Environmental impacts of ocean disposal of CO2

Environmental impacts of ocean disposal of CO2

Energy Convers. Mgmt Vol. 37, Nos 6-8, pp. 999-1005, 1996 Pergamon 0196-8904(95)00289-8 Copyright © 1996 Elsevier Science Ltd Printed in Great Brit...

412KB Sizes 0 Downloads 57 Views

Energy Convers. Mgmt Vol. 37, Nos 6-8, pp. 999-1005, 1996

Pergamon

0196-8904(95)00289-8

Copyright © 1996 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0196-8904/96 $15.00 + 0.00

ENVIRONMENTAL IMPACTS OF OCEAN DISPOSAL OF CO2

HOWARD J. HERZOG, E. ERIC ADAMS, DAVID AUERBACH, JENNIFER CAULFIELD Massachusetts Institute of Technology Cambridge, MA 02139 USA

Abstract. This paper analyzes one of the most important environmental impacts of ocean disposal of CO:, the acidification around the release point. We present a methodology which allows us to quantify the effects of lower pH on marine organisms. Preliminary results show that some impacts are inevitable around the release point, but their severity will depend on the release technology.

OVERVIEW

One option to reduce atmospheric CO: levelsis to capture and sequesterpower plant CO:. Commercial CO2 capture technology, though expensive, existstoday. However, the abilityto dispose of large quantities of CO2 ishighly uncertain. The deep ocean is one of only a few possible CO: disposal options (othersare depleted oil and gas wells or deep, confined aquifers)and is a prime candidate because the dccp ocean is vast and highly unsaturated in CO:. The term disposal isreallya misnomer because the atmosphere and ocean eventually equilibrateon a time scale of 1000 years regardlessof where the CO: is originallydischarged. However, peak atmospheric CO: concentrationsexpected to occur in the next few centuriescould be significantlyreduced by ocean disposal. The magnitude of thisreductionwill depend upon the quantityof CO2 injectedin the ocean, as well as the depth and locationof injection. Ocean disposal of C O 2 will only make sense ifthe environmental impacts to the ocean are significantly lessthan the avoided impacts of atmospheric release. We are currentlyexamining these ocean impacts through a multi-disciplinaryeffortfunded by the U.S. Department of Energy designed to summarize the current stateof knowledge. The end-product will be a report issued during the summer of 1996 consistingof two volumes: an executive summary (Vol. I) and a seriesof six,individuallyauthored topicalreports CVol II). The six topicalreports cover the foUowing subjects: (I) (2) (3) (4) (5) (6)

CO: loadings and scenarios Near fieldpcrfurbations Far fieldperturbations Impacts of CO: transport Environmental impacts of CO: release Policyand legalimplicationsof CO: release

In this paper, we review our methodology and present some preliminary results for the near field perturbation modeling and its related environmental impact.

999

10o0

HERZOG et al.: ENVIRONMENTAL IMPACTS OF OCEAN DISPOSAL OF CO2

NEAR FIELD PERTURBATIONS Five injection scenarios will be evaluated using separate models to describe three-dimensional concentration distributions of excess CO= and trace gases such as SO= and NOx. Currently, the following three injection scenarios are being modeled: Dry ice injection from the ocean surfac,, Our model incorporates the experimental dissolution rate of dry ice measured by the Japanese Central Research Institute for the Electric Power Industry (CRIEPI) [1] and a diffusion coefficient that includes both wake effects from the f~lling cube and ambient turbulence [2]. Unconfined release at 1000-1500 m forming a droplet plum,,- A schematic of a plume is shown in Fig. 1. We model the near-field of the droplet plume by modifying the model developed by Liro, et al. [3]. Lateral spreading due to plume intrusion in the stratified ambient was the calculated based on Jirka et al. [4,5]. Downstream ambient diffusion calculations are based on Brooks' model [6] using data from Okubo [2]. Confined release at 500-1OOO m forming a dense plum~ Dissolution of CO2 increases seawater density [7]. It has been shown that with a confined mixing vessel, concentrations can be increased sufficiently to generate a gravity current [8]. The mixing and ultimate depth reached by the plume are governed by the release depth, the initial CO: concentration, the bottom slope, the ambient stratification, the Coriolis force, and the entrainment and drag coefficients [7,8].

Modeling of the other two scenarios -- confined release below 500 m forming a hydrate plume and very deep release below 3000 m forming a "CO2 lake" -- is scheduled for later this year.

a

300 200 100 0

-ioo -200 -300

b

Intrusion Model

-960

Diffusion Model c

-980 -1000

g-lo2o ~'-1040

O

ambient

-1060

current

density profile

-1080 -1100 Plume Model

Fig. 1. Intrusion due to the droplet plume release of emissions from one standard power plant from 10 ports at a depth of 1100 meters in a 5 cm/s current: a) plan view at a depth of 1000 m, b) elevation view along centerline.

I001

HERZOG et al.: ENVIRONMENTAL IMPACTS OF OCEAN DISPOSAL OF CO2

For each scenario, two different loadings are considered: emissions from a single standard plant and emissions from ten standard plants. A standard plant is defined as a 500 MWo (net) coal-fired power plant capturing 90% of its CO2, resulting in 130 kg/s of CO2 being injected into the ocean. Also, the values of engineering and environmental parameters must be set. For example, engineering parameters for the droplet plume scenario include number of diffuser ports and initial droplet size, with the ambient ocean current and initial pH profile being environmental parameters. For our purposes, we chose reasonable values of the parameters to form our base case and then perform sensitivity studies on key parameters. For example, at a depth of 1000 m, our models use a base case value for ocean currents of 5 cm/s, but we also model cases with 2 and 10 cm/s currents. For a given scenario and set of parameters, the models are run assuming steady-state to yield a spatial distribution of CO2 concentration around the injection point. These concentrations can then be converted to pH and compared with ambient pH profiles. The result is a map of the change in pH near the injection point, as illustrated in Fig. 2 for the unconfined release scenario.

2000 1500 ._1000

8

"~

500 ~6

-1000 -1500 -2000

0

5

10 15 Distance (kilome~m)

20

25

Fig. 2. pH distribution at a depth of 1000 m for scenario illustrated in Fig. 1.

Next, the range of time-histories of passive organisms traveling through the plume can be determined. The average ocean CO2 concentration experienced by an organism passing through the region can be calculated by assuming that the organism is subject to the same diffusivity as the CO2. At each depth, several lateral sections are selected and a representative experience is found. Fig. 3 reports pH-time experiences for organisms at different distances from the plume centedine at a depth of 1000 m for the dry ice scenario, while Fig. 4 compares the experience at the eenterline for three injection scenarios modeled to date.

1002

HERZOG et al.: ENVIRONMENTAL IMPACTS OF OCEAN DISPOSAL OF CO2

7.6

pH

7.4 7.2

20~enterline

7.0 6.8 6.6 0

. . . .

~

. . . .

J . . . . I 4 Time (hours)6

. . . .

i 8 ....

1'0

Fig. 3. pH-time experience of organisms entering the plume at different distances from the centerline at a depth of 1000 m. The scenario illustrated is dry ice disposal of emissions from one standard power plant in a 5 cm/s current. Dry lee

7.5 7.0

Droplet Plume f

6.5

pH

6.0 5.5 5.0 Dense Plume

4.5 4-0 0

'

'

' 5b

. . . .

16o

. . . . .

Time (hours)

. . . .

260

. . . .

Fig. 4. pH-time experience of organisms entering on the centerline of the CO,_distribution at 1000 m. All scenarios assume emissions from one standard power plant in a 5 cm/s current.

ENVIRONMENTAL IMPACTS To determine the near field environmental impacts, the pH-time experiences discussed in the previous section must be combined with a model that predicts impacts on marine organisms from reduced pH. Marine life can be classified into four groups: phytoplankton, zooplankton, nekton, and benthos. The groups that will be affected by a CO2 plume depend on the injection scenario. In this paper, we will not be discussing benthic impacts. Further, since phytoplankton live above the depths affected by the plume and nekton will be able to avoid the plume, the environmental impact assessment presented here focuses on zooplankton. We assume conservatively that they are passive and have no CO2 avoidance ability. Knowing the volume of water that will experience a given time history of pH, and knowing the population density of organisms in the water is sufficient to estimate the numbers of zooplankton affected by the CO2. Most studies on the effects ofpH have been on fish in acidified lakes in response to acid rain concerns. Still, several useful studies have been performed on marine zooplankton in which the mortality was assessed for animals exposed to a constant pH for differing times or animals exposed to various pHs for a given time [9,10,11,12]. Results from these studies have been superimposed on Fig. 5 to give an approximate map of expected mortality as a function ofpH and exposure time. Isomortality curves arc drawn for 0%, 50%, and 90% mortality as guides.

HERZOG et al.: ENVIRONMENTAL IMPACTS OF OCEAN DISPOSAL OF CO2

1003

3 erustace .ans [ 121

O0

(~'*"~'

3.5. 4, 4.$-

crustace'a~2]l 5-

O% mortality

S.$~

. . . . . . 50% mortality *Is =%mortality CZ:::::~ = marine

fi.he

"l'w~l | i claoocerans [21] t II ayfi~[18]

, i ,2~-'.,~h 1

i*lli I [0or~

_ ggs~l'5j |",,

Ida. s_~.! ]]

I, , I"'ll

", ~

i

~ lOOl

i

i

i

i fish eggs [17]

i

~.

X~h*° ! t ~ s.J!°*It :~)'~*_-~,~J '~l L--u I - ~..

6°~[-

,:~~ 7-

= freshwater species

copepods'

~ oyster 1~rvae [10]

7.$8

1 time (!~) Fig. 5. Mortality of aquatic animals (zooplankton and benthos) due to time-exposure to low pH. Based on laboratory or field studies taken from the marine and freshwater literature. 'Isomortality' curves are approximated at 0% and 50% mortality (LCs0) using the more sensitive species. The near field perturbation results shown in Fig. 4 can be plotted as discrete points to yield the number of hours over which various sections of the plume experience different values ofpH (see Fig. 6). However, note that the pH varies with time in these exposure curves, while the data points in the literature studies are all obtained at constant pH. Therefore, a method must be devised to estimate the cumulative impact of varying pH exposure. One approach is to use the isomortality curves to convert exposures into a common metric. For example, on Fig. 6, an exposure of 50 hours at pH 6.0 causes roughly the same 10% mortality as an exposure of 15 hours at pH 5.5. Hence, we can approximate an exposure of 50 hours at 6.0 plus 10 hours at 5.5 as 25 hours at 5.5. For such an exposure, the graph gives a mortality of 50%. Alternatively, converting the exposures to a metric of hours at 6.0 gives a total exposure of 90 hours corresponding to about 40°/. mortality. The table below illustrates the results from this procedure and compares it with two alternative methods described in the literature [ 13,14].

Table 1. 'Worst case' mortalities for affected organisms entrained at center of plume Method Scenario Dry Ice Droplet Plume Dense Plume

This Paper 0% 40% >90%

Schubel [ 13] <0.1% 32% 100%

Mattice and Zittel [ 14] 0% 45% >90%

1004

HERZOG et al.: ENVIRONMENTAL IMPACTS OF OCEAN DISPOSAL OF CO2

The mortality figures in Table 1 are estimates of the percentage loss of organisms experiencing the most severe pH conditions for each scenario. The number of organisms experiencing these worst case conditions, as well as the less extreme conditions further from the center of the plume, vary with the scenarios. Generally, the most severe impacts are associated with the smallest number of organisms. To convert these mortality percentages to real numbers of organisms affected per unit time, we will analyze plume volumes and survey "known information on the densities and kinds of fauna at projected disposal sites.

3.5-

44.$55,5-

6-

~)

6.5"

®

~

....

7-

®

7.58

I

g

I

: : ::::l

-

I

l

: " ::::I

:

[] :

: : ':::i

..

:

:

:

: ::::!

!

l

time (hr.)

Fig. 6. Estimated exposures of zooplankton and benthos for several CO,_disposal scenarios. Continuous exposures to gradually varying pH from Fig. 4 have been approximated as discrete points. These point exposures are superimposed onto mortality curves fi'om Fig. 5.

CONCLUSIONS While there are several important environmental impacts of ocean disposal of CO2, the acidification around the release point may be the most important. Our preliminary results indicate that impacts around the release point arc inevitable. However, the size and severity of the impacted area will depend on the exact release technology. For example, the dry ice release scenario has negligible impacts, while the dense plume scenario has large impacts, but only over a limited area. The results presented in this paper arejust one piece of information that is required in doing a comprehensive analysis of the CO2 ocean disposal option. Other important pieces include: I) effects of the localized impacts presented hero on larger ocean ecosystems, ii) benefits of reduced atmospheric CO2 emissions in terms of reduced risk of possible climate and ocean ecosystem change [22,23], and iii) cost of this mitigation option compared to alternate options. By doing objective and comprehensive studies of these issues, we are building a "lmowledge base that will allow informed decisions to be made if and when more stringent CO2 emission controls are required.

H E R Z O G et al.: E N V I R O N M E N T A L IMPACTS OF O C E A N D I S P O S A L O F CO2 REFERENCES I.

N. Nakashiki, T. Ohsumi and K. Shitashima, "Sequestering of CO,. in the Dccp Ocean: Fall Velocity and Dissolution Rate of Solid CO,. in the Ocean," Technical Report EUglO03, Central Ra'Jearch Institute of Electric Po~,~erindustry, Abiko, Japan (1991).

2.

A. Okubo, "Oceanic Diffusion Diagrams," Deep-Sea Research, 18, pp. 789-802 (1971).

3.

C.R. Lira, E.E Adams and H.J. l"lerzog,"Modeling the Release of CO: in the Deep Ocean," Energy Convers. Mgmt, 33(5-8), pp. 667-674 (1992).

4.

G.H.Jirka, J.M. Jones and F.E. Sargent, "Th~x~retiealand Experimental Study of the Intermedial~ Field Dynamics of Ocean Thermal Energy Conversion Plants," (Progress report for 1978-1979), CorneU University, Ithaca. NY (1980).

5.

R.A. Paddock and J.D. Ditmars, "Initial Screening of License Applications for Ocean Thermal Energy Conversion OTEC Plants with Regard to Their Interaction with the Environment," Technical Report ANIJOTEC-EV-2, Argonne National Laboratory, Chicago, IL (1983).

6.

N,H.Brooks, Diffuser of sewage effluent in an ocean current. In Proc. 1st International Conference on Waste Disposal in the Marine Environment, pp. 246-267. Pergamon Press (1960).

7.

H. Drange and P.M. Haugan, "Carbon Dioxide Sequestration in the Ocean: The Possibility of Injection in Shallow Water,"Energy Convers. Mgmt, 33, pp. 697-704 (1992).

g.

E.E.Adams, D. Golomb, X.Y. Zhang and H.J. Herzog, Confined release of COz into shallow seawater. In Ocean Disposal of Carbon Dioxide (Edited by N. Honda and T. Ohsumi), pp. 153-161. Terratub, Tokyo (1995).

9.

ILN.Bamber, "The Effects of Acidic Seawater on Young Carpet-Shell Clams Venernpis Docussata (~) (Mollusca: Vaneracae)," Journal of,Experimental Marine Biology and Ecology, 108, pp. 241-260 (1987).

10. A. Ca•abrese and H.•. Davis• ``Th• pH T••eran•e •f Embry•s and Larva• •f Mercenaria mercenaria and Crassotrea virginica,'" Biological Bulletin, 131, pp. 42%436 (1966). I 1. G.D. Grice, E. Haogland and P.H. Wiebe, "Acid-Iron Waste as a Factor Affecting the Distribution and Abundance of Zooplankton in the New York Bight," Estuorine and Coastal Marine Science, 1, pp. 45-50 (1973). 12. M. Havas and T.C. Hutchinson, "Aquatic Invertebrates from the Smoking Hills, N.W.T.: Effect of pH and Metals on Mortality," Can..I. Fish. Aquat Sci., 39, pp. 890-903 (1982). 13. J.R.. Schubel, C.C. Coutant and P.M.J. Woodhead, thermal effects of entrainment. In Power Plant Entrainment. A BiologicalAssessment (Edited by J.R. Schubel and B.C. Mac),, Jr.), pp. 19-95. Academic Press Inc., New York (1978). 14. J.S. Mortice and H.E. Zittel, "Site Specific Evaluation of Power Plant Chlorination," dour. Water Poll. Control. Fed, 48(10), pp. 2284-2307 (1976). 15. K.E. Holtze and N.J.Hutehinsnn, "Lethality of LOw pH and AI to Early Life Stages of Six Fish Species Inhabiting Pre-Cambrinn Shield Water in Ontario." Can. d. Fish. Aquat. SoL, 46, pp. 1188-1202 (1989). 16. D.C. Rose, G.W. Williams, T.A. Hollister and P.R. Parrish, "Method for Determining Acute Toxicity of an Acid Waste and Limiting Permissible Concentration at Boundaries of an Oceanic Mixing Zone," Environmental Science and Technology, I 1(4), pp. 36%371 (1977). 17. b.l. Mount, "Chronic Effect of Low pH on Fathead Minnow Survival, Growth, and Reproduction," Water Research, 7, pp. 987-993 (1973). 18. D.F. Malley, "Decreased Survival and Calcium Uptake by the Crayfish Orconectes Virilis in Low pH," Can. ,~ Ffsh. Aquat. Sci.. 37, pp. 364-372 (1980). {9. M. Stephenson and G.L. Mackie, "Lake Acidification as a Limiting Factor in the Distribution of the Freshwater Amphipod Hyalella Azteca," Can..Z Fish. Aquat. Set., 43, pp. 288-292 (1986). 20. R. Singer, cffocts of acidic precipitation on benthos. In Acid Precipitation: Effects on Ecological Systems (Edited by Frank M. D'Itri), pp.329-363. Ann Arbor Science (1953). 21. W.E. Walton, S.M. Compton, J.D. Allan and R.E. Dnnieis, "The Effect of Acid Stress on Survivorship and Reproduction ofDaphnia Pulex (Crustacea: Cladocera)," Can. ,Z ZooL. 60, pp. 573-579 (1982). 22. D. Rocmmich and J. McGowan, "Climatic Warming and the Doclin¢ of Zoopinnkton in the California Cun'ent,'" Science, 267, pp. 1324-1326 (1995). 23. J.P. Barry., C.H. Baxter, R.D. Sagarin and S.E. Gilman, "Climate-Ralatod, Long-Ten'n Faunal Changes in the California Rocky Intertidal Community," Science, 267, pp. 672-675 (1995).

1005