GLOBAL BIOGEOCHEMICAL

CYCLES, VOL. 12, NO. 2, PAGES 259-276, JUNE 1998

Dynamics of fossil fuel CO2 neutralization by marine CaCO3 David

Archer

Department of the Geophysical Sciences, University of Chicago Chicago,Illinois

Haroon Kheshgi Exxon Researchand EngineeringCompany,Annandale,New Jersey

Ernst

Maier-Reimer

Max-Plank-Institutftir Meteorologie,Hamburg, Germany

Abstract. A detailedmodelof the oceancirculationand carboncyclewas coupledto a mechanisticmodel of CaCO3 diagenesisin deepseasedimentsto simulatethe millennium-scale responseof the oceansto futurefossilfuel CO2 emissionsto the atmosphereand deepsea. Simulationsof deepseainjectionof CO2 showthat CaCO3 dissolutionis sensitiveto passageof high-CO2 watersthroughthe Atlantic Ocean,but CaCO3 dissolutionhas a negligible impact on atmosphericpCO2 or the atmosphericstabilizationCO2 emissionin the coming centuries. The ultimate fate of the fossil fuel CO2 will be to react with CaCO3 on the seafloorand on land. An initial CaCO3 dissolutionspikereversesthe net sedimentationrate in the oceanuntil it is attenuatedby an enhancedverticalgradientof alkalinityafterabout1000 years. The magnitudeof the initial spike is sensitiveto assumptionsabout the kinetics for CaCO3 dissolution,but subsequent behaviorappearsto be lessmodeldependent.Neutralizationby seafloorCaCO3 occurs on a timescaleof 5-6 kyr, and is limited to at most 60-70% of the fossilfuel release,even if the fossil fuel releaseis smallerthan the seafloorerodibleinventoryof CaCO3. Additional neutralizationby terrestrialCaCO3 restoresa balancebetweenCaCO3 weatheringand seafloor accumulationon a timescaleof 8.5 kyr, while the deficit of seafloorCaCO3 (the lysocline) is replenishedwith an e-foldingtimescaleof approximately18 kyr. The final equilibriumwith CaCO3 leaves7-8% of the fossilfuel CO2 remainingin the atmosphere, to be neutralizedby the silicaterock cycle on a time frame of hundredsof thousandsof years.

1. Introduction

Mankind releasesCO2 to the atmosphereby combustionof fossil fuels and by deforestation, which converts relatively high biomass forests, mostly in the tropics, into lower biomassgrasslandsand farmlands,resulting in a net release of CO2. Currently, the rate of atmospheric CO2 increase corresponds to less than half of the anthropogenic release from fossil

fuels and deforestation.

The sink for the other half

appearsto be a combination of dissolutioninto the ocean and uptake by the terrestrial biosphere in regions other than regions of active deforestation [Tans et al., 1990; Schimel et al., 1994; Cias et al., 1995; Jain et al., 1996; Keeling et al., 1996; Kheshgi et al., 1996]. Future uptake by the terrestrial biosphere is difficult to predict and is not the topic of this paper. In contrast,uptake by the oceansis describableby the physics of ocean circulation and the chemistry of CO 2 and

C aCO 3, which we are beginning to be able to describe quantitatively.For our discussion of the oceanicCO2 uptake, it will be convenientto combineanthropogenic emissionand biosphericuptake into a quantity called the net terrestrial emission,suchthat future biosphereuptake acts to directly counteract fossilfuel anddeforestation CO2 emission. The first and quantitatively most significant step in the processingof the terrestrial CO2 releasewill be dissolutionin the oceans, which we will call invasion.

The timescale for

invasion is determined by the circulation timescale of the ocean and by the buffer capacity of subducting surface seawater. Following the equilibration of the atmospherewith

the water in the ocean,the CO2 will react with solid CaCO3 (a reaction called neutralization), decreasing the atmospheric component of the CO2 release still further. Two processes contribute to neutralization: (1) the transfer of carbonate ion

[CO3 =] from dissolving CaCO3 on the seafloor to the ocean (seafloor neutralization) and (2) the imbalance between the rate

of chemical weathering (dissolving) of CaCO3 on land and accumulationof solid CaCO3 on the seafloor,which resultsin a net dissolutionflux of CaCO3 on land to the ocean(terrestrial neutralization). Finally, on time frames of hundreds of

Copyfight1998 by the AmericanGeophysicalUnion. Papernumber98GB00744 0886-6236/98/98GB-00744512.00

259

260

ARCHER ET AL.' FOSSIL FUEL CO2NEUTRALIZATION BY MARINE CaCOs

In this paper, we subject a detailed model of ocean and sediment carbon chemistry to a range of fossil fuel release scenariosto predict the impact of CaCO3 dissolutionon time

sedimentarydiagenesismodel for organiccarbonand CaCO3 in deep sea sediments[Archer, 1991, 1996b]. The oceancarbon cycle model advects dissolved chemical tracers using the steadyflow field from the large scale geostrophiccirculation model, tuned to represent the present-day ocean [MaierReirner, 1993b]. The model flow field was invariant during the model integration,neglectingclimate feedbacksto oceanflow [Manabe and Stouffer, 1993]. A single nutrient, scaled to

frames

represent PO43-,limitsorganic carbon exportfromthesurface

thousandsof years the pCO2 of the atmosphere(and hence the ocean) is thought to be controlled by the balance between volcanic degassingof CO2 and its consumptionby reaction with basic componentsof silicaterocks [Walker and Kasting, 1992].

of centuries

to millennia.

Section

2 describes

the

model in detail. Sections3 and 4 analyze the effect of CaCO3 on estimates

of the net terrestrial

emission

which

result

ocean accordingto the Michaelis-Menton rate expression

in

p2 specified time-evolution of atmospheric CO2 concentration (stabilization scenarios), and results of deep sea CO2 direct injection experiments. Section 5, the longest section, P is describesthe sequenceof invasion and seafloor and terrestrial where rlat is a functionof latitude(light) and temperature, neutralization to the year A.D. 10,000 and beyond in response the nutrient concentration, and PHS is the half-saturation setat a valueof 0.01 gmolL-l. Thisproduction rate to variousCO2 releasescenarios.We explorethe dynamicsof constant, fossil fuel neutralization to understand the details of the has been adjusted relative to the formulation of Maierkinetics of neutralization in its various regimes, an Reimer[1993b] to slow the biological uptake of nutrients in understandingwhich may also guide future research into the the euphotic zone, decreasing the rate of biological transitions in global carbon cycle accompanying the productivity in high-nutrient areas such as the equatorial Pacific, allowing transport of nutrients in surface currents to transitionsbetween glacial and interglacial states. the oligotropic subtropical gyres. This was necessaryto attenuate the equator/oligotrophic contrast in the delivery 2. Model Description rates of organic carbon and CaCO3 to the seafloor. The We simulate the physics and chemistry of the ocean and resulting export productivity field of the ocean is shown in sediment response to anthropogenic CO 2 release using a Plate l a. The disadvantageof this adjustmentwas an increase previouslydocumentedmodel of ocean circulation and water in the nutrient concentrationand pCO 2 of the sea surface, of the model atmosphere by column chemistry [Maier-Reimer, 1993b], coupled to a increasing the pCO2

Rate =rla t P+PHS

L' • , (.•

• •

-0.05 -o,1

-0.15 -0.2

-0.25 -0,3 -0,35

o

10000

20000

30000

40000

o. 15

Figure 17. (a) A comparison of the full time dependent model results(A22) againstpredictionsof the dissolutionflux as a function of the time-evolving values of deep Pacific

[CO3 =] and bioturbated layer CaCO3 inventory. The parameterized function describes the time evolution of the model well, exceptthat it underestimates the magnitudeof the time dependent model dissolution flux in the millennium following CO2 release. (b) Detail of the first 5000 yearsof the experimentwith the same comparisonas (a) but also showing the dissolution spike, defined as the difference between the time dependentmodel results and the parameterizedfunction values. The timescale for decay of the dissolution spike resembles the behavior of the model in response to a homogeneousperturbation(Figure 10).

with a fit to the individual model resultsshownin Figure 16. The accumulationfluxes expectedfrom the model evolution of ACO3 and bioturbatedCaCO3 inventoryare comparedwith the actual model values in Figure 17. It can be seen that the empirically predicted accumulation rate corresponds well except for a significant spike in dissolution immediately following the fossil fuel invasion. Similar behavior was observedin the parametric model runs, a simple example of which was shown in Figure 16. The magnitude of the transitional dissolution spike was calculated by subtracting the model accumulation rate from the empirical prediction, in Figure 17(b). Both the magnitude and the decay time of the transient dissolution spike correspondwell to the initial spike of dissolutionfrom the model run in Figure 15. This observation motivates a closer inspection of the transient behavior of the parametric model run in Figure 15. Over 1000 years following the homogeneousaddition of CO2, the dissolutionflux decreasedmore than can be explained by the changing values of A CO3 = or bioturbated CaCO 3 inventory. This behavior was causedby a rearrangementof the distribution of ACO3 = in the water column, by a process analogousto the formation of a whole ocean boundary layer, causedby surface/ deep fractionationof alkalinity in the ocean analogousto the biological pump. The global productionrate

of CaCO3in themodelis approximately 1.6Gt C yr-•. Most of this CaCO 3 redissolves in the water column or at the

ARCHER ETAL.'FOSSILFUELCO2NEUTRALIZATION BY MARINECaCOs •co3 =

of 200-450 yearsdependingon the magnitudeof the fossil fuel release. (2) A fast seafloor dissolution spike after CO2 dissolvesin the oceansbut before alkalinity has a chance to

gM -lOO

-50

0

275

50

o-

reorganize accounts for half of the neutralization in the coming millennium and 10% of the total CaCO3 dissolutionin

responseto fossil fuel CO2. Beginningin this stage,mankind will have reversed the sedimentation rate of the ocean by dissolving CaCO3 faster than it is depositedon the seafloor; this is projectedto peak by the year A.D. 3000 and last up to 5000 years. (3) Steady seafloordissolutioncontinuesuntil the lysocline reaches local equilibrium with the water column saturation state by about the year A.D. 10,000. Seafloor

E

Initial 500

neutralization

years

Figure 18. A comparisonof profiles of ACO3= immediately followingand 500 yearsafter a homogeneous additionof CO2 to the water column, from 120øE, 0øN.

seafloor, concentrating alkalinity in the deep waters. The

dissolution flux from the sediments reached0.5 Gt C yr'], which acts as an added sourceof alkalinity to the deep waters comparable to the flux attributable to the biological pump. Over the courseof the 1000 year circulationtime of the ocean, the alkalinity of deep waters increaseduntil a balancebetween dissolution and deep ocean circulation was reached. This relaxation is analogous to the formation of a whole ocean boundary layer. Profiles of ocean alkalinity during the dissolutionspike and 500 years later are shown in Figure 18. The largest [CO3=] changesare found immediately above high %CaCO 3 areason the seafloor. In the IPCC model runs, the transient dissolution spike is responsible for 50% of the dissolutionflux to year 3000 and roughly 10% of the flux to the year 40,000. This transientdissolutionspike is the only case where the approximation of the ocean carbonatesystem as representableby two simple variables,deep Pacific [CO3=] and bioturbatedlayer CaCO3 inventory,breaksdown.

overall

accounts

for 60-70%

of the fossil

fuel

CO 2, and hasan e-folding timescaleof 5-6 kyr. (4) Terrestrial weatheringreplenishesocean [CO3=] on a timescaleof 8 kyr after A.D. 10,000 and restoresthe lysocline to its original depth on a timescale of 18 kyr. Completion of this stage leaves 7-8% of the fossil fuel CO 2 remaining in the atmosphere in a new local and global CaCO3 steady state. (5) Neutralization by the silicate rock cycle (not included in the model) restoresatmosphericCO2 to some unknown "set point", at which metamorphicdecarbonationbalances silicate weathering, with a time constant of several hundred thousand years.

Acknowledgments. This paper benefittedby reviews by Eric Sundquist,Scott Doney, and anotheranonymousreviewer, and was supportedby the Petroleum ResearchFund and the David and Lucille Packard Foundation.

References

Archer,D. E., The dissolutionof calcitein deepseasediments: An in situmicroelectrode study,Ph.D. thesis,Univ. Wash.,Seattle,1990. Archer,D., Modelingthecalcitelysocline,J. Geophys. Res.,96, 17,03717,050, 1991. Archer, D., An atlas of the distribution of calcium carbonate in

sediments of the deepsea,Global Biogeochem.Cycles,10, 159-174, 1996a.

Archer,D., A data-drivenmodelof the globalcalcitelysocline,Global Biogeochern.Cycles,10, 511-526, 1996b.

Archer,D., S. Emerson, andC. Reimers,Dissolution of calcitein deepsea sediments:pH and 0 2 microelectroderesults, Geochim.

6. Summary

Cosmochim.Acta, 53, 2831-2846, 1989.

Accordingto our currentunderstanding, CO2 uptakeby rock cyclesof the Earth will be slow, with no significantimpact on the pCO 2 of the atmospherefor the next thousandyears. The net

terrestrial

emission

calculated

to

meet

an

IPCC

atmospheric stabilization scenario (s750) is only a few percent higher when CaCO 3 is included in the model simulation,which is negligible comparedwith uncertaintiesin biosphericuptake or other aspectsof the ocean carbon cycle. Direct ocean injection experiments demonstrate a coupling between dissolution and ocean circulation, as CaCO 3 dissolutionproceedssomewhatmore rapidly for atmospheric and Atlantic injection cases than for Pacific injection. However, CaCO3 dissolutionwill play only a minor role in determining the fate of fossil fuel CO2 in the coming millennium.

On longer timescales,neutralization of fossil fuel CO2 by the oceans and by CaCO3 takes place in five stages. (1) Invasion of fossil fuel into the oceans will sequester7080% of the net terrestrialCO2 releaseon an e-folding timescale

Archer,D., H. Kheshgi,and E. Maier-Reimer,Multiple timescalesfor neutralizationof fossilfuel CO2, Geophys.Res. Lett., 24, 405-408, 1997.

Bacastow, R. B., andR. K. Dewey,Effectiveness of CO2 sequestration in the post-industrial ocean,Energy Conversi.and Manage.,37, 1079-1086, 1996.

Berelson, W. M., D. E. Hammond, and G. A. Cutter, In situ measurements of calciumcarbonatedissolutionrates in deep sea sediments,Geochim.Cosmochim. Acta, 54, 3013-3020, 1990.

Berner,E. K., andR. A. Berner,TheGlobalWaterCycle.,PrenticeHall, EnglewoodCliffs, N.J., 1987. Berner, R. A., and J. W. Morse, Dissolution kinetics of calcium

carbonatein sea water, IV, Theory of calcite dissolution,Am. J. Sci., 274, 107-134, 1974.

Berner, R. A., A. C. Lasaga, and R. M. Garrels, The carbonate-silicate geochemicalcycle and its effect on atmosphericcarbondioxide over the past 100 million years,Am. J. Sci., 283, 641-683, 1983. Boyle, E. A., Chemical accumulationvariationsunder the Peru Current

duringthe past130,000 years,J. Geophys.Res.,88, 7667-7680, 1983. Broecker,W. S., and T. H. Peng,The role of CaCO3 compensation in the glacial to interglacial atmospheric CO2 change, Global Biogeochern.Cycles,1, 15-29, 1987.

Broecker,W. S., andT. Takahashi, Neutralization of fossilfuel CO2 by marinecalciumcarbonate,in The Fate of FossilFuel CO 2 in the

276

ARCHER ETAL.'FOSSIL FUELCO2 NEUTRALIZATION BYMARINE CaCO3

Oceans,editedby N. R. Andersen,and A. Malahoff,pp. 213-248, Plenum, New York, 1978.

Cai, W.-J., In situ microelectrode studiesof the early diagenesisof organiccarbonand CaCO3 in hemipelagicsediments of the Pacific Ocean,Ph.D. thesis,Univ. Calif., SanDiego, 1992.

Cias,P., P. P. Tans,M. Trolier,J. W. C. White,andR. J. Francey,A large northernhemisphereterrestrialCO2 sink indicatedby the

13C/12C ratio ofatmospheric CO2,Science, 269,1098-1102, 1995. Cole, K. H., G. R. Stegen,and D. Spencer,The capacityof the deep oceansto absorbcarbondioxide,Energy Conv.and Manage.,34, 991-998, 1993.

Cole, K. H., G. R. Stegen,and D. Spencer,The capacityof the deep oceansto absorbcarbondioxide,in Direct OceanDisposalof Carbon Dioxide,editedby N. Handa,andT. Ohsumi,pp. 143-152,TerraSci.,

national CO2 emissions,in Trends93: A Compendiumof Data on Global Change,editedby T. A. Bodenet al., Rep. ORNL/CDIAC-65, pp. 505-584, Oak Ridge Nat. Lab., Oak Ridge, Tenn., 1993. Martin, J. H., G. A. Knauer, D. M. Karl, and W. M. Broenkow, VERTEX: Carbon cycling in the northeastPacific, Deep Sea Res. Part A, 34, 267-285, 1987. Milliman, J. D., Production and accumulation of calcium carbonate in the ocean:Budgetof a nonsteadystate,Global Biogeochem.Cycles, 7, 927-957, 1993. Morse, J., and F. T. MacKenzie, Geochemistry of Sedimentary Carbonates. Elsevier, New York, 1990. Nihous, G. C., S. M. Masutani, L. A. Vega, and C. M. Kinoshita, Projected impact of deep ocean carbon dioxide discharge on

atmospheric CO2 concentrations, Clim. Change,27, 225-244, 1994. Panel on Policy Implications of Greenhouse Warming (National Academy of Sciences, National Academy of Engineering, and and concentrations of carbondioxide:Key ocean/atmosphere/land Instituteof Medicine),Policy Implicationsof GreenhouseWarming: Mitigation, Adaptation, and the Science Base, Nat. Acad. Press, analyses,120 pp., Commw.Sci. and Ind. Res.Organ.,Melbourne, Victoria, Australia, 1994. Washington,D.C., 1992. Hales,B., Calcitedissolutionon the seafloor:An in situ study,Ph.D. Sarmiento, J. L., J. C. Orr, and U. Siegenthaler, A perturbation simulationof CO2 uptakein an oceangeneralcirculationmodel,J. thesisUniv. Wash., Seattle, 1995. Geophysi.Res.,97, 3621-3645, 1992. Hales,B., S. Emerson,andD. Archer,Respiration anddissolution in the sediments of the western North Atlantic: Estimates from models of in Schimel,D., I. Enting,M. Heimann,T. Wigley, D. Raynaud,D. Alves, and U. Siegenthaler, CO2 andthe carboncycle,in Climate Change situ microelectrode measurements of porewateroxygenand pH, Deep Sea Res.,41 Part I, 695-719, 1993. 1994: RadiativeForcingof Climate Changeand an Evaluationof the IPCC IS92 EmissionScenarios,editedby J. T. Houghtonet al., pp. Haugan, P.M., and H. Drange, Disposal optionsin view of ocean 35-71, CambridgeUniv. Press,New York, 1994. circulation, in DirectOceanDisposalof CarbonDioxide,editedby N. Sundquist,E. T., Geological perspectiveson carbon dioxide and the HandaandT. Ohsumi,pp. 123-141,TerraSci.,Tokyo, 1995. carboncycle,in The Carbon Cycleand AtmosphericC02: Natural Hoffert, M. I., Y.-C. Wey, A. J. Callegari, and W. S. Broecker, Atmosphericresponseto deep-seainjectionsof fossil-fuel carbon VariationsArcheanto Present,Geophys.Monogr. Ser., vol 32, edited dioxide,Clim. Change,2, 53-68, 1979. by E. T. Sundquist,and W.S. Broecker,pp. 5-59, AGU, Washington D.C., 1985. Houghton, J. T., G. J. Jenkins, and J. J. Ephraums (Eds.), Intergovernmental Panel on ClimateChange:The IPCC Scientific Sundquist,E. T., Geologicanalogs:Their value and limitationsin carbon Assessment, CambridgeUniv. Press,New York, 1990. dioxideresearch,in The ChangingCarbonCycle,A GlobalAnalysis, editedby J. R. Trabalkaand D. E. Reichle,pp. 371-402, SpringerHoughton, J. T., B. A. Callander, and S. K. Varney (Eds.), Verlag, New York, 1986. Intergovernmental Panelon ClimateChangeClimateChange1992.' The SupplimentaryReport to the IPCC Scientific Assessment, Sundquist, E. T., Influence of deep-sea benthic processes on atmospheric CO2, Philos.Trans.R. Soc.LondonSer.A, 331, 155-165, CambridgeUniv. Press,New York, 1992. 1990a. Houghton, J. T., L. G. Meiro Filho, B. A. Callendar, N. Harris, A. E. T., Long-termaspectsof futureatmospheric CO2 andseaKattenburg,and K. Maskell (Eds.), The IPCC SecondScientific Sundquist, level changes,in Sea-Level Change,editedby R. Revelle, pp. 193AssessmentReport, 572 pp., CambridgeUniv. Press,New York, 1996. 207, Nat. Acad. Press,Washington,D.C., 1990b. Jahnke,R. A., The global ocean flux of particulateorganic carbon: Tans, P. P., I. Y. Fung, and T. Takahashi, Observationalconstraintson the globalatmospheric CO2 budget,Science,247, 1431-1438,1990. arealdistributionandmagnitude,GlobalBiogeochem. Cycles.10, 7188, 1996. Tsunogai,S., and S. Noriki, Particulatefluxes of carbonateand organic carbonin the ocean. Is the marine biological activity working as a Jain,A. K., H. S. Kheshgi,and D. J. Wuebbles,A globallyaggregated sink of the atmosphericcarbon?,TellusSer. B, 43, 256-266, 1991. reconstruction of cyclesof carbonand its isotopes,TellusSer B 48: 583-600, 1996. Walker, J. C. G., and J. F. Kasting, Effects of fuel and forest conservation on future levels of atmospheric carbon dioxide, Keeling,R. F., S.C. Piper, and M. Helmann,Global and hemispheric Palaeogeog.,Palaeoclimatol.,Palaeoecol.,97, 151-189, 1992. CO2 sinksdeducedfrom changes in atmospheric 0 2 concentration, Walker, J. C. G., P. B. Hays, and J. F. Kasting, A negativefeedback Nature, 381, 218-221, 1996. mechamism for the long-term stabilization of Earth's surface Keir, R. S., The dissolutionkineticsof biogeniccalciumcarbonatesin temperature,J. Geophys.Res., 86, 9776-9782, 1981. seawater, Geochim. Cosmochim.Acta, 44, 241-252, 1980. Kheshgi,H. S., B. P. Flannery,M. I. Hoffert, andA. G. Lapenis,The Wigley, T. M. L., R. Richels, and J. A. Edmonds, Economic and effectiveness of marineCO2 disposal, Energy,19, 967-974, 1994. environmentalchoicesin the stabilizationof CO2 concentrations: Choosingthe "right"emissionspathway,Nature,379, 240-243, 1996. Khes.hg!, H. S., A. K. Jain,andD. J. Wuebbles, Accounting for the m•ss•ng carbonsinkwiththeCO2 fertilization effect,Clim. Change, Wolery, R. J., and N.H. Sleep, Interactionsof geochemicalcycleswith 33, 31-62, 1996. the Mantle, in Chemical Cyclesin the Evolution of the Earth, edited Maier-Reimer,E., The biologicalpump in the greenhouse, Global by C. B. Gregoret al., pp. 77-104, JohnWiley, New York, 1988. Planet. Change,8, 13-15, 1993a. Tokyo, 1995. Enting, I. G., T. M. L. Wigley, and M. Heimann(Eds.),Futureemissions

Maier-Reimer, E., Geochemical cyclesin an oceangeneralcirculation model.Preindusrial tracerdistributions, GlobalBiogeochem. Cycles, 7, 645-678, 1993b.

Maier-Reimer, E., andK. Hasselmann, Transport andstorage of CO2 in theocean:An inorganicocean-circulation carboncyclemodel,Clim. Dyn., 2, 63-90, 1987.

Manabe, S., and R. J. Stouffer, Century-scaleeffects of increased atmospheric CO2 on theocean-atmosphere system,Nature,364, 215218, 1993.

D.E. Archer,Department of theGeophysical Sciences, Universityof Chicago,5734 SouthEllis Avenue,Chicago,IL 60637. (e-mail: [email protected])

H. Kheshgi,ExxonResearch andEngineering Company, Route22E, Annandale, NJ 08801([email protected])

E. Maier-Reimer, Max-Plank-Institut fuer Meteorologie, Bundesstrasse 7, D-2000 Hamburg 54, Germany (email [email protected])

Marchetti,L., On geoengineering the CO2 problem,Clim. Change, 1, 59-68, 1977.

Marland,G., R. J. Andres,and T. A. Boden,Global,regional,and

(ReceivedJanuary24, 1998; revisedFebruary2, 1998' acceptedFebruary 25, 1998.)