Global Change Biology (1995) 1, 243-274

Commissioned Review Terrestrial higher-plant response to increasing atmospheric [C0 2 ] in relation to the global carbon cycle JEFFREY S. AMTHOR Lawrence Livermore National Laboratory, L-256, PO Box 808, Livermore CA 94550 USA

Abstract Terrestrial higher plants exchange large amounts of CO 2 with the atmosphere each year; c. 15% of the atmospheric pool of C is assimilated in terrestrial-plant photosynthesis each year, with an about equal amount returned to the atmosphere as CO 2 in plant respiration and the decomposition of soil organic matter and plant litter. Any global change in plant C metabolism can potentially affect atmospheric CO 2 content during the course of years to decades. In particular, plant responses to the presently increasing atmospheric CO 2 concentration might influence the rate of atmospheric CO 2 increase through various biotic feedbacks. Climatic changes caused by increasing atmospheric CO 2 concentration may modulate plant and ecosystem responses to CO 2 concentration. Climatic changes and increases in pollution associated with increasing atmospheric CO 2 concentration may be as significant to plant and ecosystem C balance as CO 2 concentration itself. Moreover, human activities such as deforestation and livestock grazing can have impacts on the C balance and structure of individual terrestrial ecosystems that far outweigh effects of increasing CO 2 concentration and climatic change. In short-term experiments, which in this case means on the order of 10 years or less, elevated atmospheric CO 2 concentration affects terrestrial higher plants in several ways. Elevated CO 2 can stimulate photosynthesis, but plants may acclimate and (or) adapt to a change in atmospheric CO 2 concentration. Acclimation and adaptation of photosyn­ thesis to increasing CO 2 concentration is unlikely to be complete, however. Plant water­ use efficiency is positively related to CO 2 concentration, implying the potential for more plant growth per unit of precipitation or soil moisture with increasing atmospheric CO 2 concentration. Plant respiration may be inhibited by elevated CO 2 concentration, and although a naive C balance perspective would count this as a benefit to a plant, because respiration is essential for plant growth and health, an inhibition of respiration can be detrimental. The net effect on terrestrial plants of elevated atmospheric CO 2 concentration is generally an increase in growth and C accumulation in phytomass. Published estimations, and speculations about, the magnitude of global terrestrial-plant growth responses to increasing atmospheric CO 2 concentration range from negligible to fantastic. Well-reasoned analyses point to moderate global plant responses to CO 2 concentration. Transfer of C from plants to soils is likely to increase with elevated CO 2 concentrations because of greater plant growth, but quantitative effects of those increased inputs to soils on soil C pool sizes are unknown. Whether increases in leaf-level photosynthesis and short-term plant growth stimula­ tions caused by elevated atmospheric CO 2 concentration will have, by themselves, significant long-term (tens to hundreds of years) effects on ecosystem C storage and atmospheric CO 2 concentration is a matter for speculation, not firm conclusion. Long­ term field studies of plant responses to elevated atmospheric CO 2 are needed. These will be expensive, difficult, and by definition, results will not be forthcoming for at least decades. Analyses of plants and ecosystems surrounding natural geological CO 2 Correspondence:

J.5.

Amthor; fax +1-510-422-6388, e-mail: Amthor@LLNLgov

([) 1995 Blackvvell Science Ltd.

243

244

J. S. A M THO R degassing vents may provide the best surrogates for long-term controlled experiments, and therefore the most relevant information pertaining to long-term terrestrial-plant responses to elevated CO 2 concentration, but pollutants associated with the vents are a concern in some cases, and quantitative knowledge of the history of atmospheric CO 2 concentrations near vents is limited. On the whole, terrestrial higher-plant responses to increasing atmospheric CO 2 concentration probably act as negative feedbacks on atmospheric CO 2 concentration increases, but they cannot by themselves stop the fossil-fuel-oxidation-driven increase in atmospheric CO 2 concentration. And, in the very long-term, atmospheric CO 2 concentration is controlled by atmosphere-ocean C equilibrium rather than by terrestrial plant and ecosystem responses to atmospheric CO 2 concentration. Keywords: carbon dioxide (C0 2 ), global carbon cycle, global environmental change, photosyn­ thesis, plants, respiration. Received 11 May 1995; revision accepted 20 July 1995

Introduction The atmospheric [C0 2 l (i.e. [C02 l a ) of Earth increased from c. 280 ppmv (c. 28 Pa CO 2 at sea-level) to c. 360 ppmv (c. 36 Pa CO2 at sea-level) during the past 200 years (Friedli et al. 1986; Keeling & Whorf 1994). The increase in [C0 2 l a is due mainly to fossil fuel burning and land-use change processes such as deforestation (Plass 1956; Houghton et al. 1983; Siegenthaler & Oeschger 1987). Atmospheric [C02 l continues to increase (Fig. 1) and might exceed 500 ppmv (or more) within the next 100 years. Although the present increase in [C02 l a is rapid - the annual rate of increase during the past several decades may be unprecedented in Earth's history - present [C0 2 l a is low compared with the concentrations existing during much of Earth's history (Berner 1992, 1994; Kasting 1993). The primitive (c. 4.6 X 109 years ago) atmosphere might have contained as much as 1 MPa CO 2, or c. 25 X 103 times the present atmospheric level (PAL). Atmospheric [C0 2l may have been l4-18 times PAL 450-550 X 106 years ago. About 300 X 106 years ago-during a glacial period of c. 50 X 106 years-[C0 2 la may have been as low as PAL, but by c. 200 X 106 years ago, [C0 2l a might have reached four times PAL. Later, during the last glacial maximum (LGM; c. 18 X 103 years ago), [C0 2l a was as low as 0.5 PAL. So, with a 105 year time perspective, [C0 2l a is now high (Fig. 2), but with a 108 year time perspective, [C02 l a is now low (Fig. 3). Consequences of the present rapid increase in [C0 2 l a have been speculated on at length. The main concern is that increased [C0 2 l a, and increased concentrations of other radiatively active gases and aerosols in the atmo­ sphere, will result in an enhanced global 'greenhouse effect' and global warming (Plass 1956; Manabe 1983). For example, c. 100 years ago, Arrhenius (1896) calculated that a 50'Yo increase in [C0 2 l a - from a then ambient level of c. 300 ppmv - would cause a 3-4 °C increase in Earth's

>­ E

0.. 0..

.~'

360

c 0

.~

350

c Q) u c

340

0

~~

!~ INV~!

U

0

N

0

330

,~

u .;:: Q)

.I:::

0..

320

C/)

0

E

«

310

vNIM* I

1960

I

I

I

1970

I

1980

I

I

I

1990

Year AD Fig. 1 Monthly mean atmospheric CO 2 concentration (ppmv) at Mauna Loa Observatory (Hawaii; c. 3400 m above MSL) from March 1958 through December 1994 (Keeling & Whorf 1994; Keeling et al. 1995). This is the longest continuous record of atmospheric CO 2 concentration and is a robust indicator of the monthly regional trend in atmospheric CO2 concentration in the middle troposphere. These data reflect the global atmospheric C balance on the annual time scale. The abscissa tick marks correspond to 1 January of the year indicated.

surface temperature, other factors remaining constant. Using a modern coupled atmosphere-ocean general circu­ lation model (AOGCM), Manabe & Stouffer (1994) pre­ dicted that a doubling of [C0 2 l a would increase Earth's surface temperature by c. 2.5-3.5 0c. Both Arrhenius and modern climate models predict that warming due to elevated [C0 2 l a is greater at high latitudes compared with equatorial regions. An enhanced greenhouse effect © 1995 Blackwell Science Ltd., Global Gange Biology, 1, 243-274

T ERR EST R I A L - P LAN T RES rON SET 0 AT M 0 S P HER I C [C 0

/000

400~--------------------------.

c

o .~

300

c

Q.J

()

c o

()

200

N

o

•

-r

•.¥ ~ ,.

•

.t-......

.

~

·-.

---t ­

280 ppm v

>

.~

~

0

N

()

.;:: Q.J

.s::;

()

a.

100

a. (J)

o

E ~

10 3 10 2 10 1 10 4 Years before 1995 AD Fig. 2 Approximate global atmospheric CO 2 concentration (ppm v) during the past 220,000 years (note log[lI abscissa). Pre­ 1956 AD values are from gases extracted from Antarctic ice cores (Vostok: Barnola l't 1I1.19S7; Jouzel et al. 1993; Siple Station: Neftel et al. 14H5; Friedli t'I ill. 19H6; South Pole: Friedli et III. 1984; Neftel <'I ill. 19H5; Adt;lie Land: Barnola et al. 1995). Post­ AD 1956 values are from atmospheric measurements made at Mauna Loa Obsen'atory (see Fig. 1). A reanalysis of the Vostok ice core for the period 0-160,O()() years ago yielded values similar to those shown here, but with somewhat lower CO 2 concentrations (several ppmv) during the period 10,000-50,000 years ago, about 120,000 years ago, and during the period 140,000-150,000 years ago (Barnola et ill. 1991). The dashed line at 280 ppmv indicates the often cited approximate preindustrial [C0 2L, but [C02 l a s varied by c. 10 ppmv during the period 1000-1700 AD (Barnola et al. 1995).

may also cause increased global precipitation and increased global evaporation, i.e. an intensification of the global hydrological cycle (Manabe 1983; Manabe & Stouffer 1994). Changes in regional hydrology resulting from changes in global atmospheric chemistry are largely unknown. So far, however, most predictions of warming exceed the observed global warming of c. 0.5 °C during the past 100 years (Jones et al. 1994). Several factors may contribute to this discrepancy; e.g. an oscillation in the global climatic system of period 65-70 years that presently favours cooling (Schlesinger & Ramankutty 1994), strato­ spheric ozone destruction (Toumi et al. 1994), and atmo­ spheric sulfur pollution (Taylor & Penner 1994). Indeed, when sulfate aerosols are included in an AOGCM, pre­ dicted global warming matches closely the observed warming of the past 100 years (Dickson 1995; Murphy 1995). Moreover, recent increases in temperature are well­ related to increases in [C0 2L (Thomson 1995). In any tD 1995 Blackwell Science Ltd., G/ohal Change Biology, 1, 243-274

16

0

()

N

14

a. en 0 E

cti

• •

• •.,.

•• •

8 6

0

en

E ~

a.

2

Q.J

~

.8

o 600

• ,.......

•

~

c

245

.­ • ... ,. .

.s::;

4

en

.c

E Q.J E

"E 0 Q.J () 12 "E () 0 () .~ 10 ()

o

Q.J

18 Q.J

2l

•

•

I

I

500

I

I

400

I

I

300

......

•

- - - - - - -.~

---~ I

I

200

I

I

100

I

o

10 6 years before 1995 AD Fig. 3 Mass of CO 2 in the atmosphere, relative to the present mass of atmospheric CO 2, during the past 600 X 106 years from the long-term geochemical global C cycle model of Berner (1444; estimated error is on the order of :::- 50'\,). The most reliable estimates of past atmospheric CO 2 content based on, e.g. analyses of ancient soils, sediments, and minerals, are in good agreement with these model predictions (RA Berner personal communication). The first major plant migration from aquatic environments to land might have occurred c. 400 X 1()6 years ago. The relatively low atmospheric CO 2 content [' 300 x 10" years ago coincides with a long (on the order of 50 X 106 years) glacial period. The dashed line approximates present atmospheriC CO 2 content.

case, based on physical principles and available data, it seems likely that global warming will continue as Earth's [C0 2L continues to increase, although climatic change at any location will probably differ from global mean changes. Because [C0 2l a is important to Earth's energy balance and climate, and because [C02 l a is increasing rapidly, the global C cycle has become a topic of vigorous study. Although fossil fuel burning by humans is driving the present increase in [C0 2l a , several other CO2 fluxes into and out of the atmosphere are larger in magnitude. These larger fluxes include those associated with terrestrial­ plant metabolism (Table 1). The terrestrial C cycle is central to ecosystem function and its study in relationship to the productivity and health of ecosystems is important in its own right, but recent interest in the terrestrial C cycle follows mainly from concern about increasing [C02l a and the global C cycle. It is noteworthy that only about half the CO2 released in fossil fuel burning during the past 100 years can be accounted for in the observed increase in [C02]a' indicating that large amounts of C are being stored on land and (or) in the oceans. Knowledge of the

246

J. S. AM THO R

Table 1 Selected estimates of major contemporary C fluxes from land and ocean to the atmosphere, from land to ocean, and from ocean water to ocean sediments. Uncertainties associated with many of these values are large relative to the values themselves.

Carbon exchange

Global flux a(Pg C per year)

Reference

-90 to -130

40-60

18

-38 to -56

-45

-52

---tiO -48 ---ti2 50

68

76.5

Bolin & Fung 1992

Bolin & Fung 1992

Raich & Schlesinger 1992

Box 1975

Lieth 1975

Whittaker 1975

Ajtay et al. 1979

Potter et al. 1993

Foley 1994

Raich & Schlesinger 1992

Raich & Schlesinger 1992

Raich & Potter 1995

Whittaker 1975

Crutzen & Andreae 1990

Crutzen & Andreae 1990

Schlesinger 1982

Meybeck 1993

Cole et al. 1994

SN Williams et al. 1992

Leavitt 1982

Gerlach 1991

Meybeck 1993

Amiotte Suchet & Probst 1995

Siegenthaler & Sarmiento 1993

Meybeck 1993

Terrestrial

Terrestrial-plant photosynthesis (gross primary production)b Terrestrial-plant respiration Root and mycorrhizal respiration Terrestrial-plant photosynthesis + respiration (NPP)

Litter and soil organic matter decomposition Litter and soil organic matter decomposition + root respiration Herbivory of terrestrial plants (excluding crops) Burning of terrestrial plants from human and natural causes Conversion of phytomass to charcoal in fires c Caliche formation Sedimentation on continents Lake surface = water CO2 release Terrestrial volcanosd Erupting lava Subaerial volcanos Chemical weathering of continental rocks River transport from land to ocean e Dissolved organic C Particulate organic C Dissolved inorganic C Net land-atmosphere exchange

1980-1994

1982-1992

1982-199z! 1990

1991

1992

3.5 1.8-4.7 -D.2 to -0.6 -D.02 -0.07 0.14 0.02 0.002 0.01-D.02 -0.24 -D.26 -0.8 -0.20 -0.10 -0.24 :+: 2.5 -D.3 to -2.8 -D.76 to -1.16 -2.2

Keeling et al. 1995

Francey et al. 1995

Fig. 4 herein

Ciais et al. 1995

-1.8 -1.4

Oceanic

Ocean uptake Ocean release Submarine volcanosd Sedimentation Net ocean-atmosphere exchange 1982-1992

1990

1991

1992

-92 90.6 0.D1-D.02 -D.2 -l.4g

-3.5 to 0.8 -1.0 -2.4

Siegenthaler & Sarmiento Siegenthaler & Sarmiento Gerlach 1991

Siegenthaler & Sarmiento Siegenthaler & Sarmiento Francey et al. 1995

Ciais et al. 1995

1993

1993

1993

1993

-3.1

Anthropogenic Fossil fuel burning 1950

19hO

1.62 2.54

Marland ct Ill. 1994

© 199'1 Blackwell Science Ltd., Global CllIlIlge Biology. 1, 243-274

TERR EST R I A L-PLAN T RES PONSE TO AT MOSPHE RIC [C 0

2

1

247

Table 1 (ont. Carbon exchange

1970 19HO 1991

Oil well fires lit in Kuwait in 1991

Global flux a(Pg C per year)

Reference

·+.01 5.17 6.03

0.04-0.09

Watsun et Ill. 1992

a Positive fluxes are into the atmosphere and negative fluxes are out of the atmosphere b The balance of photosynthetic CO 2 assimilation with photorespiratory CO 2 release C Charcoal has a very long half-life; its formation removes C from the C 'cycle' and places it in long-term storage d Positive values indicate transfer into the atmosphere or oceans l' Negative values indicate transfer from land to ocean. 'Dissolved inorganic C' means dissolved inorganic C coming from atmospheric and soil CO 2, rather than from carbonate rocks f For gross (one-way) terrestrial C flux of 100-120 Pg y-l g Includes 0.6 Pg C y-l flux from ocean to atmosphere to balance the estimated difference between river input to oceans and sedimentation out of oceans

(i) magnitude of fluxes of CO 2 from the atmosphere into oceans and onto land and (ii) responses of the processes underlying those fluxes to the increase in [C021a itself, are both needed to accurately predict future [C02 la's and therefore future climate. Terrestrial-plant responses to [C021a as they relate to fluxes and pools of C in terrestrial plants, soils, and the atmosphere are considered in this review. The emphasis is on feedbacks on the increase in global [C0 21a and their role in the global C cycle. Several previous reviews of terrestrial-plant and terrestrial-ecosystem responses to [C02 la can be consulted for additional background and analysis (e.g. Strain & Cure 1985; Eamus & Jan·is 1989; Jarvis 1989; Bazzaz 1990; Field ct al. 1992; Bowes 1993; Ceulemans & Mousseau 1994; Rogers ct al. 1994; Allen & Amthor 1995; Koch & Mooney 1996; Wullschleger ct al. 1995b).

Terrestrial plants in the global carbon cycle Plants and microbes in terrestrial ecosystems exchange large amounts of CO2 with the atmosphere each year (Table 1), and relative to the amount of C in the atmo­ sphere, the terrestrial biosphere stores large quantities of C (Table 2). Thus, terrestrial plants are important to [C0 2la and the global C cycle over the course of decades to centuries, and by extension, global climate. (Although more C is storcd in soils than in plants, the pathway of present C transfer from the atmosphere to the soil is almost entirely through plants.) For example, terrestrial­ plant gross primary production [CPP; photosynthesis less photorespiration (NB photorespiration is not a component of 'normal' respiration; the two are different processes)l may consume 12-17';<, of the atmospheric C pool each year (i.e. [90-130 Pg C y-l]/[760 Pg C); from Tables 1 and 2). Carbon that is assimilated by terrestrial plants is either © 1995 Blackwell Science Ltd., Global Change Biology, 1, 243-274

(i) released as CO 2 in subsequent plant respiration (ii) released as CO2 during decomposition of dead plant parts (heterotrophic respiration), or (iii) stored in peren­ nial plant tissues or in soils or in sediments in various forms for various periods of time. And, because the associated fluxes are so large, even small changes in CPP, plant respiration, or plant growth could significant! y affect the global C cycle and the rate of increase of [C0 21a . That said, and acknowledging that uptake and release of CO 2 by the terrestrial biota might affect [C02 la over the course of decades to centuries, I add that on longer time scales [C021a is determined by atmosphere-ocean C equilibrium rather than biotic activity in the terrestrial biosphere (see, e.g. Walker 1991). (In addition to their direct effects on CO2 exchange between land and the atmosphere, terrestrial plants also affect the terrestrial hydrologic cycle, land-surface sensible heat exchange, land-surface momentum exchange, and land-surface radi­ ation balance, all of which are important to atmospheric circulation and climate.) Estimates of fossil fuel burning (Keeling 1973; Marland ct al. 1994), which are thought to be accurate, combined with measurements of [C0 21a changes and the isotopic composition of atmospheric CO2 (i.e. the ratio of 13C0 2 to 12C02) can be used to estimate the amount of CO 2 taken up by (or released from) land and taken up by (or released from) oceans (Keeling ct al. 1995). Using this method, Ciais ct al. (1995) estimated that, on average, C was stored on land during the 1980s at a rate of c. 1.5 Pg per year (and see Table 1). Francey ct al. (1995) estimated that on average c. 1-2 Pg C were stored on land each year during the period 1982-1992. Our (C Rau, K Caldeira, J Amthor) work in progress (Fig. 4), however, indicates that the estimate of Francey ct al. (1995) may be too large, and Keeling ct Ill. (1995) estimated that the terrestrial biosphere net exchange of C was ± 2.5 Pg y~l during the

248

J . S. A M THO R

Table 2 Selected estimates of contemporary global C pool si7es. Soil includes litter and mineral soil, but most estimates of soil organic C are for the top I (or less) m of soil only, whereas l\epstad d al. (1994) report that stores of C below 1 m depth in an Amazonian forest exceed C stored in the top 1 m of soil and also exceed C stored in above-ground vegetation. Uncertainties in C content of all pools except the atmosphere are large with respect to the estimates themselves. Carbon pool

C Content (Pg)

Atmosphere (1994 estimate)

Plants

Ierne-strial

760

lake

Ocean

WI5 560 558 <1 1

2

(756)'

(490)'

(502)'



:\

::>

Soil organic matter (excluding li\-e roots) + litterb

litter only

litter + standing dead plants

Coarse woody debris in forests d

Peat

Animals Terrestrial (1970s estimate)

Humans (1995 estimate) Ocean Clobal, including protozoa and invertebrates

Ca liche in desert soils

Fossil (recoverable as fuel)

Ocean water f

Reactive marine sediments

Earth's crust

1636-2070

1395c

1511

1576

50

60

90

25-180

> 165

455 c

O.b 0.04

0.5

1.8 > 800

4000

37300

3000

75000000

Reference

Whittaker 1975 Ajtav ct al. 1979 Olson el al. 1983 Whittaker 1975 Olson et al. 1(1)1 Whittaker 1975 Olson elili. 19W1 Siegentha ler & Sarmiento 1993 Ajtay et al. 1979 Post et al. 1982 Sch lesinger 1991 Eswaran et al. 1993 Whittaker 1975 Ajtay et al. 1979 Ajtay et al. 1979 Harmon & Hua 1991 Ajtay ('I 111.1979 Corham 1995 Ajtay et III. 1979 JS Amthor unpublished Whittaker 1973 Bowen 1966 Schlesinger 1982 Sundquist 1993 Sundquist 1993 Sundquist 1993 Lasaga et al. 1985

, Values in parentheses account for a reduction in boreal forest phytomass based on more recent and statistically appropriate measurements (Botkin & Simpson 1990) of above-ground phvtomass in l\orth American boreal fOfl'sts b Many estimates of soil organic matter and litter exclude coarse woody debris and peat C Excludes fOfl'St litter d Only crude l'stimates of global coarse woody debris C content are available e Northern peatlands only f DissolVl'd inorganic C dissolved organic C and particulate C

period 1978-1994, rather than being a net uptake each year (d. Francey et al. 1995). The analysis of Francey et al. (1995) indicated that terrestrial uptake of CO 2 was not a regulator of [C0 21a during the period 1982-1992, but that ocean CO 2 exchange did control changes in [C0 21a . This perhaps reflects the nature of oceanic control of [C02l a on eyen short timescales (d. long-term regulation of [C02 l a by ocean-atmosphere C equilibrium [Walker 1991]), although Keeling et al. (1995) suggest that the terrestrial biosphere caused most of the slowdown in global [C02 l a increase during 1989-1993. The slowdown in [C02l a increase ended in 1993, howeyer, so if the terrestrial biosphere was regulating [C0 2l a , that regula­ tion may have been short-lived.

A complication for estimating annual net storage of C on land at the global scale is that some ecosystems may be net sources of CO 2 , e.g. disturbed tropical forests (Houghton et al. 1987; Houghton 1995) and 'warm' arctic tundra (Oechel ct al. 1993). The discrepancy in balancing known rates of CO 2 release from fossil fuel combustion (Marland et al. 1994), best estimates of net CO 2 releases from land-use changes (Houghton 1995), assumed (mod­ elled) rates of oceanic CO 2 uptake (see, e.g. Sarmiento & Sundquist 1992), and measured increases in [C0 2 l a (Keeling & Whorf 1994) is called the 'missing' sink for anthropogenic CO 2 (e.g. Sarmiento & Bender 1994). The concept of a missing sink arises, in my opinion, from an indefensible notion that ecosystems free from human 1995 Blackwell Science Ltd., Global Chollge Biology, 1, 243-274

TERRESTRIA L-PLANT RESPONSE TO ATMOSPHERIC [C0 2 l

-

l

ctI

3.0

•

~~ .~ fl!~~

(I)

>. OJ

-

Il...

2.5

(I)

~

ctI 2.0 c.. ::::l

()

Q)

1.5

c

'0

II)

1.0

+

'E

ctI c.. 0.5

(ij .0

0

a

0.0

"' ' / . \"

./ \ Y . /~.

.~

I

1982

I

I

1984

I

I

1986

'...:. I

I

1988

1990

I

1992

Year AD Fig. 4 Global plant + soil net uptake of e during the period 1982-1992 from the budget approach of Francey et al. (1995) for three different values of the gross flux of e out of (- into) the terrestrial biosphere, Cb (e, 80 Pg e y-l; ., 100 Pg e y-\ and +, 120 Pg e y-l), using a copy of the spreadsheet u5ed by Francey et al. (1995) sent to GH Rau by RJ Francey. Predicted net uptake of e onto land is significantly different across this range of CbS. Francey et al. (1995) used the value 80 Pg e y-l for each year of the analysis (personal communication to GH Rau) whereas terrestrial e flux estimates summarized in Table 1 indicate larger CbS. Also, C b might change from year to year due to regional and global variation in temperature, precipita­ tion, and other environmental factors. With Cb = 100 Pg e y-l, our copy of Francey's spreadsheet predicts a net uptake on land of 1.16 Pg e y-l for the period 1982-1992, and with Cb = 120 Pg e y-l, the value is 0.76 Pg e y-l. For these calculations we have not altered the 13e disequilibria terms of Francey et al. (1995). Francey et al. (1995) ascribe typical uncertainties of:+: 1.7 Pg e y-l to their estimates of annual net fluxes. NB We do not obtain the values shown in Francey et al. (1995)'s fig. 2b with our copy of their spreadsheet. In any case, I suggest that Francey et aI, (1995) overestimate net uptake of e by land because Cb in their analysis is too small. A reduction in e uptake on land is exactly matched by increased e uptake by oceans in this analysis.

disturbance are in an annual steady state with respect to C. This results in an underestimation of recent net C storage by undisturbed terrestrial ecosystems. The ana­ lyses of land-use change by, e.g. Houghton (1995), only apply to areas that are altered by human land uses, not the entire terrestrial biosphere. That estimates of C exchanges associated with land-lise change - which include C uptake during regrowth of previously dis­ turbed forests - do not balance modelled ocean C uptake, fossil fuel C releases, and atmospheric C accumulation is not evidence that the land-use related C releases are in 1LJLJ5 Bidckwell Science Ltd., C/alml Change Biology, 1, 243-274

249

error (or that ocean C exchange models are wrong, for that matter). The 'imbalance' might be accommodated by C storage in undisturbed ecosystems brought about by several mechanisms, most notably 'C02 fertilization' and 'N fertilization' resulting from anthropogenic air pollution (see More on missing sinks below). The few Pg C y-l apparently stored (net) in the terrest­ rial biosphere during recent years (see estimates in Table 1) is a large fraction of the C released from fossil fuels, but it is a small fraction of global terrestrial net primary production (NPP; GPP less plant respiration) and would be difficult to measure directly if all the C in question was stored in plants. Moreover, if the net C stored on land is sequestered in litter and soil organic matter, that C pool would be increasing only c. 0.1-0.2% per year. This would be impossible to measure directly in the short term (a year to a decade). It is often assumed that in the absence of human activities such as deforestation, the terrestrial biosphere would be in a steady state with respect to C. This assumption, however, 'is far from self-evident' (Hampicke 1980). Even if this were the case over the short term (decades), it is almost certainly untrue over recent millen­ nia. Between the LGM and c. 1700 AD, the terrestrial biosphere apparently accumulated large amounts of C [Bird et al. (1994) suggest 310--550 Pgl while temperature and [C02la increased. Indeed, northern ecosystems may be still recovering (storing C) from the last ice age. Undisturbed northern peatlands, for example, are appar­ ently accumulating C today, although at rates that are much less than fossil fuel CO2 emissions (Harden et al. 1992; Gorham 1995). The rate of growth of northern­ forest soil C pools, and the rate of production of elemental C (charcoal that has a very long lifetime) during fires in northern forests (see, e.g., Seiler & Crutzen 1980), are not known, but might make a significant long-term contribution to storage of C on land. In any case, oceans were probably a considerable source of C to the atmo­ sphere and to land since the LGM; perhaps c. 650 Pg C were released from oceans. Accumulation of C on land even while atmospheric [C02 la increased since the LGM reflects a lack of strong control on [C02l a by terrestrial ecosystems; the atmosphere--{)cean C equilibrium was the main regulator of [C0 2l a over the course of the last several millennia (and see Walker 1991). The amount of new land made available for terrestrial ecosystems during the last deglaciation was perhaps about equal to the amount of land inundated by ocean expansion. One assumption that follows from this is that the amount of C stored on land did not increase due to the availability of new land area for plant growth because the net change in land area was near zero. This assump­ tion is not true if a significant fraction of the biomass C on the inundated land is still stored on the continental

250

J. S. A M THO R

shelves in, e.g. sediments, or if the new land was more productive since the LGM than the inundated land was during the last glacial period. In those cases, the inundated land need not be a large C source, but the new land has clearly been a large C sink, with the net result being a large new land sink for C as a direct result of deglaciation. Though some understanding of the past and the present C balances of terrestrial plants and ecosystems exists, the question at hand is: How will terrestrial plants respond to elevated [C0 2 l a in the future? The answer to this question has implications for the net effect of human activities on [C0 2 la and global climate during the coming decades to centuries. Assessments of plant responses to increasing [C0 2la must be based on either (i) experimental treatment of plants or ecosystems with elevated [C02la or (ii) mathematical models of plant-C0 2 interactions. Observed relationships between [C0 2l a and plant pro­ cesses during the past 200 years might be used to assess plant-C0 2 interactions, but it is unknown whether the 'shape' of COz-plant relationships is the same below and above present [C02l a . Also, few quantitative measures of plant processes spanning the past 200 years are available. In any case, the focus herein is on experimental treatmen ts of plants and ecosystems with elevated [C0 2 la . Explana­ tions of experimental observations, and therefore a pre­ dictive capability with respect to future higher-[C0 2 las, come from knowledge of underlying biochemical, physio­ logical, and ecological processes and linkages.

Methods of studying plant responses to [C0 2 ]a Experimental techniques used to study plant responses to [C0 2 la were outlined by, e.g. Allen ci al. (1992). In brief, individual-leaf chambers are commonly used, often within the laboratory, to study short-term (seconds to hours) photosynthetic responses to elevated [C02 l a , but they (and branch chambers) also can be used for long­ term elevated-C0 2 treatments. Most whole-plant experi­ ments are conducted in laboratory controlled-environ­ ment chambers ('growth chambers') and glasshouses with varying degrees of control on, and monitoring, environmental conditions. Most of those studies involve potted-plants, and light levels are often low compared with conditions out-of-doors. Open-top field chambers, such as those used to study plant responses to air pollution, also are used to treat whole plants with elevated [C0 2 l a . Often, but not always, open-top field chamber experiments use plants growing in the ground rather than in pots. Portable and permanently situated closed­ top chambers -'small greenhouses'- also can be used to treat plants growing in the ground with elevated [COzla. During the past few years, free-air CO2 enrichment methods were used to elevate [COzla across large areas

(e.g. 500 mZ) of plants and soil not encumbered in chambers or other structures. As with open-top chambers, the use of free-air CO 2 enrichment techniques follow methods used earlier to study effects of air pollutants on plants and ecosystems (McLeod 1993). Each of the methods now in use (e.g. leaf cuvettes, branch chambers, growth chambers, glasshouses, open­ top field chambers, and free-air CO2 enrichment facilities) has advantages and disadvantages compared with the other techniques. That is to say, no single present method is best for addressing all types of research related to plant and ecosystem responses to elevated [C0 2]a' Indeed, a combination of methods is often desirable. For example, long-term free-air CO 2 fumigation experiments can include short-term measurements of leaf photosynthesis and respiration using leaf cuvettes and short-term meas­ urements of canopy CO 2 exchange rate using portable closed-top chambers (e.g. Hileman clal. 1994). In general, it is important to focus attention on studies in the field with unchambered plants and ecosystems because 'chamber effects' can be large relative to effects of elev­ ated-C0 2 on plants (e.g. Owensby et al. 1993a). Chamber experiments are, nonetheless, important in the study of mechanisms underlying plant responses to elevated [C0 2 la , and in any case, most data are from chamber experiments. Plants might acclimate and (or) adapt to changes in [C02 l a . Acclimation is a phenotypic adjustment to a short­ term, e.g. seasonal, change in the environment whereas adaptation is a genotypic adjustment to a change in prevailing environmental conditions, e.g. a millennial­ scale increase in [C02 l a or enhanced greenhouse warm­ ing. Acclimation may not become apparent for days to weeks or even months, so very short experiments may miss important acclimatory responses. Adaptation is a (very) long-term process, so experiments with a duration less than many plant life cycles will be unsuccessful in assessing adaptation to elevated rC0 2 la . On the other hand, concern for effects of elevated [C02 la during the next several decades to a century lessens somewhat the need for immediate knowledge of potential adaptation because adaptation will not occur in nature for many long-lived plants for some time. A related experimental problem is that [C0 2 l a is increasing gradually (c. 1.21 ppmv per year during the period 1958-1994; see Fig. 1) whereas experimental treatments involve large and instantaneous increases in [C0 2la. It is unknown whether plants respond differently to the continuing gradual increase in [C02 la compared with the rapid CO 2 increase used in experiments. Questions of acclimation and, especially, adaptation to elevated [COzL raise the issue of truly long-term elevated­ [C0 2l a treatments. A 25-year experiment started today, for example, obviously would not be completed for © 1995 Blackwell Science Ltd., Global Change Biology, 1, 243-274

T ERR EST R TAL - P LAN T RES P 0 N SET 0 AT M 0 S P HER TC [C 0 25 years. The need exists therefore for elevated-[C02 la treatments that were started long ago. Although far from ideal 'experiments,' natural geological CO2 springs or degassing vents have in some cases exposed terrestrial ecosystems to elevated [C021a for centuries or even millennia. The main benefit of natural CO 2 springs as research tools is the long history of elevated [C02 la in an otherwise (mostly) natural environment (Miglietta et al. 1993b). This provides the potential to study (i) plants acclimated and adapted to elevated [C0 21a (ii) interspecies competitive outcomes within elevated [C0 2 la, and (iii) changes in ecosystem biogeochemistry in response to elevated [C02 1a (Koch 1993). Drawbacks of natural CO 2 springs are the possibility of large fluctuations in [C0 2la on many time scales, and the lack of quantitative know­ ledge of past [C0 2 l a s. Nonetheless, natural CO 2 springs have the potential to provide information concerning long-term plant and ecosystem responses to elevated [C0 2L that cannot be obtained in any other way. Although natural CO2 springs are imperfect as experi­ mental tools, and the 'results' from long-term exposures of plants to elevated [C021a s near CO 2 springs are only beginning to be analysed, Nature has nonetheless 'spoken' about terrestrial-plant response to long-term elevated [C0 21a . I visited several natural CO 2 springs, and 1 collected first-hand accounts concerning plant and ecosystem processes and states around other CO 2 springs, in an attempt to understand what it is that Nature has 'said' about modern-plant responses to long-term CO 2 enrichment. In all cases with which I am familiar, plant communities surrounding CO2 springs are not signific­ antly more productive than communities more distant from CO2 springs (the 'controls'). Elevated [C02la - more than twice present [C02 1a in many cases - did not result in 'super vegetation' forming dense jungles of phytomass (see, e.g., Korner & Miglietta 1994). Carbon dioxide springs can enhance tree and herb growth under some circumstances, as indicated by comparison of tree ring widths and plant sizes near to and distant from CO 2 springs (e.g. Miglietta et al. 1993a; F Miglietta, C Korner & S Hattenschwiler, personal communication), and elevated [C0 2 L might enhance soil C pool sizes (this has yet to be quantified near CO 2 springs), but clearly, a long-term doubling (or more) of [C0 2 L near CO 2 springs does not cause a doubling of general plant size. Indeed, long-term growth stimulations near CO 2 springs may be modest in comparison with growth responses obtained in short­ term (~ 10 years) experiments using chambers or free­ air CO2 fumigation methods. But, even modest changes in global terrestrial NPP might be equivalent to significant fractions of C released during fossil fuel burning (Table 1). In any case, studies of plants near natural CO 2 springs have important implications for knowledge of overall long-term terrestrial-plant responses to elevated [C0 21a s. © 1995 Blackwell Science Ltd., Global Change Biology, 1, 243-274

2

1

251

The corollary is that not too much should be made of short-term elevated-C0 2 experiments-especially those involving chambers and (or) young plants-with respect to rates of and amounts of long-term ecosystem-level C flux and C storage, respectively.

Photosynthesis (C0 2 assimilation) Plants assimilate atmospheric CO2 in the process of photosynthesis, which uses solar radiation in the c. 400­ 700 nm wave band as its source of energy (McCree 1981). Terrestrial higher plants can be divided into three broad categories with respect to their biochemical pathways of photosynthesis: (i) C 3, (ii) C~, and (iii) crassulacean acid metabolism (CAM) (see, e.g., Edwards & Walker 1983; Ting 1985). All three types of plant use the C 3 pathway of photosynthesis - which forms two molecules of glycer­ ate-3-phosphate from one molecule each of ribulose-l,5­ bisphosphate and CO 2 - but C 4 and CAM plants use additional, preliminary CO2 assimilation reactions that concentrate CO 2 inside plants. In general, C 3, C 4, and CAM plants are adapted for different environments (Osmond ct al. 1982; Ehleringer & Monson 1993). Plant species numbers are dominated by C1 plants - on the order of 95°,.{, of higher-plant species are C 3 - and C 1 plants are generally more sensitive to [C02 1a than are C 4 .and CAM plants (Bowes 1993). Nonetheless, C 4 plants make a significant contribution to photosynthesis in grasslands (and a few other ecosystems), and grasslands are globally widespread and important to the global C cycle (Ajtay et Ill. 1979) so C 4 photosynthesis is more significant to the contemporary global C cycle than is indicated by species counts. The enzyme ribulose-l,5-bisphosphate carboxylase­ oxygenase (rubisco) is central to C 3 photosynthesis. It is a large, sluggish enzyme with a low affinity for CO 2 (Miziorko & Lorimer 1983; Bowes 1991). It is generally believed that rubisco is the most abundant enzyme on Earth, at least in terms of mass (Ellis 1979), and it is an important global biospheric pool of N. More importantly, the carboxylase function of rubisco catalyses the carb­ oxylation reaction of the photosynthetic reductive pentose phosphate (PRPP) cycle, globally involving more than 100 Pg C each year. In the present atmosphere, perhaps 10-30'X, of the C assimilated in photosynthesiS by C 3 plants is almost immediately released as CO 2 in the process of photores­ piration 1, which involves the oxygenase activity of rubi­ sco. An increase in [C021a has the potential to stimulate net CO 2 assimilation (photosynthesis less photorespira­ tion) in C 3 plants because (i) the Michaelis constant of rubisco for CO 2, KM (C02 ), is high relative to present [C0 2l a (ii) CO2 competitively inhibits photorespiration, and (iii) CO 2 is required for the activation of rubisco

252

J. S. A M THO R

(C02 molecules used in rubisco activation are not the CO 2 molecules assimilated during photosynthesis by that activated rubisco) (Sage & Reid 1994). The first two points, (i) and (ii), may be the most significant with respect to a stimulation of global photosynthesis by elevated [C0 2la in C 3 plants. But note that the response of CTplant CO2 assimilation rate to [C0 2 la is concave down, i.e. the first derivative of the photosynthetic rate decreases as [C0 21a increases. Many factors such as temperature, light, and water supply, affect the response of leaf and canopy photosyn­ thesis to [C0 2 la (e.g. Radin et al. 1987). Based on the biochemical kinetics of rubisco, an increase in [C0 21a from 360 to 720 ppmv might be expected to enhance CO2 uptake by C 3 plants on the order 10-100°/r" with a small relative response at low temperature and a large relative response at high temperature (Long 1991; Kirschbaum 1994). The interaction with temperature is related to the positive relationship between r. and temperature (see footnote 1) and has important implications for global photosynthesis with concomitant global increases in [C0 21a and temperature. Predictions of photosynthetic stimulation by elevated [C0 2 la are consistent with the few canopy-level Car exchange measurements made to date (e.g. Drake & Leadley 1991; Pinter et al. 1995). Although elevated [C0 21a stimulates 2 photosynthesis under many circumstances, the fate of the 'extra' C taken up by plants in elevated [C02 1a is not obvious. The extra C could be added to plant structure in additional growth, or respired by the plant, or added to soil C pools via lThe ratio of photorespiration (P R) to photosynthesis (P) can be estimated from PI{ / P = r, / [C021e where r, is the [C0 21 in chloroplasts at which photosynthetic carboxylation matches photorespiratory decarboxylation and [C0 21e is the [C021inside chloroplasts. (In this notation, CO 2 concentrations in chloroplasts are the equilibrium gas phase concentrations.) The value of r, depends on temperature. Based on Jordan & Ogren (1984), r, is c. 20 ppmv CO2 at 5°C and c. 54 ppmv CO 2 at 30°C (see also Woodrow & Berry 1988). At present [C0 21a , [C021e may be in the range 180-270 ppm CO 2 during much of the day for many C 3 plants (see, e.g., Caemmerer & Evans 1991; Lloyd et al. 1992; Loreto ct al. 1992). Thus, the ratio P R / P may normally be in the range 0.10-0.30 at present [C0 21a in C 3 plants. Due to their COz-concentrating processes, C 4 and CAM plants have low rates of photorespiration at present [C0 2L. 2Contrary to a positive response of leaf CO2 assimilation rate to increased [C0 2L, there are several well documented cases of short-term inhibition of photosynthesis caused by 'supraoptimal' CO2 levels (e.g., Woo & Wong 1983 and other references in Harley & Sharkey 1991). Such an inhibition of photosynthesis caused by CO2 may occur at [C0 2la's only slightly greater than present global [C0 2la. It is possible that this response to elevated CO 2 is related to reduced recycling of C in the photorespiratory cycle, i.e., a lack of glycerate entry into chloroplasts (Harley & Sharkey 1991). The significance of such a phenomenon to the global C cycle has not been assessed.

enhanced exudation from roots or accelerated root growth and death, i.e. root turnover. Some combination of these fates may be the most likely outcome, although the fate of extra assimilated C may change over time in response to acclimation, adaptation, and feedback processes. And, the photosynthetic response to elevated [C0 21a and the assimilation of extra C may itself change over time, owing to acclimation of photosynthesis to elevated [C0 21a .

Photosynthetic acclimation to elevated [C02]a Long-term (weeks to years) Car enrichment often, but not always, results in reduced amounts of photosynthetic pigments and enzymes such as rubisco in leaves of C 3 plants (Bowes 1991, 1993; Rowland-Bamford et al. 1991; Van Oosten et al. 1992; Webber et al. 1994; Wilkins et al. 1994; Jacob et al. 1995). A downregulation (acclimation) of photosynthetic capacity in response to long-term elevated [C0 21a can limit the photosynthetic response of C 3 plants to elevated [C0 2 1a (Sage et al. 1989; Gunderson & Wullsch­ leger 1994; Sage 1994). Downregulation of photosynthesis has important positive implications for resource-use effi­ ciency by plants; that is to say, downregulation may have more positive impacts on plant N balance and growth than it has a negative effect on photosynthetic CO 2 assimilation (Sage 1994). Plants grown in small pots or with low nutrient availability, i.e. plants having significant limitations to growth other than photosynthetic C assim­ ilation, may show the greatest degree of photosynthetic downregulation when grown in elevated [C021a (Arp 1991; Hogan et al. 1991; Thomas & Strain 1991; Sage 1994). This is a potential artifact of some studies using potted plants and is probably related to the balance between C source activity and C sink activity, although results to date are somewhat contradictory (Pettersson & McDonald 1994). Even plants growing in the ground, however, may generally show some degree of photosyn­ thetic downregulation in response to long-term elevated [C0 21a (Webber et al. 1994; Jacob et al. 1995). It is significant that recent biochemical studies show possible photosyn­ thetic acclimation in plants grown in elevated [C021a but for which previous reports suggested a lack of photosynthetic downregulation. For example, leaf Car exchange measurements indicated a lack of photosyn­ thetic acclimation in Citrus aurantium trees grown in elevated compared with ambient [C0 2 la (Idso & Kimball 1991), but later biochemical studies revealed a decrease in leaf soluble protein levels - an indication of downregul­ ation - in those same trees (Webber et al. 1994; and see Jacob et al. [19951 for a similar chain of events with Scirpus olneyi elevated-[C0 21a research). And, leaves of Quercus trees growing near natural CO 2 springs, which are pre­ sumably adapted to a higher-[C02 1a environment, may also downregulate photosynthetic capacity compared © 1995 Blackwell Science Ltd., Global Change Biology, 1, 243-274

T ERR EST R I A L - P L i\ N T RES P 0 N SET 0 AT M 0 S P HER I C [C 0 21 with leaves on trees growing more distant from CO 2 springs (CL Hinkson, WC Oechel, F Miglietta & A Raschi personal communication). The downregulation of Quercus-leaf photosynthesis near CO 2 springs is observed in old leaves rather than in young leaves. In some cases (Curtis & Teeri 1992), elevated [C0 2 la may increase leaf longevity, which will in part mitigate effects of photosynthetic downregulation on seasonal CO 2 assimilation in elevated [C0 2 l a , but in other cases (Pinter et al. 1996) elevated [C0 2 la may reduce leaf life span. And, an exception to photosynthetic downregul­ ation may exist in Nrfixing plants. For example, rubisco capacity was apparently incrcased in leaves of black alder (Alnus glutino~a) plants grown in elevated compared with present ambient [C0 2l a (Vogel & Curtis 1995). Photosynthetic acclimation to elevated [C0 2 la might be temperature dependent, but this has been rarely studied. In Medicago sativa leaves, downregulation of photosynthesis in elevated [C0 2 la was observed at 30°C but not at lower temperatures, whereas in Daclylus glomer­ ala leaves, photosynthetic acclimation was apparent at 15 and 20°C but not at 25 or 3() °C (Ziska & Bunce 1994a). This is an important aspect of plant response to combined increases in global [C0 2 la and temperature that needs further study. A common result of exposure of plants to elevated [C0 2 la is increased accumulation of nonstructural carbo­ hydrates in leaves (Wong 1990; Hendrix et al. 1994; Korner & Miglietta 1994), which results from enhanced photosynthesis. The accumulation of nonstructural carbo­ hydrates, and in particular soluble sugars, may be the trigger for acclimation of photosynthesis in response to elevated [C0 2la . Sugars, e.g. glucose and sucrose, can repress photosynthetic gene expression and lead to reduc­ tions in the amount of photosynthetic pigments and PRPP cycle enzymes, including rubisco (Sheen 1990,1994; Stitt et al. 1990; Krapp cl al. 1991, 1993; Cheng et al. 1992; Schafer et al. 1992; Harter 1'1 al. 1993; Van Oosten el al. 1994; Webber el al. 1994). Repression of photosynthetic genes by sugars (i.e. photosynthetic end-product repres­ sion) may underlie photosynthetic acclimation to elevated [C0 2la, and probably plays a role in maintaining a balance between carbohydrate source and sink activities (Sage 1994; Webber et al. 1994). Repression of photosynthetic genes by sugars might be most extensive when N supply is limited, in part as a mechanism of increasing carbohyd­ rate sink activity by mobilizing N in mature leaves and making it available to growing organs (Paul & Stitt 1993; Webber ct al. 1994). Two of the most important questions relating leaf physiology to the global C cycle are: Will enhanced photosynthesis in a future, higher-[C0 2 la world lead to downregulation of photosynthesis at the ecosystem, regional, and global scales? If so, to what extent? General © 1995 Blackwell Science Ltd., Global Chal1ge Biology, 1, 243-274

253

answers to these questions, i.e. answers that can be applied globally, may be slow in coming. In the meantime, the weight of the evidence indicates that photosynthetic acclimation to elevated [C0 2 la is common, but that the acclimation is rarely complete.

Feedback inhibition of photosynthesis In addition to long-term acclimatory responses to an accumulation of nonstructural carbohydrates in leaves, i.e. downregulation of photosynthesis, an increase in the size of leaf nonstructural carbohydrate pools mav result in short-term end-product feedback inhibition of photo­ synthesis (Azc6n-Bieto 1983). This end-product inhibition of photosynthesis may be related to inorganic phosphate cycling in leaves (Herold 1980; Foyer 1988). It may be most important in the short term, e.g. during the course of a single day, as compared with seasonal photosynthesis. Thus, both short-term and long-term negative feedbacks on photosynthesis from leaf nonstructural carbohydrate pools can exist in an elevated-[COzl environment. So although the stimulation of photosynthesis caused by elevated [C0 2 l a is a negative feedback on the increase in global [C02 la , there are also negative feedbacks on the response of photosynthesis to [C0 2 l a . The magnitudes of each of these feedbacks are unknown, although present annual CO 2 emissions associated with fossil fuel use are equivalent to only c. 5°It, of the annual balance of terrestrial higher-plant photosynthesis and photorespiration, i.e. GPP, so small changes in global photosynthesis might have large effects on fossil-fuel-driven [C0 2la increases. But, many long-term and multi-life-cycle experiments must be cond ucted in order to sort out likely higher­ plant photosynthetic responses to increasing [C0 2le over the decadal to century time scale, and again, over the very long-term, ocean-atmosphere C equilibrium will control [C0 2 la .

Growth One potential result of stimulated photosynthesis is enhanced plant growth. That is, 'additional' photo­ synthate can be used in the biosynthesis of 'additional' plant structural dry matter (SCI1SU Penning de Vries cl al. 1979) such as membranes, enzymes, and cell walls. A change in plant growth or C accumulation over a year or 'season' is in many ways the most meaningful measure of plant response to elevated [C0 2la . It is a more signific­ ant assessment of the effect of plants on [C0 2la than measurements of either photosynthesis or respiration. Effects of elevated [C0 2la on plant growth have been studied, and reviewed, frequently (Lawlor & Mitchell 1991; Ceulcmans & Mousseau 1994; Rogers ct al. 1994). In general, growth of C 3 plants is positively related to

254

J. S. A M THO R

[C0 2 l a . Growth of C 4 plants can also be enhanced by elevated [C0 2l a , but the magnitude of response is gener­ ally less than that observed with C 3 species. The differen­ tial growth response of C 3 compared with C 4 plants is presumably related to the stronger response of C 3 photosynthesis to elevated [C0 2 1a . Forest-tree species mean response to a doubling of [C0 2 la was a 32% increase in dry mass accumulation (Wullschleger el al. 1995b). (Forest growth, and any changes in forest growth due to increasing [C0 2la, are especially important to the global C cycle; Ajtay ct Ill. [19791 suggest that as much as 40'Yo of global terrestrial NPP occurs in forests.) Whereas ele\'ated [C0 2 la has the potential to enhance forest tree growth, it often has been suggested that when N or water are limited, plant growth will not be significantly stimulated by elevated [C0 2 1a (e.g. Kramer 1981; who is frequently cited in support of this view). Available data, however, indicate that the reilltive increase in tree growth brought about by elevated [C02 1" is about the same with and without N deficiency and with and without water limitation (Wullschleger el Ill. 1995b). But, available data are all of a short-term nature, i.e. from experiments lasting a few years at most. Long-term enhancement of forest tree growth with N deficiency is only possible if plant N / C ratio decreases, which may in fact occur in eleyated [C0 2 L (Field ct Ill. 1992). Unfortunately, almost all studies with forest tree species are limited to seedlings or saplings growing in chambers, often growing in pots, rather than large trees growing naturally for long periods of time. It is not known whether seedling/ sapling responses to elevated [C0 21a reflect the response of large trees to [C0 2la. Moreover, it has been suggested that plants grown in pots have a limited growth response to elevated [C0 2la beclluse they are grown in pots. To the extent that this is true, pot studies underestimate growth response to elevated [C0 2 la by plants in the field. In at least one case (one of only two direct studies of pot-size effects on growth response to elevated [C0 2l a 7), however, the relalive stimulation of growth by elevated [C0 2 1a was greater in 'small' com­ pared with 'large' pots (Thomas & Strain 1991). But, in a comparison of growth in 270 vs. 350 ppmv CO 2 , growth of plants in large pots responded positively to [C0 2 1a whereas growth of plants in small pots did not (Thomas & Strain 1991). The (only?) other direct study of pot-size effects on growth responses to elevated [C0 2 1a indicated that pot size per sc need not influence growth in ele\'ated [C0 21a and that soil nutrient concentrations may be more important than pot size in affecting growth response to [C0 21a (McConnaughay el Ill. 1993b; Berntson et al. 1993). The issue of pot size is far from resolved, however. Both studies addressing it were of short duration (4 weeks and 10 weeks, respectively). Both studies used what I would call small and very small pots, not small and large

pots. Thus, the long term effects of truly large (field­ like) and small rooting volumes on growth responses to elevated [C0 2 l a have not been addressed. As previously noted, however, plants in nature 'do not have unlimited below-ground resources with which to maximize growth in a CO 2-rich world' (McConnaughay el al. 1993a), although they commonly have access to at least more volume than they do in most pots used for research. And, the assertion that small pots l1eed /Jot limit growth responses to elevated [C0 2L is not the same assertion as small pots do nol limit growth responses. There is special concern for the effects of increasing [C02 1a on the growth of crops. Both the potential stimula­ tion of crop growth by elevated [C0 2 la and the effects of COrrelated climatic change on crop productivity are of interest. Most data indicate that crop yield is enhanced by elevated [C0 2 t" but only a relatively few experiments were conducted under field conditions. Economic yield in C 3 crop species responds to a doubling of [C0 21a with on average an c. 33% increase - remarkably similar to the relative stimulation of dry mass accumulation by forest tree species in doubled [C0 2 la and C 4 crop species respond with an c. 100!,, increase in economic yield (Rogers et Ill. 1994; see also Lawlor & Mitchell 1991; Pinter et al. 1996). But, I suggest that effects of future increases in [C0 2 1a will have only a minor impact on overall agricul­ tural productivity. The reason for this is that develop­ ments in technology and management have resulted in large increases in productivity over time and large increases due to technology (which includes breeding) might be anticipated in the future, whereas by compar­ ison, effects of increasing [C021a on crop growth are minor. An estimate of the relative effect of [C0 2 la com­ pared with technology on wheat yield can be obtained by comparing wheat growth in different [C0 2l as with historic changes in wheat yield. Using data presented by Polley el al. (1993), I calculated that wheat (cv. Yaqui 54) above-ground dry mass at maturity was increased c. 44% by growth in 350 compared with 2HO ppmv CO 2 in COT depletion experiments. By contrast, using U.K. mean wheat yield data for the period AD 1200-1993 summarized in Amthor & Loomis (1996, especially our fig. 2), I estimate that wheat yield has increased c. 8Q(l'Yo (!) as [C0 21of the actual atmosphere increased from 280 to 350 ppmv during the past c. 300 years. Although this analysis is not overly rigorous (e.g. Polley et al. [19931 reported above-ground biomass and we summarized grain yield), it indicates that only c. 5% of the increase in U.K. wheat yield that has occurred during recent centuries can be attributed to increasing [C02 1a . That is to say, increases in crop productivity are driven mostly by technology and management (see, e.g. Evans 1993) rather than by environmental changes, and effects of continued increases in [C02 1a may have small effects on productivity com­ rD 1995 Blackwell Science Ltd., Global Chilllge Biology, 1, 243-274

T ERR EST R TAL - P LAN T RES P 0 N SET 0 AT M 0 S P HER [C [C O 2 ] pared with continued improvements in management and breeding. For example, according to Evans (1993 [po 307]), 'there is no indication that the genetic yield potential of any of the major crops is reaching its limit. Indeed, for wheat [Triticum aestivlIm], barley [HordclIlIl vulgare] and maize [Zca mays] in some environments, the rate of improvement is accelerating.' At the same time, the relative crop growth response to increasing [C0 2la will diminish as [C0 2Ls increase, i.e. the relationship between crop growth and [C02 ]a is concave down. A similar analysis can be carried out for rice (Orl/2a sativa) using growth, yield, and [C0 2 L data in Baker ct al. (1992) and Amthor &. Loomis (1996). Moreover, from the perspective of the global C cycle, much of the C assimilated in crops is oxidized and released as CO 2 within a year of assimilation during feeding of humans and livestock (this is not, however, true for lint harvested from cotton [COsslfpilim hirsutum]). And, crops probably contain less than 1% of the C in terrestrial plants (Ajtay ct al. 1979), so even a doubling of crop phytomass would have a negligible impact on the global plant-C pool size. Agricultural-soil C pools may be enhanced by elevated [C0 2 ]a due to increased organic-C inputs from roots and stubble, but conversion of land to agriculture from more natural vegetation in the first place generally results in a loss of soil C (Houghton 1995). On the whole, increased crop productivity due to increasing [C02 ]a is likely to have no more than minor effects on the global C cycle; changes in land use associated with changes in geograph­ ical patterns of agriculture due to continued increasing human population and to climatic changes will probably have more significance effects on global terrestrial C pools and fluxes than will changes in crop productivity. A simple way to summarize the effects of [C02 la on plant growth, i.e. NPP, is the empirical coefficient ~ used as follows (e.g. Gifford 1980; Goudriaan 1993) NPP a = NPP o [1 + ~ In (Ca/C II )] NPPo ?e 0; Ca , Co > 0,

(1)

where NPP" is NPP at a given atmospheric CO2 concentra­ tion Ca , NPPo is NPP at a reference or baseline atmo­ spheric CO 2 concentration Co, and In (x) is the natural logarithm of x. Note that ~ applies to the primary effects of a change in [C02L on plant growth, not any secondary effects mediated through climatic change. This ~, and similar parameters, is called the 'biota growth factor' and the 'biotic growth factor' (Bacastow & Keeling 1973; Gifford 1980). The logarithmic response of NPP to [C0 2 L used in (1) was questioned by Gates (1985), who sug­ gested that ~ might be better calculated with a Michaelis­ Menten expression, but Cifford (1980) and Coudriaan ct al. (1985) adequately defended the use of (1), at least for moderate (a few hundred ppmv) changes in atmospheric [C02 ]•. © 1995 Blackwell Science Ltd., Global Challge Biology, 1, 243-274

255

A ~ value [sensii (1)] of unity represents a 69% increase in NPP with a doubling of [C0 2 ]a whereas ~ is c. 0.43 if doubled [C0 2 ]a increases plant growth by 30% (d. summaries for forest tree seedlings and crops outlined above). Several values of ~ were used in assessing the effects of increasing [C0 2 ]a on the global C balance with simple models of the terrestrial biosphere (summarized in Wullschleger et al. 1995b). For example, Rotmans & den Elzen (1993) were able to 'balance' the global C budget with ~ = 0.4 in their model when effects of temperature on NPP and heterotrophic respiration (decomposition) were also considered. Other typical values of ~ used in global C cycle models are in the range 0.2-0.5 (Wullschleger et al. 1995b), which are consist­ ent with experimental data summarized in Amthor & Koch (1996) and Wullschleger et al. (1995b). In a recent survey of the literature (and some unpublished datasets), however, we could find no measurements of NPP in intact, non-crop ecosystems treated with elevated [C0 2]a; above-ground NPP was estimated a few times, and whole­ ecosystem CO 2 exchange (including decomposition) was measured in a few cases, but effects of elevated [C0 2 ]a on NPP in natural ecosystems have not been rigorously studied (Amthor & Koch 1996). Thus, we are mostly in the dark with respect to knowledge of NPP responses by natural ecosystems to increasing [C0 2L. To put into perspective any plant growth (NPP) stimu­ lation that might occur with increasing [C0 2 ]" a 10% increase in present global terrestrial NPP is about equiva­ lent to present fossil fuel C emissions (though opposite in 'sign'-see Table 1). This means that a 10(10 increase in present global terrestrial NPP would about balance pre­ sent fossil fuel C emissions. But, could the C associated with extra NPP be stored in the biosphere over the long term, or would it be returned to the atmosphere in short order during subsequent litter production and decomposition? Some of the C associated with increased NPP resulting from increased [C0 2L would probably remain in the biosphere, but not all of it. The nature of the terrestrial C cycle is such that NPP does not represent permanent C storage; only a small fraction of the C in terrestrial-plant NPP subsequently enters C pools with lifetimes of 1000 or more years. Most of the C in NPP is much more labile. The negative feedback on the increase of [C0 2 L medi­ ated through enhanced photosynthesis (CPP) and plant growth (NPP) cannot, by itself, reverse the upward trend in [C02 ]a because increased CPP and NPP depend on increasing [C02 ]a. It can, however, slow it. Increased area of land covered by vegetation, or major changes in vegetation type brought about by, e.g. large-scale affor­ estation, could halt or even reverse the increase in [C0 2 la through significant increases in C storage on land if fossil C releases are significantly reduced (see also Sarmiento

256

J. S. A M THO R

et al. 1995), but this is unlikely for at least many decades, and if oceans do not then become significant sources of atmospheric CO2 , Most estimates are that the oceans are now sinks for atmospheric CO 2 (Sarmiento & Bender 1994) so they are absorbing fossil C and limiting the increase in [C0 2 L,.

An extreme view A minority view (Idso 1991a) was that the 'real-world ~ factor may well be three to four or possibly even five times greater than what has heretofore been believed.' Idso (1991a) claimed that large values for ~ can be found from 'all the available physical and biological evidence,' but in point of fact, the lion's share of the relevant data were summarized in Wullschleger et al. (1995b) and Amthor & Koch (1996) and these summaries indicate that real-world ~ is probably about 'what has heretofore been believed [i.e. ~ = 0.3-0.7].' Idso (1991b) also claimed that 'increases in atmospheric CO 2 result in growth rate increases of trees five times greater than growth rate increases of nonwoody plants' (italics added), yet data from hundreds of elevated-C0 2 experiments summarized in, e.g. Lawlor & Mitchell (1991), Ceulemans & Mousseau (1994), Rogers et al. (1994), and Wullschleger et al. (1995b), make it impossible to objectively conclude that trees generally respond much more positively to elevated [C02 ]a than do herbaceous plants. While I reserve judge­ ment on the ratio of tree to herbaceous plant growth responses to elevated [C0 2 ]a in nature, the data now available to Idso and to me indicate that herbs and trees differ by much less than a factor of five, if they differ at all, in their growth responses to elevated [C0 2 ]a' Kohlma­ ier et al. (1989, 1991) clearly outlined other difficulties with Idso's inflated ~ that do not need to be repeated here. Some of Idso's notions of extreme positive tree growth responses to elevated [C0 2 ]a come from his relatively long-term experiments with Citrus allrantillm growing in the ground in open-top chambers (e.g. Idso & Kimball 1993). In that experiment, photosynthesis and growth were greatly stimulated by elevated [C0 2 ]a' Is the basis for such a large positive response to [C02 ]a by Citrus known? Maybe. The mesophyll or internal leaf conduct­ ance to CO2 transfer in C. aurantium leaves can be very low compared with other plants (Loreto et al. 1992) and this tends to result in low CO 2 concentrations inside chloroplasts (see also Lloyd et al. [1992] with respect to C. paradisi and C. limon). The outcome is that the CO2 concentration 'seen' by rubisco is very low in Citrus leaves, and this leads to a relatively strong response to an incremental increase in the supply of CO2 because the photosynthesis vs. [C02 ] function is generally increasing, but is concave down. Thus, C. aurantium should respond strongly-in relative terms-to elevated [C0 2 ]a' at least

with respect to photosynthesis, and it 'is probably not a good plant from which to generalize about CO 2 responses given that its [mesophyll conductance] can be so low' (Loreto et al. 1992), but that is just what Idso (1991b) did. Although the C. allrantillm experiment of Idso & Kim­ ball (1993) was 'designed to overcome the problems that have plagued most prior tree experiments,' it unfortu­ nately succumbed to what are the most serious problems with such experiments: the plants were (i) grown outside their native habitat (Phoenix, Arizona is not by nature a forest of Citrus) so its applicability to natural ecosystems is possibly lost, (ii) flood-irrigated instead of being sub­ jected to natural precipitation, and (iii) fertilized instead of allowed to grow on naturally available soil nutrients. That is to say, the plants were grown as a horticultural crop rather than as natural vegetation. The experimental results, however, were extrapolated to "earth's 'global forest'" (Idso & Kimball 1993; p. 551), although the experiment is unlike any of Earth's forests. And, as mentioned above, Citrus is expected to respond more strongly to elevated [C02 la than do Earth's forests, especi­ ally when Citrus is grown in the warm (often hot) conditions of a plastic chamber in Phoenix, Arizona. In a more realistic experiment - with a forest tree species growing in a forest soil- elevated [C0 2 ]a compared with present ambient [C0 2 ]a did not significantly enhance tree dry mass accumulation over the course of three years (Norby et al. 1992). In another case of exaggeration, Idso & Kimball (1993) wrote that, with respect to their single experiment with irrigated and fertilized C. allrantillm and one analysis of the seasonal cycle of [C0 2 la at Mauna Loa by Pearman & Hyson (1981; d. Kohlmaier et al. 1989), 'it would appear to make it almost impossible for one to derive any other conclusion than that which looms so obvious from this comparison, namely, that all of Earth's trees, in the mean, likely respond to atmospheric CO2 enrichment to the same phenomenal degree [2.8 times as much growth in 655 compared with 355 ppmv CO2 ] that sour orange trees [did in our single experiment], (italics added). Not only is it possible, but it is necessary to come to a much different conclusion. Based on available data, forest tree species respond to atmospheric CO2 with a ~ of perhaps 0.3-0.7, or c. 1.2-1.5 times as much growth at 700 com­ pared with 350 ppmv CO 2 (Amthor & Koch 1996). That is what the data say. Although, as stated above, only a poor understanding of the effects of increasing [C0 2 ]a on NPP in the 'real world' exists, this is not justification to disregard the facts that are at hand in favour of wild claims of 'phenomenal' growth responses by all of Earth's forests. The honest scientific view must be more guarded. For as Idso (1991c) wrote, 'when faced with a conflict between someone's theory and many other people's measurements, it is usually wisest to go with the measure­ © 1995 Blackwell Science Ltd., Global Change Biology, 1, 243-274

T ERR EST R I A L - P LAN T RES P 0 N SET 0 AT M 0 S P HER I C [C 0 ments.' Idso's (1991b) published theory was that ~ was 2.9 [from (1): ~ = 1.8/1n (655/355) = 2.9] for 75% of Earth's terrestrial vegetation, i.e. the trees, whereas a summary of all the measurements with forest tree species indicates a ~ of c. 0.3--0.7 (Wullschleger et al. 1995b), which is about the estimate of ~ obtained for nonwoody plants and ecosystems (Amthor & Koch 1996); so, the wild theory that 'increases in atmospheric CO2 result in growth rate increases of trees five times greater than growth rate increases of nonwoody plants' (Idso 1991b) is gone and 'many other people's measurements' (sum­ marized in, e.g. Amthor & Koch 1996; Wullschleger et al. 1995b) remain.

Respiration The rate of plant respiration (i.e. glycolysis, oxidative pentose phosphate network, tricarboxylic acid cycle, and related mitochondrial electron transport and oxidative phosphorylation) is presumably often controlled by the rate of growth and maintenance processes through classic respiratory control mechanisms (Beevers 1970, 1974). That is to say, respiration slows when respiratory products (i.e. ATP, NJ\D(P)H, and C-skcleton intermediates) accu­ mulate in cells, and respiration proceeds when respiratorv products are consumed, through a series of feedback mechanisms. This implies that respiratory responses to elevated [C0 2l a may result mainly from changes in plant growth (and therefore metabolic costs of growth) and plant size (and therefore metabolic costs of maintenance); bigger plants are generally more expensive to grow and to maintain, which results in faster whole-plant respiration (Beevers 1970; McCree 1970; Thornley 1970). This is not, however, the same as faster whole-plant specific (per unit dry mass) respiration rate.

Indirect effects of elevated [CO:>]a on respiration One link between [C0 2L and respiration rate is likely to be via nonstructural carbohydrates. An increase in the amount of nonstructural carbohydrates may stimulate growth and related activities, and this would likely pull growth respiration along at a faster pace (Am thor 1994a). But in order for respiration (and growth) to proceed at a faster rate, respiratory capacity might need to increase as well. Tndeed, it appears that (i) growth processes, (ii) respiratory capacity, (iii) futile cycling and respiratory activity that may consume 'excess' nonstructural carbo­ hydrates, and (iv) related reactions, may be positively related to tissue sugar concentrations (Baysdorfer & Van Der Woude 1988; Bingham & Farrar 1988; Wenzler ct al. 1989; Stitt et a!. 1990; Williams & Farrar 1990; Farrar & Williams 1991; Geigenberger & Stitt 1991; Cheng et a!. 1992; Schafer ct al. 1992; JHH Williams et al. 1992; Bingham © 1995 Blackwell Science Ltd., Global Change Biology, 1, 243-27-+

2l

257

& Stevenson 1993; Krapp & Stitt 1994). These responses

to carbohydrate concentration, which all are directly or indirectly related to respiratory metabolism, would be expected to contribute to a balance between carbohydrate production and carbohydrate use in different [C02]aS (or across other environmental gradients that affect the rate of carbohydrate production). Thus, an increase in [C02 la can stimulate photosynthesis (see above), which can enhance levels of nonstructural carbohydrates, which can stimulate (or induce) growth and other processes consuming respiratory products, which vvould tend to then stimulate respiration. This chain of events is an indirect effect of elevated [C0 2]a on respiration because any factor other than [C0 2 ]a that also stimulated photo­ synthesis and increased the rate of production of non­ structural carbohydrates might elicit the same respiratory response (Amthor 1991). A change in plant (bio)chemical composition brought about by elevated [C0 2]a might affect rates of respiration independent of any changes in plant growth and size. For example, plant soluble protein concentration may be negatively related to [C02]a (Ziska & Bunce 1994a; Jacob et al. 1995) and this might be expected to cause a reduction in spccifi'c maintenance respiration rate because respiration associated with protein turnover can be a significant fraction of total maintenance respiration (Penning de Vries 1975; De Visser et al. 1992; see also Amthor 1994b; Bouma et al. 1994). There is evidence that elevated [C0 2L can in fact reduce specific maintenance respiration rate (Silsbury & Stevens 1984; Bunce & Caulfield 1991; Wullschleger & Norby 1992; Wullschleger et al. 1992; Ziska & Bunce 1993; Bunce 1995a). There are also reports of increased leaf specific maintenance respiration rate (Thomas cI al. 1993; Thomas & Griffin 1994), but I have shown (Amthor 1996) that, in those particular cases, the stimulation of leaf respiration caused by eleyated [C0 2 la was probably due to increased leaf respiration to support increased phloem loading and transport, resulting from increased photosynthesis, rather than being due to an increase in the rate of leaf maintenance processes. Growth costs (i.e. CO 2 released in processes associated with growth per unit C added to new plant structure) also may be affected by a change in plant composition. When an increase in [C0 2]a causes a reduction in plant protein concentration, or an increase in carbohydrate concentration, it might then reduce growth costs and therefore growth respiration per unit of plant growth (based on Penning de Vries et a!. 1974). Data from the few experiments addressing this issue indicate that elevated [C0 2]a does not affect, or causes a slight decline in, the C costs of plant growth (Silsbury & Stevens 1984; Loomis & Lafitte 1987; Baker et al. 1992; Wullschleger & Norby 1992; Wullschleger et al. 1992; Griffin ct al. 1993;

258

J . S. A M THO R

Thomas ct al. 1993; Amthor et al. 1994; Thomas & Griffin 1994; Wullschleger et al. 1995a). Changes in respiration brought about by changes in plant composition also are indirect effects of [C0 2 l a because any other factor that caused the same change in composition would be expected to cause a similar change in respiration. A simultaneous increase in plant growth and decrease in plant protein concentration (not necessar­ ily whole-plant protein content) might cause a decline in whole-plant specific respiration rate (mol CO 2 kg-I s-I) without a change in, or perhaps even an increase in, respiration per plant (mol CO2 s-1 per plant) or plant respiration per unit ground area (mol CO2 m-2 s-I).

Direct effects of [C0 2Ia on respiration In addition to the indirect effects of [C0 2 l a on respiration mediated through growth and maintenance processes, there are several reports of a direct inhibition of respiration by elevated [C0 2 l a that may be independent of respiratory control processes (Gale 1982; Reuveni & Gale 1985; Bunce 1990, 1995a; Amthor ct al. 1992; Byrd 1992; Mousseau 1993; Qi ct al. 1994; Downton & Grant 1994; Thomas & Griffin 1994; Villar ct al. 1994; Ziska & Bunce 1994b; Teskey 1995). The apparent direct inhibition of respiration by elevated [C0 2L may, however, be due to effects on nonrespiratory metabolism, e.g. stimulated dark CO 2 fixation and organic acid syntheSis (Amthor 1996). In a simple analysis, a reduction in respiration improves a plant's C balance, but it must be remembered that plant growth and maintenance each require respiration. Thus, a limitation on respiration might in turn limit growth and the maintenance processes required for plant health. Indeed, it was suggested on several occasions (Gale 1982; Reuveni & Gale 1985; Amthor 1991; Wullschleger et Ill. 1992) that to the extent that elevated [C0 2 la causes a direct inhibition of respiration, elevated [C0 2 L may be detrimental to growth and (or) maintenance processes. This view is supported by the short term experiments of Gale (1982) and also by the 21-day experiments in which soybean seedlings were exposed to elevated [C0 2 la dur­ ing the night, but not the day (Bunce 1995b). A direct inhibition of respiration - or more correctly, CO2 efflux in the dark - does not always, however, occur as a result of elevated [C0 2l a (Bunce 1990; Byrd 1992; Ryle et al. 1992a,b; Mousseau 1993; Ziska & Bunce 1994b). Available evidence indicates that the ratio of plant respiration to photosynthesis usually remains constant or declines in elevated [C0 2 l a (Hughes & CockshuIl1972; Gifford 1977; Silsbury & Stevens 1984; Du Cloux et al. 1987; Gaudillere & Mousseau 1989; Bunce 1990; Ryle et al. 1992a,b; Reid & Strain 1994; but see Nijs t'I al. 1989 and Den Hertog et al. 1993 for reports of increases in the ratio respiration/ photosynthesis caused by elevated [C0 2LJ

This implies that the C-use efficiency or growth efficiency (sensu Yamaguchi 1978) is unaffected or increased by elevated [C0 2 L. Either of these responses to elevated [C02 l a cause a carry through from stimulation of photo­ synthesis by elevated [C02l a to a comparable stimulation of growth by elevated [C02 l a because growth is related to the difference between photosynthesis and respiration. Thus, much of the pertinent literature indicates that plant respiratory responses to elevated [C0 2 L could reinforce a negative feedback on the rate of increase of [C02 l a that might be brought about by faster terrestrial higher­ plant photosynthesis. But, increasing temperature has the potential to stimulate plant respiration rate at any given [C0 2 l a - although plant respiration can acclimate and adapt to short-term and long-term changes in temperature (Amthor 1994a) - and the impact on global plant respira­ tion of increasing [C0 2l a in combination with global warming is unclear. Any effects of increasing [C0 2l a and (or) increasing temperature on the amount of CO 2 released to the atmosphere during plant respiration at the global scale is important to the global C cycle because about 10 times as much CO2 is released each year by plant respiration as is released by fossil fuel burning (Table 1). Moreover, changes in plant respiration carrv with them significant consequences for all other aspects of plant metabolism because plant health and growth depend on plant res­ piration.

Stomatal conductance and plant water use A short-term (minutes to hours) doubling of [C0 2 l a often causes a reduction in stomatal conductance (i.e. the conductance of gas flux through leaf surfaces via stomatal pores) of order 30-50% due to stomatal pore closure (Morison 1987). Stomatal conductance is not always, however, markedly affected by elevated [C02 la (Radin I't ill. 1987; Eamus & Jarvis 1989; Bunce 1992; Gunderson ('/ ill. 1993), although the ratio of water transpired to CO 2 assimilated by individllall('[lvc~ is nearly always positively related to [C0 2la (e.g. Eamus 1991). When elevated [C0 2L causes stomatal closure, it is apparently the intercellular [C0 21 (i.e. [C02L) that is of significance, rather than [C02 la or the [C0 21 at the leaf surface (Mott 1988). Stomata in C 3 and C 4 leaves may be about equally sensitive to a change in [C0 2L (Morison & Gifford 1983). In addition to stomatal closure following short-term increases in [C02 l a , stomatal density (stomata per m 2 leaf) can be affected by long-term changes in [C0 2la. An inverse relationship between [C02l a and stomatal density was observed over the 200-350 ppmv CO2 concentration range in (i) experiments (Woodward 1987), (ii) leaves of herbarium specimens collected during the last 200-250 years (Woodward 1987; Peii.uelas & Matamala 1990; Van CD 1995 Blackwell Science Ltd., Global Chill/gl' /JillloilY, 1, 243-274

T ERR EST RIA L - P LAN T RES P 0 N SET 0 AT M 0 S P HER I C l CO 2 ] der Burg ct al. 1993; but K(irner [1988] did not find an effect of [C02 le, during the past 100 years on stomatal density), (iii) leaves preserved in pack rat middens for thousands of years (Van de Water ct al. 1994), and (iv) Quaternary fossil leaves (Beerling & Woodward 1993). Thus, stomatal density might have declined as [C0 21a increased since the LGM. Such a response could be related to optimization of water lost per C gained by individual leaves. When [C0 2 L is increased above 350-360 ppmv in experiments, stomatal density remains constant or decreases (Woodward 1987; Woodward & Bazzaz 1988; Estiarte I't al. 1994; ~erris & Taylor 199.:1; Seerling & Woodward 1995). These experiments were short-term ­ less than the life cycle of the plants studied - so effects of long-term elevated [C02]a on adaptation processes that might influence stomatal density were not addressed. In any case, effects of future increases in [C02 ]a on stomatal density are unclear. But, stomatal conductance may be reduced in the future by increasing [C02L, perhaps in part as a result of smaller stomata (Miglietta & Raschi 1993) and due to stomatal closure. One result of a negative relationship between [C0 2 ]a and stomatal conductance is a somewhat conservative [C0 2L/lC0 21a ratio. Thus, across a range of [C02]aS, [CO zLllC0 21a may be about constant, although many exceptions are known. Data available indicate no general acclimation of stomata to elevated [C0 2]a during the course of days to months, i.e. [C02 L1[C0 21a does not change consistently with time (Sage 1<)<)4). Applicable data are limited, however, and Ehleringer & Cerling (1995) concluded that lC0 2UlC0 2la 'will remain constant in some species and vary in others' as [C0 2]a continues to increase during the coming decades to centuries. Although stomatal closure can reduce transpiration at the single leaf level, effects of stomatal closure on regional and global transpiration are unclear because of feedbacks from the atmosphere - specifically, the planetary bound­ ary layer (PBL) - to canopy transpiration (Jarvis & McNaughton 1986; Jacobs & De Bruin 1992). When stomatal closure occurs at the regional scale, as might be the case for regional and global [C0 2Ia increase, the tendency for reduced transpiration would itself tend to reduce the vapour pressure of the PEL. This in turn would tend to stimulate transpiration because of an increased vapour pressure gradient from inside leaves to the atmosphere. This feedback on regional transpiration is an important, but often overlooked, aspect of terrestrial­ plant and terrestrial-ecosystem responses to elevated [C02]a' Regional transpiration therefore may be less affected by an increase in global [C02 ]a than is transpir­ ation bv small field plots treated with elevated [C02] because the PBL vapor pressure is unaffected by elevated [C0 2]a at the plot scale. The same applies to plants treated © 1995 Blackwell Science Ltd., Global Change Biology, 1, 243-274

259

with elevated [C0 2] in controlled environment chambers with constant vapor pressure. These facts are important in the interpretation of measured water use in elevated­ CO2 experiments conducted in chambers and free-air CO 2 enrichment plots. Another result of reduced stomatal conductance, brought about by reduced stomatal aperture and (or) reduced stomatal density in elevated [C02]a, is the poten­ tial for elevated leaf and canopy temperature due to reduced latent heat exchange. Thus, even without a general increase in global temperature, canopy temper­ ature may be increased bv elevated [C0 2 la. This would increase the vapour pressure gradient from inside leaves to the atmosphere and therefore stimulate transpiration (and limit the effect of elevated-[C02]a-induced stomatal closure on transpiration). Stomatal responses to [C0 2 ]a are important to the C cycle in part because CO 2 assimilation (photc ,ynthesis) is controlled by stomatal conductance, at least slightly (Farquhar & Sharkey 1982). Changes in the ratio of C assimilated to water transpired, however, may be the most important effect of stomatal responses to [C0 2L on the C cycle. The general decline in stomatal conductance and increase in photosynthesis with a short-term increase in [C02 1a leads to a marked increase in water-use effici­ ency by C 3 plants. Water-use efficiency of C 4 plants is also increased by CO 2 enrichment, but mainly due to reduced transpiration rather than the combination of stimulated photosynthesis and reduced transpiration. In addition, over the long term, several non-stomatal plant responses to elevated [C02 ]a may influence stomatal conductance. For example, increased whole-plant growth or altered root I shoot ratios can affect plant and soil water status and therefore influence stomatal conductance via whole-plant water relations and water use. To the extent that elevated [C0 2l a enhances the depth of root growth and thus the depth in the soil from which water can be extracted, ele\'ated [C02]a may ill crease water use, especially in dry soils for which all (most) water would be extracted from the surface soil in any [C02 la. Even though effects of increasing [C02 ]a on plant­ community water usc have been speculated on at length for many years, there is only one set of field experiments not involving chambers that directly addresses this issue: the free-air CO 2 fumigations (at c. 550 ppmv CO 2) of cotton and wheat in Arizona (Pinter et al. 1996). (Chambers have significant effects on canopy energy exchange. In particular, the ratios of longwave radiation exchange, sensible heat exchange, and latent heat exchange are usually different inside chambers conpared with natural conditions. Chambers will generally have larger effects than elevated [C02 L docs on energy exchange, and the effects are confounded through a series of interactions with transpiration and wind. Thus,

260

J . S.

AM THO R

transpiration within chambers is unlikely to reflect trans­ piration outside chambers and effects of elevated [C0 2 la on water use by chamber-grown plants may bear little relevance to global plant responses to increasing [C02 la .) In cotton amply supplied with water in free-air CO2 enrichment experiments, elevated [C02la had no effect on transpiration compared to ambient [C02 l a plots (Dugas et al. 1994). With either ample irrigation or deficit irrigation, cotton-crop water use, i.e. transpiration plus soil evaporation, was not affected (Hunsaker et al. 1994), or perhaps even increased (Kimball et al. 1994), by elevated [C0 2 la compared with present ambient [C0 2 la . For wheat amply supplied with water, elevated [C02 la decreased water use over two seasons by c. 4.5%, but with limited irrigation, elevated [C0 2la increased crop water use c. 3% (Pinter et al. 1996). So in spite of significant speculation concerning important impacts of elevated [C02 l a on water use, the only data directly addressing the issue indicate no, or only small, effects of [C02la on whole-plant transpiration and whole-crop water use (volume of water evaporated per unit area of crop). Nonetheless, water-use efficiency (C02 assimilated per unit water evaporated) was increased in both cotton and wheat by elevated [C0 2 la . Effects of [C0 2la on the ratio of plant growth to water transpired has implications for the global C cycle. In elevated [C0 2la , more C can be assimilated for a given amount of water transpired, and this has the potential to enhance plant growth per unit of precipitation. Increased plant growth for a given amount of precipitation may lead to increased C storage - in both living plants and soil C pools - in terrestrial ecosystems that are presently 'limited' by water availability. This could result in a net transfer of C from the atmosphere to arid and semi-arid land now occupied by plants. Moreover, increased plant growth per unit precipitation in elevated [C02 l a has the potential to increase the geographical range of plants in arid and semi-arid regions. This in turn could lead to C storage in plants and soils in areas not now occupied by plants, leading to further net transfers of C from the atmosphere to land. Thus, increased plant growth per unit precipitation may function as a negative feedback on the increase in [C02l a , although the magnitude of such a negative feedback has not been properly evaluated. To the extent that stomatal closure increases canopy temperature, stomatal response to elevated [COzla may represent a positive feedback to warming and climatic change. And, although the onlv appropriate data available indicate that elevated [COzla does not affect water use by crops, climatic change - in particular warming - has the potential to increase water use, although precipit­ ation may also be intensified by an enhanced green­ house effect.

C3 vs. C4 plants The time of the evolution of C4 photosynthesis is not precisely known, but it may have been as recently as 15 X 106 years ago (Morgan et al. 1994a), as compared with the far more distant past for the emergence of C3 photosynthesis. It was suggested that low [C02 la (e.g. near PAL) was 'the primary selective factor influencing the evolution of C4 photosynthesis' (Ehleringer et al. 1991) because C3 photosynthesis is limited by [C02la near PAL, but other factors may also have been important. The global expansion of C4-dominated ecosystems 5-7 X 106 years ago may have occurred in response to declining [C02 l a (Cerling et al. 1993), but again, other factors may also have been important (Quade et al. 1989; Morgan et al. 1994b). Since C4 photosynthesis is nearly COrsaturated at present [C0 2l a whereas C 3 photosynthesis is operating well below optimal [C0 2l at PAL of CO 2, it is often suggested that the present increase in [C02l a should favour C 3 plants compared with C4 plants. To quote Kirschbaum (1994)

Enhanced photosynthetic capacity of C3 plants [in elevated CO 2] is of immediate significance for the competition between C3 and C4 plants. At a particular location where C3 and C4 plants co-exist, they must be competing for other limiting resources, such as water, nutrients or access to light. Increasing CO 2 concentration confers a selective advantage on the C3 plants, and puts them into an increasingly favourable competit­ ive position. Increased C gain by C3 plants would allow them to either increase root growth and compete more successfully with their C4 neighbours for nutrients, or increase foliage production to compete more successfully for available light. This leads to the typical experimental observations on mixed C,/C4 stands that the C3 components gain an increasing biomass share with increasing CO 2 concentration ...although the complexities of plant/plant interactions are such that results are not always consistent... Where differences are observed within a single generation, these are likely to be further compounded over successive generations. The continuously improving photosynthetic performance of C3 plants...should put great competitive pressure on neighbouring C4 plants, especially in warmer regions where the improvements in the performance of C3 plants should be most marked. As expected, several experiments support this straight­ forward conclusion concerning effects of elevated [COzL on mixtures of C3 and C4 plants. For example, in a Chesapeake Bay (MD, U.s.A.) wetland mixed community of C , sedge and C4 grasses, elevated [C0 2 l a resulted in an increase in C 3-plant above-ground dry mass and a concomitant decrease in Crplant above-ground dry mass at 'mid-season' (Drake 1992). From the LGM to several hundred years ago, C3 Cic)1995 I3lackwell Science Ltd. C/O/ill/ Chollge Ri%gy, 1, 243-274

T ERR EST R I A L - P LAN T RES PO N SET 0 AT M 0 S P HER I C [C 0 21 vegetation replaced C~ vegetation in some locations while [C02 1a increased from c. 180 to c. 280 ppmv. It was suggested 'that the increase in global CO2 concentra­ tion...contributed directly to the shift from C4 to C3 vegeta­ tion' at one such site (italics added; Cole & Monger 1994), but other environmental factors must also be considered and the replacement of C~ vegetation with C, vegetation was not global as was the increase in [C021a (Boutton et al. 1994). Woody C 3 vegetation 'invaded' Crdominated grass­ lands in some locations during the past 200 years when global [C021a increased from c. 280 to c. 360 ppmv. This was attributed by some (see references in Archer et al. 1995) to the increase in [C0 21a itself or to recent climatic change, but the most objective analyses of this phenom­ enon indicate that the main cause of recent C 3 invasions was li\'estock grazing (Archer et al. 1995). But even though C 3 invasions of C 4-dominated communities during the past 200 years were probably unrelated to global changes in [C021a , effects of further increases in [C021a might significantly influence future C 3 /C 4 competitions. Although C 4 photosynthesis is nearly COz-saturated at PAL of CO2, C 4 plants can respond positively to elevated [C0 2 L· As mentioned above, stomatal closure in C 4 plants exposed to elevated [C0 2 L leads to improved water-use efficiency, and there are direct observations that Crplant growth can be stimulated by elevated [C021a (Rogers et al. 1983; Amthor et al. 1994; reviewed by Poorter 1993). Moreover, in mixed C T C 4 communities, growth of C 4 plants may be stimulated by elevated [C0 2L when at the same time growth of C 3 plants is not affected, at least during drought (Owensby ct al. 1993b; d. Drake 1992). As stated by Henderson et al. (1994), 'precisely which trait, or which combination of traits, is most relevant to the relative fitness of C 3 and C 4 species remains baffling.' Indeed, Henderson ct al. (1994)-focusing on Australian plants-suggest that because additional warming and other climatic changes are likely to accompany the con­ tinuing increase in [C021a

...it is by no means clear that C3 plants will uniformly exploit acceleratl'd CO 2 fixatioll in the long tam. Tn fact C3 plants face 1111 array of limitations 011 translocation and utilization of assimilates, of respiration at elevated temperatllres and of unpredictable reproductive outcomes... that may constitllte larger threats to fitness than they do to C4 plants. We {Henderson et al.] are tempted to conclude that the outcomes of the contest between plants with the C3 and C4 photosynthetic pathways [inl elevated CO 2 will depend more on the extent of advantages gained by the former than on those surrendered by the latier. Tndced, when considerillg the increase in temperature occur­ rillg with increasing CO 2... lUe {Henderson et al.] conclude that there may well be a significant increase in the representation of © 1995 Blackwell Science Ltd., Global Change Biology, 1, 243-274

261

c~ grasses in the Australian flora. The southerly and easterly expansion in the range of C4 grasses in Australia may match the northerly expansion of the range of North American grasses such as Echinochloa... [Andl it may be more appropriate to ask Ilot whether C4 grasses will be endangered by global climatic cilangc, but whether C 1 grasses will be endallgered, especially in Australia.

There is divergence of thought concerning relative effects of global environmental change on C 3 and C 4 plants. With other factors unchanged, increasing [C0 2 L tends to enhance the competitive ad vantage of C 3 plants over C 4 plants, except perhaps in dry environments, but overall effects of global environmental change on the C 3 /C 4 balance is not obvious. Time, and discerning observation and experimentation, may tell. It also is unclear how a change in the C 3 /C 4 balance per se would affect the global C cycle and [C02 la.

Plant-litter production and decomposition Both the rate of transfer of C from plants to soils and the quality (chemical composition) of plant litter affect C cycling in soils, and key links, perhaps till' key links, between terrestrial-plant responses to [C0 21a and the global C cycle are the amount and the quality of plant litter produced. This is because the amount of plant­ derived C in soil and litter is thought to significantly exceed the amount of C in living plants (Table 1) and, depending in part on chemical composition, some frac­ tions of soil C have lifetimes in excess of 1000 years (Trumbore 1993). Moreover, the potential for additional C storage on land as a result of increasing [C0 21a is probably greater in the litter and soil organic matter pools compared with the living-plant C pools because organic C can accumulate on (in) soils without significant requirements of heat, nutrients, light, water, and structure; these are critical requirements for the maintenance of large pools of living-plant C. The amount of plant litter produced is clearly related to plant growth, at least in the long term. Thus, inputs of organic C to soil are expected to be positively related to [C021a to the extent that elevated [C0 21a enhances plant growth. Present ecosystem soil and litter C content, however, may be unrelated to rates of C input from plants, at least according to data from the forests and grasslands summarized by Cebrian & Duarte (1995). instead, soil organic-C pool size may be negatively related to plant turnover rate, i.e. positively related to plant lifetime (Cebrian & Duarte 1995). It is unknown if such a relationship exists in other terrestrial ecosystems or if increasing [C0 2L will affect such a relationship. Since the N/C ratio (and related properties) of leaves was altered in several elevated-C02 experiments (e.g.

262

J. S. A M THO R

Wong 1979; Norby et al. 1986, 1992; Williams et al. 1986; Curtis et al. 1989; Kuehny et al. 1991), there was concern that litter decomposability will be reduced as a result of an increase in [C0 2]a (see O'Neill & Norby 1996). A related concern is that reduced litter decomposability could reduce the rate of nutrient cycling in soils which could in turn affect plant physiology and growth. Litter quality (and amount) could also be affected by elevated [C0 2 ]a to the extent that species composition of terrestrial ecosystems is changed because different species can differ with respect to qualities, rates of decomposition, and amounts of litter produced (Kemp et al. 1994; O'Neill & Norby 1996). But what evidence addresses the concerns about effects of elevated [C0 2 la on litter quality? The short answer to this question is: Very little. O'Neill & Norby (1996) reviewed the limited data and made the important conclu­ sion that, with respect to litter quality - as with many other factors - plants growing in pots, often in 'growth chambers,' differ from plants growing in the 'field.' Moreover, it is the quality of naturally abscised leaves and dead stems and roots that is relevant to decomposi­ tion. Most speculation concerning effects of elevated [C0 2 la on decomposition, however, is based on green rather than senesced and naturally abscised leaves and the chemical composition of green leaves may differ significantly from the composition of naturally abscised leaves (Curtis et al. 1989). Generalizations that can be made presently for plants growing in the ground with normal senescence processes are (see O'Neill & Norby 1996): (i) leaf litter chemical composition is unaffected, or only slightly altered, by elevated [C0 2 ]a (ii) decomposability of leaf litter is unaf­ fected by the [C0 2 ]a to which plants producing the litter were exposed, and (iii) amount of litter production can be positively related to [C0 2 ]a. (Pot-grown plants can give different results-O'Neill & Norby 1996.) Nonetheless, Drake (1992) reported that elevated [C0 2 la slowed decom­ position in a C 3 (Scirpus olncyi) community but did not affect decomposition in a C 4 (Spartina patens) community in the field. It is unknown whether elevated [C0 2 ]. has a direct effect on decomposition per se in intact ecosystems. Additional effects of [C0 2 ]a on decomposition rate might be mediated through climatic change. Both soil temperature and moisture affect decomposition, so changes in either factor brought about by COrinduced climatic change could influence decomposition rate (And­ erson 1992; Schlesinger 1995). 'First principles' and avail­ able data indicate that warming might stimulate decomposition rates and increase CO 2 release from soils in a positive feedback on the [C0 2 ]a increase (e.g. Jenkin­ son et al. 1991), but the N cycle may modulate such a response (e.g. Schimel ct al. 1994). As outlined by, e.g.

Field et al. (1992), warming that increases the rate of decomposition can release to the soil 1 mol N for every c. 10-20 mol of CO 2 produced. For woody vegetation, 1 mol N might support the storage of c. 100-200 mol C, because of the large C/N ratio of wood. Thus, increased decomposition of organic matter in forest soils can poten­ tially store more C in wood than is lost from soil because of N cycling. For nonwoody vegetation, such a link between soil organic matter decomposition and storage of additional C in plants does not exist or is not as strong. And even in forests, which may contain c. 25°/', of global soil organic matter (Ajtay et al. 1979), 'this mechanism can persist only until the pool of soil organic matter reequilibrates at a new size' (Field ct al. 1992). The global terrestrial (plant plus soil) C pool will be affected by such a mechanism to the extent that 'additional' C stored in wood offsets 'additional' C lost from all soils, not just forest soils. Carbon dioxide and N are not the only products of decomposition that are important to global environmental change. Although C released in CH 4 during organic matter decomposition plays only a small role in the present global C cycle - c. 0.5')';, of atmospheric C is in CH 4 (Dlugokencky et al. 1994) - CH 4 is a much stronger greenhouse gas than is CO 2 (Lelieveld & Crutzen 1992), so any changes in CH 4 production resulting from increasing [C0 2 ]. might have significant implications for climatic change. Natural and agricultural wetlands are major sources of atmospheric CH 4, and primary production in wetlands is positively related to CH 4 production (Whiting & Chanton 1993). Thus, increasing [C0 2 ]a may stimulate CH 4 production by stimulating wetland productivity. Indeed, for Scirpus olneyi communities grown in elevated [C02 ]a in the field, CH 4 production was stimulated, at least during the four nights of measurements (Dacey et al. 1994). It has yet to be determined whether an incremental increase in wetland primary production caused by an incremental increase in [C0 2 ]. will weaken or strengthen global greenhouse warming; stimulated primary produc­ tion will function as a negative feedback on the [C0 2 la increase, but enhanced CH 4 production may accelerate warming. Drying of wetlands with global warming might reduce CH 4 production, if drying occurs. The link between elevated [C0 2]a and litter production and decomposition needs much more study because it is quantitatively important to the global C cycle and because so few data are available from plants growing in the ground in different [C0 2 las. Analyses of soil C pool sizes and qualities near natural CO 2 springs may be the best available measures of effects of long-term elevated [C0 2 ]a on litter production and decomposition. Any experiment that can be conducted, using any technique, is of short duration compared to the time scale of soil C turnover in many (or perhaps all) ecosystems. And, the 'step © 1995 Blackwell Science Ltd., Global Change Biologl/, 1, 243-274

TERRESTRIAL-PLANT RESPONSE TO ATMOSPHERIC [C0 2 ) change' in [C0 2la that is used in experiments may cause different plant and ecosystem responses to [C0 2 la than the responses that are occurring with the present gradual increase in global [C0 2 la. More on missing sinks There is confusion in the literature concerning the defini­ tion of the 'missing' sink of anthropogenic C. The annual global anthropogenic-C balance is often written in a form equivalent to F+ 0 - 0 - A - M

=

0,

where F is CO 2 released in fossil fuel burning (Pg C y~l), is net CO 2 released from land disturbed by humans and includes CO 2 released from wood removed from forests (Pg C y~l), 0 is oceanic net CO 2 uptake (Pg C y~l), A is atmospheric CO2 increase (Pg C y~l), and M is the missing C sink (Pg C y~l). Values of F are summar­ ized in Marland et al. (1994; and see Table 1), Houghton (1995) gives 1.6 ::+: 0.7 Pg C y~l for the 1980s value of 0, estimates of 0 are given in Table 1 and Francey et al. (1995), and A cal1 be derived from Fig. 1 along with the relationship 1 ppnw CO 2 = 2.12 Pg C. In most analyses M is greater than zero. The anthropogenic-C balance equation above is not in the correct form, however. It should be

°

F + D - 0 - A - U

=

0,

where U is net CO2 uptake by all land not dim-Ill! disturbed by humans (Pg C y~l). The difference U - 0 is annual global terrestrial ecosystem net CO2 uptake (Pg C y~ 1 ) (it could be negative). A comparison of these two anthropo­ genic-C balance equations indicates that M = U, although this point is often unappreciated. Recent year-to-year variation in U - 0 inferred by, e.g. Francey el al. (1995) and Keeling ct al. (1995), mav be related to interannual variation in weather, but there are mostly questions rather than answers - concerning year-to-year changes in rates and controls of global CO 2 exchange by terrestrial ecosystems. Two assumptions commonly made when considering the missing C sink M are that (i) regrowth of disturbed forests, either from logging or previous conversion to agriculture, is the missing C sink and (ii) the annual net C uptake by land not disturbed by humans U is zero. Assumption (i) is false by definition, and assumption (ii) may be false in actuality. The missing C sink, and distinctions among 0, M, and U, are discussed further by Houghton (1995; and references therein). It is well known that many forests, particularly those in northern temperate and boreal regions, are accumulating phytomass C. The northeastern U.s. is a good example: it contains extensi\'e area of aggrading (sensu Bormann © 1995 Blackwell Science Ltd., Clobal Change Biology, 1, 243-274

263

& Likens 1979) forest now accumulating C (e.g. Armen­ tano & Ralston 1980). Although this forest growth has at

times been assigned to the missing sink M (assumption riD, much of that C accumulation occurs in previously cleared or logged forests and is therefore a component of D. Indeed, Houghton (1995) includes that forest regrowth in his estimates of D. and it cannot be simultan­ eously assigned to U or M. In short, C accumulating in previously disturbed forests is a component of 0, not of M (see Houghton 1995 for further discussion), and it is included in estimates of global D. A point worth consideration with respect to increasing [C0 2L and refor­ estation and afforestation programmes - reforestation and afforestation are human activities included in 0 - is that in many areas Eucalyptus is planted (e.g. Seiler & Crutzen 1980) and the growth response of Eucalyptus to elevated [C0 2 )a may be particularly strong (WuUschleger et al. 1995b). Assumption (ii) can be evaluated by solving the anthro­ pogenic-C balance equation for U. To the extent that 0 is known (see Francey ct al. [1995] & Keeling ct al. [1995) for estimates of 0 that vary considerably from year to year), U can be found because F and A are well quantified and 0 can be taken from Houghton (1995). Available data indicate that the terrestrial biosphere is presently a net sink for on the order of 0-3 Pg C y~l (e.g. Conway ct al. 1994; Winn et al. 1994; Francey ci al. 1995; but see also Keeling cl al. 1995). If U - D is 1.5 Pg C y~l (see Table 1) and 0 is 1.6 ::+: 0.7 Pg C y-l (Houghton 1995), U is 3.1 ::+: 0.7 Pg C y~l. Can this be the case? Not if ecosystems undisturbed by humans are in steady state with respect to C. As mentioned above, however, some undisturbed northern forests and peatlands may be accu­ mulating C, primarily in peat and soil C pools, as they continue to 'recover' from the last glacial cvcle. Changes in natural disturbance cycles might also affect Li, and there is evidence that during recent decades Canadian forests accumulated C for that reason (Kurz 1'/ al. 1995). These mechanisms may not be sufficient to support the required value of U, however. Other causes of positive U (as compared with an undisturbed terrestrial biosphere that is in steady state with respect to C content) that have been proposed include global warming, increased N deposition on land, and CO2 fertilization of terrestrial plants. Potential effects of warming on increased N mineraliz­ ation, associated with increased soil organic matter decomposition, and subsequent net C storage in wood were outlined by Field ct al. (1992). Some evidence indicates that warming during the past 200 years stimu­ lated boreal forest productivity relative to preindustrial times (Jozsa & Powell 1987). Warming might have increased growing season length, and therefore annual

264

J . S. A M THO R

plant growth, in temperate, boreal, and tundra eco­ systems. Plant growth responses to increased N deposition, resulting from air pollution, might also enhance C storage in terrestrial ecosystems (Field et al. 1992). Increased N deposition is probably greatest in northern temperate ecosystems, where air pollution is greatest, and its effects might therefore be largest in northern ecosystems. It was proposed that 0.6 Pg C might be stored in terrestrial ecosystems each year due to fertilization from anthropo­ genic N releases (Schimel 1995). This is not sufficient to account fully for the estimate of U above. How and where might increasing [C02l a increase U? The main mechanism is probably stimulated photosyn­ thesis, although increased water-use efficiency may also be important. The interaction between temperature and [C02l a with respect to C 3 photosynthesis indicates that increasing [C02l a -induced photosynthetic enhancement might be most significant in tropical (and other warm) areas (Kirschbaum 1994). Direct effects of [C02l a on respiration might increase or decrease U, depending on which metabolic processes are affected and whether they are essential for plant growth and health. As outlined ab()\"e, plant growth is probably stimulated, to some degree, by increasing [C02l a , and plant growth responses to increasing [C0 2la could be spread throughout the earth because CO 2 is well mixed in the atmosphere; increasing [C0 2l a is not limited to areas of CO 2 emission. The parameter U includes CO2 exchange by plants and soils, however, so knowledge of plant response to [C02l a is not sufficient to quantify the response of U to [C02l a . Carbon dioxide flux measurements made over an undisturbed tropical rain forest in Brazil during 55 days were used to estimate annual whole-ecosystem C exchange (Grace et 111.1995). The estimated annual whole­ ecosystem C exchange rate so obtained was c. 0.24 !lmol CO2 m- 2 s-1 (units used by leaf physiologists) or c. 91 g C m-2 y-l (units used by ecologists). Carbon uptake cannot be so precisely measured with the methods used, i.e. the annual rate obtained cannot be distinguished from zero with great confidence, and that value was obtained by extrapolation from 55 measurement days to 365 days. Nonetheless, the result indicates annual net C uptake by an undisturbed forest. If those measurements represent an actual net uptake of C by the forest that is sustained over many years, they could be indicative of a response to increasing [C0 2la. Or, the C gain might be due to normal year-to-year fluctuations between C net loss and C net gain with a net-gain period studied by chance, because it is unlikely that the C balance of any ecosystem is exactly zero during any 365-day period. Unfortunately, there is little direct data concerning annual net C exchange by other undisturbed terrestrial ecosystems. Other factors may also be important to the present

apparent C sink activity of the terrestrial biosphere. Deep roots (up to 18 m or more) in Amazonian forests (Nepstad et al. 1994) not only increase the size of the soil (and plant) pools of C compared with values listed in Table 2, but deep roots also increase the potential C sink strength of tropical forests in response to increasing [C0 2 la. This may be particularly important if increasing [C0 2l a is causing an increase in root/shoot ratio (see Rogers f'I al. 1994 for qualifications). Carbon also may be accumulating below-ground, independently of any change in [C02l a , in tropical savannas that were subject to the introduction of deep-rooted grasses (Fisher ('t Ill. 1994) - a land-use change factor not included in D by Houghton (1995). In addition, increasing [C0 2l a may be enhancing water-use efficiency, NPP, and C storage in both managed and unmanaged tropical savannas. Carbon that accumulates in undisturbed terrestrial ecosystems may do so in living plants, in litter and dead wood, or in soil humus. Some 'additional' C storage in all of these C pools is likely. After analysing soil chronosequences, Schlesinger (1990) concluded that only c. 0.4 Pg C y-l are likely to be stored in soil humus under natural vegetation globally, and noted that agriculture generally results in a net loss of soil C compared with the same soil prior to cropping. This implies that soil organic matter is not a likely sink for large quantities of C; 0.4 Pg C y-l is much less than the 3.1 Pg C y-l estimate of U above. But, Schlesinger'S (1990) analysis is based on historic trends, not on ecosystem processes occurring with present, relatively high [C02l a , with present N deposition, and with present, relatively warm climate. Moreover, all these factors are likely to continue to change for decades. Thus, a convincing case has not been made for a low C storage potential of soil today and into the future when environmental conditions, and plant productivity, may differ from those of the past several thousand years; a possible rapid soil-C accumulation rate should not be ignored. Also, litter and dead wood globally contain large amounts of C (not considered by Schlesinger [1990]), and those pools could be increasing in size as a result of recent stimulations of plant photosynthesis and growth. The limits to global living wood C content are unknown, and could be greater than present stocks, so this C pool too could increase in size over the coming decades. Increased monitoring of spatial and temporal gradients of [C02l a and the atmospheric l3C02/l2C02 ratio are key to resolving general locations of today's (probably broadly distributed) terrestrial net sink for C (Ciais et al. 1995; Keeling et al. 1995). To the extent that C is accumulating on land across large areas, it cannot be accurately and directly measured at the individual ecosystem scale. In the extreme case that 3.1 Pg C was added uniformly to Earth's land surface each year, C accumulation rates would be only 0.055 !lmo! CO 2 m·· 2 s-1 or 21 g C m-2 y-l ([J

1995 Blackwell Science Ltd., Glo/Jal Cizange Biology, 1, 243-274

T ERR EST R I A L - P LAN T RES P 0 N SET 0 AT M 0 S P HER I C [C 0 2]

. c4

Elevated CO2 ~

Increased water-use

•. .

L

(short term)

-

C02~

More growth

respiration

Stimulated

More C storage in

presently vegetated

~""'" ~ ...... -~~--

J 1 I

Photosy(ntlonhegtictennrepression Tn)

Feedback InhibitIOn

___

265

Exp8nslOn of vegetatIOn Into, and subsequent C storage

In,

nonvegetated and and semi-arid areas

Photosfthesls

More nonstructural carbohydrate

Mobilization of N In

~!

photosynthetic apparatus

~ More

growth

I

~ Transport

of N

to C sinks

~

~ CO 2-

More maintenance Larger respiration (?) - - Plats

~

More C input to soli

'J

More water used (7)

. More C stored In plants. most Importantly in wood

Fig. 5 Proposed main terrestrial higher-plant responses to elevated [C0 2l a that may be of quantitative importance to the global C cycle. Elevated [C02l a stimulates photosynthesis, much more so in C 3 plants compared with C 4 plants, which in turn may increase plant nonstructural carbohydrate concentration (Bowes 1993). Accumulation of nonstructural carbohydrates can inhibit (Azcon-Bieto 1983; Foyer 1988) or repress (Sheen 1994; Webber ct al. 1994) photosynthesis, but it might also enhance growth and growth respiration (Farrar & Williams 1991). Repression of photosynthesis mav be related to the mobilization of N in photosvnthetic enzymes and pigments, and that mobilized N might be used for additional plant growth (Webber et al. 1994), which could be significant for plants growing in N-deficient soils. Increased maintenance respiration, normallv resulting from larger plant size, may be mitigated by changes in plant composition, i.e. reduced N/C ratio, that in turn reduce specific maintenance costs (Amthor 1991). And, respiration mav be directly inhibited by elevated IC0 2 l a (not shown; Amtho~ 1996). Enhanced plant growth in turn could lead to enhanced C storage in perennial plant tissues, most importantly wood, and enhanced inputs of litter (both above-ground and below-ground) to soil C pools. An increase in root growth or size might also increase C inputs to the soil via exudation. Enhanced input of C to soil could lead to an increase in soil C content. In addition to stimulated photosynthesis prr se, elevated [C02 la can potentially increase whole-plant water-use efficiency, which in turn can enhance growth of plants in arid and semi-arid areas now containing some vegetation. Increased water-use efficiency might also allow the expansion of vegetation into presently un vegetated arid and semi-arid areas. Increased plant growth in arid and semi­ arid areas would enhance organic-C inputs to soil in those areas, which might lead to larger soil C pools. Plant responses to elevated [C02 la shown here tend to act as negative feedbacks on the rate of increase of [C02 l a , but all the links in this scheme are subject to modification by climatic change accompanying the increase in [C02 l a . And, effects of human activities (other than fossil fuel burning) have a greater impact on the C balance of many individual ecosystems than do increasing [C0 2l a and global warming.

(four times smaller than the annual value proposed by Grace et al. [19951 for an undisturbed tropical rain forest). Such a rate of C accumulation would be deep within the noise of measurements of ecosystem C exchange. Ecosystem and global C cycle models now in use are inadequately tested (or not tested at all!) against measured terrestrial C fluxes - so little significance can be attached to their predictions - and point measurements of CO 2 exchange in the field may be too geographically specific to be useful in quantifying regional and global scale processes. Thus it may be some time before a reasonable assessment of the C sink activity of the terrestrial biosphere can be obtained by ground-based meas­ urements. An issue related to the future global C cycle is the potential displacement (or migration) of biomes caused by climatic change and (or) elevated [C0 2]a' Changes in the area of individual biomes caused by environmental © 1995 Blackwell Science Ltd., Global Change Biology, 1, 243-274

change might alter terrestrial ecosystem C fluxes propor­ tionally, at least in the long term. For example, if boreal forest replaces some tundra as a result of warming, CO 2 assimilation and C storage in the 'new' boreal forest might exceed values in the 'old' tundra because boreal forest photosynthesis can exceed tundra photosynthesis. On the other hand, warming of tundra could cause significant releases of CO 2 due to changes in the water table leading to accelerated decomposition (Oechel et al. 1993) if drying is not prevented by increased precipitation due to an enhanced global greenhouse effect (Manabe & Stouffer 1994). Such changes are difficult to evaluate because the early phase of transition from one biome to another at a particular site might result in a net release of CO2 due to reduced NPP and continuing decomposi­ tion, but this might be followed in time by increased NPP and soil C storage. The extent and timing of future biome migrations (if any) is unknown.

266

J. S. A M THO R

Summary Aside from its potentially negative impact on climate, it is easy to think of elevated [C02 l a as a good thing for plants. The most obvious and immed iate effect of elevated [C0 2 l a on terrestrial C 3 higher plants is a stimulation of photosynthesis. And because terrestrial-plant photosyn­ thesis assimilates about 20 times as much CO 2 as is released in fossil fuel burning each year, even a small increase in photosynthesis has the potential to signific­ antly affect the fate of fossil C released by human activities. Stimulated photosynthesis initiates a string of process responses that can affect nearly all aspects of the C cycle of terrestrial higher plants and ecosystems (Fig. 5). Most immediately, stimulated photosynthesis may lead to the accumulation of nonstructural carbohydrates, which may reduce photosynthetic capacity (downregulation) and photosynthetic rate (inhibition) in a series of negative feedbacks. The interaction between N availability (or N concentration in a plant) and [C02 l a may playa role in the degree of photosynthetic acclimation to elevated [C02l a although this issue remains unresolved. Accumu­ lation of nonstructural carbohydrates may also stimulate capacities and rates of respiration, biosynthesis, and related reactions that consume photoassimilate in a series of positive feedforwards in plant metabolism. A common response of C 3 plants to elevated [C02l a in experiments is enhanced growth, but growth of C4 plants also can be stimulated bv elevated [C0 2l a . In addition to effects on terrestrial-plant photosynthetic metabolism and subsequent growth, elevated [C02 L can cause partial stomatal closure, although whole-ecosystem water use may, at the same time, be unaltered. In any case, elevated [C0 2 l a generally increases the ratio of the amount of CO 2 assimilated (or amount of plant grown) to the amount of water transpired. As a result, more growth is possible for a given amount of precipitation or soil moisture. One possible effect of this increased water­ use efficiency is an expanded range of plants in arid and semi-arid regions as well as enhanced growth by plants already occupying arid and semi-arid areas. This could lead to the storage of additional C in plants and soils of these regions. Increasing [C0 2 L may benefit C 3 plants more than it benefits C 4 plants, but climatic change accompanying increasing [C02 l a may confer relative advantages to C 4 plants, so the overall impact of global environmental change on competition between C 3 and C 4 plants is unknown. And, the long term (decadal) growth responses of perennial plants to elevated [C0 2l a are unclear. To the extent that an increase in [C02 l a causes an increase in plant growth and (or) an increase in the geographical range of plants, the amount of plant litter

produced will increase globally, i.e. the input of plant C to the global soil C pool will increase. Because the chemical composition of dead leaves allowed to senesce and abscise naturally was not significantly affected by [C02 l a in the few experiments germane to this issue, it must presently be concluded that the net effects of increased leaf growth (if any) caused by elevated lC0 2L will result in larger soil C pools if other factors such as climate remain sensibly constant. Little is known of the effects of elevated [C0 2L on root or (woody) stem litter quality, so although it is likely that rates of C input to soil from roots and stems will increase with an increase in [C02 ]." effects of elevated [C0 2 L, on whole-plant litter decomposition rate are unknown. In general, increasing [C0 2L may lead to greater input of organic C to soils, but warming might cause increased rates of soil organic matter decomposition. 3 Although most plant responses to [C0 2l a act as negati\'e feedbacks on the increase in lC0 2 L, those plant responses cannot end the increase in [C02 la in the next several decades. Inputs of CO 2 to the atmosphere from fossil fuel combustion are too large, and it is unlikely that reductions in fossil fuel use will be significant - or occur at all ­ during the next several decades. Moreover, further cli­ matic change associated with increasing [C0 2 L may be more of a detriment to C storage on land than a benefit, at least in the long term. Therefore, whole-ecosystem responses to continued global environmental change and human land-use changes may be as likely to stimulate the future increase in [C02 L as they are to dampen it.

Acknowledgements This paper is dedicated to Spike Goodbody He made critical contributions to the manuscript, but died just as it was com­ pleted. Mv financial support for the preparation of this paper came from I ,awrcnce Livermore National Laboratory's I.aborat­ ory Directed I,t'search and Dl'\'elopment Program (9'1-Dl-OO'i) under the auspices of the U.S, Department of Energy, Environ­ mental Sciences Division (contract No. W-7405-Eng-48).

References Ajtay GL, KclJll'r P, Duvigneaud P (1979) Terrestrial primary production and phytomass. In:lhc Global Carbon eljele (eds Bolin B, Degens ET, Kempe S, Ketner P), pp. 129-181. Wiley, Chichester. Allen LH Jr, Amthor JS (1995) Plant physiological responses to elevated CO 2 , temperature, air p(lllution, and UV-B radiation.

30 ne of the anonymous reviewers of this paper commented that "some [readers] might consider [my] conclusions [about global terrestrial higher-plant response to incrcc1sing atmospheric CO 2 concentration to be] a little wishy-washy" That reviewer went on to say that "this cautious approach is the only intellectuc11ly honest one to a question of this complexity" © 1995 Blackwell Science Ltd., Global Challge Riologl/, 1,243-274

T ERR EST R I A L - P LAN T RES P 0 N SET 0 AT M 0 S P HER 1C [C 0 Tn: Biotic Feedbacks ill the Clohal Climatic SYSIt'III (eds Woodwell GM, Mackenzie FT), pp. 51-84. Oxford University Press, New York. Allen LH Jr, Drake BG, Rogers HH, Shinn JH (1992) Field techniques for exposure of plants and ecosvstems to elevated CO 2 and other trace gases. Critical Reviews ill [,lallt Sciences, 11, 8.~-119.

Amiotte Suchet P, Probst)L (lLJ95) A global model for present-day atmospheric/soil CO2 consumption by chemical erosion of continental rocks (GEM-C02), Tel/lis, 47B, 273-280. Amthor JS (1991) Respiration in a future, higher-C02 world. Plant, Cell Wid Ellviro/lll/rnt, 14, 13-20. Amthor jS (1994a) Plant respiratory responses to the environment and their effects on the carbon balance. In: I'lllllt EIlZ'ironment interactions (ed. Wilkinson RE), pp. 50]-554. Marcel Dekker, New York. Amthor JS (1994b) Respiration and carbon assimilate use. In: Physiology and Determination ofCrop Yield (eds Boote KJ, Bennett JM, Sinclair TR, Paulsen C;M), pp. 221-250. American Society of Agronomy, Madison, WI. Amthor jS (1996) Plant respiratory responses to ele\'ated CO2 partial pressure. In: Advances in Carhon Dioxide Effects Resmrch (eds Allen LH Jr, Kirkham MB, Olszyk DM, Whitman C), pp., in press. American Society of Agronomy, Madison, Wisconsin. Amthor JS, Koch GW (1996) Biota growth factor~: stimulation of terrestrial ecosystem productivity by elevated atmospheric CO 2 , In: Carbon Dioxide lind Terrestrial EUNlstems (eds Koch GW, Mooney HA), pp. 399-414. Academic Press, San Diego. Amthor JS, Loomis RS (1996) Integrating knowledge of crop responses to elevated CO2 and temperature with mechanistic simulation models: model components and research needs. In: CariJon Dioxide and Terrestrial Ecosystellls (eds Koch GW, Moonev HA), pp. 317-345. Academic Press, San Diego. Amthor jS, Koch GW, Bloom Aj (1992) CO 2 inhibits respiration in leaves of RlImex crispus L. Plant Physiology, 98, 757-760. Amthor JS, Mitchell RJ, Runion GB, Rogers HH, Prior SA, Wood CW (1994) Energy content, construction cost and phytomass accumulation of GllIcine max (L.) Merr. and Sorgh1lm hicolor (L.) Moench grown in elevated CO 2 in the fkld. NCiI' Phytologist, 128,443-450. Ande'rson jM (1992) Responses of soils to climate' change'. Advances in Ecological Resmrch, 22, 163-210. Archer S, Schimel DS, Holland EA (1995) Mechanisms of shrubland expansion: land use, climate or CO2? Climatic Challge, 29, 91-99. Armentano TV, Ralston CW (1980) The role of temperate zone forests in the global carbon cycle. Canadillll IUlImal of Forest Research, 10, 53-60. Arp WJ (1991) Effects of source-sink relations on photosynthetic acclimation to elevated CO2 , Plant, Cdl and Environment, 14, 869-875. Arrhenius S (1896) On the influence of carbonic acid in the air upon the temperature of the ground. London, Edinburgh, lind Dublin Philosophical Magazine and JOllrnal of Science (Series 5), 41,237-276. Azc6n-Bieto J (1983) Inhibition of photosynthesis by carbohydrates in wheat Imws. Plant PhysioloSlj, 73, ii81-686. Bacastow R, Keeling CD (1973) Atmospheric carbon dioxide and

© 1995 Blackwell Science Ltd., Global Challge Biology, 1, 243-274

2

1

267

radiocarbon in the mtural carbon cycle: II. Changes from A.D. 1700-2070 as deduced from a geochemical model. In: Carbon and the Biosphere (eds Woodwell eM, Pecan EV), pp. 86-135. Atomic Energy Commission, National Technical Information Scrvice, U.s. Department of Commercc, Springfield, Virginia. Baker jT, Laugel F, Boote KJ, Allen LH Jr (19lJ2) EHects of daytime carbon dioxide concentration on dark respiration in rice. Plant, Cell and Environment, 15, 231-239. Barnola JM, Raynaud D, Korotkevich YS, Lorius C (1987) Vostok ice core provides, 160,000-year record of atmospheric CO 2 , Nature, 329, 408--414. Barnola J-M, Pimienta P, Ravnaud D, Korotkevich YS (199]) COc­ climate relationship as deduced from the Vostok ice core: a re­ examination based on new measurements and on a re­ evaluation of the air dating. Telllls, 43B, 83-LJO. Barnola JM, Anklin M, Porcheron f, Raynaud D, Schwander J, Stauffer B (1995) CO2 evolution during the last millennium as recorded by Antarctic and Greenland ice. Tel/us, 47B, 264-272. Bavsdorfer C, Van Der Woude W (1988) Carbohydrate responsive proteins in the roots of Pcnllisetumamericlllllllll. Plant Physiology, 87,566-570. Bazzaz FA (19LJO) The response of natural ecosystems to the rising global CO 2 levels. Annual Revif(c of Ecology and Systelllatics, 21, 167-196. Benling DJ, Woodward FT (1993) Ecophvsiological responses of plants to global em'ironmental change since the Last Glacial Maximum. NezI' I'hytologist, 125, 641-648. Beerling DJ, Woodward Fl (1995) Stomatal responses of variegated. leaves to CO2 enrichment. Annals of Botany, 75,507-511. Beevers H (1970) Respiration in plants and its regulation. In: Prediction alld Measurement of I'hotosllnthetic Prodllctivity (ed. Setlik I), pp. 209-214. rudoc, Wageningen, The Netherlands. Beevers H (1974) Conceptual developments in metabolic control, 1924-1974. Pia II t PllIlsiology, 54, 437-442. Berner RA (1992) Palaeo-C02 and climate. Nature, 358,114. Berner RA (1994) GEOCARB II: a re\'ised model of atmospheric CO 2 over phanerozoic time. AmCYican Journal of Science, 294, ')6-9].

Berntson GM, McConnaughay KDM, Bazzaz FA (1993) Elevated CO 2 alters deployment ot" roots in 'small' growth containers. Oecologia, 94, 558-Sii4. Bingham If, Farrar JF (1988) Regulation of respiration in roots of barley. Physiologia Plantarulll, 73, 278-285. Bingham If, Stevenson EA (1993) Control of root growth: effects of carbohydrates on the extension, bnmching and rate of respiration of different fractions of wheat roots. Physiologia Plllntarum, 88, 149-]58. Bird MI, Lloyd J, Farquhar GD (] 994) Terrestrial carbon storage at the LGM. Nature, 371, 566. Bolin B, Fung I (1992) Report: the carbon cycle revisited. In: Modeling the Earth System (ed. Ojima D), pp. 151-164. UCAR/ Office for Interdisciplinary Earth Studies, Boulder, Colorado. Bormann FH, Likens GE (1979) Pattem and Process in a Forested Ecosystem. Springer-Verlag, New York, 253 pp. Botkin DB, Simpson LC (1990) Biomass of the North American boreal forest: a step toward accurate global measures. Biogeochemistry, 9, ]iil-174. Bouma TJ, De Visser R, Janssen JHJ A, De Kock MJ, Van Leeuwen

268

J . S. A M THO R

PH, Lambers H (1994) Respiratory energy requirements and rate of protein turnover in vivo determined by the use of an inhibitor of protein synthesis and a probe to assess its effect. Physiologia Plantarum, 92, 585-594. Boutton TW, Archer SR, Nordt LC (1994) Climate, CO2 and plant abundance. Nature, 372, 625-626. Bowen HJM (1966) Trace Elements in Biochemistry. Academic Press, London, 241 pp. Bowes G (1991) Growth at elevated CO2 : photosynthetic responses mediated through Rubisco. Plant, Cell and Environment, 14, 795-806. Bowes G (1993) facing the inevitable: plants and increasing atmospheric CO 2 , Annual Review of Plant Physiology lind Plllnt Molecular Biology, 44, 309-332. Box E (1975) Quantitative evaluation of global primary productivity models generated by computers. In: Primary Productivity of the Biosphere (eds Lieth H, Whittaker RH), pp. 265-283. Springer-Verlag, New York. Bunce JA (1990) Short- and long-term inhibition of respiratory carbon dioxide efflux by elevated carbon dioxide. Annals of Botany, 65, 637-{'42. Bunce JA (1992) Stomatal conductance, photosynthesis and respiration of temperate deciduous tree seedlings grown outdoors at an elevated concentration of carbon dioxide. Plalll, Cell lind Environment, 15, 541-549. Bunce JA (1995a) The effect of carbon dioxide concentration on respiration of growing and mature soybean leaves. Plllnt, Cl'il /llld Em'irol1lllcnt, 18, .:;75-58l. Bunce J A (I 995b ) EHects of elevated carbon dioxide concentration in the dark on the growth of soybean seedlings. Annals of Botany, 75, 365-368. Bunce JA, Caulfield F (1991) Reduced respiratory carbon dioxide efflux during growth at elevated carbon dioxide in three herbaceous perennial species. Annals of Botany, 67, 325-330. Byrd GT (1992) Dark respiration in C 3 and C 4 species. PhD thesis, University of Georgia, Athens, Georgia, USA, 116 pp. Caemmerer S von, Evans JR (1991) Determination of the average partial pressure of CO 2 in chloroplasts from leaves of several C" plants. Australian Journal of Plant Physiology, 18, 287-30::;. Cebrian J, Duarte CM (1995) Plant growth-rate dependence of detrital carbon storage in ecosystems. Science, 268, 1606-1608. Cerling TE, Wang Y, Quade J (1993) Expansion of C4 ecosystems as an indicator of global ecological change in the late \1iocene. Nature, 361, 344-345. Ceulemans R, Mousseau M (1994) Effects of elevated atmospheric CO2 on woody plants. New Phytologist, 127, 425-446. Cheng C -L, Acedo GN, Cris ti nsin M, Con kling MA (1992) Sucrose mimics the light induction of Arabidopsis nitrate reductase gene transcription. Proceedings of the National Academy of Sciences, USA,89,1861-1864. Ciais P, Tans PP, White JWC et al. (1995) Partitioning of ocean and land uptake of CO2 as inferred by ODC measurements from the NOAA Climate Monitoring and Diagnostics Laboratory Global Air Sampling Network. Journal of Geophysical Research, 100,5051-5070. Cole DR & Monger HC (1994) Influence of atmospheric CO2 on the decline of C4 plants during the last deglaciation. Nature, 368, 533-536.

n,

Cole Caraco NF, Kling GW, Kratz TK (1994) Carbon dioxide supersaturation in the surface waters of lakes. Sciena, 265, 1568-1570. Conway TJ, Tans PP, Waterman LS, Thoning KW, Kitzis DR, Masarie KA, Zhang N (1994) Evidence for interannual variability of the carbon cycle from the National Oceanic and Atmospheric Administration/Climate Monitoring and Diagnostics Laboratory global air sampling network. Journal of Geophysical Research, 99, 22, 831-22, 855. Crutzen PI. Andreae MO (1990) Biomass burning in the tropics: impact on atmospheric chemistry and biogeochemical cycles. Science, 250,1669-1678. Curtis PS, TeeriJA (1992) Seasonal responses of leaf gas exchange to elevated carbon dioxide in Popuills gmndidcntllta. Canadian Journal of Forest Research, 22, 1320-1325. Curtis PS, Drake BG, Whigham OF (1989) Nitrogen and carbon dynamics ir C 3 and C 4 estuarine marsh plants grown under elevated CO 2 in situ. Oecologia, 78, 297-30l. Dacey JWH, Drake BG, Klug MJ (1994) Stimulation of methane emission by carbon dioxide enrichment of marsh vegetation. Nature, 370, 47-49. Den Hertog J, Stulen I, Lambers H (1993) Assimilation, respiration and allocation of carbon in Plantago major as affected by atmospheric CO 2 levels. Vcgetatio, 104/105, 369-378. De Visser R, Spitters CJT, Bouma TJ (1992) Energy cost of protein turnover: theoretical calculation and experimental estimation from regression of respiration on protein concentration of full­ grown leaves. In: Molecular, Biochemical IIlld Physiological Aspects of Plant Respiration (eds Lambers H, van der Plas LHW), pp. 493-508. S1'B Academic Publishing, The Hague, The Netherlands. Dickson 0 (1995) Aerosols' role simulated in new global warming model. Nature, 374, 487. Dlugokencky EJ, Lang PM, Masarie KA, Steele LP (1994) Atmospheric CH 4 records from sites in the NOAA/CMDL air sampling network. In: Trends '93: A Compendium of Data on Global Change (eds Boden TA, Kaiser Of', Sepanski RJ, Stoss FW), pp. 274-350. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, Oak Ridge, TN. Oownton WJS, Grant WJR (1994) Photosynthetic and growth responses of variegated ornamental species to elevated CO2 , Australian Journal of Plant Ph1l';iology, 21, 273-297. Drake BG (1992) A field study of the effects of elevated CO 2 on ecosystem processes in a Chesapeake Bay wetland. Australian Journal of Botany, 40, 579-595. Drake BG, Leadley PW (1991) Canopy photosynthesis of crops and native plant communities exposed to long-term elevated CO 2 , Plallt, Cell alld Environment, 14, 853-860. Du Cloux He, Andre M, Daguenet A, Massimino J (1987) Wheat response to CO2 enrichment: growth and CO2 exchanges at two plant densities. Journal of Experimental Botany, 38,1421-1431. Dugas WA, Heuer ML, Hunsaker 0, Kimball BA, Lewin KF, Nagy J, Johnson M (1994) Sap flow measurements of transpiration from cotton grown under ambient and enriched CO 2 concentrations. Agricu !tural and Forest Meteorology, 70, 231-245. Eamus 0 (1991) The interaction of rising CO2 and temperatures with water use efficiency. Plant, Cell and Environment, 14, 843­ 852. 1995 Blackwell Science Ltd., Global Clillilge Biology, 1, 243-274

T ERR EST R I A L - P LAN T RES P 0 N SET 0 AT M 0 S P HER 1 C [C 0 2l Eamus D, Jarvis PG (1 Y8Y) The direct effects of increase in the global atmospheric CO 2 concentration on natural and commercial temperate trees and forests. Advances in Ecological Research, 19, 1-55. Edwards G, Walker D (1983) C3, C4: Mechal1isms, and Cellular al1d Environmental Regulation, of Photosynthesis. University of California Press, Berkeley, 542 pp. Ehlcringer JR, Monson RK (1993) Evolutionary and ecological aspects of photosynthetic pathway variation. Anllual Review of Ecology allli Systell/atics, 24, 411-439. Ehleringer JR, Cerling TE (1Y%) Atmospheric CO 2 and the ratio of intercellular to ambient CO 2 concentrations in plants. Tree Physiology, 15, lO5-111. Ehleringer JR, Sage RF, Flanagan I.B, Pearcy RW (1991) Climate change and the evolution of C 4 photosynthesis. Trends in Ecology & Emlutioll, 6, 95-99. Ellis TJ (1979) The most abundant protein in the world. Trends in Biochemical Sciences, 4, 241-244. Estiarte M, Penuelas J, Kimball BAct al. (1 YY4) Elevated CO 2 effects on stomatal density of wheat and sour orange trees. Journal of Experimental Botany, 45, 1665-1668. Eswaran H, Van Den Berg E, Reich P (1993) Organic carbon in soils of the world. Soil Science Society of America JOllrnal, 57, 192-194. Evans LT (19Y3) Crop E1'ollition, Adaptation and Yield. Cambridge University Press, Cambridge, 500 pp. Farquhar GD, Sharkey TD (1 Y82) Stomatal conductar.ce and photosynthesis. Allllulll R.eview of Plant Physiology, 33, 317-345. Farrar JF, Will iams JHH (I Y91 ) Control of the rate of respiration in roots: compMtmentation, demand and the supply of substrate. Society for Experimental Biology seminar series, 42, 167-188. Ferris R, Taylor G (1994) Stomatal characteristics of four native herbs following exposure to elevated CO 2 , Allnals of Botany, 73, 447-453. Field CB, Chapin FS III, Matson PA, Mooney HA (1 Y92) Responses of terrestrial ecosystems to the changing atmosphere: a resource-based approach. Annual Review of Ecology and Systematics, 23, 201-235. Fisher MJ, Rao 1M, Ayarza MA, I .ascano CE, Sanz JI, Thomas RJ, Vera RR (1994) Carbon storage by introduced deep-rooted grasses in the South American savannas. Nature, 371, 236-238. Foley JA (1994) Net primary productivity in the terrestrial biosphere: the application of a global model. Journal of Geophysical RcsCllrcli, 99 (20), 773-20, 783. Foyer CH (1988) Feedback inhibition of photosynthesis through source-sink regulation in leaves. Plant Phl/siology and Biochemistry, 26, 483-492. Francey RJ, Tans PP, Allison CE, Enting IG, White JWc, Trolier M (1995) Changes in oceanic and terrestrial carbon u pta ke since 1982. Nature, 373, 326-330. Friedli H, Moor E, Oeschger H, Siegenthaler U, Stauffer B (1984) llC I 12C ratios in CO 2 extracted from Antarctic ice. Geopizysical Research Lellers, 11, 1145-1148. Friedli H, l.otscher H, Oeschger H, Siegenthaler U, Stauffer B (1986) Ice core record of the 13C/12C ratio of atmospheric CO2 in the past two centuries. Nature, 324, 237-238. Gale J (1982) Evidence for essential maintenance respiration of !eaves of Xanthium strumarium at high temperature. Jounwi of ExperimelitaIHotmll/, 33, 471-476. © 1995 Blackwell Science Ltd., Glohal Change Biology, 1, 243-274

269

Cates DM (1985) Global biospheric response to increasing atmospheric carbon dioxide concentration. In: Direct Effects of Increasing Carbon Dioxide on Vegetation (eds Strain BR, Cure JD), pp. 171-185. Department of Energy, National Technical Information Service, U.s. Department of Commerce, Springfield, Virginia. Gaudillere J-p, ~ousseau M (1989) Short term effect of CO 2 enrichment on leaf development and gas exchange of young poplars (Populus eummericalla cv r 214). Acta Oecoiogical Oeeologia Plantarum, 10,95-105. Ceigl'l1berger P, Stitt M (1991) A 'futile' cvcle of sucrose synthesis and degradation is involved in regu lating partitioning between sucrose, starch and respiration in cotvledons of germinating Ricinus communis L. seedlings when phloem transport is inhibited. Planta, 185, 81-YO. Gerlach TM (1991) Present-day CO 2 emissions from volcanos. EOS, 72, 24Y, 254-255. Gifford RM (1 Y77) Growth pattern, carbon dioxide exchange and dry weight distribution in wheat growing under differing photosynthetic environments. Australian Journal of Piant Physiology, 4, 99-11 O. Gifford RM (1980) Carbon storage by the biosphere. In: Carboll Dioxide and Climate: Australian R.esearch (ed. Pearman GI), pp. 11>7-181. Australian Academy of Science, Canberra. Corham E (1995) The biogeochemistry of northern peatlands and its possible responses to global warming. In: Hiotie Feed/Jacks in the Glohal Climatic System (eds Woodwell CM, Mackem:ie FT), pp. 169-187. Oxford University Press, New York. Goudriaan J (1993) Interaction of ocean and biosphere in their transient responses to increasing atmospheric CO 2 , Vegetatio, 104/105,329-337. Goudriaan J, van Laar HH, van Keulen H, Louwerse W (1985) Photosynthesis, CO 2 and plant production. In: Wheat Growth and Modelil1g (eds Day W, Atkin RK), pp. 107-122. Plenum Press, New York. Grace L Lloyd J, McIntyre J et al. (1995) Fluxes of carbon dioxide and water vapour over an undisturbed tropical forest in sou th­ west Amazonia. Global Change Biologl/, 1, 1-12. Criffin KL, Thomas RB, Strain BR (1993) Effects of nitrogen supply and e!e\'ated carbon dioxide on construction cost in leaves of Pinus taClia (L.) seedlings. Oecologia, 95, 575-580. Cunderson CA, Wullschleger SD (1994) Photosynthetic acclimation in trees to rising atmospheric CO 2 : a broader perspecti\·e. Photosynthesis Research, 39, 369-388. Gunderson CA, Norby RJ, Wullschleger SD (1993) Foliar gas exchange responses of two deciduous hardwoods during 3 years of growth in elevated COi no loss of photosynthetic enhancement. Plant, Cell and Environment, 16, 797-807. Hampicke U (1980) The effect of the atmosphere-biosphere exchange on the global carbon cycle. Experientia, 35, 776-781. Harden JW, Sundquist ET, Stallard Rf, Mark RK (1992) Dvnamics of soil carbon during deglaciation of the Laurentide Ice Sheet. Science, 258,1921-1924. Harley Pc, Sharkey TD (1991) An improved model of Cl photosynthesis at high CO 2 : reversed O 2 sensitivity explained by lack of glycerate reentry into the chloroplast. Photosynthesis Research, 27, 169-178. Harmon ME, H ua C (1991) Coarse woody debris dynamics in two old-growth ecosystems. BioScience, 41, 604-610.

270

J . S.

AM THO R

Harter K, Talke-Messerer C, Barz W, SchJfer E (1993) Light and sucrose-dependent gene expression in photomixotrophic cell suspension cultures and protoplasts of rape. Plant Journal, 4, 507-516. Henderson S, Hattersley P, von Caemmerer S, Osmond CB (1994) Are C 4 pathway plants threatened by global climatic change? In: Ecophysiology of Photosynthesis (eds Schulze E-D, Caldwell MM), pp. 529-549. Springer-Verlag, Berlin. Hendrix DL, Mauney JR, Kimball BA, Lewin K, Nagy J, Hendrey CR (1994) Influence of elevated CO2 and mild water stress on nonstructural carbohydrates in field-grown cotton tissues. Agricultural alld Forest Meteorology, 70, 153-162. Herold A (19S0) Regulation of photosynthesis by sink activity ­ the missing link. New Phytologist, 86, 131-144. Hileman DR, Huluka C, Kenjige PK et al. (1994) Canopy photosynthesis and transpiration of field-grown cotton exposed to free-air CO2 enrichment (FACE) and differential irrigation. Agricultural and Forest Meteorology, 70, 189-207. Hogan KP, Smith AI', Ziska LH (1991) Potential effects of elevated CO 2 and changes in temperature on tropical plants. Plant, Cell and Environment, 14, 763-778. Houghton RA (1995) Land-use change and the global carbon cycle. Global Change Biology, 1, 275-287 (this issue). Houghton RA, Hobbie JE, Melillo JM, Moore B, Peterson BJ, Shaver CR, Woodwell CM (1983) Changes in the carbon content of terrestrial biota and soils between IS60 and 1980: a net release of CO2 to the atmosphere. Ecological MOllographs, 53, 235-262. Houghton RA, Boone RD, Fruci JR et al. (1987) The flux of carbon from terrestrial ecosystems to the atmosphere in 19S0 due to changes in land use: geographic distribution of the global flux. TeIlLls, 39B, 122-139. Hughes AP, Cockshull KE (1972) Further effects of light intensity, carbon dioxide concentration, and day temperature on the growth of CimjsantilemLlm morifoliIlIll c\'. Bright Golden A II lie in controlled environments. Annals of Botany, 36, 533-550. Hunsaker OJ, Hendrey CR, Kimball BA, Lewin KF, Mauney JR, Nagy J (1994) Cotton evapotranspiration under field conditions with CO2 enrichment and variable soil moisture regimes. Agricultural and Forest Metearologlj, 70,247-258. Idso SB (1991a) Comment on 'Modelling the seasonal contribu tion of a CO2 fertilization effect of the terrestrial vegetation to the amplitude increase in atmospheric CO2 at Mauna Loa Observatory' by C.H. Kohlmaier et al. Tel/us, 43B, 338-341. Idso SB (1991b) The aerial fertilization effect of CO 2 and its implications for global carbon cycling and maximum greenhouse warming. Bulletin of the American Meteorological Society, 72, 962-965. ldso SB (1991c) Reply to comments of L.D. Danny Harvey, Bert Bolin, and P. Lehmann. Bulletin of the American Meteorological Society, 72, 1910-1914. Idso SB, Kimball BA (1991) Downward regulation of photosynthesis and growth at high CO2 levels. Plant Physiology, 96,990-992. Idso SB, Kimball BA (1993) Tree growth in carbon dioxide enriched air and its imp1icationsfor global carbon cycling and maximum levels of atmospheric CO 2 , Global Biogeochemical Cycles, 7, 537-555. Jacob L Creitner C Drake BC (1995) Acclimation of photosynthesis

in relation to rubisco and nonstructural carbohydrate contents and in situ carboxylase activity in Scirpus olneyi grown at elevated CO2 in the field. Plant, Cell and EnvironmCllt, 18, 875­ 884. Jacobs CMJ, de Bruin HAR (1992) The sensitivity of regional transpiration to land-surface characteristics: significance of feedback. jOllmal of Climate, 5, 6S3-698. Jarvis PC (1989) Atmospheric carbon dioxide and forests. Philosophical Transactiolls of the Royal Society, B, 324, 369-392. Jarvis PC, McNaughton KC (1986) Stomatal control of transpiration: scaling up from leaf to region. Advances in Ecological Research, 15, 1-49. Jenkinson OS, Adams DE, Wild A (1991) Model estimates of CO 2 emissions from soil in response to global warming. Natllre, 351, 304-306. Jones PO, Wigley ML, Briffa KR (1994) Clobal and hemispheric temperature anomalies-land and marine instrumental records. In: Trends '93: A Compendium of Data 011 Global Change (eds TA Boden, DP Kaiser, RJ Sepanski, FW Stoss), pp. 603­ 608. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, Oak Ridge, TN. Jordan DB, Ogren WL (1984) The CO2 /02 specificity of ribulose-I, 5-bisphosphate carboxylase/oxygenase. Planta, 161, 308-313. Jouzel J, Barkov NI, Barnola JM et al. (1993) Extending the Vostok ice-COTe record of palaeoclimate to the penultimate glacial period. Nature, 364, 407-412. Jozsa LA, Powell JM (1987) Some climatic aspects of biomass productivity of white spruce stem wood. Canadian journal of Forest Research, 17,1075-1079. Kasting IF (1993) Earth's early atmosphere. Scicnce, 259, 920-926. Keeling CD (1973) Industrial production of carbon dioxide from fossil fuels and limestone. Tel/liS, 25, 174-198. Keeling CO, Whorf TP (1994) Atmospheric CO 2 records from sites in the S10 air sampling network. In: Trends '93: A COlllpendiulll of Data on Global Change (eds Boden TA, Kaiser OP, Sepanski RJ, StossFW), pp.16-26. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, Oak Ridge, TN. Keeling CD, Whorf TP, Whalen M, van der Plicht J (1995) Interannual extremes in the rate of rise of atmospheric carbon dioxide since 1980. Nature, 375, 666-670. Kemp PR, Waldecker DC, Owensby CE, Reynolds )F, Virginia RA (1994) Effects of elevated CO 2 and nitrogen fertilization pretreatments on decomposition on tallgrass prairie leaf litter. Plant and Soil, 165,115-127. Kimball BA, LaMorte RL, Seay RS et til. (1994) Effects of free-air CO 2 enrichment on energy balance and evapotranspiration. Agricultllral lind Forest Meteorology, 70, 259-278. Kirschbaum MUF (1994) The sensitivity of C 3 photosynthesis to increasing CO2 concentration: a theoretic a I analysis of its dependence on temperature and background CO2 concentration. Plant, Cell and Environment, 17,747-754. Koch CW (1993) The use of natural situations of CO2 enrichment in studies of vegetation responses to increasing atmospheric CO 2 , In: Design and Execution of Experiments 0/1 CO 2 Enrichment (eds Schulze ED, Mooney HA), pp. 3SI-391. Office for Official Publications of the European Communities, Luxembourg. Koch CW, Mooney HA (eds) (1996) Carbon Dioxide and Terrestrial Ecosystems. Academic Press, San Diego. © 1995 Blackwell Science Ltd., Global Changc Biology, 1, 243-274

T ERR EST R I A L - P LAN T RES rON SET 0 AT M 0 S P HER 1C [C 0 Kohlmaier GH, Sire E-O, Janecek A, Keeling CD, Piper SC, Revelle R (1989) Modelling the seasonal contribution of a CO2 fertilization effect of the terrestrial vegetation to the amplitude increase in atmospheric CO2 at Mauna Loa Observatory. Tellus, 41B,487-510. Kohlmaier GH, Revelle R, Keeling CD, Piper SC (1991) Reply to Idso. Telius, 43B, 342-346. Kiirner C (1988) Does global increase of CO 2 alter stomatal density? Flora, 181, 251--257. KbrnerC, Miglietta F (1994) Long term effects of naturally elevated CO2 on mediterranean grassland and forest trees. Oecologill, 99, 343--351 . Kramer PJ (1 Y81) Carbon dioxide concentration, photosynthesis, and dry matter production. BioScience, 31, 29-33. Krapp A, Stitt M (1994) lnlluence of high carbohydrate content on the activity of plastidic and cytosolic isoenzyme pairs in photosynthetic tissues. Plant, Cell alld Environment, 17, 861-866. Krapp A, Quick WP, Stitt M (1991) Ribulose-l,5-bisphosphate carboxylase-oxygenase, other Calvin-cycle enzymes, and chlorophyll decrease when glucose is supplied to mature spinach leaves via the transpiration stream. Planta, 186, 5S-69. Krapp A, Hofmann B, Schafer C, Stitt M (1993) Regulation of the expression of rbcS and other photosynthetic genes by carbohydrates: a mechanism for the 'sink regulation' of photosyntlwsis 7 Plant lou,.,1I11, 3, H17-828. Kuehny JS, Peet MM, Nelson PV, Willits DH (1991) Nutrient dilution bv starch in CO 2-enriched Chrysanthemum. IOllrll1l1 of Experimental Botlllllj, 42, 711-716. Kurz WA, Apps ML Beukema SJ, Lekstrum T (1995) 20th century carbon budgl'l of Canadian forests. Tellus, 47B, 170-177. Lasaga AC, Berner RA, Garrels RM (1985) An improved geochemical model of atmospheric CO 2 fluctuations over the past 100 million years. In: The Carbon Cycle IIlId Atnwspheric CO 2: Natural Variations Archean to Presellt (cds Sundquist ET, Broecker WS), pp. 397--411. American Ceophysical Union, Washington, DC. Lawlor OW, Mitchell RAC (1991) The effects of increasing CO2 on crop photosynthesis and productivity: a review of field studies. Plant, Cell alld T:I1('iroIlI1lCllt, 14, 807-818. Leavitt SW (19H2) Annual volcanic carbon dioxide emission: an estimate from eruption chronologies. Environmental Gcologlj, 4, 15-21. Lelieveld J, Crutzen PJ (1 Y92) Indirect chemical effects of methane on climate warming. Natllre, 355, 339-342. Lieth H (1975) Modeling the primary productivity of the world. In: Primary Productivity ofthe Biosphere (eds Lieth H, Whittaker RH), pp. 237-263. Springer-Verlag, New York. Lloyd J, Syvertsen JP, Kriedemann PE, Farqu har GO (1992) Low conductances for CO 2 diffusion from stomata to the sites of carboxylation in leaves of woody species. Plant, Cell and Ellvironment, 15, 873-899. Long SP (1991) Modification of the response of photosynthetic productivity to rising temperature by atmospheric CO2 concentrations: has its importance been underestimated? Plant, Cell and Environment, 14, 729-739. Loomis RS, Lafitte HR (1987) The carbon economy of a maize crop exposed to elevated CO2 concentrations and water stress, as determined from elemental analyses. Field Crops Researcil, 17, 63-74.

© 1995 Blackwell Science Ltd., Glo/Jal Change Biology, 1, 243-274

21

271

Loreto F, Harley pc, Di Marco G, Sharkey TD (19Y2) Estimation of mesophyll conductance toC0 2 1lux by three different methods. Plant Physiologl/, 98, 1437-1443. Manabe 5 (1983) Carbon dioxide and climatic change. Advances in Geophysics, 25, 39-82. Manabe 5, Stouffer RJ (1994) Multiple-century response of a coupled ocean-atmosphere model to an increase of atmospheric carbon dioxide. JOllmal of Climate, 7, 5-23. Marland G, Andres RJ, Boden TA (1994) Global, regional and national CO 2 emissions. In: Trends '9.1: A Cl/llpelidilll1l of Data 0/1 Global Change (eds Boden TA, Kaiser DP, Scpanski RJ, Stoss FW), pp. 508-584. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratorv, Oak Ridge, TN. McConnaughay KDM, Berntson GM, Bazzaz f'A (I YY3a) Plant responses to carbon dioxide. Naturc, 361, 24. McConnaughay KDM, Berntson GM, Bazzaz f'A (1993b) Limitations to COTinduced growth enhancement in pot studies. Occologia, 94, 550-557. McCree KJ (1970) An equation for the rate of respiration of white clover plants grown under controlled conditions. In: Prediction and Measurement ofPllOtosl/ntizetic Productivity (ed. Setlik I), pp. 221-229. Pudoc, Wageningen, The Netherlands. McCree KJ (1981) Photosynthetically active radiation. In: Encyclopcdia of Plant Physiology, Vol. 12A, 1'11l/Siologiml Plant Ecologv I (eds Lange OL, Nobel PS, Osmond CB, Ziegler H), pp. 41-55. Springer-Verlag, Berlin. McLeod AR (1993) Open-air exposure systems for air pollutant studies their potential and limitations. In: Dcsigll and T:xcclitioll of Experiments on CO 2 Eliridllllcllt (eds Schul/c E-D, Mooney HA), pp. 353-365. Office for Official Publications of the European Communities, Luxembourg. Meybeck M (1993) Riverine transport of atmospheric carbon: sources, global typology and budget. Water, Air, and Soil Pollution, 70,443--463. Miglietta F, Raschi A (1 Y93) Studying the effect of elevated CO 2 in the open in a naturally enriched environment in Central Italy. Vegetatio, 104/105, 391--400. Miglietta F, Raschi A, Bettarini 1, Resti R, Selvi f' (J 993a) Natural CO 2 springs in Italy: a resource for examining long-term response of vegetation to rising atmospheric CO2 concentrations. Plant, Cell and Envirolll11cllt, 16, 873-87H. Miglietta ~~ Raschi A, Bettarini I, Badiana 1\1, van Gardingl'n P (l9Y3b) Carbon dioxide springs and their use for l'xperimentation. In: Design and Exccution ofExl'crilllclits Oil CO2 Enrichment (cds Schulze ED, Mooney HA), pp. 393--403. Office for Official Publications of the European Communities, Luxembourg. Miziorko HM, Lorimer GH (1983) Ribulose-1,5-bisphosphate carboxylase-oxygenase. Anllllal Review of Biochemistry, 52, 507-535. Morgan ME, Kingston JD, Marino BD (1994a) Carbon isotopic evidence for the emergence of C4 plants in the Neogene from Pakistan and Kenya. Natllre, 367,162-165. Morgan ME, Kingston JD, Marino BD (1994b) Expansion and emergence of C4 plants. Nlltllre, 371, 112-113. Morison JIL (1987) Intercellular CO 2 concentration and stomatal response to CO 2 , Tn: Stomatal Function (eds Zeiger E, Farquhar GO, Cowan IR), pp. 229-251. Stanford University Press, Stanford, CA.

272

J . S. A M THO R

Morison JIL, Cifford RM (1983) Stomatal sensitivity to carbon dioxide and humidity. A comparison of two C 3 and two C 4 grass species. Plant Physiology, 71, 789-796. Mott KA (1988) Do stomata respond to CO2 concentrations other than intercellular? Plant Physiology, 86, 200-203. Mousseau M (1993) Effects of elevated CO 2 on growth, photosynthesis and respiration of sweet chestnut (Castanea sativa Mill.). Vegetatio, 104/105, 413-419. Murphy JM (1995) Transient response of the Hadlev Centre coupled ocean-atmosphere model to increasing carbon dioxide. Part Ill: analvsisof global-mean response using simple models. journal ofClill1ale, 8, 496-514. Neftel A, Moor E, Oeschger H, Stauffer B (1985) Evidence from polar ice cores for tlw increase in atmospheric CO2 in the past two centuries. Nature, 315,45-47. Nepstad DC, de Carvalho CR, Davidson EA et Ill. (1994) The role of deep roots in the hydrological and carbon cycles of Amazonian forests and pastures. Nature, 372, 666-669. Nijs I, Impens I, Behaeghe T (1989) Leaf and canopy responses of Lolium perenne to long-term elevated atmospheric carbon­ dioxide concentration. Planta, 177, 312-320. Norby RI, O'Neill EG, Luxmoore RJ (1986) Effects of atmospheric CO 2 enrichment on the growth and mineral nutrition of Quercus III/m seedlings in nutrient-poor soil. Plant I'/ll/siologlf, 82,83-1)LJ. Norbv RJ, Gunderson CA, Wulischleger SD, O'Neill EG, McCracken MK (1992) Productivity and compensatory responses of vellow-poplar trees in elevated CO 2, Nlltllrl', 357, 322-324. Oechel WC, Hastings Sj, Vourlitis G,Jenkins M, Riechers G, Grulke N (1993) Recent change of Arctic tundra ecosystems from a net carbon dioxide sink to a source. Naftlre, 361, 520-523. Olson JS, Watts JA, Allison LJ (1983) Carbon in Live Vegetation of Major World Ecosystems, ORNL-5862. Oak Ridge National Laboratory, Oak Ridge, TN, 164 pp. O'Neill EG, Norby RJ (1996) Litter quality and decomposition rates of foliar litter produced under CO2 enrichment. Tn: Carbon Dioxide and Terrestrial Ecosystems (eds Koch GW, Moonev HA), pp. 87-103. Academic Press, San Diego. Osmond CB, Winter K, Ziegler H (1982) Functional significance of different pathways of CO2 fixation in photosynthesis. In: Encyclopedia of I'llInt Physiology, Vol. 12B, Physiological Plant Ecologl/ II (eds Lange OL, Nobel PS, Osmond CB, Ziegler H), pp. 479-547. Springer-Verlag, Berlin. Owensby CE, Coyne PI, Auen LM (lLJ93a) Nitrogen and phosphorus dynamics of a tallgrass prairie ecosystem exposed to elevated carbon dioxide. Plant, Cell and Environment, 16, 843-850. Owensby CE, Coyne PI, Ham JM, Auen LM, Knapp AK (1993b) Biomass production in a tallgrass prairie ecosystem exposed to ambient and elevated CO 2 , Ecological Applications, 3, 644-{)53. Paul MJ, Stitt M (1993) Effects of nitrogen and phosphorus deficiencies on levels of carbohydrates, respiratory enzymes and metabolites in seedlings of tobacco and their response to exogenous sucrose. Plant, Cell lind Environment, 16, 1047-1057. Pearman GI, Hyson P (1981) The annual variation of atmospheric CO2 concentration observed in the northern hemisphere. journal of Ceoplil/sical Research, 86, 9839-9843.

Penning de Vries FWT (1975) The cost of maintenance processes in plant cells. Annals of Botany, 39, 77-92. Penning de Vries FWT, Brunsting AHM, van Laar HH (1974) Products, requirements and efficiency of biosynthesis: a quantitative approach. journal ofTheoretical Biology, 45, 339-377. Penning de Vries FWT, Witlage JM, Kremer D (1979) Rates of respiration and of increase in structural dry matter in young wheat, ryegrass and maize plants in relation to temperature, to water stress and to their sugar content. Amwls of Botany, 44, 595-609. Pefiuelas I, Matamala R (1990) Changes in Nand S leaf content, stomatal density and specific leaf area of 14 plant species during the last three centuries of CO 2 increase. Journal of Experimental Botany, 41, 111LJ-1124. Pettersson R, McDonald AJS (1994) Effects of nitrogen supply on the acclimation of photosynthesis to elevated CO 2. PlwtoSljlltilesis Research, 39, 389-400. Pinter JP Jr, Kimball BA, Garcia RL, Wall GW, Hunsaker OJ, LaMorte RL (1996) Free-air CO2 enrichment: responses of cotton and wheat crops. In: Carbon Dioxide and Terrestrial Ecosystems (eds Koch GW, Mooney HA), pp. 215-249. Academic Press, San Diego. Plass GN (1956) The carbon dioxide theory of climatic change. Tel/us, 8, 140-154. Polley HW, Johnson HB, Marino BD, Maveux HS (1993) Increase in C3 plant water-use efficiency and biomass over Glacial to present CO2 concentrations. Nature, 361, 61-64. Poorter H (1993) Interspecific variation in the growth response to an elevated ambient CO 2 concentration. Vegctatio, 1041105, 77-97. Post WM, Emanuel WR, Zinke PI, Stangenberger AG (1982) Soil carbon pools and world life zones. Nature, 298, 156-159. Potter CS, Randerson JT, Field CB, Matson PA, Vitousek PM, Mooney HA, Klooster SA (1993) Terrestrial ecosystem production: a process model based on global satellite and surface data. Global Biogeochemical Cycles, 7, 811-841. Qi JE, Marshall JD, Mattson KG (1994) High soil carbon dioxide concentrations inhibit root respiration of Douglas fir. New Phylologist, 128,435-442. Quade I, Ceding TE, Bowman JR (1989) Development of Asian monsoon revealed by marked ecological shift during the latest Miocene in northern Pakistan. Nature, 342,163-166. Radin JW, Kimball BA, Hendrix DL, Mauney JR (1987) Photosynthesis of cotton pbnts exposed to elevated levels of carbon dioxide in the field. Photosynthesis Research, 12,191-203. Raich JW, Potter CS (lLJLJS) Global patterns of carbon dioxide emissions from soils. Global Biogeochemical Cycles, 9, 23-36. Raich JW, Schlesinger WH (1992) The global carbon dioxide flux in soil respiration and its relationship to vegetation and climate. TeUus, 44B, 81-LJLJ. Reid CD, Strain BR (1994) Effects of CO2 enrichment on whole­ plant carbon budget of seedlings of Fagus grandifolia and Acer sacchaYllIll in low irradiance. Oecologia, 98, 31-39. Reuveni I, Cale J (1985) The effect of high levels of carbon dioxide on dark respiration and growth of plants. Plant, Cell and El1l'iromnent, 8, 623-628. Rogers HH, Bingham GE, Cure JD, Smith JM, Surano KA (1983) Responses of selected plant species to elevated carbon dioxide in the field. Journal of Enl'ironlnental Qualitl{, 12, 569-574. (cj

1995 Blackwell Science Ltd., Glo/Jal Change Biology, 1, 243-274

T ERR EST R I A L - P LAN T RES P 0 N SET 0 AT M 0 S P HER I C [C 0 Eogers HH, Runion CB, Krupa SV (1994) Plant responses to atmospheric CO 2 enrichment with emphasis on roots and the rhizosphere. Environmental Pollution, 83, 155-189. Rotmans I, den Elzen MGJ (1993) Modelling feedback mechanisms in the carbon cycle: balancing the carbon budget. Tdlu5, 45B, 301-320. Rowland-Bamford AI, Baker JT, Allen LH Jr, Bowes G (1991) Acclimation of rice to changing atmospheric carbon dioxide concentration. Plant, Cell and Environment, 14, 577-583. Ryle GJA, Powell CE, Tewson V (1992a) Effect of elevated CO2 on the photosynthesis, respiration and growth of perennial ryegrass. Journal of Exprrilll1'lltal Botany, 43, 811-818. Ryle GJA, Woledge I, Tewson V, Powell Cr: (1992b) Influence of elevated CO2 and temperature on the photosynthesis and respiration of white clover dependent on N2 fixation. Annals of Botany, 70, 213-220. Sage RF (1994) Acclimation of photosynthesis to increasing atmospheric COi the gas exchange perspective. Photosljnthesis Re:;mrch, 39, 351-368. Sage RF, Reid CD (1994) Photosynthetic response mechanisms to environmental change in C3 plants. In: f'lant-Eni'ironment Interactions (ed. Wilkinson RE), pp. 413-499. Marcel Dekker, New York. Sage RF, Sharkey TD, Seemann JR (1989) Acclimation of photosynthesis to ele\'ated CO2 in five C J species. Plant Phlf:;iology, 89, 590-596. Sarmiento JL, Bender M (1994) Carbon biogeochemistry and climate change. Photo:;ynti1csis Research, 39, 209-234. Sarmiento JL, Sundquist ET (1992) Revised budget for the oceanic uptake of anthropogenic carbon dioxide. Nature, 356, 589-592. Sarmiento J1, Le Quere C, Pacala SW (1995) Limiting future atmospheric carbon dioxide. Global BiogeochclIliml Cycles, 9, 121-137. SchMer C, Simper H, Hofmann B (1992) Glucose feeding results in coordinated changes of chlorophyll content, ribulose-1,5­ bisphosphate carboxylase-oxygenase activity and photosyn­ thetic potential in photoautrophic suspension cultured cells of Chenopodium rubrllln. Plant, Cell and Environment, 15, 343-350. Schimel DS (1995) Terrestrial ecosystems and the carbon cycle. Glt,lilll Change Biology, 1, 77-91. Schimel DS, Braswell BH, Holland EA et al. (1994) Climatic, edaphic, and biotic controls over storage and turnover of carbon in soils. Global Biogeochemical Cycles, 8, 279-293. Schlesinger WH (1982) Carbon storage in the caliche of arid soils: a case study from Arizona. Soil Science, 133, 247-255. Schlesinger WH (1990) Evid enee from chronoseq uence studies for a low carbon-storage potential of soil. Nature, 248, 232-234. Schlesinger WH (1991) Biogeochemistry: An Analysis of Global Change. Academic Press, San Diego, 443 pp. Schlesinger WH (1995) Soil respiration and changes in soil carbon stocks. In: Biotic Feedlmcks ill the Global Clilllatic Sl/stem (eds Woodwell GM, Mackenzie FT), pp. 159-168. Oxford University Press, New York. Schlesinger MW, Ramankutty N (1994) An oscillation in the glob a I climate system of period li5-70 years. Nature, 367, 723-726. Seiler W, Crutzen PJ (1980) Estimates of gross and net fluxes of carbon between the biosphere and the atmosphere from biomass burning. Clilllatic Change, 2, 207-247. © 1995 Blackwell Science Ltd., Global Change Biology, 1, 243-274

2

1

273

Sheen J (1990) Metabolic repression of transcription in higher plants. Plant Cell, 2, 1027-1038. Sheen J (1994) Feed back control of gene expression. PlwtosYIl thesis Research, 39, 427-438. Siegenthaler U, Oeschger H (1987) Biospheric CO 2 emissions during the past 200 years reconstructed by deconvolution of ice core data. Tel/liS, 39B, 140-154. Siegenthaler U, Sarmiento JL (1993) Atmospheric carbon dioxide and the ocean. Nature, 365, 119-125. Silsbury JH, Stevens R (1984) Growth efficiency of Trifolium subterrnneuln at high [C0 2l. In: Advances in Photosynthesis Research, Vol. IV (ed. Sybesma C), pp. 2.122-2.136. Martinus Nijhoff/Dr W Junk Publishers, The Hague. Stitt M, Schaewen A nll1, Willmitzer L (1990) 'Sink' regulation of photosynthetic metabolism in transgenic tobacco plants expressing yeast invertase in their cell wall involves a decrease of the Cal\'in-cycle enzymes and an increase of glycolytic enzymes. Planta, 183, 40-50. Strain BR, Cure JD (eds) (1985) Direct Effect;; oj Increasing Carbon Dioxide on Vegetllti(lll (DOE/ER-0238). U.s. Department of Energy, Washington, DC, 286 pp. SundqUist ET (1993) The global carbon budget. Science, 259, 934­ 941. Taylor KE, Penner JE (1994) Response of the climate system to atmospheric aerosols and greenhouse gases. Natwe, 369, 734­ 737. Teskey RO (1995) A field study of the effects of elevated CO2 on carbon assimilation, stomatal conductance and leaf and branch growth of Pinus taeda trees. Plallt, Cell and Environlllent, 18, 565-573. Thomas RB, Griffin KL (1994) Direct and indirect effects of atmospheric carbon dioxide enrichment on leaf respiration of Glycine max (L.) Merr. Plant Physiology, 104, 355-361. Thomas RB, Strain BR (1991) Root restriction as a factor in photosynthetic acclimation of cotton seedlings grown in elevated carbon dioxide. Plant Physiology, 96, 627-634. Thomas RB, Eeid CD, Ybema R, Strain BR (1993) Crowth and maintenance components of leaf respiration of cotton grown in elevated carbon dioxide partial pressure. I'lant, Cell and En()ironment, 16, 539-546. Thomson DJ (1995) The seasons, global temperature, and precession. Science, 268, 59-68. Thornley JHM (1970) Respiration, growth and maintenance in plants. Natllre, 227, 304-305. Ting IP (1985) Crassulacean acid metabolism. Annual Review oj Plallt Physiology, 36, 595-622. Toumi R, Bekki S, Law KS (1994) Indirect intluence of ozone depletion on climate forcing by clouds. Nature, 372, 348-351. Trumbore SE (1993) Comparison of carbon dynamics in tropical and temperate soils using radiocarbon measurements. Global Biogeochemical Cye/cs, 7, 275-290. Van der Berg I, Visscher H, Dilcher 01, Kiirschner WM (1993) Paleoatmospheric signatures in Neogene fossil leaves. Science, 260, 1788-1790. Van de Water PK, Leavitt SW, Betancourt JL (1994) Trends in stomatal density and l1C / 12C ratios of Pillll; flcxilis needles during last glacial-interglacial cycle. Sciencc, 264, 239-243. Van Oosten J-I, Afif 0, Dizengremel P (1992) Long-term effects of

274

J. S. AM THO R

a CO 2 enriched atmosphere on enzymes of the primarv carbon metabolism of spruce trees. Plant Physiology and Biochelllistry, 30,541-547. Van Oosten J-I, Wilkins D, Besford RT (1994) Regulation of the expression of photosynthetic nuclear genes by CO2 is mimicked by regulation by carbohydrates: a mechanism for the acclimation of photosynthesis to high CO 2 ? Plant, Cell and Environment, 17, 913-923. Villar R, Held AA, Merino J (1994) Comparison of methods to estimate dark respiration in the light in leaves of two woody species. Plant Physiology, 105, 167-172. Vogel CS, Curtis PS (1995) Leaf gas exchange and nitrogen dynamics of 'lrtixing, field-grown Alnus glutinosa undt'r ele\'ated atmospheric CO 2, Global Change Biology, 1, 35-1l1. Walker JCG (1991) Feedback processes in the biogeochemical cycles of carbon. In: Scientists on Gaia (eds SchneiderSH, Boston PJ), pp. 183-190. MIT Press, Cambridge, MA. Watson RT, Meira Filho LG, Sanhueza E, Janetos A (1992) Greenhouse gases: sources and sinks. In: Climate Change 1992 (eds Houghton JT, Callander BA, Varney SK), pp. 25-466. Cambridge University Press, Cambridge. Webber AN, Nie G-Y, Long SP (1994) Acclimation of photosynthetic proteins to rising atmospheric CO 2 , ['izMoslllltizesis Researciz, 39,413-425. Wenzler He, Mignery G!\, Fisher LM, Park WD (1989) Analvsis of a chimeric c1ass-1 patatin-GUS gene in transgenic potato plants: high-level expression in tubers and sucrose-inducible expression in cultured leaf and stem explants. Plallt Molecular Biology, 12, -II-50. Whiting Gj, Chanton JF (1993) Primary production control of methane emission from wetlands. Nature, 364, 794-795. Whittaker RH (1975) Comnllmities and Ecosystems, 2nd edn. Macmillan, New York, 385 pp. Wilkins D, Van Oosten J-J, Besford RT (1994) Effects of elevated CO 2 on growth and chloroplast proteins in Primus avium. Tree Physiology, 14, 769-779. Williams JHH, Farrar JF (1990) Control of barley root respiration. Phljsiologia Plantarmn, 79, 259-266. Williams JHH, Winters AL. Farrar JF (1992) Sucrose: a now I plant growth regulator. In: Molecular, Biochemical and 1'11l/siologiml As/,ects olPlant Res/,iration (eds Lambers H, vander Flas U IW), pp. -163-469. 51'B !\cademic Publishing, The Ilague, The Netherlands. Williams SN, Schader SJ, Calvache ML, Lopez D (1992) Global carbon dioxide emission to the atmosphere by volcanoes. Geochimica et Cosmochimica Acta, 56, 1765-1770. Williams WE, Garbutt K, Bazzaz F!\, Vitousek PM (1986) The response of plants to elevated CO 2, IV. Two deciduous-forest tree communities. Oecologia, 69, 454-459. Winn CD, Mackenzie FT, Carrillo CJ, Sabine CL, Karl DM (1994) Air-sea carbon dioxide exchange in the North Pacific

Subtropical Gyre: implications for the global carbon budget. Global Biogrochemical Cycles, 8,157-163. Wong SC (1979) Elevated atmospheric partial pressure of CO 2 and plant growth. I. Interactions of nitrogen nutrition and photosynthetic capacity in C 3 and C 4 plants. Oecologia, 44, 68-74. Wong SC (1990) Elevated atmospheric partial pressure of CO 2 and plant growth. II. Non-structural carbohydrate content in cotton plants and its effect on growth parameters. PllOtoStjlltizesis Research, 23, 171-180. Woo KC, Wong SC (1983) lnhibition of CO 2 assimilation by supraoptimal COi effect of light and temperature. Alistralian Journal of Plant Physiology, 10, 75 8.~. Woodrow IE, Berry JA (1988) Enzymatic regulation of photosynthetic CO 2 fixation in C, plants. ;IIlIllIllI Review of Plallt Physiology alld Pialltlvluicclilar BiuluSY, 39, 533-594. Woodward Fl (1987) Stomatal numbers are sensitive to increases in CO 2 from pre-industrial levels. Nature, 327, 617-1l18. Woodward Fl, Bazzaz F!\ (1988) The responses of stomatal denSity to CO 2 partial pressure. Journal of Experimental Botallt;, 39, 1771-1781. Wullschleger SD, l\orby RJ (1992) Respiratory cost of leaf growth and maintenance in white oak saplings exposed to atmospheric CO 2 enrichment. Canadiall 1[1111'11111 [,f Forest Rest'l7rciz, 22, 1717­ 1721. Wullschleger SD, Norby RJ, Gunderson CA (1992) Growth and maintenance respiration in k'aves of Liriod<'lldnm tlllipifera L. exposed to long-term carbon dioxide enrichment in the field. New Phytologist, 121, 515-321. Wullschleger SD, Norby 11.], Hanson PJ (1995a) Growth and maintenance respiration in stems of Quercus alba after four years of CO 2 enrichment. Physiologia Plalltarllm, 93, 47-54. Wullschleger SD, Post WM, King AW (1995b) On the potential for a CO 2 fertilization effect in forests: estimates of the biotic growth factor based on 58 controlled-exposure studies. In: Biotic Feedbacks ill the Global Climatic System (eds Wood well GM, Mackenzie FT), pp. 85-107. Oxford University Press, New York. Yamaguchi J (1978) Respiration and the growth efficiency in relation to cropproducti\'ity IOlinIal oltize FaCility of Agricliiture, Hokkaido University, 59, 59-129. Ziska LH, Bunce JA (1993) Inhibition of whole plant respiration by elevated CO 2 as modified bv growth temperature. Physiologia Plantarum, 87, 459-466. Ziska LH, Bunce JA (1994a) Increasing growth temperature reduces the stimulatory effect of elevated CO 2 on photosynthesis or biomass in two perennial species. Physi%gia Plantarlml, 91, 183-190. Ziska LH, Bunce JA (1994b) Direct and indirect inhibition of single leaf respiration by elevated CO 2 concentrations: interaction with temperature. Physiologia Plantamm, 90,130-138.

(01995 Blackwell Science Ltd., C/o/'al Change Biology, 1, 243-274