1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Natural Migration Rates of Trees: Global Terrestrial Carbon Cycle Implications
EP A/600/A-96/032
Allen M. Solomon
National Health and Environmental Effects Research Laboratory
Environmental Protection Agency
200 S.W. 35th Street
Corvallis, OR 97333 USA
INTRODUCTION
Migration of populations or species of trees ("tree migration") in response to climate change
is of interest both to paleoecologists who assess past vegetational responses to climate change,
and to global ecologists concerned with future climate change induced by increasing greenhouse
gases (GHGs). A major difference between climate-driven tree migrations in prehistory and
those expected in the future is the high speed of the latter climate change. The 4-6 km which
temperate-zone July isotherms are predicted to move northward annually (Solomon et al, 1984)
are about an order of magnitude more rapid than prehistoric rates deduced from paleoecological
evidence. Assuming prehistoric rates of warming matched the rate of tree migration (T. Webb
1986; Prentice et al. 1991), fossil pollen data allow inference of 400 m/yr (Davis 1983) to 800
m/yr (Gear and Huntley 1991) of climate change and tree migration at most. The rate may be
even slower if tree migration includes the establishment and maturity of the tree population
(Bennett 1986) as well as the processes of seed transport, establishment, growth and seed
production, normally defined as migration (e.g., Davis, 1989; MacDonald et al. 1993).
The difference in definition is important for predicting the amount of carbon (CO2 is the
most important of the GHGs) that will reside in the atmosphere in the future. The oceans
provide the ultimate long-term control on atmospheric carbon concentrations (e.g., Sundquist
1985; Prentice et al. 1993). However, the terrestrial biosphere modulates the shorter-term
changes in carbon content, measured over a few decades or centuries (Gammon et al. 1985;
Keeling et al. 1995; Denning et al. 1995). Forests store about 2/3 of above-ground terrestrial
organic carbon and over half of the carbon present in the world's soils (Dixon et al. 1994). The
presence of a few trees on the landscape (e.g., MacDonald et al. 1993), indicated by
establishment and reproductive maturity of seed trees, contributes little carbon to terrestrial
stocks. Instead, closed-canopy stands of mature mixed or pure species provide the dense
carbon stocks of interest. These arc associated with mature, stable populations.
Projections of global terrestrial carbon cycle dynamics under warmer climates of a doubled
GHG concentration have used static vegetation models (Prentice and Solomon 1990). These
projections hinge on the critical assumption that the migration of trees and the formation of
mature, stable populations at new locations proceeds at the same rate as the climate change to
which it is responding (Sedjo and Solomon 1989; Leemans, 1989; Prentice and Fung 1990;

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Smith, et al. 1992a, b; Smith and Shugart, 1993a, b; Solomon et al. 1993). To date, these static
model exercises have projected increased global terrestrial carbon storage under future
warming, because large new land areas suitable for forest growth are created either by warming
of high latitude treeless tundra, or by increased hydrologic cycle intensity in treeless steppe.
Yet, unchanged or decreased rather than increased carbon storage may result if forests
cannot migrate and establish in the time required to attain the doubled GHG benchmark. The
objective of the current paper is to estimate the time required for forests to develop in regions
new to them, to estimate the time required for forests to die out where they become climatically
obsolete, then to calculate the impacts of those times on future terrestrial carbon stocks.
MIGRATION RATE LIMITS
For our purposes, three time-variable steps may be distinguished in the migration process:
seed transport, coupled with establishment of seedlings; reproductive maturation of individuals:
and, forest, i.e., maturation of populations in closed canopy forests.
Seed Transport. Tree seed transport, whether by wind, animals or running water, requires
little more than a day. Establishment of tree seeds (seed germination and growth of a taproot
into mineral soil) requires one to two growing seasons. Taken together, transport and
establishment are instantaneous compared with the multiple decades required for warming.
However, transport and establisment is not a singular process but rather, must be repeated
multiple times, each consisting of several time-variable processes.
Some shifting populations possess sharp boundaries which resemble a slowly moving wave
(Davis, 1987). These depend upon the regular transport of seeds 10-100 m, followed by a few
decades to centuries of tree maturity before another 10-100 m "step" is taken. Other population
boundaries consist of "infection sites," located well beyond of the main population, surrounded
by population voids (Davis, 1987, 1990), and derived from irregular transport events. The
latter pattern results from migration comprised of rare long-distance transport and establishment
events, and is followed by local population growth via transport and establishment between
founder seed sources (Bennett 1984, Davis 1987). This migration form probably produces the
most rapid migration rates (Leishman et al., 1992; Collingham et al., 1996). Multiple rare
events, by definition, form a (long) time-ordered process. Jumps of 100-200 km have been
detected in the Holocene at about 1000 year intervals (S. Webb, 1986; Davis et al., 1986).
Tree Maturation. The time required to complete tree life cycles varies considerably. Most
evergreen and deciduous conifers can reproduce within 10 years of seed germination and
deciduous hardwoods, within 15 years (Harlow et al. 1979). Yet, these times apply to trees
growing in full sunlight (i.e., without other trees nearby), and to reproduction by trees still the
size of saplings (i.e., 2-5 meters tall). Annual seed production in this case is very small (1000s
instead of 100,000s of seeds per tree; Bums and Honkala 1990), and transport distances may
be quite small because of the low stature of the seed sources (e.g., <50 m for trees <5 m tall,

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
according to experiments by Greene and Johnson 1989). The short life cycling times occur
when tree populations invade treeless areas and form clumped seed sources (with low carbon
densities), rather than when populations expand to form areas of continuous forest between
seed sources, or grow rapidly within stands already populated by mature forests.
Forest Maturation. The development of scattered individuals and stands of trees into closed
canopy forests containing maximum carbon densities takes considerably longer than maturation
of isolated trees. First, the areas between trees and stands must be occupied. If 10-15 year-old
trees each generation produce seedlings 50-100 m away, even a founder population of one
tree/km2 (not a rare event distribution but one calculated by Greene and Johnson, 1995, as rare
enough to constrain metapopulation expansion) requires 5-10 generations to plant seeds in all
parts of the square kilometer. The sequential completion of 5 or 10 generations could require
50-100 years in conifers and 75-150 years in deciduous hardwoods from dense populations,
and a millennium from sparse tree densities (ibid.). The minimum number of life cycles
obviously depends on the distances between seed sources, but would usually exceed the 5-10
generations exemplified at distances beyond about 2-10 km from the original forests (ibid.).
This is consistent with estimates of mid-Holocene hemlock occupation of northern lower
Michigan in about 500 years (Davis et al., 1986) and documentation of periods exceeding 1000
years for local forest development during the forestation of the British Isles (Bennett 1986).
Second, closed forests must develop. Tree growth rates are much slower in the shade than in
the open. In closed forests, trees must grow to reach the canopy surface before producing
significant numbers of seeds (Daubenmire, 1959; Waring and Schlesinger, 1985). This
requires a minimum of 25-50 years in most temperate and boreal regions in which closed
canopy forests exist (Harlow et al. 1979). Certain shade-tolerant species may require 150 (Acer
saccharum, Canham 1988) to 450 years (Tsuga canadensis, Godman and Lancaster 1990) to
reach the canopy and complete the cycle. Even the 25-50 years for life cycle completion in
closed canopy forests presumes optimum rates of height and diameter growth. Yet, warming is
likely to slow growth rates, as it has in the past (e.g., Fritts 1976; Briffa et al. 1995).
In sum, the minimum time required for development of mature forests by tree species that
originate elsewhere is between 100 years (conifers that spread from initial seed sources in an
arbitrary five 10-year generations and grow to canopy height in 25 years) and 200 years
(deciduous hardwoods that spread from initial seed sources in ten 15-year generations and
grow to canopy height in 50 years). This "transient response" to climate changes is consistent
with forest maturity rates measured in the past and simulated by mechanistic gap models (e.g.,
Solomon 1986; Bugmann 1993). The latter models exclude the time needed for development of
isolated tree seed sources but include crude effects of chronic climate change on tree growth. In
reality, the rate will probably be much slower, perhaps approaching the >1000 yr period
inferred by Bennett (1986, 1988) in prehistoric data describing forest initiation and maturation.
Forest Tree Mortality. The other half of the migration question for calculating carbon
stocks is the concomitant mortality of trees which have become climatically and fatally
"obsolete" (i.e., after climate variables exceed their climate tolerances). Stress induced by

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
warming and drought may directly slow growth of individual trees and thereby, induce
mortality, usually in less than a decade (Nichols, 1968; Franklin et al., 1987; Auclair, 1992),
especially among seedlings and saplings (Peet and Christiansen, 1987). However, climate
stresses more commonly predispose trees to succumb to other mortality agents (Waring, 1987),
such as air pollutants (Hinrichsen, 1987), wildfire (Payette 1992), insect infestations (Holling
1992) and windthrow (Webb 1989). Such direct and indirect stress-induced mortality can be
viewed as chronic decline over several decades if it is measured at the scale of a large region,
where new individuals, populations and stands are dying each year (Mueller-Dombois, 1992).
The amount of time required to extirpate populations from regions in which they have
become climatically obsolete is exceedingly complex to predict. The fatal obsolescence itself is a
non-linear time-transgressive property of the rate and spatial distribution of climate change.
Trees growing in areas undergoing the most rapid climate changes, growing near limits of their
distribution and growing in stressed habitats may die in only a few years, while others may just
begin to sense stress by the time GHGs have doubled. Following mortality, release of carbon
from dead trees may require additional decades. Harmon et al (1986) cite 50% volume loss
times of 14-172 years for log mineralization of softwoods under temperate climates inducing
slow decomposition, and 2-24 years for hardwoods.
Although seedlings and saplings should disappear quickly, the loss of mature trees which
store most of the carbon is more relevant. Increased, climate-induced mortality of mature trees,
and carbon losses to mortality greatly exceeding carbon gains from new growth (Kirschbaum
and Fischlin 1996), is expected to generate a future pulse of increased atmospheric CO2
(Solomon, 1986; King and Neilson, 1992; Smith and Shugart 1993a,b). Epidemics of
pathogens and insects, mortality agents through which climate change may act, have required
about 50 years in eastern North America to kill most individuals of American chestnut (Odum,
1969) and American elm (Gibbs, 1978) in the 20th century and eastern hemlock in mid-
Holocene time (Davis, 1981). Based on the foregoing, I assumed that trees would be extirpated
from areas of fatal obsolescence in the 60-70 years required to reach climate of doubled GHGs.
MIGRATION RATES APPLIED
The minimum of 100 to 200 years required to develop forests composed of new species in a
given region following the imposition of a changed climate can be compared to the expected
time climate will take to change. Frequently, the time needed to impose the climate of doubled
concentrations of CO2 or of all GHGs (e.g., Houghton et al. 1995) is used, currently expected
to occur in 60 to 70 years or by about the year 2050 (Greco et al. 1994). At that rate, forest
migration would be very incomplete at best, even if the forest development began when climate
change began, rather than following sometime after initiation of the climate change.
This lag is incompatible with the assumption of instant migration and forest maturity
incorporated by the static models discussed in the introduction, above. Consider that the static

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
vegetation models utilize a half-degree latitude and longitude grid (one degree by Prentice and
Fung 1990), with a latitudinal resolution of about 55 km at the equator. Measurement of any
tree migration which sequesters carbon by the 70 year doubling of GHGs is highly unlikely, if
not impossible, even at the normal 150-400 m/yr migration rate Davis (1983) estimated from
fossil pollen data (producing tree migration of only 11-28 km during the 70 yrs). A more
realistic assumption is that no forest migration occurs in the 70-year time span needed to reach a
doubling of CO2 or GHGs, although migration would occur eventually. Belotelov et al. (1996)
applied this model condition to the Holdridge Life Zone System to define the absolute range of
carbon values possible for specified future climate scenarios in the former Soviet Union.
Solomon and Kirilenko (1996) modified the Biome 1.1 model (Prentice et al. 1992, 1993)
to reflect this "delayed migration" condition. They assumed that tree functional types (TFTs)
which were incompatible with doubled GHG climates would disappear during the 60 or 70
years from areas acceptable under the initial climate, but that TFTs could not appear in areas in
which they were absent under the initial climate. Nonarboreal plant functional types (shrubs,
grasses; NAFTs) were modeled as migrating instantly, on the assumption that they are able to
produce seed as quickly as their first or second growing seasons, greatly reducing lags in
response to rapid climate change.
Figure 1 illustrates a Biome 1.1 model run using climate output from the UKTR coupled
ocean-atmosphere GCM for climate of current and of doubled GHG concentration (Solomon
and Kirilenko 1996). Note that differences between instant migration and no migration are
present at all latitudes. Reduced temperature constraints on tree growth in high latitudes permit
occupation of latitudinal bands of nonarboreal tundra by boreal forests under instant migration
Figure 1 . Global distribution of land occupied by forest, projected by the BIOME 1.1
model (Prentice et al., 1993) under the UKTR climate scenario (Murphy and Mitchell
1996), Areas in black show differences between assuming instant migration and absent
migration and areas in grey are forested under either instant or absent migration.

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
assumptions. Similarly, increased soil moisture and instant tree migration produces forest
occupation of longitudinally-oriented steppe and savanna areas by dry temperate forests in
North and South America and by tropical forests in Africa, India and Asia. In part the patterns
reflect the great difference in sensitivity of temperature to GHGs at high latitudes compared to
that at low latitudes which undergo little warming (Greco et al 1995).
RESULTS AND DISCUSSION
The paper is written to estimate the time required to form closed canopy, mature forests by
immigrating species and the time required to eliminate trees which have become fatally obsolete.
The objective is to define the implications of this lagged climate response to the global carbon
cycle. Prentice et al (1993) have transformed biome areas into carbon stocks by appying carbon
density estimates for aboveground and belowground carbon.
Table 1 compares terrestrial carbon storage under future climate with migration versus no-
migration assumptions. Like other static models, this exercise with instant migration generated
more terrestrial carbon under doubled GHG climate than under modern climate. Simulated
tropical forests invade large areas of grassland and savanna due to the positive moisture balance
this particular GCM projects for the tropics. The difference between modern and future carbon
with instant migration is considerably less than others have projected (Prentice and Fung 1990;
Leemans, 1989; Smith et al, 1992a, b; Smith 1993; Solomon et al. 1993). The UKTR coupled
ocean-atmosphere GCM is considerably less GHG-sensitive than those used in the past.
TABLE 1. Above ground (AG)1, below ground (BG)2 and total (TOTAL) biomass in Petagrams
(Pg) under modern and doubled GHG climate, with and without tree migration
DOUBLED GHG CLIMATE*
BIQMES
Boreal Forests
Temperate Forests
Tropical Forests
Boreal Nonforest
Temperate Nonforest
Tropical Nonforest
Forest Biomass
Total Biomass
MODERN CLIMATE3 '
AG BG	TOTAL
300	470
360	529
289	589
177	188
189	211
144	196
WITH MIGRATION
AG
BG
TOTAL
179
298
477
174
343
517
346
330
676
8
127
134
25
191
216
51
132
183
NO MIGRATION
AG BG TOTAL
165
292
457
147
336
483
287
281
568
12
151
163
33
261
294
49
128
177
170
169
300
11
22
52
639 949 1588 699 971 1670 599 909 1508
723 1459 2177 783 1420 2203 693 1449 2142
1.	from table converting area to biomass in Prentice et al. 1993, based on Olson, et al. 1983.
2.	from table converting area to biomass in Prentice et al. 1993, based on Zinke et al. 1984.
3.from	data of Leemans and Cramer (1990).
4.	from temperature and precipitation differences between 2 X C02 and 1 X C02 of Murphy and
Mitchell (1996) applied to data of Leemans and Cramer (1990).

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
In contrast to carbon increases associated with instant migration, the amount of carbon
stored in terrestrial vegetation and soils declines under GHG-induced climate when one assumes
that trees do not migrate (Table 1). All forests decline in biomass and do so both above (40 Pg)
and belowground (40 Pg). Nonarboreal biomass that replaced forest biomass (9 Pg above
ground, 30 Pg below ground) does not make up for the forest carbon loses. Despite the
moderate future climate scenario used, and hence, the moderate amount of forest response
projected, the difference in forest biomass between the two assumptions is significant. The
above-ground forest biomass difference is 100 Pg or about 16% of the initial 639 Pg, and the
total forest biomass difference is 162 Pg, or 10% of the initial 1583 Pg.
The foregoing values, for vegetation composition responses to climate change alone, exceed
the calculated amount of carbon sequestered in terrestrial vegetation from carbon fertilization, as
hypothesized from lab experiments (Melillo et al. 1993; Schimel et al. 1995), or calculated by
difference from ocean uptake (Denning et al. 1995). Although the global difference for all
biomes is considerably less (61 Pg), it is the forests, especially those at high latitudes, which are
suspected of increasing storage of the carbon not accounted for in ocean-atmosphere models.
Clearly, where this view depends on the future distribution of forests (e.g., Melillo et al. 1993;
VEMAP Participants 1996), it needs to be reexamined. In any case, the amount of carbon
tabulated under the traditional instant-migration assumption should be recalculated based on a
concept of imperceptibly slow forest immigration response to climate change in the 21st century .
ACKNOWLEDGEMENTS
I thank Joe Alcamo who inadvertently convinced me to examine the ecological implications of
the concept, and Andrew Kirilenko, who first suggested to me the idea of nonmigration of trees
under changed climate, and who modified the Biome 1.1 model to run without tree migration.
Harold Mueller provided the figure. Critical reviews by Wolfgang Cramer, David Hollinger,
George King, Andrew Kirilenko, William Leak, Dale Solomon and an anonymous reviewer
improved the paper considerably. This paper was subjected to peer and administrative review by
the U.S. Environmental Protection Agency and was approved by the Agency for publication.
REFERENCES CITED
Auclair, AND (1993) Extreme climatic fluctuations as a cause of forest dieback in the Pacific
Rim. Air Water and Soil Pollut. 66:207-229
Bennett KD (1984) The post-glacial history of Pinus sylvestris in the British Isles. Quat. Sci.
Rev. 3:133-155.
Bennett KD (1986) The rate of spread and population increase of forest trees during the
postglacial. Philos. Trans. R. Soc. London B 134:523-531.
Bennett KD (1988) Modeling changes in beech populations: A reply to Dexter et al. (1987).
Rev. Palaeobot. Palynol. 56:361-364.

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
Belotelov NV, Bogatyrev BG, Kirilenko AP, and Venevsky SV (1996) Modelling time-
dependent biome shifts under global climate changes. Ecological Modeling (in press).
Briffa KR, Jones PD, Schweingruber FH, Shiyatov SG, and Cook ER (1995) Unusual
twentieth-century summer warmth in a 1,000-year temperature record from Siberia. Nature
376:156-159.
Bugmann HKM (1993) On the ecology of mountainous forests in a changing climate: A
simulation study. Ph.D. Dissertation, Swiss Federal Institute of Technology, Zurich.
Burns RM and Honkala BH (eds) (1990) Silvics of North America. Volume 1, Conifers;
Volume 2, Hardwoods. Agr. Hndbk. 654. USDA Forest Service, Washington D.C.
Canham CD (1988) Growth and canopy architecture of shade-tolerant trees: response to canopy
gaps. Ecology 69:786-795
Collingham YC, Hill MO and Huntley B (1996) The migration of sessile organisms: a
simulation model with measureable parameters. J. Veg. Sci. (in press)
Daubenmire RF (1959) Plants and environment. J. Wiley and Sons, Inc., NY
Davis MB (1981) Outbreaks of forest pathogens in Quaternary history. Proc. IV Int. Palynol.
Conf., Lucknow (1976-1977) 3:216-227
Davis MB (1983) Quaternary history of deciduous forests of eastern North America and
Europe. Ann. Mo. Bot. Gard. 70:550-563.
Davis MB (1987) Invasions of forest communities during the Holocene: Beech and hemlock in
the Great Lakes Region. 373-393 IN Gran AJ, Crawley MJ and Edward PH (eds)
Colonization and stability. Blackwell Scientific Publications, Oxford UK
Davis MB (1989) Lags in vegetation response to greenhouse warming. Clim. Change 15:75-82
Davis MB (1990) Climatic change and the survival of forest species. 99-110 IN Woodwell GM
(ed) The earth in transition: Patterns and processes of biotic impoverishment. Cambridge
Univ. Press, Cambridge UK
Davis MB, Woods KD, Webb SL and Futyma R (1986) Dispersal versus climate: Expansion of
Fagus and Tsuga into the Upper Great Lakes region. Vegetatio 67:93-103
Denning AS, Fung IY and Randall D (1995). Latitudinal gradient of atmospheric CO2 due to
seasonal exchange with land biota. Nature 376:240-243.
Dixon RK, Brown S, Houghton RA, Solomon AM, Trexler MC and Wisneiwski J (1994)
Carbon pools and flux of global forest ecosystems. Science 263:185-190.
Franklin JF, Shugart HH and Harmon ME (1987) Tree death as an ecological process.
Bioscience 37:550-556
Fritts HA (1976) Tree rings and climate. Academic Press, NY
Gammon RH, Sundquist ET and Fraser PJ (1985) History of carbon dioxide in the atmosphere
25-62 IN Trabalka JR (ed) Atmospheric carbon dioxide and the global carbon cycle.
DOE/ER-0239, US Dept. of Energy, Washington DC.
Gear A J and Huntley B (1991) Rapid changes in the range limits of scots pine 4000 years ago.
Science 251:544-547.
Godman RM and Lancaster K (1990) Tsuga canadensis (L) Carr., Eastern Hemlock. 604-612
IN Burns RM and Honkala BH, (eds) (1990a) Silvics of North America. Volume 1,
Conifers. Agr. Hndbk. 654, USDA Forest Service, Washington D.C.
Greco S, Moss RH, Viner D and Jenne R (1995) Climate Scenarios and Socioeconomic
Projections for IPCC WG II Assessment. Material assembled for Lead Authors by IPCC
WG II TSU, Washington, D.C.
Greene DF and Johnson EA (1989) A model of wind dispersal of winged or plumed seeds.
Ecology 70:339-347.
Greene DF and Johnson EA (1995) Long-distance wind dispersal of tree seeds. Can. J. Bot.
73:1036-1045
Gibbs JN (1978) Intercontinental epidemiology of Dutch elm disease. Ann. Rev. Phytopath.
16:287-307
Harlow WM, Harrar ES and White FM (1979) Textbook of dendrology, 6th edn. McGraw-
Hill, NY
Harmon ME, Franklin JF, Swanson FJ, Sollins P, Gregory SV, Lattin JD, Anderson NH,
Cline SO, Aumen NG, Sedell JR, Lienkaemper GW, Cromack K, and Cummins KW
(1986) Ecology of coarse woody debris in temperate ecosystems. Adv. Ecol. Res. 15:133-
302
Hinrichsen D (1987) The forest decline enigma. Bioscience 37:542-546
Holling CS (1992) The role of forest insects in structuring the boreal landscape. 170-191 IN

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
Shugart, HH, Leemans R and Bonan GB (eds) A Systems Analysis of the Global Boreal
Forest. Cambridge University Press, NY.
Houghton JT, Meira-Filho LG, Bruce J, Lee H-S, Callander BA, Haites E, Harris N and
Maskell K (eds) (1995) Climate change 1994: Radiative forcing of climate change and an
evaluation of the IPCCIS92 emission scenarios. Cambridge Univ. Press, Cambridge, UK
Keeling CD, Whorf TP, Wahlen M and van der Plicht J (1995) Interannual extremes in the rate
of rise of atmospheric carbon dioxide since 1980. Nature 375:666-670.
King GA and Neilson RP (1992) The transient response of vegetation to climate change: A
potential source of CO2 to the atmosphere. Water, Air and Soil Pollut. 64:365-383
Kirschbaum M and Fischlin A (1996) The impacts of climate change on forest ecosystems,
IPCC WG II Second Assessment Report, Chapter 1. (in press)
Leemans R (1989) Possible changes in natural vegetation patterns due to a global warming 105-
122 IN Hack] A (ed.) Der Treibhauseffekt: Das Problem-Mogliche Folgen-Erforderliche
Massnahmen. Akadamie fiir Umwelt und Energie, Laxenburg, Austria.
Leemans R and Cramer WP (1991) The II AS A database for mean monthly values of
temperature, precipitation and cloudiness on a global terrestrial grid. Research Report
RR91-18. Internat. Inst, for Applied Sys. Anal. Laxenburg, Austria.
Leishman MR, Hughes L, French K, Armstrong D and Westoby M (1992) Seed and seedling
biologv in relation to modeling vegetation dynamics under global climate change. Aust. J.
Bot. 40:599-613
MacDonald GM, Edwards TWD, Moser KA, Pienitz R and Smol JP (1993) Rapid response of
treeline vegetation and lakes to past climate warming. Nature 361:243-246.
Melillo JR, McGuire AD, Kicklighter DW, Moore B, Vorosmarty CJ and Schloss AL (1993)
Global climate change and terrestrial net primary production. Nature 363: 234-239.
Mitchell JFB, Johns TC, Gregory JM and Tett SFB (1995) Climate response to increasing
levels of greenhouse gases and sulphate aerosols. Nature 376:501-504
Mueller-Dombois D (1992) Potential effects of the increase in carbon dioxide and climate change
on the dynamics of vegetation. Water, Air and Soil Pollut. 64:61-79
Murphy JM and Mitchell JFB (1996) Transient response of the Hadley Centre coupled ocean-
atmosphere model to increasing carbon dioxide. Part II: Spatial and temporal structure of the
response. Jour, of Climate (in press)
Nichols JO (1968) Oak mortality in Pennsylvania: A ten-year study. J. For. 66:681-694
Odum EP (1969) Fundamentals of Ecology, WB Saunders Co, Philadelphia PA
Olson JS, Watts J A and Allison LJ (1983) Carbon in Live Vegetation of Major World
Ecosystems. ORNL/TM-5862. Oak Ridge National Laboratory, Oak Ridge, TN.
Payette S (1992) Fire as a controlling process in the North American boreal forest. 144-169 IN
Shugart HH, Leemans R and Bonan GB (eds) A Systems Analysis of the Global Boreal
Forest. Cambridge Univ. Press, Cambridge, UK.
Peet RK and Christiansen NL (1987) Competition and tree death. Bioscience 37:586-595
Prentice K and Fung IY (1990) The sensitivity of terrestrial carbon storage to climate change.
Nature 346:48-50.
Prentice IC and Solomon AM (1990) Vegetation models and global change. 365-383 IN
Bradley RS (ed) Global Changes of the Past. ODES, UCAR, Boulder, CO
Prentice IC, Bartlein PJ and Webb T IH (1991) Vegetation change in eastern North America
since the last glacial maximum: A response to continuous climatic forcing. Ecology
72:2038-2056.
Prentice IC, Cramer WP, Harrison SP, Leemans R, Monserud RA and Solomon AM (1992) A
global biome model based on plant physiology and dominance, soil properties and climate.
J. Biogeogr. 19:117-134.
Prentice IC, Sykes MT, Lautenschlager M, Harrison SP, Denissenko O and Bartlein PJ (1993)
Modelling global vegetation patterns and terrestrial carbon storage at the last glacial
maximum. Global Ecol. and Biogeogr. Let. 3:67-76.
Schimel D, Enting IG, Heimann M, Wigley TML, Raynaud D, Alves D and Siegenthaler U
(1995) CO 2 and the Carbon Cycle. 39-71 IN Houghton JT, Meira-Filho LG, Bruce J, Lee
H-S, Callander BA, Haites E, Harris N and Maskell K (eds) Climate change 1994:
Radiative Forcing of Climate Change and an Evaluation of the IPCC IS92 Emission
Scenarios. Cambridge Univ. Press, Cambridge, UK
Sedjo RA and Solomon AM (1989) Climate and forests. 105-119 IN Rosenberg NJ, Easterling
WE, Crosson PR and Darmstadter J (eds) Greenhouse Warming: Abatement and

-------
1	Adaptation. Resources for the Future, Washington, D.C.
2	Smith TM, Leemans R and Shugart HH (1992a) Sensitivity of terrestrial carbon storage to CO 2-
3	induced climate change: Comparison of four scenarious based on general circulation
4	models. Clim. Change 21:367-384
5	Smith TM, Shugart HH, Bonan GB and Smith JB (1992b) Modeling potential response of
6	vegetation to global climate change. Adv. Ecol. Res. 22:93-116.
7	Smith TM and Shugart HH (1993a) The transient response of terrestrial carbon storage to a
8	perturbed climate. Nature 361:523-526.
9	Smith TM and Shugart HH (1993b) The potential response of global terrestrial carbon storage to
10	a climate change. Water, Air and Soil Pollut. 70:629-642.
11	Solomon AM (1986) Transient response of forests to CC>2-induced climate change: Simulation
12	experiments in eastern North America. Oecologia 68:567-79.
13	Solomon AM and Kirilenko AP (1996) Implications for terrestrial carbon stocks of delayed tree
14	species immigration under changed climate (ms. submitted to Nature)
15	Solomon AM, Tharp ML, West DC, Taylor GE, Webb JW and Trimble JL (1984) Response of
16	Unmanaged Forests to C Dioxide-Induced Climate Change: Available Information, Initial
17	Tests, and Data Requirements. TR-009, United States Department of Energy Washington,
18	DC.
19	Solomon AM, Prentice IC, Leemans R and Cramer WP (1993) The interaction of climate and
20	land use in future terrestrial carbon storage and release. Water. Air and Soil Pollution
21	70:595-614.
22	Sundquist, ET (1985) Geologic analogs: Their value and limitations in carbon dioxide research
23	371-402. IN Trabalka JR and Reichle DE (eds) The Changing Carbon Cycle; A Global
24	ANalysis. Springer-Verlag, NY
25	VEMAP Participants (1996) Comparing biogeography and biogeochemistry models in a
26	continental-scale study of terrestrial ecosystem responses to climate change and C02
27	doubling. Global Biogeochem. Cyc, (in press)
28	Waring RH (1987) Characteristics of trees predisposed to die. Bioscience 37:569-574
29	Waring RH and Schlesinger WH (1985) Forest Ecosystems: Concepts and Management.
30	Academic Press NY
31	Webb SL (1986) Potential role of passenger pigeons and other vertebrates in the rapid Holocene
32	migrations of nut trees. Quat. Res. 26:367-375
33	Webb SL (1989) Contrasting windstorm consequences in two forests, Itasca State Park,
34	Minnesota. Ecology 70:1167-1180
35	Webb T in (1986) Is vegetation in equilibrium with climate? How to interpret late-Quaternary
36	pollen data. Vegetatio 67:75-91.
37	Zinke PJ, Stangenberger AG, Post WM, Emanuel WR and Olson JS (1984) Worldwide organic
38	soil carbon and nitrogen data, ORNL TM-8857, Oak Ridge National Laboratory.

-------
f
/ 2y .* /y\^\\-Wi/	J		*ys ,s>J\s^k<-r i/^.L * ~\ \ s ^nUn \
! ^ \ ^ \ 7^S.\aX ^ji-ifc4 v
tor\	tbrVA<*?o^l \
nvV>t
\s NV \ v S
\^ .is x xj*
> il'ȣV-
\	\t&y\ -\	i	
?///// TlS/ JjJ/
r^VwV
i\	\ \ \ \ \
, \ «\—\„ \ \ \
W\ \ v 1 i
"w1^- r ; I

Hal Muallar/Mantech Environmental: 95 08-23.09:47:16Wod Robinson PUKMOPOLY P UKMO T POLY

-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing,
1. REPORT NO. 2.
EPA/600/A-96/032
3. RE"
4. title and subtitle
Natural migration rates of trees: Global
terrestrial carbon cycle implications
5. REPORT DATE
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
A.M. Solomon
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
U.S. EPA, NHEERL, Corvallis, OR,
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
US EPA ENVIRONMENTAL RESEARCH LABORATORY
200 SW 35th Street
Corvallis, OR 97333
,3 mwt mutf# ? per,od covEReo
14. SPONSORING AGENCY CODE
1i#^>'!LEMf$TA§I<5£>I£s Response to Rapid Environmental Change of the Past, Present and
Future. B. Huntley, et al, eds., Springer-Verlag
16. ABSTRACT
This paper discusses the forest-ecological processes which constrain
the rate of response by forests to rapid future environmental change.
It establishes a minimum response time by natural tree populations
which invade alien landscapes and reach the status of a mature, closed
canopy forest when maximum carbon storage is realized. It considers
rare long-distance and frequent short-distance seed transport, seedling
and tree establishment, sequential tree and stand maturation, and
spread between newly established colonies. The universally applied
assumption of instant forest migration and development must be replaced
by an assumption of no forest migration during the next century.
Previously simulated increases in terrestrial carbon stocks become
declines under doubled greenhouse gas-induced warming.
17.	KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b. IDENTIFIERS/OPEN ENOEO TERMS
c. COSATl Field/Group
Forest ecological processes,
forest migration


18. DISTRIBUTION STATEMENT
19. SECURITY CLASS (This Report)
21. NO. OF PAGES
11
20. SECURITY CLASS (This page)
22. PRICE
EPA Form 2220-1 (R»v. 4-77) previous edition is obsolete

-------