discovered in the tropics and the relative richness gap
closes, it appears that the tropics are by no means richer on the orders of mag-
nitude that apply to some other groups of organisms.Outstandingly high levels
of species richness also have been documented in basidiomycete fungi (“mush-
rooms”) with hyper-diverse floras documented in coastal rainforests of British
Columbia (Roberts et al. 2004) and over 750 macro-fungal species from a sin-
gle stand of old-growth forest on a hill in rural Victoria on Vancouver Island
(C
ˇesˇka 2009). Here, too, numbers may be far higher than in the tropics, espe-
cially of ectomycorrhizal fungal species (a type of mycorrhizae composed of a
fungus sheath around the outside of root tips—Allen et al. 1995). How these
numbers stack up in the long term against species numbers in the more poorly
knownTropics remains to be seen,but the fact that key physiological processes
for many fungal and lichen species are optimal at cool temperatures through
community adaptation (Friedman and Sun 2005) suggests that, for lichens at
least, the pattern may hold.
The generally low diversity of trees species in temperate rainforests, with
some noted exceptions such as Valdivia and Japan (see table 1-3), should not
seem too surprising, since these rainforests tend to have dense overstory
canopies and occur in cloudy climates at high latitudes,leaving little light avail-
able for understory canopy layers. Many endemic plant species are associated
with warm-temperate or seasonal rainforests, such as the forests in south-
central Chile and northern California, as well as all rainforests that occur
on islands, and other areas in the Southern Hemisphere. In addition, many
moisture-adapted taxa that provide a unique physiognomy and structure
closely tied to these rainforests, including epiphytic mosses, liverworts, and
lichens, are associated with moist rainforest climates (Goward and Spribille
2005; see table 1-3). In these groups, endemism is locally high in Tasmania and
New Zealand,Japan,Valdivia, and parts of northwest North America while it is
low to nonexistent in the isolated patches of rainforest in Europe and Eastern
Canada.This is likely correlated with the extent of glaciation and/or availability
of extensive glacial refugia, combined with a long history of good dispersal
across and between continents in these regions. Other species-rich taxa in these
rainforests include insects (mostly soil and canopy species) and gastropods
(mainly in the Pacific Northwest),with high levels of endemism in certain taxa.
Apart from that, tropical rainforests are exceptional across taxa (see table 1-3).
Disturbance Dynamics
Stand-replacing disturbances are relatively rare in temperate and boreal rain-
forests, as they are in tropical moist forests.As a result, both rainforest types are
32t e m pe ra te a n d b or e a l ra i n f o re s t s o f th e w o r l d

dominated by ancient trees that have a complex structure and pattern, due to a
long history of small patch or gap disturbances (see box 1-1).This history, along
with the evolution of tree defenses against diseases,has allowed certain tree spe-
cies to reach very old ages (Waring and Franklin 1979) in not only temperate
rainforests (see above examples for tree species) but tropical rainforest trees as
well (e.g., Hymenolobium mesoamericanum of Costa Rica can live for hundreds of
JustWhat AreTemperate and Boreal Rainforests? 33
BOX 1-1
Gap Phase Dynamics ofTemperate and Boreal Rainforests.
While most temperate and boreal rainforests are subject to various stand-
replacing disturbances such as canopy fires, hurricanes, and landslides,
forests in moist climates often have small-scale disturbances that serve to
maintain the species composition and structure of the forest over time.
Some authors have called these disturbances “maintenance dynamics”(see
Veblen and Alaback 1996; Perry et al. 2008). A key ecological conse-
quence of frequent gap disturbances is that a wide range of light environ-
ments and ecological conditions can be maintained in a forest that en-
riches its structural and compositional diversity.This also promotes a rich
assortment of plant and animal species requiring vastly different light lev-
els (e.g., both shade-tolerant and -intolerant species), and implies forest
structure and composition can be theoretically maintained indefinitely.
The extent to which a given rainforest is dominated by gap dynamics de-
pends on many factors, including susceptibility to intense windstorms or
geomorphic disturbances (landslides and flooding), as well as the suscepti-
bility of individual trees to mortality, insects, and disease.
Key disturbance features of temperate and boreal rainforests are sum-
marized as:
• Usually small-scale events affecting 1–4 percent of the forest area
annually, although these gaps are eventually filled by light-seeking
plants, creating a continuous push-pull dynamic between gap-
dependent and gap-avoiding (anti-gap) species (Nowacki and
Kramer 1998; Franklin et al. 2002).
• A small number of trees are killed in each disturbance event, usually
fewer than 10 trees (Lertzman et al. 1996; Ott and Juday 2002).
• Gaps vary widely in size and shape, creating a rich mosaic of condi-
tions in the forest (Ott and Juday 2002).

years—Fichtler et al. 2003).The infrequency of natural fires in both rainforests
adds to tree longevity (e.g., see Gavin et al. 2003).
While both tropical and temperate rainforests are affected by and in turn
affect regional climates, tropical rainforests, along with the world’s oceans, play
a major role in the planet’s climate regulation.When either rainforest type is cut
down, much of their stored carbon is released as carbon dioxide, thus con-
tributing to global warming as well as regional changes in moisture (evapora-
tive losses) and temperature (as discussed in Chapter 11). Understanding this
basic fact is key to climate change negotiations for protecting the world’s ma-
ture forests in both the tropics and temperate zones for their pivotal role in
long-term carbon storage (see chapters 10, 11).
RAINFORESTS: GOING, GOING, GONE?
Unfortunately, both temperate-boreal and tropical rainforests have been re-
duced by at least half their estimated original extent (i.e., before widespread
34t e m pe ra te a n d b or e a l ra i n f o re s t s o f th e w o r l d
BOX 1-1
Continued
• When gaps are created by wind events,root-throw can create a rich
diversity of soils and microhabitat conditions in the forest,including
“pit and mound” micro-topography (Bormann et al. 1995) and
nesting sites for birds (e.g.,winter wrens Troglodytes troglodytes often
nest in root-wads).
• Tree architecture, including rooting depth, height and exposure of
canopy, and resistance to decay fungi play key roles in determining
susceptibility to windthrow.
• Openings in canopy created by gaps promote regeneration of tree
and understory species, leading to greater diversity in the forest
(Spies et al. 1990; Franklin et al. 2002).
• While in theory gap disturbances can maintain the structure and
composition of the forest indefinitely, in practice gap dynamics can
lead to changes in forests due to changes in the environment at the
time of gap creation,including seed availability and dispersal,micro-
climate, and specific characteristics of a given gap event (see Lertz-
man et al. 1996).

human-related destruction of rainforests—see Bryant et al. 1997; Myers et al.
2000; Ritter 2008; see table 1-3). Logging in the tropics is typically accompa-
nied by the burning of vegetation and conversion of biologically rich forest to
agriculture fields also used by livestock.A recent development is the clearing of
rainforest to grow crops for biofuels (e.g.,Borneo,Malaysia, and forest thinning
in the temperate zone). In the tropics, this comes with severe depletion of al-
ready nutrient-deficient laterite (acidic) soils due to the leaching of nutrients
otherwise held in place by rainforest trees, thus hampering afforestation efforts.
Temperate and boreal forests, on the other hand,mainly have been degraded by
conversion of biologically rich, older rainforest to simplistic tree plantations, or
have been high-graded, where old high-value trees (or forest patches) are re-
moved without providing for adequate rates of regeneration of older age classes
or ecological types (as discussed throughout this book).
Notably, some researchers (Kauppi et al. 2006) contend that the world’s
forests have been increasing over a 15-year period (1990–2005) measured by
accruing wood volume, biomass, and captured carbon (growing stock).While
this is certainly a positive development,it misses the point about ongoing losses
to intact and high-quality forests such as old-growth or primary forests. Glob-
ally, very few large, intact primary forests (e.g., “frontier forests”) remain
(Bryant et al. 1997). In addition, according to estimates provided by theWorld
Wildlife Fund, approximately 13 million hectares of forests are destroyed glob-
ally each year mainly in the tropics.5But these losses are not just restricted to
the tropics. For instance, the United States was recently ranked seventh in the
world in deforestation, an annual rate of 215,000 hectares (FAO 2005).These
alarming losses come at a time when deforestation (including forest conversion
as used here) was second only to fossil-fuel emissions in global contributions to
greenhouse-gas pollutants, although growth in emissions from forestry slowed
from 1970 to 2004 (IPCC 2007).These forests are not equated by tree farms
achieved through planting, as the difference in terms of quality of forest com-
position, genetics, function, structure,and long-term storage of carbon (and its
release by forestry operations) is hard to measure at a global scale, but such
comparison is certainly feasible at regional scales through measures of forest
quality, remote sensing, and landscape change-detection analysis.
Ongoing consumption of wood products, particularly in the United States,
Canada, Japan, and Europe (where per capita consumption levels are highest),
will continue this alarming trend of forest conversion in the temperate zone
and complete deforestation in the tropics. Recycling, the use of alternative
JustWhat AreTemperate and Boreal Rainforests? 35
5www.worldwildlife.org/climate/northsouthpartner.html

fibers, and improvements in manufacturing technologies are offsetting this
trend somewhat. Greater interest in the conversion of cellulosic fiber from
forests to liquid fuel (biofuels), however, will put more pressure on the world’s
forests, both tropical (UNEP 2009) and temperate/boreal (Searchinger et al.
2009). Afforestation cannot keep pace with ongoing demand without further
degradation of rainforest biota from the loss of primary forests and the suite of
ecosystem services they uniquely provide.
ABOUT THIS BOOK
This book, while focused primarily on the ecology of temperate and boreal
rainforests,is intended as a rallying call for global action to conserve these rain-
forests, which, like so many of the world’s rainforests, are at a critical juncture.
Each of the regional chapters is a closer examination of the history and ecolog-
ical characteristics of the largest remaining examples of temperate and boreal
rainforest, and provides essential information that can be used to make clearer
global priorities for the conservation of these important rainforests.
The regional chapters (chapters 2–9) largely maintain a consistent structure
throughout that includes basic information on rainforest location and types,cli-
matic conditions, significant ecological attributes of regional and global impor-
tance, ecological processes such as natural disturbances and forest succession,
keystone or exemplary rainforest species,regional rainforest classifications (zones
or subtypes), threats, and conservation priorities. In Chapter 10, we summarize
key findings from each of the rainforest regions in order to stitch together a uni-
fying vision, based on fundamental concepts of conservation biology, for con-
serving the world’s temperate and boreal rainforests.We end the book in Chapter
11 with a call for an international accord to prepare these rainforests for the in-
evitable consequences of climate change. Most important, we hope that the
principles and concepts outlined in this book provide a scientific foundation for
expanding rainforest protections around the globe, so that these remarkable rain-
forests will continue to meet the growing demands of human communities for
the life-giving services that these forests have provided to us for millennia.
LITERATURE CITED
Alaback, P. 1991.Comparative ecology of temperate rainforests of the Americas along anal-
ogous climatic gradients. Revista Chilena de Historia Natural 64:399–412.
36t e m pe ra te a n d b or e a l ra i n f o re s t s o f th e w o r l d

———. 1993. North-South comparisons: Managed systems. Pp. 347–48 in Contrasts in
global change responses between North and South America. Ed. H. A. Mooney, E. Fuentes,
and B. I. Kronberg. New York: Academic Press.
———. 1996. Biodiversity patterns in relation to climate in the temperate rainforests of
North America. Pp. 105–33 in High-latitude rain forests and associated ecosystems of the west
coast of the Americas: Climate, hydrology, ecology and conservation. Ed. R. G. Lawford, P. B.
Alaback, and E. R. Fuentes. Ecological Studies 116. Berlin: Springer-Verlag.
———, and J. Pojar.1997.Vegetation from ridgetop to seashore. Pp. 69–88 in The rainforests
of home. Ed. P. K. Schoonmaker, B. von Hagen, and E. C.Wolf.Washington DC: Island
Press.
Allen, E. B.,Allen, M. F., Helm, D. J.,Trappe, J. M., Molina, R., and E. Rincon. 1995. Pat-
terns and regulation of mycorrhizal plant and fungal diversity. Plant and Soil 170:47–
62.
Armesto, J. J., C. Smith-Ramirez, and C. Sabag. 1996.The importance of plant-bird mutu-
alisms in the temperate rain forest of southern South America. Pp. 248–65 in High-
latitude rain forests and associated ecosystems of the west coast of theAmericas: Climate,hydrol-
ogy, ecology and conservation. Ed. R. G. Lawford, P. B. Alaback, and E. R. Fuentes.
Ecological Studies 116. Berlin: Springer-Verlag.
Arroyo, M. T. K., M. Riveros, A. Penaloza, L. Cavieres, and A. M. Faggi. 1996. Phytogeo-
graphic relationships and regional richness patterns of the cool-temperate rainforest
flora of southern South Amer ica. Pp. 134–72 in High-latitude rain forests of the west coast
of theAmericas: Climate, hydrology, ecology and conservation. Ed. R. Lawford,P.Alaback, and
E. R. Fuentes. Ecological Studies 116. Berlin: Springer-Verlag.
———,C. Marticorena,O.Matthei, and L. Cavieres. 2000.Plant invasions in Chile:Present
patterns and future predictions. Pp. 385–421 in Invasive species in a changing world. Ed.
H.A. Mooney and R. J. Hobbs.Washington, DC: Island Press.
Axelrod, D. I. 1976. History of coniferous forests. Berkeley, CA: Univ. of Califor nia Press.
Baker,T. R., E. N. Coronado, O. L. Phillips, J. Martin, G. M. F. van der Heijeden, M. Garcia,
and J. S. Espejo. 2007. Low stocks of coarse woody debris in a southwest Amazonian
forest. Oecologia 152:495–504.
Bormann, B.T., H. Spaltenstein, M. H. McClellan, F. C. Ugolini, J. K. Cromack, and S. M.
Nay. 1995. Rapid soil development after windthrow disturbance in pristine forests.
Journal of Ecology 83:747–56.
Bryant, D., D. Nielsen, and L. Tangley. 1997. Last frontier forests: ecosystems and economies on
the edge. World Resources Institute,Washington, DC. www.wri.org/publication/last-
frontier-forests
Burnham, K. P., and D. R.Anderson. 2002. Model selection and multimodel inference a practical
information-theoretic approach. 2nd edition. New York: Spr inger.
C
ˇesˇka, O. 2009. A survey of macrofungi on Observatory Hill: Fall 2008 and winter
2008/2009. Unpublished report available at www.geog.ubc.ca/biodiversity/eflora/
macrofungi_observatory_hill.html
Cook,J.,A. L.Bidlanck, et al.2001.A phylogenetic perspective on endemism in theAlexan-
der Archipelago of the North Pacific.Biological Conservation 97:215–27.
JustWhat AreTemperate and Boreal Rainforests? 37

Craighead,L., and B. Cross.2007. Identifying core habitat and connectivity for focal species
in the interior-cedar hemlock forests of North America to complete a conservation
area design.Pp. 1–16 in USDA Forest Service Proceedings RMRS-P-49.
Ezcurra, C., N. Baccala, and P.Wardle. 2008. Floristic relationships among vegetation types
of New Zealand and the Southern Andes:Similarities and biogeographic implications.
Annals of Botany 101:1401–12.
FAO. 2005. The global forest resources assessment 2005. www.fao.org/forestry/fra2005/
en/
Farjon,A. 2005.Monograph of Cupressaceae and Sciadopitys. Royal Botanic Gardens,Kew, UK.
Fichtler, E., D. A. Clark, and M. Worbes. 2003. Age and long-term growth of trees in an
old-growth tropical rain forest, based on analyses of tree rings and 14C1.Biotropica 35
(3):306–17.
Franklin, J. F.,T. A. Spies, R.Van Pelt,A. B. Carey, D. A.Thornburg, D. Rae Berg, D. B. Lin-
denmayer, et al. 2002.Disturbances and structural development of natural forest eco-
systems with silvicultural implications, using Douglas-fir forests as an example. Forest
Ecology and Management 155:399–423.
Friedman, E. I., and H. J. Sun 2005. Communities adjust their temperature optima by shift-
ing producer-to-consumer ratio, shown in lichens as models: I. Hypothesis. Microbial
Ecology 49:523–27.
Fujimori,T.1971. Primary production of a young Tsuga heterophylla stand and some specu-
lations about biomass of forest communities on the Oregon coast. USDA Forest Ser-
vice Research Paper PNW-123.
Gallant, A. L. 1996. USGS ecoregions of Alaska map. www.explorenorth.com/library/
maps/ecoreg-alaska.html
Gavin, D. G., L. B. Brubaker, and K. P. Lertzman. 2003. Holocene fire history of a coastal
temperate rain forest based on soil charcoal radiocarbon dates.Ecology 84 (1):186–201.
———, and F. S. Hu. 2006. Spatial variation of climatic and non-climatic controls on spe-
cies distribution: the range limit of Tsuga heterophylla.Journal of Biogeography 33:1384–
96.
Goward, T. and T. Spribille. 2005. Lichenological evidence for the recognition of inland
rainforests in western North Amer ica. Journal of Biogeography 32:1209–19.
Hickey, J. E. 1990. Change in rainforest vegetation in Tasmania. Tasforest 2:143–49.
Hijmans, R. J., S. E. Cameron,J. L. Parra, P. G. Jones,and A. Jarvis. 2005.Very high resolution
interpolated climate surfaces for global land areas. International Journal of Climatology
25:1965–78.
Hinojosa, L. F., J. J. Armesto, and C. Villagran. 2006. Are Chilean coastal forests pre-
Pleistocene relicts? Evidence from foliar physiognomy, palaeoclimate, and phytogeog-
raphy. Journal of Biogoegraphy 33:331–41.
Holderidge, L. R.,W. C. Grenke,W. H. Hatheway,T. Liang, and J.A. Tosi. 1971. Forest envi-
ronments in tropical life zones. Oxford, UK: Pergamaon Press.
Intergovernmental Panel on Climate Change (IPCC). 2007. Synthesis report. Geneva,
Switzerland. www.ipcc.ch/contact/contact.htm
Jarmon, S. J.,and M. J. Brown. 1983.A definition of cool-temperate rainforest in Tasmania.
Search 14:81–87.
38t e m pe ra te a n d b or e a l ra i n f o re s t s o f th e w o r l d

Kauppi, P. E., J. H.Ausubel, J. Fang,A. S. Mather, R.A. Sedjo, and P. E.Waggoner. 2006. Re-
turning forests analyzed with the forest identity. Proceedings of the National Academy of
Sciences 103:17574–79.
Keith, H., B. G. Mackey, and D. B. Lindenmayer. 2009. Re-evaluation of forest biomass car-
bon stocks and lessons from the world’s most carbon dense forests.Proceedings of the Na-
tional Academy of Sciences 106 (28):11635–40.
Kellogg, E. L., ed. 1992. Coastal temperate rain forests: Ecological characteristics, status and distribu-
tion worldwide. Portland, OR: Ecotrust.
———, ed. 1995. The rainforests of home:An atlas of people and place. Part 1: Natural forests
and native languages of the Coastal temperate rain forest. Portland, OR: An Interrain
Publication.
Kirk, R. and J. Franklin. 1992. The Olympic rainforest. Seattle: Univ. of Washington Press.
Kirkpatrick, J. B., and K. J. M. Dickinson. 1984.Vegetation of Tasmania 1:500,000. Forestry
Commission, Hobart.
Köppen,W.1918. Klassifikation der Klimate nachTemperatur, Niederschlag, und Jahreslauf.
Petermann’s Mitteilungen 64:193–203.
Kricher, J. 1997. A neotropical companion. Princeton, NJ: Princeton Univ. Press.
Lawford, R., P. Alaback, and E. R. Fuentes, eds. 1996. High-latitude rain forests and associated
ecosystems of the west coast of the Americas: Climate, hydrology, ecology and conservation. Eco-
logical Studies 116. Berlin: Springer-Verlag.
Lertzman, K., G. D. Sutherland, A. Inselberg, and S. C. Saunders. 1996. Canopy gaps and the
landscape mosaic in a coastal temperate rain forest.Ecology 77:1254–70.
Losos, E. C., and E. G. Leigh Jr. 2004. Tropical forest diversity and dynamism. Chicago: Univ. of
Chicago Press.
Lusk, C. 2008. Constraints on the evolution and geographical range of Pinus. New Phytolo-
gist 178:1–3.
Miyawaki, A., ed. 1980. Vegetation of Japan. Band 1:Yakushima.Tokyo: Shibundo. (In Japa-
nese with German summary.)
———, ed. 1981. Vegetation of Japan. Band 2: Kyushu.Tokyo: Shibundo. (In Japanese with
German summary.)
———, ed. 1982. Vegetation of Japan. Band 3: Shikoku.Tokyo: Shibundo. (In Japanese with
German summary.)
———, ed. 1983. Vegetation of Japan. Band 4: Chugoku.Tokyo: Shibundo. (In Japanese with
German summary.)
———, ed. 1984. Vegetation of Japan. Band 5: Kinki. Tokyo: Shibundo. (In Japanese with
German summary.)
———, ed. 1985. Vegetation of Japan. Band 6: Chubu.Tokyo: Shibundo. (In Japanese with
German summary.)
———, ed. 1986. Vegetation of Japan. Band 7: Kanto.Tokyo: Shibundo. (In Japanese with
German summary.)
———, ed. 1987. Vegetation of Japan. Band 8:Tohoku. Tokyo: Shibundo. (In Japanese with
German summary.)
———,ed. 1988. Vegetation of Japan. Band 9:Hokkaido.Tokyo:Shibundo.(In Japanese with
German summary.)
JustWhat AreTemperate and Boreal Rainforests? 39

———, ed. 1989. Vegetation of Japan. Band 10: Okinawa/Ogasawara.Tokyo: Shibundo. (In
Japanese with German summary.)
Moen, A. 1999. National atlas of Norway: Vegetation. Hønefoss: Norwegian Mapping
Authority.
Moore,K. 1990.Where is it and how much is left? The state of the temperate rainforest in
British Columbia. Forest Planning Canada 6:15.
Myers,N.,R.A. Mittermeier,C. G. Mittermeier,G.A. B.da Fonseca, and J. Kent. 2000.Bio-
diversity hotspots for conservation priorities. Nature 403:853–58.
Nadkarni, N. M.,T. J. Matelson, and W. A. Haber. 1995. Structural characteristics and flor-
istic composition of a neotropical cloud forest, Monteverde, Costa Rica. Journal of
Tropical Ecology 11:481–95.
Nowacki, G. J., and M. G. Kramer. 1998.The effects of wind disturbance on temperate rain
forest structure and dynamics of southeastAlaska. USDA Forest Service,Pacific North-
west Research Station. PNW-GTR-421.
Olson, D. M., and E. Dinerstein. 1998.The global 200: A representation approach to con-
serving the earth’s most valuable ecoregions. Conservation Biology 12 (3):502–15.
Omernik, J. M. 1987. Ecoregions of the conterminous United States. Map (scale
1:7,500,000). Annals of the Association of American Geographers. 77:118–25.
Ott,R.A., and G. P. Juday. 2002.Canopy gap characteristics and their implications for man-
agement in the temperate rainforests of southeastAlaska. Forest Ecology and Management
159:271–91.
Perry, D.A., R. Oren, and S. C. Hart. 2008. Forest ecosystems. Baltimore, MD: Johns Hopkins
Univ. Press.
Phillips, S. J., R. P. Anderson, and R. E. Schapire. 2006. Maximum entropy modeling of spe-
cies geographic distributions. Ecological Modelling 190:231–59.
Premoli,A. C.,T. Kitzberger, and T.T.Veblen. 2000. Isozyme variation and recent biogeo-
graphical history of the long-lived conifer Fitzroya cupressoides. Journal of Biogeography
27:251–60.
Read, J., and R. S. Hill. 1985. Dynamics of Nothofagus-dominated rain forest on mainland
Australia and lowland Tasmania. Vegetatio 63:67–78.
Richards, P. W. 1996. The tropical rainforest. Cambridge, UK: Cambridge Univ. Press.
Ritter, M. 2008.The forest biome. www.uwsp.edu/geo/faculty/ritter/geog101/textbook/
climate_systems/tropical_rainforest_1.html
Roberts, C., O. C
ˇesˇka, P. Kroeger, and P. Kendrick. 2004. Macrofungi from six habitats over
five years in Clayoquot Sound,Vancouver Island. Canadian Journal of Botany 82:1518–
38.
Schoonmaker, P. K., B. von Hagen, and E. C.Wolf , eds. 1997. The rainforests of home: Profile of
a North American bioregion. Washington, DC: Island Press.
Searchinger,T. D., S. P. Hamburg, J. Melillo,W. Chameides, P. Havlik, D. M. Kammen, G. E.
Likens, et al. 2009. Fixing a critical climate accounting error. Science 326:527–28.
Shanks, R. E. 1954. Climates of the Great Smoky Mountains. Ecology 35:354–61.
Sheard,J.W.1995. Disjunct distributions of some North American,corticolous,vegetatively
reproducing Rinodina species (Physciaceae, lichenized Ascomycetes). Herzogia 11:115–
32.
40t e m pe ra te a n d b or e a l ra i n f o re s t s o f th e w o r l d

Simenstad, C.A., M. Dethier, C. Levings, and D. Hay. 1997.The ter restrial/marine ecotone.
Pp. 149–88 in The rainforests of home. Ed. P.K. Schoonmaker, B. von Hagen, and E. C.
Wolf. Washington, DC: Island Press.
Smith-Ramirez,C. 2004.The Chilean coastal range:A vanishing center of biodiversity and
endemism in South American temperate forests. Biodiversity and Conservation 13:373–
93.
Smithwick, E.A. H., M. E. Harmon, S. M. Remillard, S. A.Acker, and J. F. Franklin. 2002.
Potential upper bounds of carbon stores in forests of the Pacific Northwest. Ecological
Applications 12:1303–17.
Spies,T.A., J. F. Franklin, and M. Klopsch. 1990. Canopy gaps in Douglas-fir forests of the
Cascade Mountains. Canadian Journal of Forest Research 20:649–58.
Spribille,T., S. Pérez-Ortega,T.Tønsberg, and D.Schirokauer.2010. Mining lichen diversity
in the Klondike. The Bryologist. Forthcoming.
Stephenson, N. L. 1990.Climatic control of vegetation distribution:The role of the water
balance. American Naturalist 135 (5):649–70.
Terborgh, J. 1992. Diversity and the tropical rainforest. New York: Scientific American Librar y.
Thornthwaite,C.W.1948.An approach toward a rational classification of climate.Geograph-
ical Review 38:55–94.
Tuhkanen, S. 1984. A circumboreal system of climatic-phytogeographical regions. Acta
Botanica Fennica 127:1–50.
Tønsberg,T. 1992.The sorediate and isidiate,corticolous, crustose lichens in Norway.Som-
merfeltia 14:1–31.
———. 1993.Additions to the lichen flora of North America. Bryologist 96:138–41.
United Nations Environment Programme (UNEP). 2009.Toward sustainable production
and use of resources: assessing biofuels. UNEP headquarters, Nairobi, Kenya.
Veblen, T. T. 1985. Forest development in tree-fall gaps in the temperate rain forests of
Chile. National Geographic Research. Spring: 162–83.
———, and P. B.Alaback. 1996.A comparative review of forest dynamics and disturbance
in the temperate rainforests in North and South America. Pp. 173–213 in High-latitude
rain forests of the west coast of the Americas: Climate, hydrology, ecology and conservation. Ed.
R. Lawford, P. Alaback, and E. R. Fuentes. Ecological Studies 116. Berlin: Springer-
Verlag.
———, R. S. Hill, and J. Read. 1996. Ecology and biogeography of Nothofagus forests. New
Haven:Yale Univ. Press.
———, and T. Kitzberger. 2002. Interhemispheric comparison of fire history: The Col-
orado Front Range, USA, and the Northern Patagonian Andes,Argentina. Plant Ecol-
ogy 163:187–207.
Waring, R. H., and J. F. Franklin. 1979. Evergreen coniferous forests of the Pacific North-
west. Science 204:1380–86.
Whittaker, R. H. 1975. Communities and ecosystems. NewYork: Macmillan and Co.
Willson, M. F., B. L. Rice, and M.Westoby. 1990. Seed dispersal spectra: A comparison of
temperate plant communities. Journal ofVegetation Science 1:547–62.
JustWhat AreTemperate and Boreal Rainforests? 41