Ecosystem CO2 Exchange and Variation in the 5lsO of Atmospheric CO2

Lawrence B. Flanagan

Introduction

Changes in the composition of the atmosphere have important effects on the functioning of terrestrial ecosystems (Canadell et al, 2000). Elevated carbon dioxide levels cause changes to ecosystem photosynthesis and water flux (Drake etal., 1997), and in turn ecosystem activity influences the atmosphere by causing seasonal fluctuations and latitudinal gradients in the concentration of atmospheric CO2 (Flanagan and Ehleringer, 1998; Fung, 2000). Atmospheric monitoring programs have developed as a tool to measure and document changes in the activity of terrestrial ecosystems on large spatial scales (Tans and White, 1998; Canadell et al, 2000). A working hypothesis in earth system science is that variation in the uptake and release of carbon dioxide in terrestrial ecosystems is largely responsible for the substantial year-to-year variation in the annual rate of increase in atmospheric CO2 concentration (Fung, 2000). Measurements of the stable carbon isotope composition of atmospheric CO2 provide additional information to help separate out the relative contribution of oceanic and terrestrial processes to seasonal and interannual variation in atmospheric CO2 concentration (Tans and White, 1998; Chapters 13 and 14). In addition, the oxygen isotope ratio of CO2 has been promoted as a tool to differentiate terrestrial ecosystem photosynthesis and respiration for their effects on the composition of atmospheric CO2 (Farquhar et al, 1993). The independent information recorded by carbon and oxygen isotopes of atmospheric CO2 reflect their different mechanisms of isotope fractionation during photosynthesis and respiration (see Chapter 14) and this is illustrated by the contrasting patterns of seasonal change for these two isotopic species (Fig. 10.1).

One important pattern that has been recorded in data collected by the Climate Monitoring and Diagnostics Laboratory (National Oceanic and

172 10. Ecosystem CO2 Exchange and Variation in the <5 ^ O of Atmospheric CO 2 370

O 350 O

Alert, Canada (1992-97)

172 10. Ecosystem CO2 Exchange and Variation in the <5 ^ O of Atmospheric CO 2 370

Alert, Canada (1992-97)

O 350 O

■ } } i i i

i ;

i i

i

i

i

* i i ■ i

i

. i

i i

}

Figure 10.1 Average seasonal changes in the concentration and stable isotope composition of atmospheric carbon dioxide at Alert, Canada during 1992-1997. The data were obtained from the Climate Monitoring and Diagnostics Laboratory, National Oceanic and Atmospheric Administration, United States Department of Commerce (available at: http://www.cmdl.noaa.gov/index.html).

Atmospheric Administration, United States Department of Commerce) flask network is the systematic decline of 0.08%oyr-1 in the <5180 value of atmospheric CO2, at least during 1993-1997 (Fig. 10.2). A number of mechanisms can contribute to this pattern. First, the combustion of biomass and fossil fuel releases carbon dioxide into the atmosphere with

Introduction 173 Alert, Canada

—•- Mean 1992-97

-0.5

—1993

—*- 1996

0

-1.0

K

0

-1.5

A,

00

-2.0

-2.5

i , i . 1 ^ 1

Figure 10.2 Decline in the oxygen isotope composition o, PDB) of atmos pheric C02 at Alert, Canada during 1993-1996. The monthly values for 1993 and 1996 are compared to the average seasonal pattern during 1992 through 1997. The data were obtained from the Climate Monitoring and Diagnostics Laboratory, National Oceanic and Atmospheric Administration, United States Department of Commerce (available at: http://www.cmdl.noaa.gov/index.html).

a 5lsO (PDB) value of —17%o, and with a total atmospheric forcing of approximately — 0.05%oyr-1 (Ciais etal, 1997a; Stern etal., 2001). Second, Gillon and Yakir (2001) suggested that increased conversion of tropical forests (C3 vegetation) to pasture grasses (C4 vegetation) could cause a decline of 0.02%oyr_1 in the ¿lsO of atmospheric CO2. This change is a result of the low carbonic anhydrase activity in C4 plants that limits the equilibration of CO2 and water in chloroplasts and, therefore, reduces the apparent discrimination against CO2 molecules containing lsO during photosynthetic gas exchange (Williams et al., 1996; Gillon and Yakir, 2001; Ometto et al., 2004). Stern et al. (2001) proposed that other land use changes associated with the expansion of agriculture have increased CO2 fluxes from the soil to the atmosphere and that this might account for the remaining negative forcing on the atmosphere of approximately 0.01 %oyr-1. There is, however, a great deal of uncertainty in these estimates of the effect of land-use change on the <5lsO of atmospheric CO2.

An alternative approach to the problem was taken by Ishizawa et al. (2002), who used model calculations to illustrate that an increase in both photosynthesis and total ecosystem respiration in the northern hemisphere could cause the observed decline in ¿lsO value of atmospheric CO2, if this change was accompanied by lower C1800 discrimination during photosynthesis.

A decline in photosynthetic discriminadon can occur because of larger stomatal limitation of photosynthesis and the associated reduction in the ratio of CO2 concentration in the chloroplast and that in the ambient atmosphere. Alternatively, a reduction in photosynthetic discrimination could be caused by lower leaf water <518C) values. While it is possible to conceive of a set of environmental conditions that might result in higher photosynthesis and respiration in northern hemisphere land ecosystems and result in lower photosynthetic discrimination, a limitation of the Ishizawa et al (2002) analysis is that they do not relate their model calculations to actual shifts in environmental conditions that were apparent during the period 1993 to 1997. In this sense there is no mechanistic connection between the current understanding of controls on isotopic fractionation and ecosystem CO2 fluxes and the observed trend in the <5lsO of atmospheric CO2.

In this chapter I describe a case study of ecosystem CO2 exchange and associated isotope effects in a northern, temperate grassland that makes direct mechanistic links between interannual environmental changes and their consequences for variation in the oxygen isotope composition of atmospheric CO2. This study is complementary to that of Ishizawa et al. (2002) and consistent with their general conclusions, but includes a more detailed mechanistic analysis of the interactions involved at the ecosystem level.

Oxygen Isotope Effects during Ecosystem CO2 Exchange

Several recent studies have documented that terrestrial ecosystems exert the dominant influence on the oxygen isotope composition of atmospheric C02 (Ciais et al, 1997a,b; Peylin et al, 1999; Cuntz et al, 2003a,b). The previous chapter of this book described the relevant isotope effects involved. The 180/160 composition of atmospheric CO2 is dependent on these isotopic fractionation processes, as well as the magnitude of the one-way CO2 fluxes, gross photosynthesis, and total ecosystem respiration that occur during ecosystem-atmosphere CO2 exchange. In this study I made use of net ecosystem C02 exchange (NEE) data that were collected using the eddy covariance technique (Flanagan et al, 2002). Using the micrometeoro-logical sign convention, gross primary production (GPP) photosynthesis is negative and represents removal of CO2 from the atmosphere, while total ecosystem respiration (TER) is positive, and represents a gain of carbon dioxide by the atmosphere.

Both of the gross fluxes (GPP and TER) have associated isotope compositions that normally represent opposite effects on the 180/160 composition of atmospheric CO2. The net isotope effect on the atmosphere is termed a net isoflux and was calculated as:

where ¿a is the <5180 value of atmospheric CO2 (this parameter is held constant at 0%o [PDB] for this analysis); A is discrimination against ClsOO that occurs during photosynthetic gas exchange; and <5r is the <5lsO value ofC02 respired by plants and soil in the ecosystem. In order to estimate seasonal and interannual changes in the isofluxes for C1800, the continuous eddy covariance measurements of NEE were used to calculate daily-integrated values of GPP and TER, as described in detail by Flanagan et al. (2002). The daily average isotope effect associated with GPP and TER was estimated as described below. The apparent discrimination against c18oo during photosynthesis (A) was calculated as:

A = a + £[0eq(5e - 5a) - (1 - 0eq)a/(e + 1)1 (10-3)

where a is the average fractionation during diffusion of C1800 from the air into the chloroplast (assumed constant at 7.4%o); e = c;/(ca — q), ca being the concentration of atmospheric CO2 (assumed constant at 370pmolmol_1) and c; the concentration of CO2 inside the chloroplast; 0eq is the proportional extent of isotopic equilibrium between oxygen in CO2 and oxygen in chloroplast H20 (assumed constant at 0.7); and Se is the 8lsO value ofC02 in the chloroplast. Measurements of the 513C of plant biomass samples were used to estimate seasonal changes in the integrated value of Cj by using Eqs 6 and 8 in Farquhar et al. (1989) and assuming that discrimination by Rubisco was 27%o, fractionation during diffusion of C02 into the leaf was 4.4%o, and the 513C of source C02 was —8%o. Linear interpolation was used between measurements of plant biomass ¿13C made at regular intervals during each growing season in order to estimate a daily-integrated value for c\. The Craig and Gordon (1965) model of evaporative enrichment was used to estimate the value of chloroplast water as described by Flanagan et al. (1991, 1997). The leaf water model required input values of stem water S180, which were estimated from measurements of the S180 of soil CO2 (see below), and the <5lsO of atmospheric water vapor, which was assumed to be constant at —28.3%o (Standard and Mean Ocean Water, SMOW). The leaf water SlsO values were calculated for 1-hour intervals during the growing season, and weighted by gross photosynthesis values to obtain daytime averaged leaf water 5lsO values. The <518O value of CO2 in equilibrium with chloroplast water was calculated using the temperature-dependent fractionation factor for CO2-H2O exchange as described by Flanagan et al. (1997). The daily noon air temperatures and the daytime average (gross photosynthesis weighted) S180 value of leaf water were used to calculate the daytime average S l8C) of chloroplast CO2.

The oxygen isotopic composition of CO2 released during ecosystem respiration (<5r) was calculated as the <5180 value of soil CO2 minus 7.3%o (Miller et al., 1999), to account for fractionation that occurs during diffusion of ClsOO out of the soil. We made regular measurements of the 5lsO value of soil CO2 at 10 cm depth during each growing season. This approach assumes that during the day when photosynthesis is active, respiratory CO2 is only released from the soil. In addition, it assumes that in the dark when photosynthesis is not active, CO2 released from plant tissues has the same oxygen isotopic composition as that released from the soil. This would require that leaf water (and other plant water) would have the same <5lsO value as stem and soil water at night.

Variation in Environmental Conditions and Associated Changes in Ecosystem Isofluxes

Comparisons are made here among three years with progressively lower amounts of precipitation. Ecosystem C02 fluxes and productivity in grassland ecosystems are strongly dependent on growing season precipitation, with the growing season defined as April through August in this system (Flanagan et al., 2002). During our study, precipitation received during 1999 (268 mm) was quite close to the 30-year average for the site (236 ± 86, mean ± SI), n = 30). However, the subsequent 2 years received summer rainfall that was significantly below normal (123 mm in 2000 and 107 mm during 2001). Drought has the potential to influence ecosystem isofluxes in several ways. First, during water stress the magnitude of both ecosystem photosynthesis and respiration is reduced during the growing season. Second, increased stomatal limitation of photosynthetic gas exchange during water stress can result in lower CO2 concentrations in the chloroplast and consequently reduced discrimination against ClsOO during photosynthesis. Potentially counteracting this mechanism for drought-induced decline in C1800 discrimination is an increase in the enrichment of 180 in leaf water caused by higher leaf-air vapor pressure deficits during warmer and drier environmental conditions. Apparent discrimination against C1800 during photosynthesis will increase in association with the lsO content of leaf water. Analyses in our grassland ecosystem suggest that higher leaf water <5lsO values, apparent under drought conditions, have a stronger positive effect on photosynthetic discrimination than the reduction in discrimination caused by lower chloroplast CO2 concentrations (Fig. 10.3). So there was a progressive increase in 18C) discrimination during ecosystem photosynthesis from 1999 through the 2001 growing seasons.

Variation in Environmental Conditions and Ecosystem Isofluxes 177 1999 2000 2001

Variation in Environmental Conditions and Ecosystem Isofluxes 177 1999 2000 2001

150 210 270 90 Time, day of year

Figure 10,3 (Top) Comparison of daily-average oxygen isotope composition (¿180%o, SMOW) of leaf water. (Bottom) Comparison of daily-average discrimination against C1800 molecules during photosynthetic gas exchange (A%o). Measurements were made in a grassland ecosystem near Lethbridge. Canada during 1999 through 2001.

150 210 270 90 Time, day of year

Figure 10,3 (Top) Comparison of daily-average oxygen isotope composition (¿180%o, SMOW) of leaf water. (Bottom) Comparison of daily-average discrimination against C1800 molecules during photosynthetic gas exchange (A%o). Measurements were made in a grassland ecosystem near Lethbridge. Canada during 1999 through 2001.

Isotope effects during ecosystem respiration can also be altered by drought conditions. Evaporation of water from the soil can be a larger fraction of total ecosystem evapo-transpiration and as a consequence soil water in the shallow depths (0-10 cm) can become enriched in lsO. As a result the isotope ratio of soil respired CO2 could have higher S1H O values relative to times when soil moisture is abundant. Our measurements illustrated higher 8 lsO values for soil CO2 during 2000 and 2001 compared to 1999, consistent with this mechanism (Fig. 10.4). The evaporative enrichment of 180 in soil water would also alter the source water taken up by plants and contribute to higher leaf water 5180 values observed under the drier conditions of 2000 and 2001 (Fig. 10.3).

Comparison of the net isofluxes, and the associated photosynthesis and respiration isofluxes, revealed significant differences among the three study years (Fig. 10.5). During the period of active plant growth when ecosystem photosynthetic CO2 uptake exceeded the loss of CO2 by respiration, the ecosystem net isoflux had a positive forcing that increased the <51H(.) of atmospheric CO2 in this ecosystem. The magnitude of the positive forcing was increased as water availability declined during 2000 and 2001 because of the higher leaf water ¿lsO values. In addition, the ecosystem respiration isoflux was higher in drought years despite the lower respiration rates

Time, day of year

Figure 10.4 Comparison of seasonal changes in the oxygen isotope composition (<5180 %o, PDB) ofsoilCC>2 (at 10 cm soil depth) in a grassland ecosystem near Lethbridge, Canada during 1999 through 2001.

Time, day of year

Figure 10.4 Comparison of seasonal changes in the oxygen isotope composition (<5180 %o, PDB) ofsoilCC>2 (at 10 cm soil depth) in a grassland ecosystem near Lethbridge, Canada during 1999 through 2001.

because of the increase in ¿180 of soil water. So the higher positive net forcing on the atmosphere under dry environmental conditions was caused by an increased positive photosynthetic isoflux and a less negative respiration isoflux. The less-negative respiration isoflux was a consequence of the reduced respiration rates under drought and the associated higher 5180 values of ecosystem respired CO2. The cumulative effect of these changes in net isoflux illustrates that drought should cause a positive forcing and increase the 5180 value of atmosphere during ecosystem CO2 exchange (Fig. 10.6).

Implications for the Declining Trend in ¿180 of Atmospheric CO2

While it is unwise to generalize the conclusions of this case study in a single grassland ecosystem to a global pattern in atmospheric CO2, there are useful mechanistic insights obtained from this study. This analysis showed that reduced ecosystem CO2 exchange, coupled with a range of associated isotope effects, resulted in an increased positive forcing on the <5lsO value of atmospheric CO2 as water stress increased (Fig. 10.6). As a consequence it is reasonable to predict that an increase in water availability would result in higher ecosystem productivity and a decline in the positive forcing, which in turn could contribute to a reduction in the <5180 value of atmospheric CO2. This is consistent with the global model analysis developed by Ishizawa et al. (2002), who suggested that higher ecosystem productivity, higher ecosystem respiration, and lower photosynthetic discrimination in

150 210 270 90 150 210 270 90

Time, day of year

Figure 10,5 Comparison of seasonal variation in the gross one-way isofluxes (for photosynthesis and respiration) and the net isofluxes for lsO in CC>2- Measurements were made in a grassland ecosystem near Lethbridge, Canada during 1999 through 2001.

1 j

lk

I

150 210 270 90 150 210 270 90

Time, day of year

150 210 270

Figure 10,5 Comparison of seasonal variation in the gross one-way isofluxes (for photosynthesis and respiration) and the net isofluxes for lsO in CC>2- Measurements were made in a grassland ecosystem near Lethbridge, Canada during 1999 through 2001.

the northern hemisphere could explain the approximately 0.5%o reduction in S180 value of atmospheric CO2 observed during 1993-1997. However, Ishizawa et al. (2002) suggested that the reduced photosynthetic discrimination was caused by lower chloroplast CO2 concentrations, while our analysis for the grassland ecosystem indicated that change in the oxygen isotope ratio of leaf water was primarily responsible for the lower photo-synthetic discrimination. Normally shifts in environmental factors that cause an increase in photosynthesis also cause an associated rise in stomatal conductance and higher chloroplast CO2 concentration, although nitrogen fertilization may be an exception to this generalization. The difference in the mechanism proposed here relative to that suggested by Ishizawa et al. (2002) is significant for the application of atmospheric measurements as a tool in earth system science. Further analysis of the problem is warranted and can be productively approached using global-scale models that include all the a? 150

1 100 x

90 120 150 180 210 240 270 300 Time, day of year

Figure 10,6 Comparison of the cumulative net isofluxes for I80 in CC>2- Measurements were made in a grassland ecosystem near Lethbridge, Canada during 1999 through 2001.

important mechanistic details revealed from theoretical and experimental studies in physiological ecology (Riley et al, 2002; Chapter 9).

Figure 10,6 Comparison of the cumulative net isofluxes for I80 in CC>2- Measurements were made in a grassland ecosystem near Lethbridge, Canada during 1999 through 2001.

References

Canadell J., Mooney H. A., Baldocchi D. D„ Berry J. A., Ehleringer J. R„ Field C. B., Gower S. T., Hollinger D. V., HuntJ. E., Jackson R. B., Running S. W., Shaver G. R., Steffen W., Trumbore S. E., Valentini R. and Bond B. V. (2000) Carbon metabolism of the terrestrial biosphere: A multitechnique approach for improved understanding. Ecosystems 3: 115-130.

Ciais P., Denning A. S.. Tans P. P., Berry J. A., Randall D. A., Collatz G. J., Sellers P. J., White J. W. C., Trolier M., Meijer H. A. J., Francey R.J., Monfray P. and Fleimann M. (1997a) A three-dimensional synthesis study of <5180 in atmospheric CO9 1. Surface fluxes. J Geophys Res 102: 5857-5872.

Ciais P., Tans P. P., Denning A. S., Francey R. J., Trolier M., Meijer H. A. J., White J. W. C., Berry J. A., Randall D. A., Collatz G. J., Sellers P. J„ Monfray P. and Heimann M. (1997b) A three-dimensional synthesis study of <5180 in atmospheric CO2 2. Simulations with the TM2 transport model. J Geophys Res 102: 5873-5883.

Craig H. and Gordon L. I. (1965) Deuterium and oxygen-18 variations in the ocean and the marine atmosphere. In Proceedings of a Conference on Stable Isotopes in Océanographie Studies and Paleotemperatures (E. Tongiorgi, ed.) pp. 9-130. Spoleto, Italy. Lischi & Figli, Pisa.

Cuntz M., Ciais P., Hoffmann G. and Knorr W. (2003a) A comprehensive global 3D model of S180 in atmospheric CO9: Validation of surface processes. J Geophys Res 108: doi: 10.1029/2002JD003153.

Cuntz M„ Ciais P., Hoffmann G„ Allison C. E„ Francey R.J., Knorr W., Tans P. P., WhiteJ. W. C. and Levin I. (2003b) A comprehensive global 3D model of (5^80 in atmospheric CO2: 2. Mapping the atmospheric signal. J Geophys Res 108: doi: 10.1029/2002JD003154.

Drake B. G., Gonzalez-Meier M. A. and Long S. P. ( 1997) More efficient plants: A consequence of rising atmospheric CO2? Annu Rev Plant Physiol Plant Mol Biol 48: 609-639.

Farquhar G. D., Ehleringer J. R. and Hubick K. T. (1989) Carbon isotope discrimination and photosynthesis. Annu Rev Plant Physiol Plant Mol Biol 40: 503-537.

Farquhar G. D., Lloyd J., Taylor J. A., Flanagan L. B., Syvertsen J. P., Hubick K. T., Wong S. C. and Ehleringer J. R. (1993) Vegetation effects on the isotope composition of oxygen in atmospheric C02. Nature 363: 439-443.

Flanagan L. B., Comstock J. P. and Ehleringer J. R. (1991) Comparison of modeled and observed environmental influences on the stable oxygen and hydrogen isotope composition of leaf water in Phaseolus vulgaris L. Plant Physiol 96: 588-596.

Flanagan L. B. and EhleringerJ. R. (1998) Ecosystem-atmosphere CO2 exchange: Interpreting signals of change using stable isotope ratios. Trends EcolEvol 13: 10-14.

Flanagan L. B., BrooksJ. R., Varney G. T. and EhleringerJ. R. (1997) Discrimination against q18q!6q dur;ng photosynthesis and the oxygen isotope ratio of respired CO2 in boreal forest ecosystems. Global Biogeochem Cycles 11: 83-98.

Flanagan L. B., Wever L. A. and Carlson P. J. (2002) Seasonal and interannual variation in carbon dioxide exchange and carbon balance in a northern temperate grassland. Global Change Biol 8: 599-615.

Fung I. (2000) Climate change: Variable carbon sinks. Science 290: 1313.

Gillon J. and Yakir D. (2001) Influence of carbonic anhydrase activity in terrestrial vegetation on the lsO content of atmospheric COz. Science 291: 2584-2587.

Ishizawa M., Nakazawa T. and Higuchi K. (2002) A multi-box model study of the role of the biosphere metabolism in the recent decline of 5lsO in atmospheric COg. Tellus 54B: 307-324.

Miller J. B„ Yakir D., White J. W. C. and Tans P. P. (1999) Measurement of 180/I60 in the soil-atmosphere CO2 flux. Global Biogeochem Cycles 13: 761-774.

OmettoJ. P. H. B., Flanagan L. B., Martinelli L. A. and EhleringerJ. R. (2004) Oxygen isotope ratios of waters and respired C02 in Amazonian forest and pasture ecosystems. Ecol Appl 14: in press.

Peylin P., Ciais P., Denning A. S., Tans P. P., BerryJ. A. and WhiteJ. W. C. (1999) A 3-dimensional study of the <5I80 in atmospheric C02: Contribution of different land ecosystems. Tellus 51B: 642-667.

Riley W. J., Still C. J., Torn M. S. and BerryJ. A. (2002) A mechanistic model of H2180 and C1800 fluxes between ecosystems and the atmosphere: Model description and sensitivity analyses. Global Biogeochem Cycles 16: doi: 10.1029/2002GB001878.

Stern L. A., Amundson R. and Baisden W. T. (2001) Influence of soils on oxygen isotope ratio of atmospheric CO2. Global Biogeochem Cycles 15: 753-759.

Tans P. P. and WhiteJ. W. C. (1998) In balance, with a little help from the plants. Science 281: 183-184.

Williams T. G., Flanagan L. B. and Coleman J. R. (1996) Photosynthetic gas exchange and discrimination against 13C02 and ClsOO in tobacco plants modified by an antisense construct to have low chloroplastic carbonic anhydrase. Plant Physiol 112: 319-326.

Was this article helpful?

0 0

Post a comment