SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methods
  5. 3. Results and Discussion
  6. 4. Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[1] The global distribution of potential wetlands and their methane (CH4) emissions at the present-day and the Last Glacial Maximum (LGM) are estimated using a GCM simulation of LGM climate, a vegetation model, and simple algorithms for determining wetland area based on topography and soil moisture, and CH4 emissions based on ecosystem carbon turnover in wet soils. LGM wetland area was 15% larger than present, but CH4 emissions were 24% less. Extensive wetlands were simulated on the exposed continental shelves. The soil CH4 sink was simulated as 14 Tg now but <0.5 Tg at the LGM. CH4 emissions at LGM were limited by substrate availability, in turn due to low atmospheric CO2. The glacial-interglacial change in atmospheric CH4 concentration cannot be completely attributed to changes in the wetland source.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methods
  5. 3. Results and Discussion
  6. 4. Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[2] Atmospheric methane (CH4) concentrations varied between ca. 350 and 700 ppb over the last 400 000 years [Petit et al., 1999]. Between the Last Glacial Maximum (LGM, ca. 21 000 yr BP) and pre-industrial Holocene (to ca. 1850), CH4 concentrations increased nearly 100% and more closely parallel the fast variations of polar temperature records than any other measured gas [Chappellaz et al., 1990, 1993a; Raynaud et al., 1988]. Recent studies agree that CH4 emissions from wetlands drove prehistoric changes in ice-core CH4, but conflict as to the location of wetlands and the environmental controls of CH4 emission [Chappellaz et al., 1993b; Dällenbach et al., 2000; Worthy et al., 2000]. It seems likely that there are strong feedbacks between temperature and wetland CH4 emissions [Schimel et al., 1996].

[3] I present here new estimates of wetland area and CH4 emissions for the present-day and the LGM, using “bottom-up” ecosystem modeling. I used a digital elevation model (DEM) to determine suitable low-relief areas for extensive wetland formation. I ran the global biogeography and biogeochemistry model BIOME4, forced by climate and atmospheric CO2 concentration, to determine the soil wetness, substrate availability (NPP), and potential CH4 production rate (as a fraction of heterotrophic respiration, Rh). Potential wetland area and CH4 emissions are shown for the present-day, forced by modern observed climate, and the LGM, driven by a coupled atmosphere-vegetation general circulation model (AVGCM).

2. Methods

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methods
  5. 3. Results and Discussion
  6. 4. Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[4] Climatology and soils datasets for the present-day and LGM drove the BIOME4 global vegetation model [Kaplan, 2001], simulating fields of vegetation type, soil moisture, NPP, and Rh. I estimated wetland area and extent based on these outputs plus topographic information. I combined the vegetation model output with wetland area information to simulate CH4 emission.

[5] A mean climatology for the late 20th century (CLIMATE v2.2) provided a 0.5° baseline for the experiments, extrapolated over as necessary shelf areas [W. Cramer, pers. comm., 1998] (http://www.pik-potsdam.de/cramer/climate.htm). The FAO digital soil map of the world [FAO, 1995] provided information on soil texture and depth. I used BIOME4 to simulate monthly soil wetness, NPP, and potential CH4 emission. A DEM [GETECH, 1996] was used to identify areas flat enough to support wetlands. Potential CH4 emissions were calculated where wetland areas were identified. Wetland areas were identified on the 5′ grid of the digital terrain model, and methane emissions were calculated for these wetland gridcells using the BIOME4 output of the nearest 0.5° grid node.

[6] For the present-day simulation I ran BIOME4 with the climatology described above and a CO2 concentration of 324 ppm (the mean [CO2] during the period of the climatology). For the LGM experiment I used a climatology derived from the GENESIS/IBIS AVGCM [Levis et al., 1999] and an atmospheric [CO2] of 211 ppm (glacial minimum) and 324 ppm (control). The boundary conditions used in the GENESIS/IBIS simulations were sea surface temperatures (SST) in both control and LGM simulations, insolation, greenhouse gas concentrations, and a simple parameterization of tropospheric aerosols. Modern SSTs were from Shea et al. [1992] and LGM SSTs from CLIMAP [1981]. GENESIS/IBIS ran at a R15 spectral resolution. I used the GENESIS/IBIS LGM-RPV model simulation, which included vegetation-atmosphere coupling and the physiological effects of low CO2.

[7] The wetland location algorithm selects gridcells sufficiently flat and with high enough soil moisture on a monthly basis. By comparing fields of slope and wetness to maps of wetland areas, I empirically defined threshold values for slope (<0.3%) and volumetric soil wetness (>65%).

[8] Following Christensen et al. [1996], CH4 emission was estimated as a fraction of Rh. At equilibrium, Rh is a function of NPP, soil temperature, and vegetation type. I considered water table depth implicitly in simulating soil moisture on a monthly time step. Extending the Christensen et al. [1996] method, I considered vegetation type and structure to account for differences in the CH4 oxidizing capacity of ecosystems. Wetland ecosystems dominated by grasses and other herbaceous plants were assumed to have a greater propensity for direct transport of CH4 than woody ecosystems. In mixed ecosystems, I varied the CH4 oxidizing capacity as a function of the tree-grass ratio. The model of Ridgwell et al. [1999] was used to estimate the CH4 sink in upland soils.

3. Results and Discussion

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methods
  5. 3. Results and Discussion
  6. 4. Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

3.1. Wetland Areas

[9] Global potential natural wetland area was simulated as 11.0 × 106 km2: though larger than previous estimates, this is plausible based on comparison to wetland datasets and recent drainage (Plate 1a). Seasonal wetlands accounted for 6.8 × 106 km2, or 61% (Figure 1). Present wetland estimates from maps, satellite remote sensing, and field observation range from 4.6 to 9.5 × 106 km2 (Table 1). These datasets disagree and it is unclear which is accurate [Darras et al., 1999; Hagemann and Dümenil, 1997].

image

Figure Plate 1.. Potential natural wetlands and CH4 emission simulated for (a) the present day and (b) the LGM.

Download figure to PowerPoint

image

Figure 1. Simulated longevity of potential natural wetlands for the present-day.

Download figure to PowerPoint

Table 1. Wetland Areas for Present-Day, Pre-Industrial Holocene (PIH), and LGM in 106 km2
StudyPresent-dayPIHLGM
This study (potential natural)11.0 12.9
Chappellaz et al. [1993]5.26.32.6
Aselmann and Crutzen [1989]5.77.0 
Cogley [1994]4.6  
Darras et al. [1999]9.5  
Matthews and Fung [1987]5.3  

[10] The model failed to predict some known wetlands, notably in N. Alaska, the Niger Bend, and the Sudd. In in the far north, permafrost and lateral transport of surface and groundwater may be responsible. In Africa the wetlands are formed by flooding river water from a distant source. The model also understandably failed to predict widespread, discontinuous wetlands (e.g. Scandinavia and the central USA) that are smaller than 5′ in extent.

[11] At the LGM, wetland area was nearly 15% larger than in the potential natural present-day simulation (Table 1). There are large shifts in the simulated distribution (Plate 1b). The low relief of the continental shelves promoted wetland formation. The wetlands of Europe, northern Siberia, and Hudson Bay were covered by ice sheets, but large wetland areas were present in Beringia, on the Sunda and New Guinea Shelves, and the Atlantic coastal shelves of North and South America. The wetlands of the Yellow-Yangtze River delta were greatly increased. These model results are supported by evidence from shallow marine cores suggesting that extensive river and wetland systems existed on the Sunda shelf [Hanebuth et al., 2000], while paleovegetation data indicate wetland vegetation in Beringia and southeastern North America [Edwards et al., 2000; Webb et al., 1993].

3.2. CH4 Emissions

[12] Simulated present CH4 emissions for are 140 Tg yr−1: within the range of several other studies despite the wide discrepancy in wetland area (Table 2) [Matthews, 2000]. The soil sink is calculated as 14 Tg yr−1, similar to other results [Chappellaz et al., 1993b; Fung et al., 1991; Ridgwell et al., 1999].

Table 2. Estimated Global Natural CH4 Source in Tg yr−1, Net of the Soil CH4 Sink
StudyPresent-dayPIHLGM
  1. a

    Bottom-up methods refer to process-based estimates of the CH4 source, top-down studies use measured tropospheric or ice-core CH4 concentrations, and may also use isotopic composition, to infer source strengths.

wetland CH4source simulated using bottom-up methods
This study140 107
Aselmann and Crutzen [1989]80  
Bartlett and Harriss [1993]109  
Cao et al. [1996]92  
Chappellaz et al. [1993b]11513676
Matthews and Fung [1987]110  
Walter [1998]263  
 
wetland CH4sourceinferred using top-down methods
Hein et al. [1997]227  
Houweling [1999]131163 
 
total CH4 source inferred using top-down methods
Brook et al. [2000] 159111
Crutzen and Brühl [1993]57022595
Dällenbach et al. [2000]  106
Martinerie et al. [1995]496187115
McElroy [1989]  180
Pinto and Khalil [1991] 17095
Valentin and Crutzen [1990] 252175

[13] Despite the sensitivity of methanogenesis to O2 exposure, seasonal wetlands are an important component of the simulated global CH4 flux. Over half of the seasonal wetlands in this simulation persist for 5–9 months and appear in the tropics where microbial turnover and recovery are fast enough to allow CH4 emission (Figure 2) [T. Christensen, pers. comm., 2001].

[14] CH4 emission rates at LGM were typically lower than in the present-day simulation, with tropical wetlands reaching peak emissions of only 25 g m−2 yr−1 (Plate 1b). Temperate emissions ranged between 1 and 7.5 g m−2 yr−1. The global simulated net flux of CH4 was 107 Tg yr−1. Though the total wetland area was 15% greater than present, global CH4 emissions were 25% less. The soil sink for CH4 is simulated as only 0.6 Tg yr−1.

[15] The increase in modeled wetland CH4 production after the LGM is in part due to the limiting effects of low atmospheric CO2 concentrations. In a sensitivity test, I used the vegetation-wetland-CH4 emissions model with the simulated LGM climate scenario but mid-20th century CO2 concentration, giving a CH4 source of 140 Tg yr−1, identical to the pre-industrial simulated source. In contrast to CO2 enrichment sudies which have not produced a consistent increase in wetland CH4 emissions, the low atmospheric [CO2] of the LGM has been demonstrated to have lowered vegetation productivity [Cowling and Sykes, 1999] and is virtually certain to have also affected CH4 emissions by limiting the production of substrates.

[16] The simulated glacial-interglacial change in CH4 source strength cannot fully account for the 350 ppb change measured in ice cores. The terrestrial sink for CH4 cannot be invoked as the sign of the change is wrong. At the low CH4 concentrations of the LGM, methanotrophs in upland soils were barely able to metabolize CH4 from the atmosphere. Changes in other CH4 sources and sinks may therefore be required to explain the increase in atmospheric CH4 since the LGM. This result is in contrast to earlier studies which suggested that the long-term increases in atmospheric CH4 concentration were the effect of changing temperature and precipitation patterns on wetlands [Chappellaz et al., 1993b; Crutzen and Brühl, 1993; Martinerie et al., 1995; Petit-Maire et al., 1991; Pinto and Khalil, 1991; Thompson et al., 1993]. Atmospheric OH concentrations are strongly regulated by the concentration of CH4 and other biogenic trace gases (chiefly CO and NOx). It is expected that source strength and emission patterns of CO and NOx were very different at the LGM due to alterations in vegetation distribution, climate, and fire regimes. To better constrain LGM OH concentrations will require a comprehensive model of reactive trace gas sources. Other natural terrestrial and marine CH4 sources and sinks (e.g. termites, slow outgassing of clathrates, the atomic Cl sink) may also have been different at the LGM, but relevant data are lacking.

4. Conclusions

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methods
  5. 3. Results and Discussion
  6. 4. Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[17] While the location of wetlands since the Last Glacial Maximum has changed greatly, neither the total wetland area nor CH4 emissions changed in proportion with the observed changes in atmospheric CH4. CH4 emissions appear to have been controlled by substrate availability, which at the LGM was strongly limited by low atmospheric CO2 concentrations. These results place the close covariance of the polar temperatures and atmospheric [CH4] into a new light, as the mechanisms controlling the global CH4 source over glacial-interglacial time scales may be less sensitive to climate change than previous studies suggested [Thompson et al., 1993]. On shorter timescales, CH4 from boreal wetlands may respond rapidly to climate change, as recent work has suggested [Christensen, 1999; Worthy et al., 2000]. Future studies should include simulation of the lateral tranport of water and may wish to investigate the sensitivity of wetland CH4 emissions to rapid climate changes at the end of the last glacial period.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methods
  5. 3. Results and Discussion
  6. 4. Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[18] I thank S. Levis for providing the GCM simulation output and W. Cramer for the 20th century climatology. I. C. Prentice, S. Shafer, P. J. Bartlein, J. W. Williams and several others contributed to the development of the BIOME4 model. I.C. Prentice, S. Houweling and three anonymous reviewers made valuable comments on the manuscript.

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methods
  5. 3. Results and Discussion
  6. 4. Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information
  • Aselmann, I., and P. J. Crutzen, Global distribution of natural freshwater wetlands and rice paddies, their net primary productivity, seasonality and possible methane emissions, J. Atmos. Chem., 8(4), 307358, 1989.
  • Bartlett, K. B., and R. C. Harriss, Review and assessment of methane emissions from wetlands, Chemosphere, 26(1–4), 261320, 1993.
  • Brook, E. J., S. Harder, J. Severinghaus, E. J. Steig, and C. M. Sucher, On the origin and timing of rapid changes in atmospheric methane during the last glacial period, Global Biogeochem. Cycles, 14(2), 559572, 2000.
  • Cao, M., S. Marshall, and K. Gregson, Global carbon exchange and methane emission from natural wetlands: Application of a process-based model, J. Geophys. Res., 101(D9), 14,39914,414, 1996.
  • Chappellaz, J., J. M. Barnola, D. Raynaud, Y. S. Korotkevich, and C. Lorius, Ice-core record of atmospheric methane over the past 160,000 years, Nature, 345, 127131, 1990.
  • Chappellaz, J., et al., Synchronous Changes in Atmospheric CH4 and Greenland Climate Between 40-Kyr and 8-Kyr Bp, Nature, 366, 443445, 1993a.
  • Chappellaz, J. A., I. Y. Fung, and A. M. Thompson, The atmospheric CH4 increase since the Last Glacial Maximum (1). Source estimates, Tellus, 45B(3), 228241, 1993b.
  • Christensen, T. R., Potential and actual trace gas fluxes in Arctic terrestrial ecosystems, Polar Res., 18(2), 199206, 1999.
  • Christensen, T. R., I. C. Prentice, J. Kaplan, A. Haxeltine, and S. Sitch, Methane flux from northern wetlands and tundra — An ecosystem source modelling approach, Tellus, 48B(5), 652661, 1996.
  • CLIMAP project members, Seasonal reconstructions of the Earth's surface as the Glacial Maximum, Geological Society of America Map Chart Series 36, Boulder, 1981.
  • Cogley, J. G., GGHYDRO—Global Hydrographic Data, Release 2.1, 23 pp., Trent Climate Note 91-1, Department of Geography, Trent University, Peterborough, 1994.
  • Cowling, S. A., and M. T. Sykes, Physiological significance of low atmospheric CO2 for plant-climate interactions, Quat. Res., 52, 237242, 1999.
  • Crutzen, P. J., and C. Brühl, A model study of atmospheric temperatures and the concentrations of ozone, hydroxyl, and some other photochemically active gases during the glacial, the preindustrial Holocene and the present, Geophys. Res. Lett., 20(11), 10471050, 1993.
  • Dällenbach, A., Changes in the atmospheric CH4 gradient between Greenland and Antarctica during the Last Glacial and the transition to the Holocene, Geophys. Res. Lett., 27(7), 10051008, 2000.
  • Darras, S., M. Michou, and C. Sarrat, IGBP-DIS wetland data initiative: A first step towards identifying a global delination of wetlands, IGBP-DIS Working Paper 19, IGBP-DIS, Tolouse, 1999.
  • Edwards, M. E., et al., Pollen-based biomes for Beringia 18,000, 6000 and 0 14C yr B. P. J. Biogeog., 27, 521554, 2000.
  • FAO, Digital Soil Map of the World and Derived Soil Properties, CD-ROM, Food and Agriculture Organization, Rome, 1995.
  • Fung, I., et al., 3-dimensional model synthesis of the global methane cycle, J. Geophys. Res., 96(D7), 13,03313,065, 1991.
  • GETECH, Global DTM5, CD-ROM, Geophysical Exploration Technology, Leeds, 1996.
  • Hagemann, S., and L. Dümenil, Comparison of two global wetlands datasets, in Earth Surface Remote Sensing, edited by G. Cecchi, E. T. Engman, and E. Zilioli, pp. 193201, 1997.
  • Hanebuth, T., K. Stattegger, and P. M. Grootes, Rapid flooding of the Sunda Shelf: A late-glacial sea-level record, Science, 288, 10331035, 2000.
  • Hein, R., P. J. Crutzen, and M. Heimann, An inverse modeling approach to investigate the global atmospheric methane cycle, Global Biogeochem. Cycles, 11(1), 4376, 1997.
  • Houweling, S., Global Modeling of Atmospheric Methane Sources and Sinks, Ph.D. thesis, University of Utrecht, Utrecht, 1999.
  • Kaplan, J. O., Geophysical Applications of Vegetation Modeling, 128 pp., Ph.D. Thesis, Lund University, Lund, 2001.
  • Levis, S., J. A. Foley, and D. Pollard, CO2, climate, and vegetation feedbacks at the Last Glacial Maximum, J. Geophys., 104(D24), 31,19131,198, 1999.
  • Martinerie, P., G. P. Brasseur, and C. Granier, The chemical composition of ancient atmospheres: A model study constrained by ice core data, J. Geophys., Res., 100(D7), 14,29114,304, 1995.
  • Matthews, E., Wetlands, in Atmospheric Methane, edited by M. A. K. Khalil, Springer- Verlag, Berlin, 2000.
  • Matthews, E., and I. Fung, Methane emission from natural wetlands: Global distribution, area, and environmental characteristics of sources, Global Biogeochem. Cycles, 1(1), 6186, 1987.
  • McElroy, M. B., Studies of Polar Ice: Insights for atmospheric chemistry, in The Environmental Record in Glaciers and Ice Sheets, edited by H. Oeschger, and C. C. Langway, pp. 363377, John Wiley and Sons, New York, 1989.
  • Petit, J. R., et al., Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica, Nature, 399, 429436, 1999.
  • Petit-Maire, N., M. Fontugne, and C. Rouland, Atmospheric methane ratio and environmental changes in the Sahara and Sahel during the last 130 kyrs, Palaeogeogr. Palaeoclimatol. Palaeoecol., 86(1–2), 197204, 1991.
  • Pinto, J. P., and M. A. K. Khalil, The stability of tropospheric OH during ice ages, inter-glacial epochs and modern times, Tellus, 43B(5), 347352, 1991.
  • Raynaud, D., J. Chappellaz, J. M. Barnola, Y. S. Korotkevich, and C. Lorius, Climatic and CH4 Cycle Implications of Glacial Interglacial CH4 Change in the Vostok Ice Core, Nature, 333, 655657, 1988.
  • Ridgwell, A. J., S. J. Marshall, and K. Gregson, Consumption of atmospheric methane by soils: A process-based model, Global Biogeochem. Cycles, 13(1), 5970, 1999.
  • Schimel, D., et al., Radiative forcing of climate change, in Climate Change 1995: The Science of Climate Change, edited by J. T. Houghton, et al., pp. 65131, Cambridge University Press, Cambridge, 1996.
  • Shea, D. J., K. E. Trenberth, and R. W. Reynolds, A global monthly sea surface temperature climatology, J. Climate, 5, 9871001, 1992.
  • Thompson, A. M., J. A. Chappellaz, I. Y. Fung, and T. L. Kucsera, The atmospheric CH4 increase since the Last Glacial Maximum (2). Interactions with oxidants, Tellus, 45B(3), 242257, 1993.
  • Valentin, K. M., and P. J. Crutzen, A two-dimensional, global photochemical study on the influences of increasing atmospheric chemistry since the Last Glacial Maximum, Abstracts volume, Sixth CACGP Symposium, Chemistry of the Global Atmosphere, Chamrousse, 1990.
  • Walter, B., Development of a Process-Based Model to Derive Methane Emissions from Natural Wetlands for Climate Studies, 160 pp., Ph.D. thesis, Universität Hamburg, Hamburg, 1998.
  • Webb, T.III, P. J. Bartlein, S. P. Harrison, and K. H. Anderson, Vegetation, lake levels and climate in eastern North America for the past 18,000 years, in Global Changes Since the Last Glacial Maximum, edited by H. E. Wright Jr, et al., pp. 415467, University of Minnesota Press, Minneapolis, 1993.
  • Worthy, D. E. J., I. Levin, F. Hopper, M. K. Ernst, and N. B. A. Trivett, Evidence for a link between climate and northern wetland methane emissions, J. Geophys. Res., 105(D3), 40314038, 2000.

Supporting Information

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methods
  5. 3. Results and Discussion
  6. 4. Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.