Net ecosystem production in the great lakes basin and its implications for the North American missing carbon sink: A hydrologic and stable isotope approach

https://doi.org/10.1016/j.gloplacha.2007.08.004Get rights and content

Abstract

This study resolves the Great Lakes Basin (GLB) hydrologic cycle into its components using stable isotopes of hydrogen and oxygen together with long term meteorological data. The results are used to calculate basinwide Net Primary Production (NPP). Stoichiometric relations of carbon and water during photosynthesis form the basis of computing NPP. Mean annual discharge from the GLB is 29% of the precipitation flux and in δD–δ18O space it has a slope of 5.6, distinctly lower than the 7.5 average slope for precipitation in the basin. Mean annual evaporation flux to the atmosphere from water bodies, soils, and wet canopies is 24% of the precipitation flux. Transpiration however, is the strongest pathway for loss of water from the basin, annually returning 342 Pg (petagram = 1015 g) or 47% of the precipitation flux to the atmosphere.

Transpiration and carbon assimilation during photosynthesis are coupled processes. For the GLB vegetation, approximately every mole of CO2 fixation requires 850 mol of H2O loss by transpiration. Therefore, basinwide annual NPP is 0.268 Pg C. Heterotrophic respiration in soils and herbivory annually release 0.138 Pg C to the atmosphere. The surplus NPP or 0.130 Pg C year 1 Net Ecosystem Productivity (NEP) is stored in GLB vegetation, consistent with the postulated missing North American carbon sink.

Introduction

The annual cycling of water is the largest movement of a chemical substance at the surface of the earth. Evaporation from the tropics and transport to higher latitudes not only transfers water, but also much of the absorbed heat energy (Schlesinger, 1997). Besides a major role in the redistribution of the earth's surface energy, water vapor plays a key role in regulating the earth's climate. It is the dominant greenhouse gas and has the strongest known feedback mechanism for amplifying climate change (Cubasch and Cess, 1990). With increasing temperature the ability of the atmosphere to hold water increases and leads to increased evaporation, and may consequently increase cloudiness and greater radiative absorption (Wetherald and Manabe, 1988, Rind, 1998, Soden et al., 2002). Increasing air temperature may also cause decreased snow cover and consequently lower surface albedo. All these factors contribute to amplified global warming. It is these feedbacks in climate models that result in predicted model temperature rise between 1.5 and 5.5 °C by the end of the century (Houghton et al., 1992). Thus, the earth's surface energy and water cycles are intimately interconnected and exert a strong influence on one another (cf. Oort, 1970, Ramanathan, 1987).

The earth's surface energy and water cycles are tightly coupled to the global carbon cycle through photosynthesis and respiration and together fuel most biotic processes on earth. The terrestrial biosphere exerts a significant influence on global carbon balance and hence climate change (Bolin et al., 1986, Schimel, 1995). Annually it sequesters roughly 10–30% of the global annual emissions from fossil fuels burning and industrial activities (Schimel et al., 2001). It is not possible to balance the global carbon cycle without allocating an additional 30–60% of these emissions to the terrestrial biosphere. Hence the term missing sink was coined for the unaccounted carbon (Siegenthaler and Oeschger, 1978, Broecker et al., 1979).

Although it is mostly agreed that multiple mechanisms (regrowth of abandoned farmland and logged forests, anthropogenic N deposition, CO2 fertilization, and global warming) are responsible for enhanced terrestrial NPP (the balance of annual C fixation and autotrophic respiration) during the past two decades (Fan et al., 1998), geographical distribution and nature of the missing sink remains controversial. Alternative ecosystems, such as northern boreal and temperate forests, peatbogs, tropical rainforests, and savannah grasslands have been proposed to harbor the missing sink (Post, 1990, Fischer e t al., 1994, Phillips et al., 1998). This issue has scientific, environmental, and political significance and is pertinent to emission reduction quotas under the Kyoto Convention.

Several techniques have been used to establish the nature and distribution of the missing sink. For example, the tower-based eddy covariance technique for net CO2 flux between forest canopy and the atmosphere (Baldocchi et al., 1988); inversion techniques (Tans et al., 1989, Fan et al., 1998) employ CO2 concentrations and atmospheric transfer models; models of forest inventory and land use statistics (Dixon and Brown, 1994, Houghton, 1999); balances of the isotopic composition in atmospheric CO2 (Ciais et al., 1995); quantification of O2/N2 gradient in the atmosphere (Keeling et al., 1996); and satellite-derived primary production (Myneni et al., 2001, Hicke et al., 2002, Nemani et al., 2003). All these methods have strengths and weaknesses. Some provide improved accuracy, but are limited by small temporal and spatial scales of measurement and by uncertainties associated with scaling up from point measurements to landscape or regional level. Forest inventories or direct field measurements are labor intensive, and destructive. Other methods rely on surrogate or indirect estimates of carbon fluxes. In addition, they yield only aboveground carbon stocks. In a comparison of 17 models of NPP for the terrestrial biosphere, Cramer et al. (1999) pointed out that modeled regional and global NPP estimates were sensitive to the simulation method for the water balance. This observation highlights the role of water in the global carbon cycle and the need for reliable estimates for water balance components.

In this paper we use long term meteorological, hydrological, and land cover data together with stable isotopes of hydrogen and oxygen to resolve water balance and use the results to calculate basinwide annual NPP for the Great Lakes. This method has produced fast and economical estimates for the Ottawa (Telmer and Veizer, 2000), the Mississippi (Lee and Veizer, 2003), the Saskatchewan (Weinrauch, 2003), and the Volta river basins (Freitag et al., 2008-this issue).

We chose the Great Lakes Basin, the source of the second largest river in North America, because it is located completely within the boreal-temperate forest realm and has several decades of continuous meteorological and hydrological data. Straddling the border between Canada and the United States (Fig. 1), the basin lies within two provinces and eight states. Owing to the large size of the basin, 774,174 km2, of which the lakes cover roughly one-third, a great diversity of climate, soil, and vegetation, characterizes the region. It is affected by warm and humid air masses originating in the Gulf of Mexico as well as cold, dry air from the Arctic. Mean annual temperature varies between 0–2 °C in the northern and 8–10 °C in the southern parts of the basin (National Atlas of Canada, 1978–1995). Similarly, mean annual precipitation is < 700 mm in the NW increasing sharply to > 1000 mm in the southern parts of the basin. Thin acidic soils overlie granitic rocks of the Canadian Shield in the northern part of the basin and deeper glacio-fluvial sediments occur in the south. Consequently conifers dominate northern parts of the basin and deciduous forests and agricultural and residential areas the southern parts.

We sampled all the Great Lakes, except Lake Michigan, aboard C.C.G.S. Limnos (Canada Centre for Inland Waters, Burlington, Ontario). Sample locations are shown in Fig. 1. Most water samples were collected by lowering a 2.5 L, pre-washed, glass bottle to a depth of 3 m. In addition, at least three samples from each lake were collected at mid-thermocline and mid-hypolimnion using a Seabird® profiler and a General Oceanics® Rosette sampler. The St. Lawrence River was sampled bimonthly, from a boat at mid river, for 1 year (except during the Jan–Mar 1996 freeze) at Cornwall. Samples were filtered through 0.45 μm, cellulose acetate membrane filters and stored in 30 mL Nalgene®, low density polyethylene bottles. In order to minimize post-sampling evaporation, the bottles were filled completely, with lids tightly closed and kept refrigerated. Isotope analyses were carried out within six months of sampling.

All isotope analyses were performed at the G.G. Hatch Isotope Laboratory, University of Ottawa. Hydrogen gas was prepared by catalytic reduction of water with zinc (Coleman et al., 1982). About 110 mg of granular zinc was heated at 500 °C for 30 min with 3 μL of each sample in separate glass tubes. The H2 gas was analyzed with an automated double collector VG 602D mass spectrometer. Samples were processed in batches of 8, with two additional internal laboratory standards (− 84 and − 227.6‰ VSMOW) included to correct for analytical drift. Analytical precision for D/H ratios was ± 1.3‰ for the entire procedure.

The 18O/16O ratios were determined by the CO2–water equilibration method of Epstein and Mayeda (1953). Three milliliter samples were loaded in glass vessels on an automated shaker, mounted on a water bath, maintained at 25 °C. About 160 μM CO2 was fed into each vessel and allowed to equilibrate for 12 h. The equilibrated CO2 was purified cryogenically and analyzed on a triple collector VG SIRA 12 mass spectrometer. Analytical precision was ± 0.1‰ for the analyses.

Section snippets

Water balance

Water balance is the most fundamental aspect of water resources and the hydrologic cycle and entails quantification of its four major components. These are, precipitation (P), river discharge (Q), evapotranspiration (E), change in soil, surface, and groundwater storage reservoirs (ΔS), and they are related by the following continuity equation.P=Q+E+ΔS

When observed for more than one year, seasonal fluctuations in the various terms cancel each other and the storage term (ΔS) becomes negligible,

Water use efficiency

The basis for calculating CO2 fixation by vegetation from water balance is the Transpiration Ratio (TR), the inverse of Water Use Efficiency (WUE), defined as the net carbon uptake per unit water transpired, both quantities expressed in mass or molar units. These can be established at various spatial scales, such as, leaf, whole plant, or entire ecosystem. In addition, WUE can be expressed as long term (intrinsic) or short term (instantaneous). In contrast to instantaneous WUE, intrinsic WUE

Basinwide carbon fixation and storage

With the water balance for the GLB resolved for transpiration and the water–carbon relationship for vegetation established in the preceding sections, now we estimate NPP. Terrestrial vegetation in the basin annually fixes 0.268 Pg C or 519 g C m2. Note that 257,800 km2, or about one-third, of the GLB is covered with water and that the NPP per unit area is based on the terrestrial part of the basin. In addition, NPP reported here is based on long term WUE that takes into account autotrophic

Implications for the “missing carbon sink”

With regard to C fluxes between the atmosphere and GLB, discussed above, two questions can be raised:

  • 1)

    Does the GLB represent a net source or sink to atmospheric CO2 and

  • 2)

    What will be the impact of global climate change on the status of carbon sink?

In order to answer the first question, it is important to mention uncertainties associated with our NPP estimates. Possible sources of errors in our estimates could be: (1) the water balance and isotope mass balance, (2) the interception flux, and (3)

Acknowledgements

Fieldwork was carried out with the assistance of the Canada Centre for Inland Waters, Burlington, Ontario. Many thanks to Steve Smith, Barry Moore, and support staff aboard the C.C.G.S. Limnos for technical assistance and hospitality. We thank Wendy Abdi, Paul Middlestead, Gilles St-Jean, and Patricia Wickham for assistance with the isotope analyses. Paul Ferguson provided valuable conceptual suggestions. Costs of field and analytical work were defrayed by the Natural Sciences and Engineering

References (103)

  • AlexanderW.J.R.

    Development of a multi-year climate prediction model

    Water S.A.

    (2005)
  • AllenR.G. et al.

    Crop evapotranspiration — guidelines for computing crop water requirements

    FAO Irrigation Drainage Papper

    (1998)
  • BaldocchiD.D. et al.

    Measuring biosphere-atmosphere exchanges of biologically related gases with micrometeorological methods

    Ecology

    (1988)
  • BarbourM.G. et al.

    Terrestrial Plant Ecology

    (1987)
  • BarbourM.G. et al.

    Terrestrial Plant Ecology

    (1999)
  • BolinB. et al.

    Global perspective, in land use, land use change and forestry

  • BowenG.J. et al.

    Interpolating the isotopic composition of modern meteoric precipitation

    Water Resour. Res.

    (2003)
  • BroeckerW.S. et al.

    Fate of fossil fuel carbon dioxide and the global carbon budget

    Science

    (1979)
  • BrownR.M.

    Distribution of hydrogen isotopes in Canadian waters

  • CarslawK.S. et al.

    Cosmic rays, clouds, and climate

    Science

    (2002)
  • CiaisP. et al.

    Partitioning of ocean and land uptake of CO2 as inferred by d13C measurements from the NOAA Climate Monitoring and Dioagnostics Laboratory Global Air Sampling Network

    J. Geophys. Res.

    (1995)
  • ColemanM.L. et al.

    Reduction of water with zinc for hydrogen isotope analysis

    Anal. Chem.

    (1982)
  • CoplenT.B. et al.

    Stable hydrogen and oxygen isotope ratios for selected sites of the U.S. Geological Survey's NASQAN and Benchmark surface-water networks

  • CraigH.

    Isotopic variations in meteoric waters

    Science

    (1961)
  • CraigH. et al.

    Deuterium and oxygen-18 variations in the ocean and the marine atmosphere

  • CramerW. et al.

    Comparing global models of terrestrial net primary productivity NPP: overview and key results

    Glob. Chang. Biol.

    (1999)
  • CroleyT.E. et al.

    Great Lakes monthly hydrologic data

  • CubaschU. et al.
  • DangerfieldJ.M. et al.

    Overcompensation by Acacia erubescens in response to simulated browsing

    J. Trop. Ecol.

    (1996)
  • DincerT.

    The use of oxygen-18 and deuterium concentrations in the water balance of lakes

    Water Resour. Res.

    (1968)
  • DingmanS.L.

    Physical Hydrology

    (2002)
  • DixonR.K. et al.

    Carbon pools and flux of global forest ecosystems

    Science

    (1994)
  • EpsteinS. et al.

    Variation of oxygen-18 content of waters from natural sources

    Geochim. Cosmochim. Acta

    (1953)
  • FanS. et al.

    A large terrestrial carbon sink in North America implied by atmospheric and oceanic carbon dioxide data and models

    Science

    (1998)
  • FischerM.J. et al.

    Carbon storage by introduced deep-rooted grasses in the South American savannas

    Nature

    (1994)
  • FreitagH. et al.

    Water and carbon fluxes from savanna ecosystems of the Volta River watershed, West Africa

    Glob. Planet. Change.

    (2008)
  • FriedmanI. et al.

    Compilation of stable isotope fractionation factors of geochemical interest

    U. S. Geol. Surv. Prof. Pap.

    (1977)
  • FriedmanI. et al.

    Water vapor exchange between a water droplet and its environment

    J. Geophys. Res.

    (1962)
  • GatJ.R. et al.

    The contribution of evaporation from the Great Lakes to the continental atmosphere: estimate based on stable isotope data

    Geophys. Res. Lett.

    (1994)
  • GibsonJ.J. et al.

    Estimating evaporation using stable isotopes: quantitative results and sensitivity analysis for two catchments in northern Canada

    Nord. Hydrol.

    (1993)
  • GLERL (Great Lakes Environmental Research Laboratory, accessed September, 2007)....
  • Government of Canada U.S.E.P.A., 1995. The Great Lakes Environmental Atlas and Resource Book. 3rd ed. Government of...
  • GraceJ.

    Presidential Address: understanding and managing the global carbon cycle

    J. Ecol.

    (2004)
  • GSFC, DAAC, (Goddard Space Flight Center Distributed Active Archive Center, accessed May, 2003)....
  • HaighJ.D.

    The impact of solar variability on climate

    Science

    (1996)
  • HansonP.J. et al.

    Separating root and soil microbial contributions to soil respiration: a review of methods and observations

    Biogeochemistry

    (2000)
  • HetheringtonE.D.

    The importance of forests in the hydrological regime

  • HickeJ.A. et al.

    Satellite-derived increases in net primary productivity across North America, 1982–1998

    Geophys. Res. Lett.

    (2002)
  • HickeJ. et al.

    Postfire response of North American boreal forest net primary productivity analyzed with satellite observations

    Glob. Chang. Biol.

    (2003)
  • Cited by (18)

    • Modeled ecosystem responses to intra-annual redistribution and levels of precipitation in a prairie grassland

      2015, Ecological Modelling
      Citation Excerpt :

      Evaporation rate might be relatively high in coarse textured soils, however, low total evapotranspiration amount of coarse textured soils limit transpiration amount. Low transpiration indicated plants had diminished the stomatal openings, which decreased productivities (Claesson and Nycander, 2013; Karim et al., 2008). Consequently, NPP was lower in both fine and coarse textured soils than in medium textured soils.

    • Stable isotope mass balance of the Laurentian Great Lakes

      2014, Journal of Great Lakes Research
      Citation Excerpt :

      Similarly, we used an average of localized grids on land within each Great Lake catchment to estimate δR. Because a strong evaporative signal was not evident from the d-excess in any of the rivers sampled, we assume that catchment runoff in the Great Lakes basin has not been strongly affected by evaporation and that transpiration – which does not produce an isotope effect on waters under steady-state conditions – dominates total catchment evapotranspiration (Jasechko et al., 2013; Karim et al., 2008). First, to calculate the equilibrium liquid–vapor fractionation factors (αl–v*) for δ18O and δ2H, monthly average lake surface temperatures were collected from over-lake monitoring buoy records (National Data Buoy Center: www.ndbc.noaa.gov).

    • Applications of stable water and carbon isotopes in watershed research: Weathering, carbon cycling, and water balances

      2011, Earth-Science Reviews
      Citation Excerpt :

      The knowledge of (T) obtained via the above isotope and GIS methods, enables us not only to estimate the water balance for a given catchment but also provides a first order evaluation of the large scale biological uptake of CO2. Further details of this type of basin wide water and carbon balance studies can be found in Telmer and Veizer (2000), Lee and Veizer (2003), Ferguson et al. (2007), Freitag et al. (2008), Karim et al. (2008), and Brunet et al. (2009). As an example, a detailed case study by Ferguson and Veizer (2007) is outlined in the following.

    • Carbon and oxygen dynamics in the Laurentian Great Lakes: Implications for the CO2 flux from terrestrial aquatic systems to the atmosphere

      2011, Chemical Geology
      Citation Excerpt :

      The isotopic signature of organic matter is primarily dependent on the photosynthetic pathway of the plant from which it is derived: − 27‰ for C3 vegetation and − 12.5‰ for C4 (Vogel, 1993, Fig. 3B). In the Great Lakes Basin, 86% of the vegetation is C3 (Karim et al., 2008) and therefore CO2 produced from respiration in the soil is expected to have an isotopic signature of ~−25‰. Soil CO2 may be subject to isotopic enrichment from diffusion, producing slightly more positive values (Cerling et al., 1991).

    • In situ-measured primary production in Lake Superior

      2010, Journal of Great Lakes Research
    View all citing articles on Scopus
    View full text