Ask the Scholar

Document scope · 1 page
doc
Scholar
Ask about this object, its catalog metadata, its source description, or the page inventory. For page-specific OCR and visual context, open one of the page chats.

Scholar Source Context

Document identity
localId
120359028
label
Re-Seeding of Oceans
core
doc
dtoType
document
pageCount
1
Source metadata
Source extras
naId
120359028
levelOfDescription
fileUnit
otherTitles
42-t-7432005-20120769F-Seg1-030-013-2017
recordType
description
ocrSource
nara-archive
Single page context
seq
1
pageIndex
0
type
document
mediaId
2bb5c3e3a8185b38
ocrText
FOIA Number: 2012-0769-F FOIA MARKER This is not a textual record. This is used as an administrative marker by the William J. Clinton Presidential Library Staff. Collection/Record Group: Clinton Presidential Records Subgroup/Office of Origin: Council on Environmental Quality Series/Staff Member: Kathleen (Katie) McGinty Subseries: OA/ID Number: 2615 FolderID: Folder Title: Re-Seeding of Oceans Stack: Row: Section: Shelf: Position: S 61 5 7 1 GLOBAL ENVIRONMENTAL CHANGE REPORT money for the federal government life cycle costs and are cost-effec- Copies of the OTA report, Energy immediately, and be more than paid tive." All new federal buildings Efficiency in the Federal Govern- back in a few years or sooner." must meet energy performance ment: Government By Good standards, and agencies must reduce Example? are available for US $5 On 17 April, President Bush ordered their motor vehicle fuel consump- from the Superintendent of Docu- each federal agency to develop and tions by at least 10% in 1995 from ments, USGPO, Washington, DC implement a plan to reduce its over- 1991 levels. A "maximum number 20402-9325. Tel: +1-202-783-3238. all energy use in buildings by 20% practicable" of alternative fueled Reference GPO Stock Number 052- from 1985 energy use levels "to the vehicles must be purchased by agen- 003-01242-1. extent that these measures minimize cies each year. Science Update Researchers Protest Their in GECR. Andrew Lacis of NASA the plants and animals that require Inclusion in Skeptics' Goddard, who was listed in the them. Of course, coastal areas and Agenda same proposal as co-investigator, regions of upwelling are more pro- On page 1 of Global Climatic hadn't heard of the meeting or the re- ductive, but they account for only search agenda until GECR tele- 10.1% of the ocean's surface. The Change: A New Vision for the 1990s, the US greenhouse skeptics' phoned him on 13 May. Both Kukla rest, as Colinvaux says, is "a blue and Robinson strongly objected to desert more useless to life than most alternative research agenda (see GECR, 1 March, pp. 1-3), one finds the "consensus" wording in the of Arabia." The Antarctic ocean, the following statement: "The con- prospectus; Robinson has written however, is the most enigmatic ex- sensus of the scientists in this re- to Patrick Michaels and Robert Ball- ception to that sweeping statement. ing (the organizers of the meeting Nutrients are abundant there, and search prospectus is that there is con- siderable evidence that the impact of and the authors of the prospectus' in- yet phytoplankton are comparatively troduction) to ask that the language scarce. Because of the Antarctic's future climatic change may be neutral or even beneficial.' be revised. great unrealized potential for biotic activity, scientists have looked to it But to at least three of the scientists Balling told GECR that he and as a potential modifier of CO2 con- whose proposals were included in the Michaels are polling all of the scien- centrations under different climate agenda, that claim is misleading. Ac- tists named in the prospectus and regimes hence the proposals for cording to one of them, there was no will prepare a statement on the ac- iron fertilization and the widely ac tual level of consensus within the formal consensus reached or even cepted theory that a phytoplankton called for at the meeting which pro- group. Balling says that he and bloom in the Antarctic ocean caused duced the agenda, nor was the intro- Michaels did circulate the agenda's global CO₂ concentrations to drop introduction to the conference atten- duction to the research agenda circu- during the last ice age. lated among the participants for their dees prior to publication, but Kukla approval. Finally, at least four of denies having seen it. As for the re- These ideas, however, are set spin- searchers who weren't invited to the the researchers listed were not even ning by a new study reported in the invited to the meeting, did not know meeting, Balling points out that all 16 May issue of Nature (vol. 351, their proposals were being included, of them were co-investigators. He pp. 220-222). R.A. Mortlock and and were unaware that their names says it was the responsibility of the colleagues from Lamont-Doherty would appear in the prospectus. principal investigators who attended and Columbia University (Palisades, the meeting to notify their co-inves- New York) use three lines of On 6 May, GECR received a letter tigators that their names would be in- evidence to show that phytoplankton from David Robinson of Rutgers cluded in the prospectus. productivity was in fact low, not University, urging us to print a high, during the last ice age, and the clarification. Robinson, who was Ice Ages and Plankton plants were less efficient at taking listed on a project proposal as a co- Don't Mix up nutrients. Mortlock et al. investigator with George Kukla of analyzed records of the accumula- the Lamont-Doherty Geophysical As Paul Colinvaux pointed out in tion rates of diatom shells, the ratio Observatory, was not invited to the his classic 1978 primer on ecology, of germanium to silicon in opal meeting and didn't hear about the re- Why Big Fierce Animals Are Rare, (also related to diatom production), most of the world's ocean is a search agenda until he read about it and the carbon isotope ratio in the desert. Nutrients are scarce, as are shells of foraminifera to reach their VOL. III, NO. 10 1991 5 GLOBAL ENVIRONMENTAL CHANGE REPORT conclusions. As W.H. Berger of keep declining. (Readers should Sedjo told GECR that his analysis Scripps Institution of Oceanography keep in mind GECR's 15 March "flies in the face of the conventional points out in an accompanying ar- Focus Report on the potential vul- wisdom" that Northern temperate for- ticle on pp. 191-192, this analysis nerability of nonfossil energy sys- ests are in balance with regard to CO2. should "severely damage conven- tems to climate, however. One tional belief in Antarctic modulation should also note that Ausubel's of atmospheric carbon dioxide." Upgrade on NIST analysis is based on industrialized However, Berger says the question Refrigerant Mixtures countries.) Ausubel notes that Database Available is still not totally resolved, since ecosystem vulnerability is another there could be alternative explana- question entirely, and does not sug- The US National Institute of Stand- tions for some of Mortlock's find- gest overtly that climate change is ards and Technology (NIST) has ings. nothing to worry about, but he im- released an update to its Thermo- plies that "business as usual" may dynamic Properties of Refrigerants Technologies Reduce yield a more favorable future for and Refrigerant Mixtures Database. Societies' Vulnerability to humanity than many have suggested. The database, which will run on any Climate AT- or XT-class personal computer Reforestation in the (and requires 512 K of memory) al- In the 25 April issue of Nature (vol. lows researchers to produce tables 350, pp. 649-652), Jesse Ausubel of Northern Hemisphere: of thermodynamic properties of 18 The Rockefeller University in New The Missing Sink? pure refrigerants, including ethane- York discusses evidence that tech- According to an article in the 4 May based compounds, and 24 mixtures. nology has allowed human societies issue of New Scientist, the spread of Mixtures of up to five components to become increasingly shielded from forests in the Northern Hemisphere can be analyzed. the effects of climate and weather. could account for much of the miss- Ausubel argues convincingly that ing sink for carbon. Roger Sedjo of The upgrade contains new data on there is a trend toward systems that the Washington-based Resources for mixing parameters for four blends are less vulnerable to climate, citing the Future used forestation data (including HCFC-22 and HFC- examples ranging from food preserv- from the UN Food and Agriculture 152a), and has revised Carnahan- atives to weather satellites. "In Organization to conclude that 700 Starling-DeSantis coefficients for all many respects we seem to be million metric tons of carbon are 18 refrigerants. 'climate-proofing' society, Ausubel being captured in Northern Hemi- Version 2.0 of the REFPROP says. Even agriculture may be more sphere trees and soils each year. database is available for US $340 robust than many people think: Sedjo presented his paper at this from the Standards Reference Data transportation systems allow people to rely on food sources outside their year's annual meeting of the Program, A320 Physics Building, area and thus "spread production American Association for the NIST, Gaithersburg, MD 20899, risks across more climatic zones," Advancement of Science. A second USA. Tel: +1-301-975-2208; Fax: and existing crop cultivars in the US unpublished paper by Sedjo, which +1-301-926-0416.: Upgrading to ver- considers the effect of tropical sion 2.0 from 1.0 costs $90. are generally replaced by new ones within just eight to nine years. deforestation, concludes that while the destruction of tropical forests New Climate Journal Calls Ausubel points out that the adapta- adds 2 billion tons of carbon into the for Papers tions for long-term climatic change atmosphere, much of that will be Climate Research, subtitled Interac- will probably resemble those for reabsorbed while the forests grow. tions of Climate with Organisms, Eco- other climate variations - those Sedjo says the net emission of car- systems, and Human Societies, has that "mollify the difference between bon from tropical deforestation is issued a call for papers. The new daytime and nighttime temperatures, 700 million tons, coincidentally the journal is published by Inter-Research protect against normal variability be- same amount that's being absorbed in Germany, and appears to be in the tween days, shield from storms and by Northern Hemisphere forests. same league as Climatic Change and hail, adjust to the seasons, and adapt Last year, Taro Takahashi and Pieter Global Environmental Change: to the wide range of climates where Tans reported in Science that oceans Human and Policy Dimensions. people already live." absorb only at most one billion tons of carbon, and there must be a large Climate Research invites papers on Ausubel points to the steady im- terrestrial sink (GECR, 9 March interactions of climate with or- provements in technological perfor- 1990, p. 5). Sedjo's findings could ganisms, populations, ecosystems, mance and efficiency, suggesting explain "a large amount" of the miss- and human societies; short- and that this trend will continue and that ing sink. long-term changes in climatic ele- human vulnerability to climate will ments; methodological aspects of 6 17 MAY 1991 VOL. III, NO. 10 ALBERT GORE, JR. 393 RUSSELL SENATE OFFICE BUILDING TENNESSEE PHONE: 202-224-4944 United States Senate WASHINGTON, DC 20510-4202 May 7, 1991 Discover Magazine 3 Park Avenue New York, N.Y. 10016 To the Editor: John Martin proposes iron fertilization of the Southern Ocean as at least a partial solution to the threat of global warming. ("Earth On Ice", April 1991). In my opinion, he's got it exactly wrong. Our manhandling of the global environment created the problems we face; it is not part of the solution. It would be easy to attack Martin's suggestion on technical grounds: there is evidence that iron is not the limiting nutrient in the oceans; very recent research shows that, even if adding iron would increase plankton populations, the net amount of carbon sequestered would be modest; massive amounts of money and other resources would have to be mustered if the program were to be even marginally successful in achieving its objectives. But my opposition to Martin's suggestion runs much deeper. The proposal is dangerous because it plays to wishful thinking that there is no need for fundamental change in the design of our civilization; and that, with a little creative engineering, we can out-smart nature once again. The truth, however, is that we cannot, without serious repercussions, continue our hubristic campaign of re-creating the environment. You would think we would have learned that lesson by now. This is certainly not the first time we have undertaken such technological "fixes". Our determination to produce crops in areas not suited to produce them, for example, has poisoned the soil and, in some areas, created vast deserts of salt and pesticides. And consider our desire to adorn ourselves cheaply in silk by bringing gypsy moths into the country -- vast stands of diseased and dying trees are the legacy of that venture. The kudzu vine that chokes native plants and trees in the Southeast; the rabbits that overran Australia; and the Africanized bees that have now made their way into the United States, are all testimony to our inability to gauge and control the effects of grand eco-engineering schemes. And those are nothing compared to this proposal. It is time we stop kidding ourselves. The simple truth is that it is only by changing our behavior -- and our relationship to the earth -- that we can effectively address the global environmental problems we face. Study after study tells us we can make make significant gains at minimal cost with existing technologies. The need for such action is clear. The ozone layer is being depleted at an alarming rate; we need to move quickly to end the use of ozone destroying chemicals. We continue to experience record warmth, severe drought, extensive forest fires, and other warning signals consistent with those long since predicted to accompany the onset of global warming and climate change; we need to cut our emissions of greenhouse gases in every sector of the economy. The very threads of the web of life are being severed as rampant deforestation drives thousands of species to extinction; we need to preserve and protect sensitive ecosystems and manage our forests sustainably. So let's get on with it. I challenge our scientists to use their insights and creativity to devise ways for us to live in harmony with nature -- rather than speeding us further along on the collision course we are currently travelling. al Dore Al Gore United States Senator Chairman, Senate Subcommittee on Science, Technology, and Space PHOTOCOPY MISC. HANDWRITING will send + hanks DISCOVER March 21, 1991 The Honorable Al Gore United States Senate 393 Russell Senate Building Washington, DC 20510 Dear Senator Gore: Enclosed is the most recent issue of DISCOVER. It contains the article ``Earth on Ice'' on page 54, which we thought might interest you. On our letters page we like to encourage an exchange of views about our articles. We'd be happy to consider any letter that you might want to write. Thank you for your time. Sincerely, Clarence V. Reynolds Bitoy Hanson Betsy Hanson Clarence V. Reynolds Researchers/Reporters 100% A FAMILY MEDIA PUBLICATION, 3 PARK AVENUE, NEW YORK, NY 10016 212-779-6200 FAX 212-725-3962 recycled ALBERT GORE, JR. 393 RUSSELL SENATE OFFICE BUILDING TENNESSEE PHONE: 202-224-4944 United States Senate WASHINGTON, DC 20510-4202 April 16, 1991 Discover 3 Park Avenue New York, N.Y. 10016 To the Editor: Responding to your article "Earth on Ice" (April 1991), I must express my concerns regarding the iron fertilization of phytoplankton in the Antarctic as a means of lowering atmospheric CO2 levels. The concept of iron fertilization as a "fix-it" for reversing the greenhouse effect begs the issue, and proposes a simplistic solution to a very real and complex problem. Certainly the concept may be embraced as a means of conducting business as usual, but presenting iron fertilization as a panacea for this global problem is misleading. The concept of iron fertilization in the Antarctic has generated critical research and discussion that question the plausibility and desirability of adopting such a proposal. Most notable is Peng and Broecker's (1991) research on ocean dynamics that concludes negligible to insignificant lowering of atmospheric CO2 by iron fertilization after 100 years of totally successful fertilization. Their results suggest that with a "business-as-usual" scenario of fossil fuel CO2 emissions in the next century, the CO2 content of the atmosphere would be lowered by only about 5-15% below what it would have been without fertilization. What are the costs of this concept? As you mention in the article, the costs of testing the iron fertilization of a small 60 mile X 60 mile square of ocean would run between the tens of millions of dollars and more than a hundred million. Once begun, the fertilization process would require even, annual dispersal of iron for at least 100 years to ensure the retention of carbon. The fertlization process would involve the use of hundreds of barges and ships, their transport and upkeep, the purchase of soluable iron, and the payment of labor -- at an estimate in the billions of dollars (Borchelt 1990). Given the annual costs of such an operation over a 100 year period, would a lowering of atmospheric CO2 levels by 5-15% be worth it? Furthermore, the environmental costs of the iron fertilization concept need to be fully explored and weighed. Many scientists are concerned that the long-term ecological effects of iron fertilization may be devastating. For example, successful fertilization would increase the amount of organic material in the deep ocean water. In the worst case scenario, reoxidation of these organic materials would diminish the supply of oxygen in the Antarctic deep water to the point of anoxic condition. Repercussions would include the loss of krill ALBERT GORE, JR. 393 RUSSELL SENATE OFFICE BUILDING TENNESSEE PHONE: 202-224-4944 United States Senate WASHINGTON, DC 20510-4202 eggs and krill upon which whales, seals, squid, fish, and penguins feed. Such a loss of oxygen in the deep ocean would liken the area to a devastated and less productive lake environment. The concept of iron fertilization focuses our attention on short term "fix-it" approaches in dealing with the greenhouse effect rather than on changing the attitudes and behaviors that produce CO2 emissions in the first place. If we expended as much money, time, and energy that the iron fertilization scheme proposes over the next 100 years into reducing the primary production of CO2 emissions, the problem would be solved -- or, at least, substantially reduced. Sincerely, Al Gore U.S. Senator REFERENCES: Peng, T.H., and W.S. Broecker 1991 Dynamical limitations on the Antarctic iron fertilization strategy. Nature 349: 227-229. Borchelt, Rick 1990 Enlisting Marine Algae To Ease Global Warming. NewsReport. December 1990-January 1991: 5-7. Clinton Presidential Records Digital Records Marker This is not a presidential record. This is used as an administrative marker by the William J. Clinton Presidential Library Staff. This marker identifies the place of a publication. Publications have not been scanned in their entirety for the purpose of digitization. To see the full publication please search online or visit the Clinton Presidential Library's Research Room. THE HUNT FOR BROWN DWARFS DISCOVER THE NEWSMAGAZINE OF SCIENCE APRIL 1991 $2.95 EARTH ON ICE A Bold Attack on Global Warming 04 37151 0 285567 7 NEWSAND VIEWS BIOGEOCHEMICAL CYCLES DAEDALUS Ironing out greenhouse effects In two minds Robin S. Keir EACH hemisphere of the human brain has its own specialism. The left hemisphere it has been proposed that CO. could be atmospheric and Antarctic surface water (which dominates the right side of the removed from the atmosphere and there- Peo. is maintained at & larger value. The body) is good at language and logic; the bv its greenhouse effect reduced by using consequent reduction in the rate of right hemisphere (dominating the left side) the so-called from ypothesis of Martin increase of atmospheric CO. is also some- has better spatial awareness and emotional According to this ideaviton is thellimiting what greater. intuition. Daedalus points out how our nutrient for biological growth in Antarctic Surely in looking at how to tackle the view of a knotty problem tends to shift surface waters. which have abundant threat of greenhouse warming, we should cyclically back and forth every few hours or phosphate: nitrate and silicate: By fertiliz- not focus on which model of iron fertiliza- so, This, he says, is our two hemispheres ing. this area with iron. phytoplankton non gives the best result (the possibility of looking at the matter in turn: growth would be greatly, stimulated. and undesirable ecological effects put to one He relates this alternation of mood to our the surface-water's partial pressure of CO. side for the moment). but rather we nostrils, which have to warm and humidify would lie greatly reduced - by more than should worry about the increasing rates of the air before It reaches our lungs. It's a 150 parts per million (pip.m.). But even if fossil-fuel consumption. The total scale of hard job. It can drain 5-15 watts of heat the fertilization were completely success- fossil-fuel reserves is not known with any continuously from the warming and wett- full: the effect on atmospheric CO. would certainty but most estimates are between ing surfaces, and the nostrils tend to open he much smaller according to Peng and 0 and 12 times the present amount of alternately, a few hours at a time: (You Broccker on page 227 of this issues Their carbon in the atmosphere as CO,, mostly notice this when you have a cold.) Each model calculations show that after 100 as coal Barring increased uptake by the hemisphere of the brain puts out only years of successful fertilization. the atmos- terrestrial biosphere. which seems un- about 10 watts of heat. The active 0 11. pheric CO. content would be reduced only likely because of deforestation, 70-80 per draining heat copiously from the hernis- by about 30 P.P.M. an amount which is cent of this carbon will accumulate in the phere close above it, must cool it strongly The brain is very sensitive to temperature, PRESERVATION also small in comparison to the hundreds atmosphere if it is burned over a period of PHOTOCOPY of parts per million- the concentration is a. few hundred years. regardless of the growing feverish in the warm and sleepy in projected to rise during this time owing to actual history of the CO₂ generation over the cold. So as each nostril opens up in fossil-fuel burning this period When we have consumed turn, the hemisphere on that side grows The authors point out that the problem most of the fossil fuel. the atmosphere will sleepy, and our mood swings accordingly. is that the capacity of the Antarctic surface have reached a level of at least 1.500 In this connection, Daedalus recalls the water bvitself to take up atmospheric CO P.P from. its present level of around traditional yoga exercise of breathing In is, very small. Therefore this region acts 345 p.p.m. at this point it may not mat- through one nostril and out the other. The only as an interface through which addi- ter whether a sequestering effort reduced nostril of Inflow gets no relief from warm tional Owenald be sequestered from the the atmospheric CO. by 30 p.p.m. or exhaled breath, and must cool its h atmosphere and transferred into the large 300 p.p.m. sphere very strongly - a subtle yogic way volume of intermediate and deep waters However, exactly when, the maximum of opening the mind to new percepti 3: The rateiof this transfer depends upon the CO. level will be reached can vary greatly. Daedalus's psychonasal air-conditioner' nature of the circulation the speed with For example, a slowly growing production provides. each nostril with its own air, which the upper waters are replaced and of CO. may result in only a doubling of the of defined temperature and humidity It the extent to which they remain in contact atmospheric content by the year 2100, but can fire up or chill out either hemisphere with the atmosphere afterwards expansion of CO, production at the pre- at will. Plumbed into the prototype, The model simulations on page 228 1973 rate of 4.5 per cent a year could lead DREADCO volunteers are tackling the (their Fig. 3) illustrate this aspect quite to levels over. 1:200 p.p.m. by then? usual psychiatrists' battery of insulting clearly Initially there is artransitory peri- Because fossil fuel is a finite resource, we questions and emotional ordeals. With its od of few years during which the CO. will need to find an alternative within the aid, the sternest logicians are enjoying partial pressure (P... ) in the surface water next two or three centuries (sooner if sub- flights of fancy, while poets buckle down decreases rapidly and subsequently recov stantial consumption of coal is to be to mathematical proofs. on to within 30 p.m. or so of the atmos- avoided) The slower that fossil fuel is The final 'Psyconditioner' will be a pheric value The latter evidently occurs burned in the meantime, the better. special hat with concealed plumbing and because of rapid exchange of CO: across Neither iron fertilization nor any other heat-pump, and integral nose-plece. By con- the sea surfaces After this transient, the method of CO. sequestration can be con- trolling the wearer's nostril-air, it will surface ni maintains a nearly constant sidered a serious alternative. expand his consciousness under full control. disequilibrium with the atmosphere The Modern, highly verbal society tends to put resulting decrease in atmospheric COMs Robin S: Keir Is at GEOMAR, Forschungs- the right hemisphere at a disadvantage; zentrum für Marine Geowlssenschaften an der quite sligh, the time constant is probably Christian-Albrechts-Universität zu Kiel, Wisch- the Psyconditioner will allow anyone to get of the order of a few hundred years The hofstrasse 1-3. 2300 Kiel 14, Germany in touch with his Intuitive nature. Its Antarene surface ater is Bushed everv wearer will be able to look at a problem first years In the en ulation, but for every 1 Martin - Paleocesnography 6. (1990) logically, then intuitively then (with the turnover the water takes up only. small 2. Peng: TH & Breecker. W S: Nature 349, 227-229 aid of a special superheater) with fevered part of the atmospheric CO TRO that the 1991 Wenk t & Siengenthal V In The Carbon Cycle: and imagination and even with fevered logic time constant for O. removal is about Atmospheric CO2: Natural Variations Archeen to Present At. a picture-gallery* or social function he 100/times the water residence time (SOE Sundquist. E.T. & Broecker, W. $.) 185-194 (Ameri- will engage his right hemisphere, while the can Geophysical Union. Washington DC. 1985). The result found hv Pengand Broecker 100% Sarmiento. 1. L.& Siegenthaler. U. Nature (In the small print of an insurance-contract will is Friund (4.) be somewhat model. dépen- press) engross his left one. The most inhibited or 5. dent Indeed Joes at lusing-a different Ratty, R: M. & Marland/G in Interactions of Energy and Climate. (eds Bach,, W., Penkrath J: & Williams, J.), undisciplined wearer will be able to experi- circulation pattern through the Antarche 191 212 (Reidet: Dontrecht, 1988) ence previously suppressed aspects of his surface region and darger/surface area Keeling C D. & Bascastow, R.B. in Energy and Climate 72 -95 (National Academy of Sciences. Washington CC, personality - and suppress them again as had the diséquilibrium between the 1977) necessary. David Jones, 198 NATURE VOL 349 17 JANUARY 1991 PHOTOCOPY LETTERS TO NATURE PRESERVATION duced possible source for the two-level systems would bei tipping of The CO2 partial pressure in surface ocean water is influenced occtral the central phenyl ring of terphenyl between various allowed by the extent to which plant growth reduces the total CO₂ and orientations that are made accessible by disorder in the crystal content. SCO2 of surface water. The magnitude ofthis reduction effect structure" Validation of this model requires detailed tem depends on the efficiency with which the limiting nutrients 10 be perature-dependent measurements of the jump rate for a single phosphate and nitrate are utilized In today's temperate and turba. centre over a wide range of temperature in hope of observing tropical ocean the utilization efficiency is high and therefore the the characteristic oth(E/2kT) dependence, where E is the reduction of the CO2 partial pressure. Pco. is near maximum anner. energy splitting in an asymmetric two-level model and k is By contrast. the utilization efficiency is low in the polar oceans ved to Boltzmann-s constant Hence, were a means to be found to increase the efficiency of juency ng the As the spectral jump rates and sizes that we observe are nutrient utilization in these waters. their CO. partial pressure uggest different for different defects; direct observation of this spectral and in turn that for the atmosphere, could be reduced. Martin have diffusion effect would not be possible in any spectroscopy that and his workers have shown in incubation experiments averages over many defects. Direct measurement of the that plant growth rates in waters from polar regions can be jower behaviour of individual guest molecules in a glass would lead greatly accelerated by adding trace amounts of dissolved iron show to a clearer understanding of the origins of dynamics in amor- Iron is an essential micronutrient required for the metabolism phous solids than is available from ensemble-averaged of all forms of life; its primary function being cytochrome from seem measurements. Much more intriguing is the fact that we have formation. Because iron is one of the most particle-reactive sen spectral jumping in a crystalline environment, where such elements, its concentration in the sea is very low, with the lowest width data effect is not expected. values occurring in regions like the Antarctic that are most removed from the continents (opent To date, most of the attention regarding the Manin proposal either Received 1 November accepted December 1990 has been focused on its biological aspects and little attention nearly 1. Moerner, W E: & Kador, L. Priy 5. Rev: Lett, 62, 2535-2538 (1989) usiline has been given to the dynamical aspects Even though/successful 2. Kador L: Horne D E. & Moerner W. E. phys. Chem 94, 1237-1248 (1990) iron fertilization would initially produce up to a twofold ggests 3 Orit M. & Bernard: J. Phys. Rev. Lett 65; 2716 (1990) 4. Itano W. M. Bargquist 1.C. & Wineland D. J. Science 237, 612-617 (1987) decrease in the CO₂ partial pressure in Antarctic surface waters. topose 5. de Vries, H. & Wiersma, D. A. J. chem Phys as, 897-901 (1978) this reduction might be In the absence of water where B Dison R. W& Fayer. M. D. , phys. Chem 2001-2004 (1980) lically Stoneham A. M. Rev., mod. Phys 41. 82-108 (1969). circulation; only an amount of CO. equal to that removed as a 3 Yen. W. M & Seizer: P. M. (eas) Laser. Spectroscopy of Solids (Springer: Bertin. 1981) result of the initial iron fertilization would be sequestered from 15 may 9 de Vries, M. & Wiereme 5. A , chem. Phys. 72, 1851 1863 (1980). the atmosphere. In this case, the atmosphere the Antarctic ns, by 10. de Vries H. & Wiersma, D. A. I chem. Phys 70, 5807-5822 (1979) 11. Gardiner C. W. Handbook of. Stochestic Methods 78-79 (Springer. Berlin, 1983). surface ocean and the non Antarctic surface ocean would ssisted 12 Nagoumey W Sandbarg 1 & Denmelt. H. Phys Rev. Lett. 58, 2797-2799 (1986) nd the 13 Moerner: W. É. & Carter. T. Phys Rev. Lett. 59. 2705-2708 (1987) 14. Patterson F. G: Lee. H W. H Wilson, W. L. & Fayer, M. cnem. Phys 84; 61-60 (1984) num. 15 Phillips. W.. A. (ed.) Amorphous Solids: Low Temperature Properties (Springer, Berling 1981). One 16. Sussman AS Phys. kondens Materie 2 146-160 (1964). CO₂ exchange Well mixed atmosphere CO₂ exchange 17 Baudour. L.. Delugeard, Y & Calliaau. H. Acta Crystallogy. 832, 150-154 (1978) 18: Golding B. & Graebner, J.E. in Amorphous Solids Low Temperature Properties (ed. Phillips, W. A.) 0 Surface ocean mixed layers 107-134 (Springer. Berlin. 1981). 75 19 Friedrich 1. & Hearer, D. in Optical Spectroscody of Classes (ed. 2schokke. I.) 148-198 (Reidel, Dordrecht. 1986). ACKNOWLEDGEMENTS. We thank M. Orrit for providing a copy of his manuscript prior to publication Depth ocean (m) Antarctic (10%) Diffusion-zone This work was supported in part by the US Office of Naval Research Non-Antarclic (90%) 1cm Dynamical limitations on the 2,000 Antarctic iron fertilization W strategy Well mixed sea T-H. Peng* & W. $. Broeckert 4,000 Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831. USA CO2 exchange Well.mixed atmosphere CO. exchange 1 Lamont-Doherty Geological Observatory. Columbia University, Palisades, 0 Surface ocean mixed layers New York 10964, USA 75 MARTIN et al. have proposed an ingenious means by which the rise in stmospheric CO2 content generated by the burning of fossil Antarctic could be stimulated by the addition of dissolved iron, Depth in (m) Antarclic (10%) Diffusion zone is that plant production In the nutrient-rich surface waters of the Non Antarctic 190%) Iem fuels and deforestation might be partially compensated. The idea thereby reducing the CO₂ partial pressure In these waters and 2,000 allowing CO2 to flow from the stmosphere into the Antarctic W Ocean. We have used a box model callbrated with translent tracer data to examine the dynamical aspects of this proposal, and Well mixed deep conclude that after 100 years of totally successful fertilization the CO₂ content of the atmosphere would be lowered by only 10 ± 5% 4,000 below what It would have been la the absence of fertilization So 0 F.IG. 1 Linked vertical advection and diffusion model used to ovaluate the if after 100 years the CO₂ content of the atmosphere were 500 µatm response to iron fertilization of Antarctic surface waters The upper panel without fertilization, it would be between 425 and 475 patm with shows the case where the water upwelled in the Antarctic is transferred full fertilization. In other words, If our model callbration is correct, laterally to the non-Antarctic surface ocean The lower panel shows the single even if iron fertilization worked perfectly it would not significantly case where this upwelled water is converted to deep water and transferred centre reduce the atmospheric CO2 content. directly to the model's deep reservoir 227 NATURE VOL 349 17 JANUARY 1991 PHOTOCOPY LETTERS TO NATURE PRESERVATION 3.0 10 Ocean as measured during the GEOSECS surveys of the world a 100 ocean 10 These results show that the Antarctic waters contain 2.5 far less bomb-produced radiocarbon than invaded this region from the atmosphere.' By contrast, the temperate region of the 2:0 Southern Hemisphere ocean has an inventory exceeding the 0 PO, kg amount received by invasion. This excess roughly balances the 1.5 deficiency in the Antarctic. This pattern can only be explained by upwelling in the Antarctic coupled with lateral transport to 1.0 the adjacent temperate regions!! The upwelling rate required to explain the Antarctic deficiency is 15 m yr (ref. 11). As the low bomb-radiocarbon water in the Antarctic. Ocean covers 0.5 10% of the total ocean area, this corresponds to a flux of 17 SV (17 10° -52 0.0 0 500 1,000 1,500 2.000 In addition to upwelling. account must be taken of the extent 2,400 of vertical mixing. The mean penetration depth of the bomb h radiocarbon for the fifteen GEOSECS radiocarbon profiles for 100 latitudes south of 49°S is 260 m (ref. 11): This extent of penetra- tion into the upwelling plume can be explained using a vertical 2,300 10 eddy diffusivity of 5 (the steady-state tracer penetra tion depth into such a plume would be 630 m). Hence, if this 0 CO2 (fimol interpretation of the homb-radiocarbon data is correct, the 2,200 volume of water immediately beneath the Antarctic surface available for CO., storage is only 2.2> 1016 m² (630 3.5x 10" m') or 1.8% of the ocean's volume, a volume too small to 2:100 house a significant amount of excess CO2 Thus it is the fate of the upwelled water that matters. Additional storage capacity will be provided by that portion of the upwelling water that is 2.000 converted to notiom water along the southern fringe of the 0 500 1,000 1:500 2.000 Antarctic or to intermediate waters along the northern fringe of Depth (m) the Antarctic. We devised a simple model to test our ideas Two box FIG: 2. Vertical distribution of a PO, and 5 100, in the Antarctic column before the onset of fertilization (dashed line) end" 10 and 100 years after diffusion columns' are linked by an overlying atmosphere and the onset of fertilization for the scenário where water is upwelled at the an underlying deep sea (see Fig. T): One column represents the rate of 174 Svrand transferred to the deep sed. Antarctic and the other the non-Antarctic region of the ocean Each column is capped by a mixed layer 75 m deep. These mixed layers are underlain by diffusion zones 2,000 m thick Beneath rapidly reach a new equilibrium, leaving the atmosphere with the diffusion zone is a single well mixed deep reservoir: The a CO₂ content only slightly lower than it had before fertilization. area of the Antarctic column was taken to be 10% of the ocean To achieve a significant reduction of the atmosphere CO. total The actual effective area of nutrient-rich water in the content the surface waters of the Antarctic must he replaced Antarctic is 16% of the ocean area (out to 45° S). We adopted frequently from below. a somewhat lower area to compensate for the fact that iron Thus the critical issue is the rate of vertical mixing in the fertilization cannot work during the lightless winter months. Antarctic Ocean. Traditional thinking is that deep ocean water During these periods vertical mixing will return the CO₂ partial upwells in this region A portion of this upwelled water moves pressure to its fertilization value. beneath the Antarctic's sea-ice fringe where it is densitied by We set the vertical eddy diffusivity in the Antarctic column brine release causing it to sink back into the deep sca. The at Meints and that in the non-Antarctic column at 1 cm2 S remainder moves to the north Part is converted to intermediate The upwelling rate in the Antarctic column was set at 17 $v water which penetrates northward into the Atlantic Pacific and (15 vr Because we do not have information with which to Indian oceans at a depth of 1,000 m. The remainder is mixed apportion the late of the upwelled water we studied two limiting into the temperate surface water. cases in the first the water is transferred entirely to the surface Confirmation of the upwelling hypothesis comes from the of the non-Antarctic column. In the second it: is transferred inventories of bomb produced radio arbon in the Antarctic entirely to the deen reservoir: 350 350 15 266 Peo 34 280 FIG, 3 Model runs for an Antarctic upwelling the of 4.St 300 265 300 252 Surface non Antarctic Left-hand panel shows the case where the upwelled water Is transferred to the surface of the Antara (II Orean and the right-hand panel- shows the Case 250 Atmosphere water is transferred to the deep sea Iri cach case tetails successful fertilization is conducted for years First the steady conditions before fertilization THE en has Antarctic surface waters is nearly identical 1/1 USA for the 20° 200 atmosphere Surface Antarctic 150 150 50 100 0 50 100 Time (yr) 228 NATURE VOL 349 17 JANUARY-1991 PHOTOCOPY LETTERS TO NATURE PRESERVATION vorid We started our calculation with the carbon cycle in a steady 12: Geschger, H., Siegenthaler: U., Schotterer: & Gugelman A Tellue 27.168.192.(1975) ntain state The residence times for PO₄ with respect to biological 13. Foster, T: DH& Welss R. F Antaict 74, 7611988) 14: Stouffer Manabe $ & Bryan: K Nature: 342. 660-652 (1989) gion removal from the surface reservoirs were set to vield 1.6 umol of the per kg PO₂ in the surface water above the Antarctic column and ACKNOWLEDGEMENTS Discussions with. Sarmiento prompted this study He also pointed 0.4 an important mistake in our paper We had assumed that the reductions obtained foi a steady state MI the no PO. in the surface waters of the non-Antarctic column. The atmosphere with 280 p.m. CO2 applied directly to the enthropogenically perturbed etmosphere :S the regeneration function for falling organic debris was set to yield Samiento pointed out that this is not correct. We have carried out CO, Increase models and found that the percentage change in CO₂ partial pressure and not the absolute change that remaine ained PO depth profiles similar to those observed (the choice of this approximately the same. We have done the perturbation celculations and find that this:19 if ort to regeneration function has no influence on the result of the correct This work was supported by the Lawrence Livermore National Laboratory Exxon Corporation uired calculation of the atmospheric CO₂ response to iron fertiliz- and the Carbon Dioxide Research Program (W.S.B.) and by the U.S. Department of Energy (T.H.P.) through a;joint contract Martin Marketta Energy Systems:inc is the ation). The atom ratio of carbon to phosphorus in the organic overs matter falling from the surface mixed layer was 130. We then 17 Sv adjusted the ECO-/alkalinity ratio in the model ocean to yield an atmospheric Pco of 280 platm. Control of pore-water chemistry - xtent We simulated a totally successful iron fertilization by greatly bomb decreasing the residence time (with respect to biological the base of the Florida escarpm nt es for removal) of PO. in Antarctic surface water. bringing its PO₄ netra content to near zero, and continued the iron fertilization for 100. by processes within the platform ertical vears. Figure 2 shows the evolution of the vertical distributions netra of PO. and ECO₂ in the Antarcticicolumn for the deep-transport Jeffrey P. Chanton*, Christopher S. Martens if, this case The total PO. in the water column remains unchanged but & Charles K. Paull 1, the a bulge of excess 200, appears, representing the CO2 transfer- urface red from the atmosphere to the Antarctic water column. In the Marine Science Program and Department of Geology lateral-transfer scenario (see Fig. 3), the atmospheric CO2 con- University of North Carolina Chapel Hill, North Carolina 27599. USA: nall to tent drops ever more slowly, reaching an asymptote patm Date of lower than the initial value: In the deep-transfer scenario the PORE waters collected from seep sediments hosting active ipacity decrease continues, reaching -34 µatm after 100 years. The chemosypthetic communities tend to be rich in sulphide, chloride that is reason for the difference is that in one case the surface water and ammonium and depleted in sulphate relative to the concentra- of the from the Antarctic is transferred to the surface of the non- tions in sea water. To Investigate the source of the energy-rich inge of Antarctic region, allowing the excess CO₂ to re-enter the atmos- compounds and the processes causing low sulphate concentrations phere, whereas in the other this water is removed to the deep in seep-sediment pore waters, we have measured the suiphur Isotope box sea, isolating it from the atmosphere. composition, 8³⁴S, of pore waters from seep sediments at the base :re and As a sensitivity test we have changed the upwelling rate for of the West Florida escarpment. The isotopic composition of pore- nts the the deep-transfer case. A rate of 8:7 Sv yields a drop in atmos- water sulphate remains approximately constant as Its concentration ocean pheric Peo. of 28 palm and that of 34.8.Sv a drop of 47 patm. is depleted, Indicating that processes within the Florida platform. mixed The question naturally arises as to what would happen if iron rather than microbial processes at seep sites, control pore-water Beneath fertilization were terminated Our model shows that any reduc- chemistry in these sediments. The composition of the seep brines, firs The tion in atmospheric CO: content accomplished by the fertiliz- deduced from a linear mixing model, provides Information on ocean ation would be lost, on more or less the same timescale as it processes deep within the platform. in the was gained. Dense biological communities living on highly sulphidized dopted Clearly, the key to the evaluation of. the iron-fertilization sediments found at the 3,280-m-deep base of the Florida escarp: (at iron scheme is the rate of vertical overturn in the Antarctic. If our ment are supported by chemosynthetic processes that use nonths interpretation of the bomb-radiocarbon record is correct, then reduced substrates seeping from the highly jointed limestone partial this process is too slow to create a significant decrease in the cliff forming the edge of the Florida platform (Fig. 1). These CO. content of the atmosphere. The distribution of freons in seeps occur along at least 10% of the escarpment base. Sedimen (column the Antarctic water column offers further evidence in this regard. tary pore waters beneath the large populations of mussels and cm The single published freon section11, a meridional traverse across tube worms are enriched in ammonium (up to 2mM), radio SV the Atlantic sector of the Antarctic, shows that the freon con- carbon-depieted biogenic methane (up to 10 mM) and sul which in centration for the entire water column between 500 m below the phide (up to 5.7 mM). Sediments consisting of 30% sulphide. limiting surface and 500 m above the bottom is <6% of the surface minerals have been reported⁹. The acidity generated by sea-floor surface concentration. Although we have not formally modelled freons, oxidation of dissolved sulphide has been suggested as one insferred this result is broadly consistent with the dynamics adopted here. mechanism for undercutting and croding the edge of the carbon In any case. such surveys offer constraints on the rates of both. ate continental margin²: deep. and intermediate-w formation. As such information We collected sediment cores in these chemosynthetic com IS critical to the evaluation of temperature buffering of the munities with the submersible Aluin. The pore waters were reenhouse warming by the Antarctic Ocean14, as well as to expressed from sediments by pressure filtration. Depth profiles Bustion of the iron fertilization scheme, its procurement of pore-water concentrations show chloride and ammonium 34 should be placed high on the list of environmental research enrichment and corresponding sulphate depletion relative to priorities. overlying sea water (Fig. 2). Both pore-water sulphate and 52 ammoniumare linearly correlated with pore-water chloride con centrations (Fig. 3), indicating conservative mixing between sea water and a sulphate-depleted, ammonium-rich brine The Reduced Subtember Seceived 16 November 1990 chloride-sulphate trend yields a value of 27.5% for a sulphate Begin. R Chem Enging News 24 1990). concentration of zero, suggesting that this is the chlorinity of Martin IN & Fitzwater E Nature 331. 341 343 (1988) the brine exiting the platform at the sea floor. The equation & Gardon R M Deep Sea Res. 35. 177-196,(1988) Marchdays Gordon R M Fitzwater 5. & Broenkow W. W. Deep See Res 36, 649-680 (1989). fitting the chloride-ammonium data yields a value of 2.2.mM Marger, M. Doroon R M & Fitzwater. S. Nature 346, 156-158 (1990). for the concentration of ammonium in the suiphate-depleted Mattin. FM Palaeoceanigraphy 8. 1-13 (1989). Deacon G t. R Discovery Pep 15. 125-152 (1937). brine. 6 H & Starver M Rediocarbon 22. 25-53 (1980) 00 9 Student M 6 Ostluno o Radiodarbon 22. 1:24 (1980). 10 Storer M & Outlund H.G Radiocarbon (1983). Present address: Department of Oceanography, Figrida State University Tallahassee Florida 32306 11 Brogover Wash Peng. 1.11 Ostiund G. & Stuiver, M. 2. geophys Res 90, 6953-6970 (1985) USA ARY 1991 NATURE VOL 349 17 JANUARY 1991 229 GLOBAL ENVIRONMENTAL CHANGE REPORT tion growth will be in the Third dramatically with both the changing blooms in the Adriatic, are concerned World. What is unexpected is that balance of fish species and the in- about the impact on the food chain. the data show developing countries creased water quality. Oceanographers and atmospheric are already major sources." The authors note that many other scientists say that the phytoplankton The English edition of World lakes in the Sudbury region showed may simply release most of the carb- Resources 1990-91 is available for increased pH after SO₂ emissions on back rather than sinking to the US $20.95 from World Resources In- declined, and they conclude that "con- ocean floor. And some policy ditions suitable for the survival and analysts worry that if every industry stitute Publications, PO Box 4852, could fill a tanker with iron to atone Hampden Station, Baltimore, MD reproduction of lake trout and other for its carbon sins, few would take ex- 21211, USA. For overseas orders, or acid-sensitive organisms can return for information on editions in Rus- without liming or other ameliorative pensive and restrictive measures to treatments." keep emissions down." sian, Spanish, Arabic, Japanese, Ger- man, or Chinese, contact WRI at But a number of US agencies have ex- 1709 New York Avenue NW, Algae-CO₂ Connection pressed interest in Martin's proposal, Washington, DC 20006. Tel: +1-202- Draws Ironic Attention and Anderson reports that the Nation- 638-6300; Fax: +1-202-638-0036. al Research Council will hold a con- A report in the 10 May issue of Na- ference on this topic (to follow up a ture (vol. 345, pp. 156-157) by John Reducing Industrial conference held last December) later Martin of the Moss Landing Marine Emissions Can Reverse this year. Laboratory (Moss Landing, Califor- Effects of Acid Rain on nia, USA) indicates that iron could be Lakes UK Opens New Climate a major limiting factor for phytoplank- Research Center According to a report in the 31 May ton. A lack of iron "prevents [the issue of Nature, acidified lakes are algae] from blooming and using up On 25 May, UK Prime Minister Mar- capable of rapid biological recovery the luxuriant supplies of major garet Thatcher opened the when industrial emissions are nutrients found in vast areas of the Meteorological Office's Hadley Cen- reduced. This means that simply cut- southern ocean." ter for Climate Prediction and Re- ting back on SO₂ emissions may have This in itself is not terribly controver- search, at Bracknell, UK. According direct benefits for acid-endangered to the Met Office, the immediate re- sial. Martin has found that levels of lakes; elaborate schemes for liming iron are extremely low in southern search goals of the new center are to: and otherwise neutralizing lake pH oceans, and he provides a note at the 1) predict global climate change to may be unnecessary. the year 2100; 2) provide the founda- end of his paper indicating that sub- The authors, J.M. Gunn and W. sequent experiments in the Ross Sea tion for improved predictions of have indeed stimulated phytoplank- regional climate change; and 3) pro- Keller, of the Ontario Ministry of vide a focus for national research Natural Resources and the Ontario ton productivity. programs relevant to the prediction of Ministry of Environment, respective- But Martin has gone a step further. climate change. The center will be ly, monitored water quality and Since increased phytoplankton ac- funded at £5.5 million (US $9.2 mil- biological characteristics of a lake tivity will take CO₂ out of the air, he lion) per year. near Sudbury, Ontario (Canada) from reasons, why not add a lot of iron to 1978 to 1988. Average yearly SO₂ The center will coordinate Met Of- the oceans to combat climate change? emissions from the Sudbury metal fice climate research efforts, and will This proposal, reminiscent.of smelters dropped from over one mil- act in concert with other UK research biomanipulation projects to reverse lion metric tons in 1975-79 to institutions. lake eutrophication by introducing 718,000 tons in 1988. The average new fish species, has met with a For more information, contact the pH during the study period rose from wide range of reactions, and a lot of Hadley Center, Meteorological Of- 5.4 in 1980 to 5.9 in 1988. attention. fice, London Road, Bracknell, Lake trout reproduced successfully in Berkshire, UK RG12 2SZ. Tel: +44- As Christopher Anderson notes in the the lake after pH rose above 5.5, and 344-856655; Fax: +44-344-854942. 31 May issue of Nature, "Biologists, invertebrate populations changed noting the deadly effects of algal 6 8 JUNE 1990 VOL II, NO. 11 GLOBAL ENVIRONMENTAL CHANGE REPORT ness of weatherization and commer- after 1997 while adding an additional reducing methane emissions from cial tax incentive and loan programs. 900 megawatts of renewable energy animal wastes), capturing methane sources in the region by 2005. On from landfills, reducing transporta- State transportation options include the sink side, ODOE proposes plant- tion fuel use and natural gas use, and setting (and enforcing) a 55 mile per ing large areas of unstocked or under- reducing gas emissions from coal hour speed limit statewide, convert- productive timberlands. Boosting mines. ing Oregon private and public vehicle recycling programs would also help fleets to natural gas, converting all Written comments on the strategy are decrease CO2 emissions. intra-city buses in the state to natural due by 10 December; for a copy of gas by the year 2005, and a number For other greenhouse gases, ODOE the energy plan, contact John Savage, of initiatives to reduce vehicular use proposes mandatory recycling of Administrator, Policy and Planning in the city of Portland. Electricity op- CFCs, reducing methane emissions Division, Oregon Dept. of Energy, tions include a moratorium on new from cattle (through low-methane 625 Marion Street NE, Salem, OR coal power plants in the region, and diets, reducing beef consumption, in- 97310. Tel: +1-503-378-4040; Fax: phasing out some existing coal plants creasing reproductive efficiency, and +1-503-373-7806. Science Update Northern Lake Ecology nutrient), which along with the of responsible scientists and policy- Disrupted by Climate temperature rise caused certain makers are seriously considering it as Warming phytoplankton populations to grow. an option to complement anthropo- The scientists also recorded a slight genic emissions reductions. The Some of the most severe effects of increase in the phytoplankton species NRC workshop summary emphasizes greenhouse warming are expected in diversity. On the other hand, the that there are still many unanswered high northern latitudes, where the longer ice-free season and decreased questions, and scientists will most temperature change is predicted to be snowfall, the clearer water (a result likely perform a "Transient Iron Ex- greatest and the ecosystems are easily of the lowered runoff), and the in- periment" (TIE) to settle key uncer- stressed. A new report, appearing in creased wind velocity, combined to tainties. The TIE would begin some- the 16 November issue of Science increase the depth of the warm-water time in the next 3-5 years, and would (vol. 250, pp. 967-970), gives an em- layer. This is bad news for cold- involve spreading iron over 400 pirical preview of climate change's loving species such as lake trout, square kilometers of the Antarctic impact on boreal lakes and their sur- which may die out in the more shal- Ocean. Preliminary assessments indi- rounding watersheds. low lakes as the growing warm layer cate that the TIE would have no nega- Over the past 20 years, according to compresses or eliminates the cold, tive environmental consequences. the report, average air and water tem- oxygen-rich layer near the bottom. As GECR reported in the 9 Novem- peratures in the Experimental Lakes Regardless of the cause of the warm- ber issue (vol. II, no. 19, pp. 4-5), the Area of northwestern Ontario, Canada, ing trend recorded by Schindler et al, maximum impact of a totally success- have risen 2°C. Coincident with this their study represents a landmark ful, intensive iron fertilization pro- change has been a decline in precipi- analysis of how a real ecosystem gram would be significant, but not tation and runoff, and a dramatic in- responds to a real change in climate. enormous given the scale of the crease in the time it takes for lakes to project. The NRC workshop found replenish lost water - from 4.6 years More on Iron Fertilization: that iron fertilization over an area of to over 20 years. The droughts have The NRC Workshop 40 million square kilomcters (18 mil- increased the occurrence of fires in On 29-31 October in Irvine, Califor- lion square miles) would result in an the watershed, resulting in greater wind velocities over the lakes. nia, the US National Research Coun- uptake of 1 gigaton of carbon annual- cil (NRC) held a workshop on marine ly. That's a healthy 1/6 of the total The effects of these changes on the algal productivity and carbon dioxide amount of carbon released by burn- lakes are complex. The research assimilation. NRC released a ing fossil fuels today (1/8 - 1/9 of team, led by David Schindler of the workshop summary on 16 November. total anthropogenic emissions when University of Alberta (Edmonton, Al- deforestation is factored in), though berta, Canada), found that lower While some observers have objected as one scientist pointed out, it's only water levels in the lake studied led to to the idea of enhancing CO₂ removal 1/28 of the amount some studies an increase in nitrogen (a key by phytoplankton, a growing number project for fossil carbon emissions in VOL. II, NO. 22 21 NOVEMBER 1990 5 GLOBAL ENVIRONMENTAL CHANGE REPORT the mid-21st century with uncon- policy analysts working on climate tific and Industrial Research, report strained growth. Fertilization would change issues: How does one choose that transmission of UV light through have to take place annually for at the most reliable estimate? sea ice is naturally the greatest during least 100 years to ensure that the carb- the austral spring, when the ice is Fortunately, methane experts M.A.K. on isn't just released back into the at- Khalil and R.A. Rasmussen of the relatively transparent. mosphere through upwelling, and Oregon Graduate Center (Beaverton, Unfortunately, early spring is also the such a program would require spread- Oregon, USA) have come up with time when ozone depletion is at its ing 1-5 million tons of iron each year guidelines. Presenting their analysis most severe. Trodahl and Buckley over the area. in a recent issue of Tellus (vol. 42B, say that the shallow ozone hole of Particularly interesting at the NRC pp. 229-236), Khalil and Rasmussen 1988 enhanced the net UV dose workshop was the discussion of find that a realistic budget should under the ice by a factor of ap- potential ecological side effects. Gus- portray the anthropogenic fraction of proximately four over pre-1975 tav Paffenhofer of the Skidaway In- total methane emissions at between levels. The 1987 and 1989 (and stitute of Oceanography (Savannah, 40-70%, and total current emissions presumably the 1990) holes yielded Georgia, USA) noted that krill - a at between 420-620 teragrams per about a ten-fold UV enhancement vital food for whales, squid, fish, seals, year. Only two of the 11 global over the pre-1975 levels. and penguins - might increase with methane budgets they looked at met Ken Ryan, a colleague of Buckley's the rising phytoplankton populations. these conditions. Which two? One at the Department of Scientific and But he warned that as greater amounts was their own 1983 study (Khalil and Industrial Research, told GECR that of carbon sink toward the ocean Rasmussen, Journal of Geophysical he'd just returned from a trip to Ant- floor, microbial activity in the deeper Research, vol. 88, pp. 5131-5144), arctica where he studied the response ocean might increase, diminishing and the other was a 1988 paper by of algae to the increased UV doses. the supply of oxygen. Krill eggs dev- Cicerone and Oremland (Global Ryan said he recorded depressed clop at depths of several hundred to a Biogeochemical Cycles, vol. 2, pp. algal growth rates in every experi- thousand meters, and a lack of oxy- 299-327). ment he performed, but his prelimi- gen there "could mean certain death Khalil and Rasmussen base their nary results did not show the effects [and] an eventual crash or greater criteria on observed features of the worsening significantly with higher diminution of the krill population global methane cycle. Significantly, doses. Copies of the NRC Bicoastal Session they note that methane sinks may be Summary of the Marine Algal Produc- declining somewhat (see GECR, 8 Global Average tivity and Carbon Dioxide Assimila- December 1989, pp. 4-5), which may Temperatures in 1990 tion Workshop are available from Dr. have led to a somewhat longer atmos- Could Reach Record High Oskar R. Zaborsky, Director, Board pheric lifetime for the gas. As GECR's 26 October Hotline sec- on Biology, National Research Coun- tion noted, the global average temper- cil, 2101 Constitution Avenue NW, Study Yields First ature trend for 1990 shows that - so Washington, DC 20418-0001, USA. Measurement of UV far, at least - 1990 could end up Several articles on iron fertilization Penetration of Antarctic being the warmest year since the mid- will also appear in November issue of Sea Ice 1800s. Here are a few more details: Global Biogeochemical Cycles (vol. Studies published two years ago On 11 October, Phil Jones of the 4, no. 1), published by the American show that algae living on the under- Geophysical Union. University of East Anglia (Norwich, side of sea ice surrounding Antarctica UK) told a US Senate panel that the are sensitive to ultraviolet (UV) radia- Most Methane Budgets temperature record for the first eight tion. These algae are important, Appear Unrealistic months of 1990 "should easily make since they may account for up to 30% it the warmest year yet recorded." Over the past 10 years, researchers of the primary food production in the The first few months were exception- have published nearly a dozen region; disruptions in their population ally warm, especially March. separate estimates of methane emis- could have serious effects on the An- sions from anthropogenic and natural tarctic ecosystem. And on 22 October, Jim Angell of the sources. These estimates range wide- US National Oceanic and Atmos- In the November issue of Geophysi- ly in both their figures for total emis- pheric Administration announced that cal Research Letters (vol. 17, no. 12, sions and in their ratios of the global average surface air pp. 2177-2179), H.J. Trodahl and anthropogenic emissions to total temperature in the early part of this emissions. This presents a dilemma R.G. Buckley of, respectively, New year was the warmest on record. Zealand's Victoria University and the that's quite familiar to scientists and March, April, and May temperatures New Zealand Department of Scien- 6 21 NOVEMBER 1990 VOL. II, NO. 22 A 'GERITOL SOLUTION Enlisting Marine Algae To Ease Global Warming ach year, combustion of workshop participants calls for "fer- E fossil fuels, deforestation, tilizing" large areas of the world's and other human and oceans with iron, an essential nutri- natural activities release ent for marine plants that is appar- eight or nine billion tons of carbon ently in short supply in many oceanic in the form of carbon dioxide — waters. into the Earth's atmosphere. This car- bon dioxide is a key player in current Life in the Sea concern about global warming; as a Of primary interest to the 40 scien- "greenhouse gas," it has the effect of tists and others who gathered at the trapping heat and preventing it from Academies' Beckman Center in Irv- radiating back into space. ine, Calif., were marine algae. To In addition to reducing carbon many, algae are best known as the Grant Heilman emissions, scientists have theorized stringy mats of green scum that form Sargassum (brown algae) about ways to recapture some of the in stagnant pools. But algae - and carbon that is being released. One of especially marine algae - are a California kelp, can reach lengths of the most innovative ways - boosting highly variable, often beautiful group nearly 100 feet. the amount of carbon dioxide used of plants. Other macroalgae float unattached by marine algae during photosynthe- Microalgae form the bulk of the in the sea. A good example is Sar- sis - was a topic of a workshop in- oceans' plankton, typically micro- gasso Weed. which forms dense mats October sponsored by the Research scopic plants that drift with sea cur- across much of the Sargasso Sea in Council's Board on Biology to answer rents or, sometimes, have limited the tropical Americas. key ecological and economic ques- powers of locomotion. What all these algae have in com- tions about the feasibility of this ap- At the other end of the size spec- mon is that they are plants that live proach. Francisco Ayala of the Uni- trum are the macroalgae. the familiar by photosynthesis. In the presence of versity of California, Irvine, chaired "seaweeds." These may be rooted to sunlight, water and nutrients, they the meeting. rocks and are the green, brown and are able to take up carbon dioxide Often dubbed the "Geritol solu- red masses that frequently wash from the atmosphere and use the car- tion," one technique studied by ashore when pulled loose from their bon to create sugars, starches and anchor. The largest marine alga, December 1990-January 1991 5 other carbohydrates. They release in the atmosphere, workshop partici- Ecological Impact oxygen in the process. pants concluded. If the algae in iron- The impact of iron fertilization on the poor areas of the ocean reached their The Iron Hypothesis marine environment was one of the peak productivity. they could remove The surface of the sea typically has critical questions generated by the by photosynthesis about 2 billion plenty of sunlight and carbon dioxide workshop group. The algae species tons of the 5 billion to 6 billion tons for algae to utilize in photosynthesis. that currently make up the algal pumped into the atmosphere annu- But some scientists believe that cer- populations in the open ocean are ally by the combustion of fossil fuels. tain nutrients might be lacking in species adapted to iron-poor condi- However, so little is known about parts of the ocean that limit the abil- tions; these would almost certainly be ity of algae to reach their maximum supplanted by other algae species productivity. with unknown repercussions in the Increased algal productiv- food web, said William Sunda of the John Martin of Moss Landing ity could have the benefit of National Marine Fisheries Service of (Calif.) Marine Laboratories has pro- posed that iron may be a "limiting" significantly reducing the the National Oceanic and Atmos- nutrient for almost one-fifth of the amount of carbon dioxide pheric Administration. world's oceans. While iron is one of in the atmosphere. Other organisms in the open-ocean the most abundant metals in the community also could be adversely Earth's crust, it is extremely insoluble the mechanics of the carbon cycle in affected, explained Gregory Mitchell the ocean and the biology and ecol- of the Marine Research Division of in seawater. It is carried to the ocean primarily by wind and suspended in ogy of the organisms involved that Scripps Institution of Oceanography. streams and rivers that feed into the there is no way of accurately predict- Krill, for example, are tiny crusta- sea. the or environmental con- ceans that feed on phytoplankton. In the open ocean far from land, sequences of fertilizing large areas of They make up the bulk of the diet of the ocean with soluble iron, confer- many fish and of some whales and little iron is available to the algae that grow there. ees agreed. therefore are a critical link near the If iron were introduced into these Workshop participants recom- bottom of the food chain. One might areas - principally Arctic, Antarctic mended that a pilot project move expect their populations to burgeon and equatorial waters - the result forward after about two more years as algae populations grow, he said. of laboratory research. The pilot However, the eggs of krill hatch at should be a marked increase in algal growth and productivity. Martin project would involve one research considerable depths in the ocean. If reasoned. vessel periodically spraying soluble much organic matter from increased iron on a 400-square-kilometer algal populations falls from the ocean The Carbon Cycle patch of ocean near Antarctica. surface to these middle levels and de- Increased algal productivity could Workshop attendées said there composes, these waters might be- have the benefit of significantly re- would be no major environmental come deprived of oxygen. Young krill ducing the amount of carbon dioxide consequences for a project of this might die for lack of oxygen. "We just magnitude. 6 NewsReport don't know for sure what the impact might be," he said. Even if environmental problems were somehow surmounted, impos- ing engineering problems remain. A team of researchers from the E.I. du Pont de Nemours and Co. - Ed- ward Howard and Thomas O'Brien -- calculated some of the logistics and cost. cost. For a 600-square-mile area, one would need special barges to haul the ferrous chloride (soluble iron) at a cost of $2.5 billion plus maintenance. They said a flotilla of hundreds of barges and ships would be required to transport and distribute the iron, each operating 365 days a year. Add to this the cost of the iron solution itself and the labor involved, and the proposed bill mounts rapidly. Macroalgae Grant Heilman Mass farming of macroalgae also was discussed by the meeting partici- are anchored. Fertilizers would be Kelp (brown algae) farms would pants. Compared to microalgae, sprayed over the kelp periodically to remove some atmospheric carbon. however, the contributions of larger help them grow, and the kelp plants seaweeds to increased carbon uptake themselves - composed mostly of Undaunted by the engineering diffi- would be minimal, conferees agreed. carbon in the form of starches and culties posed by both macroalgae and Kelp already has been the subject of carbohydrates - would be chemi- microalgae "farming," conferees considerable experimentation. cally converted into methane and the agreed that research should proceed. Wheeler North of the California In- remaining carbon somehow sunk to "We need to resolve questions re- stitute of Technology has been con- the ocean's depths. gardless of the ultimate utility of any ducting some of this research. However, North noted that experi- ocean fertilization scheme," they While microalgae usually are free- mental designs with which he has wrote in a summary prepared during floating, most seaweeds are attached worked have not been adequate. Of- the meeting. "The area is socially rele- to rocks or other substrates. North ten, the kelp get tangled up in the vant and intellectually engaging." envisions submerged platforms towing or anchor lines and become - Rick Borchelt towed by barges on which the kelp no more than a ball of shredded seaweed. 7 December 1990-January 1991 R1 THE WASHINGTON POST THE WASHINGTON POST INDEX SCIENCE NOTEBOOK Rising energy demand in Temple Hills man among Fighter Chavez gets little Third World threatens four area homicide victims B5 D4 Oceanography: Problems With 'Geritol Solution' recognition, money economies, report says A7 Martin Luther King remem- Preview of tonight's college Suspect in rare books theft bered for taking stand basketball games D8 to go on trial A8 L ast summer scientists proposed an growth would not be sustained because in against war B5 Bullets return home from Jurors' work frees man they Violations cited at Maryland D9 imaginative but radical solution to the Southern Ocean, the surface waters, long, tough road trip where all the action is, are not replaced found guilty of murder A10 health department B5 Notebook: Alarie does well the buildup of carbon dioxide in the TWA blames weak economy, News from around the re- playing for Grant D9 atmosphere, which many researchers quickly enough by deeper water that gulf situation for layoffs A12 Geddes overtakes Sheehan fear could lead to global warming. contains other needed nutrients such as gion, in capsule form B5 Officials encouraged by lay- Holiday schedules and for Jamaica golf victory D11 Why not, they asked, dump hundreds phosphorus and nitrogen. ing of condor eggs A12 what's open and closed B5 of thousands of tons of iron fertilizer In other words, the algae in the News from around the na- tion, in capsule form WASHINGTON BUSINESS into the ocean and SO create a giant surface water might have all the iron A12 OBITUARIES bloom of marine algae? The growing they need, but there may not be enough algae would suck carbon dioxide out of mixing of the exhausted surface water Banks and S&Ls suffering WORLD Claudio Villar, 79, former from troubled loans the atmosphere and sequester it in the and nutrient-rich deep water to sustain 1 heavyweight boxer B4 Depositors are wary of fed- sea. The algae in the Southern Ocean accelerated growth. Soviet forces attack Latvian John C. Morgan, 76, in- eral insurance system 1 around Antarctica appear to be starving Wallace Broecker of Columbia building, killing four A1 spired "Twelve O'Clock Rudolph A. Pyatt Jr. 3 for iron. Scientists half-jokingly called University and T. H. Peng of Oak Ridge Lithuanian parliament is High" B4 Wash Biz 3 their scheme the "Geritol Solution." National Laboratory conclude that "even symbol of defiance A13 Malcolm Maloof, 39, with Lawyers 5 Alas, the Geritol Solution might not if iron fertilization worked perfectly, it Talks on Soviet troop move National Rehabilitation Amtek program helps im- Hospital work. According to a computer would not significantly reduce from Poland hit snag A13 B4 migrants develop careers 6 China presses trials of pro- simulation reported in the current issue atmospheric carbon dioxide content." Regional Report 7 A14 Dividends 7 of the journal Nature, a spurt of algal -William Booth democracy activists STYLE South African judge rules on Indicators 8 death squads A18 Anger among blacks, and an GMU institute puts painting antidote C1 firm on right track 9 Entomology: A New Anti-Cockroach Weapon THE FEDERAL PAGE Opera: Menotti's "Saint of Jane Applegate 9 Bleecker Street" C1 New chief of Ticketron to Medical journals and flow of TV Preview: "Line of Fire," emphasize better service 10 n the ongoing war against information studied A19 "The Last to Go" and Appointments 11 cockroaches-which has engaged an Northeastern PBS's "The Sixties" C2 Business Law 12 endless variety of poisons, baits, governors seek release of more fed- Jonathan Yardley on colum- Power Computing 15 booby-trapped "motels" and bedroom eral energy money A19 nists and self-control C2 Personal Computing 16 slippers-scientists have found a new Early release of Medicare Opera: NYC's "Figaro" at Bankruptcy Filings 17 ally: The breeze. funds sought A19 new George Mason hall C3 Contracts 18 Dance: ABT's "Coppelia" on Insider Transactions 24 In research conducted by the Saturday C3 Business Calendar 24 Agriculture Department's insect Experts believe that roaches, which EDITORIAL/OP-ED Book World: "Rediscovering Computer Calendar 24 research lab in Florida, scientists must stay moist to survive, find that a America" C3 Washington Investing 29 installed equal numbers of roaches on constant breeze dries them out. Editorials, Letters to the Personalities: Effi Barry Selected area stocks 30 two sides of an attic. Air vents were Researchers suggest that home Editor A20 talks about her book C3 Economic index 33 installed on one side; the other was left builders make sure that attics and wall Op-ed: Evans and Novak, Style Plus: Martin Luther Money market funds 34 Washingtoon A21 unventilated. In two weeks, most of the voids are well ventilated. King Jr.'s dream C5 Foreign stocks 37 Charles McC. Mathias on insects had fled to the still side. Why? -Malcolm Gladwel Actor Robert Young's ap- Martin Luther King. parent suicide attempt C6 Charles M. Lichenstein Arts Beat: The shifting Number of Pages Today 120 and Paul M. Joyal on sin- sands of political comedy C7 Pharmacology: Alcohol and Aspirin a Bad Mix Jack Anderson B8 gling out Arab Ameri- Theater: "The Island," at Ask Beth B6 cans. Frank H. Wu gives American Showcase C7 Around the Nation A12 the perspective of a Chi- Around the Region T here are few things aspirin can't But, according to the scientists, this B5 was because the studies were nese American A21 do. In addition to fighting SPORTS Books C3 conducted on people drinking on an B6 headaches, arthritis and sundry other Bridge empty stomach. In those cases, without TERS TO NATURE ke place. In all cases no significant anisotropy is observed, carbon in the Southern Ocean, where concentrations of the indicating that melting is necessary to obtain alignment. nutrients P and N are high, can lead to alterations of atmospheric Our method has several advantages over existing techniques. CO₂ in excess of 100 p.p.m.4⁻⁷ Recently, it has been suggested The temperatures and magnetic fields required during sample that biological production in these regions may be limited by a preparation are relatively modest. The direction of the crystal restricted supply of iron8⁻¹¹. Although the hypothesis that iron orientation can be controlled carefully and easily. In addition, is the ultimate limiting factor there is controversial we adopt because of the relatively rapid cooling rates used during the it here to obtain upper-limit estimates of the possible reduction thermal treatment cycle, the sample production times are sig- in atmospheric CO2. Because the ratio of iron to carbon incor- nificantly shorter than those commonly experienced in melt- porated in plants is rather low, between 1:10,000 and texture-growth techniques If the alignment is due to the 1:100,00 the amount of iron required to carry out this presence of anisotropic magnetic particles into a liquid, the fertilization is relatively modest: ~10⁶ tons of iron per year efficiency of the process will depend on the size and the number assuming that all the iron goes into organic matter. Our model of nuclei that are aligned by the field, and consequently on the study differs from that of Peng and Broecker¹⁷ in the area heating rate, the time and the temperature of the annealing and fertilized (16% of the world ocean rather than 10%), and in that the cooling rate, all of which govern the grain growth. The we include scenarios with anthropogenic CO2 emissions for process could be adapted for industrial production of textured which we find that fertilization has a greater effect-than when samples and may prove to be generally applicable to all magnetic? we assume no man-made emissions. substances that have a residual magnetic anisotropy at high We consider there to be three factors that exert the most temperature. important control on the response of atmospheric, CO₂ to an HILDA Model Received 17 October 1990; accepted 8 January 1991. 1. Farrel, D. E. et al. Phys. Rev. B 38, 4025-4027 (1987). 2. Livingston, J. D., Hart, H. R. Jr & Wolf, W.P. J. appl. Phys. 64, 5806-5808 (1988). Atmosphere 3. Nakagawa, Y., Yamasaki, H., Obara, H. & Kimura. Y. Jap. I appl. Phys. 28, L547-L550 (1989). 4. Lees, M. et al. in Proc. Int. Conf. 'From Modern Superconductivity Towards Applications' (ed. Tournier, R. & Suryanarayanan, R.) 49-54 (IITT International Gournay-sur-Marne, 1990). 5. De Rango. P., Lees, M., Lejay, P., Sulpice. A. & Tournier, R. in Proc. Int. Conf 'From Modern Superconductivity Towards Applications' (ed. Tournier. R. & Suryanarayanan. R.) 21-26 (IITT g, gh International, Gournay-sur-Marne. 1990). 6. Miljak, M., Collin, G. & Hamzic. A. J. Mag Mater. 76 & 77, 609-611 (1988). 0 7. Jin, S. et al. Appl. Phys. Lett. 52, 2074-2076 (1988). LS HS 8. Salama, K., Selvamanickam, V., Gao, L. & Sun, K. Appl. Phys. Lett 54, 2352-2354 (1989) 9. Meng. R. L. et al. Nature 345, 326-328 (1990). 75 10. Schulz, L. G. J. appl. Phys. 20, 1030 (1949). u Estimates of the effect of Fp Fₚ Southern Ocean iron fertilization Depth (m) K W on atmospheric CO2 q concentrations F. Joos*, J. L. Sarmiento & U. Siegenthaler* Interior HD * Physics Institute, University of Bern, CH-3012 Bern, Switzerland 3,800 + Atmospheric and Oceanic Sciences Program, Princeton University, Princeton, New Jersey 08544, USA 84% 16% Area IT has been suggested¹⁻ that fertilizing the ocean with iron might FIG. 1 Structure of the model used for our study. which is based on a model offset the continuing increase in atmospheric CO2 by enhancing developed by G. Shaffer (personal communication) and one of US (J.L.S.). the biological uptake of carbon, thereby decreasing the surface- The ocean interior is resolved by a vertical stack of boxes connected by ocean partial pressure of CO₂ and drawing down CO2 from the advection (w) and diffusion (K): The high-latitude regions are simulated by atmosphere. Using a box model, we present estimates of the two well-mixed boxes which are connected to each other by mixing (u): maximum possible effect of iron fertilization, assuming that iron The parameter q is a rudimentary representation of the ven on is continuously added to the phosphate-rich waters of the Southern of the interior oceans by high-latitude waters. The model parameters Ocean, which corresponds to 16% of the world ocean surface. We obtained by fitting bomb and natural . 14C are: find that after 100 years of fertilization, the atmospheric CO₂ 465 + 7,096 exp m)/253 m² yr⁻¹; (equiva- lent to s⁻¹); (69.8 concentration would be 59 p.p.m. below what it would have been 0.00186 (79.5 10⁶ gas-exchange rate (g₁, gn) at 280 p.p.m.= with no fertilization, assuming no anthropogenic CO2 emissions, 15.1 mol yr⁻¹: ocean surface area 3.62 depth of and 90-107 p.p.m. less when anthropogenic emissions are included layer = average depth of ocean =3,800 m. We solve CO vation in the calculation. Such a large uptake of CO₂ is unlikely to be equations for the perturbation of phosphate and CO₂ = Cactual achieved in practice, owing to a variety of constraints that require with an initial condition for Cpert. of 0 everywhere. The relation between further study; the effect of iron fertilization on the ecology of the changes of 2CO₂ and Pco₂ is calculated using the equations of Peng et Southern Ocean also remains to be evaluated. Thus, the most al.²⁴. The phosphate and carbon removed from the surface high-latitude effective and reliable strategy for reducing future increases in box are regenerated to the dissolved form in the deep box below (dashed atmospheric CO₂ continues to be control of anthropogenic arrows, Fo perturbation flux of particulate carbon and phosphorus). A small amount of this excess phosphate makes its way through the to emissions. low-latitude surface waters, where it is removed and regenerated below A flux of dead biogenic organic matter from the ocean surface with an exponential scale depth of 1,160 m obtained from a fit to oceanic continuously transports carbon (and nutrients) to depth and nutrient data by G. Shaffer and J.L.S. Multiplying the resulting phosphate thus influences the surface concentration of total dissolved inor- fluxes by the C:P Redfield ratio (130) gives the corresponding carbon fl ganic carbon (ΣCO₂). Model studies have shown that variations LS, low-latitude surface; HS, high-latitude surface; and HD, high-latitude deep in the efficiency of biological uptake and export of organic boxes. 772 NATURE VOL 349 28 FEBRUARY 1991 LETTERS TO NATURE TABLE 1 Projected changes in atmospheric CO2 partial pressures resulting from various scenarios beginning in 1990 After 50 years After 100 years Effect of Effect of A B fertilization C D fertilization Scenario* Unfertilized Fertilized (B-A) Unfertilized Fertilized (D-C) High-latitude area fraction= 9.7%+ Initialized at pre-industrial levels Our model 0 -26 -26 0 -39 -39 - 0 -34 -34 Peng and Broecker - - Constant-emission 80 46 -34 156 98 -58 -68 Business-as-usual 150 114 -36 430 362 High-latitude area fraction=16.0% Initialized at pre-industrial levels 0 -41 -41 0 -59 -59 Constant-emission 79 24 -55 151 61 =90 Business-as-usual 146 88 -58 417 310 -107 The scenario which is initialized at pre-industrial levels of atmospheric CO₂ has no anthropogenic. source of CO₂ to the atmosphere, thus the CO₂ drawdown by the ocean results in a reduction below the initial pre-industrial value of 278.3 p.p.m. In the 'constant-emission' scenario, atmospheric CO₂ content is prescribed using observations until 1990²³, after which the CO₂ emission to the atmosphere is fixed at the model-determined value of in 1990. In the "business-as-usual' scenario, the atmospheric CO₂ content is prescribed until 1989 after which the annual emission increases according to the IPCC büsiness-as-usual scenario²², in which the man-made emissions increase linearly with time from in 1990 to 22.4 Gt in 2100. The 1990 atmospheric CO₂ partial pressure is 355 p.p.m. + The area fraction of 9.7% in our model corresponds to the same area in m² as chosen by Peng and Broecker¹⁷. # The Peng and Broecker¹⁷ model is initialized at 280 p.p.m. and has a phosphate reduction of 1.6 mmol m⁻³. The result given here is for their scenario with 17.5 10⁶ of water upwelling in the high latitudes and being placed directly into the deep ocean. enhanced carbon uptake resulting from iron fertilization. First from the natural ¹⁴C background; the data in the Southern Ocean is the reduction in average Σ CO₂ per unit area that occurs when are scanty, especially from pre-bomb time; and the bomb inven- fertilization is started. We take the supply of phosphate from tories south of 46°S show considerable variation. Instead, we depth as a reasonable guide to the maximum potential biological checked our model with high-latitude CFC-11 observations, uptake of carbon, although we recognize that light supply, our obtaining a standing crop of 1,976-nmol for 1984, compar- ability to spread iron efficiently over the large ocean areas able to an estimate of 1,950 ± 550 nmol obtained for the involved, the behaviour of the ocean ecosystem, and other same period from measurements south of 46°S by Weiss et al.²⁰ processes may interfere long before the phosphate is depleted. in the South Atlantic, by the Pacific Marine Environmental (Carbon cycling is linked to that of phosphate through the Laboratory of the National Oceanic and Atmospheric Redfield ratio of C: = 130 in organic matter.¹⁸) Second is the Administration in the South Pacific (R. H. Gammon and D. area over which the fertilization occurs. We use a recent analysis Weisgarver, personal communication), and by R. Weiss (per- of National Oceanic Data Center phosphate data (S. Levitus sonal communication), also in the South Pacific: and R. G. Najjar, personal communication) to estimate that the Figure 2 summarizes results from the iron fertilization assum- Southern Ocean's (>30°S) water volume in the depth range ing the business-as-usual and constant-emission CO₂ scenarios. 0-75 m with phosphate 1.0 mmol amounts to 15.8% of the Without fertilization, the increase in atmospheric CO₂ from its world ocean volume in the same depth range. The average 1990 value of 355 p.p.m. is 146 p.p.m. after 50 years, and phosphate content of this volume is 1.63 mmol m⁻³. We therefore 417 p.p.m. after 100 years in the business-as-usual scenario (Fig. take the-high-latitude box of our 'standard' model to consist of 2a). When iron fertilization is initiated, an additional flux of 16% of the world ocean area (with 9.7% as an alternative case organic carbon out of the high-latitude surface box is stimulated. corresponding to the area used by Peng and Broecker¹⁷), and Its size is approximately constant at 5.5 Gt C yr⁻¹ after an initial we determine the enhancement to the biological production by peak in the first year. This increased carbon flux leads to a forcing a reduction of phosphate by 1.5 mmol m⁻³ in the surface massive decrease in the high-latitude surface Pc₀₂ (see Fig. 2b), box of this area. Third is the reduction in CO₂ partial pressure and a resulting enhancement of atmospheric CO₂ uptake so that resulting from a given carbon removal. We have found that this the atmospheric CO₂ now increases by only 88 p.p.m. after 50 is considerably larger at high rather than at low CO₂ levels years (60% of the increase in the unfertilized scenario) and by because of the nonlinear relation between Pc₀₂ and CO2. We 310 p.p.m. after 100 years (74% of the increase in the unfertilized use three different anthropogenic CO₂ emission scenarios to scenario). The decrease in atmospheric Pc₀₂ resulting from iron examine the contribution of this nonlinearity, one in which the fertilization reduces the low-latitude air-sea difference so that ocean and atmosphere are initialized at the pre-industrial value the response of the low latitude regions is opposite in sign to of 278 p.p.m. with no anthropogenic sources, a 'constant- that of the high latitudes (Fig. 2b). After 100 years, the additional emission' scenario, and a 'business-as-usual' scenario (see Table CO₂ flux from the atmosphere to the high-latitude box in the 1 legend for descriptions of the scenarios). iron fertilization scenario is 2.65 Gt This is partly The rate at which atmospheric CO₂ is taken up by the oceans balanced by a reduction of 0.51 Gt C yr⁻¹ in the low-latitude also depends on the rate at which the ocean circulation and uptake of anthropogenic CO2, for a net annual oceanic uptake mixing transport to the abyss the excess CO₂ that invades the of 2.13 Gt C yr⁻¹ due to iron fertilization. fertilized region from the atmosphere. As discussed below, we The effect of the nonlinearity of the carbon chemistry find that this factor is less critical than those mentioned above. equations can be discerned by comparing the three scenarios Figure 1 shows the model we use to simulate the ocean circula- shown in Table 1. The constant-emission scenario with fertiliza- tion and mixing. Parameter values are obtained by fitting the tion of 16% of the world ocean shows a reduction of 90 p.p.m. model to pre-bomb ¹⁴C observations as well as the GEOSECS in the increase of atmospheric CO₂ resulting from 100 years of bomb inventory. estimates of Broecker et al.¹⁹ (U.S. and F.J., to iron fertilization, compared with 59 p.p.m. for the scenario be published elsewhere). However, the bomb inventory estimates initialized at 278 p.p.m. The further increase in atmospheric CO₂ in the Southern Ocean were not considered adequate for use in in the business-as-usual scenario has a still larger effect: the fitting exercise because the bomb signal is not easy to discern 107 p.p.m. The dependence on the prevailing atmospheric CO2 NATURE: VOL 349 28 FEBRUARY 1991 773 LETTERS TO NATURE concentrations can be understood when considering that Pco₂ iron fertilization of ~60%. Broecker²¹ argues that their choice depends in a nonlinear way on Σ CO2. The higher the atmo- of the smaller area was intended to compensate for the fact that spheric concentration of CO2, the larger the change in oceanic much of the Antarctic is in darkness during part of the year. Pc₀₂ for a given reduction of Σ CO₂ due to iron fertilization We attempted to compensate for this effect by fertilizing the (see Fig. 2c). We find that the percentage reduction in atmos- ocean for only six months of the year. The CO₂ uptake in the pheric CO₂ concentration which would occur after 100 years of business-as-usual case went down from 107 p.p.m. to 88 p.p. m. iron fertilization decreases with higher CO₂ levels (21% for far less than the drop to 68 p.p.m. which occurs when using the pre-industrial initialization, 14% for the business-as-usual smaller area. scenario). This is in contrast to the assertion of Peng and To be effective, iron fertilization has to be applied con Broecker¹ that the relative reduction is constant. tinuously so that the atmosphere-ocean system does not revert Peng and Broecker¹⁷ emphasize that the key to evaluating the to its unfertilized state. This is shown in Fig. 3, which show impact of iron fertilization is the rate of vertical exchange in what happens if iron fertilization is terminated after 50 years the Southern Ocean. We find that the uncertainty in the calibra- At 50 years the ocean in the business-as-usual iron fertilization tion of the high-latitude transport parameters u and W gives simulation has taken up 57.9 p.p.m. more CO₂ than the non error limits of +14% -24% for the absolute value of the iron-induced fertilized ocean. If fertilization is stopped-at this point, the reduction in atmospheric CO₂ for the pre-industrial scenario. difference between the fertilized and-non-fertilized scenarios Therefore, the precise magnitude of the vertical exchange seems reduces to 35.5 p.p.m. after 100 years. not to be as critical as uncertainties in other processes. This is Changes in the Earth's equilibrium temperature are also supported by the fact that Peng and Broecker's result agrees logarithmically related to the atmospheric CO₂ content, with a well with ours for the same pre-industrial scenario (Table 1), doubling of CO₂ leading to an equilibrium warming of between although their high-latitude vertical exchange, which is cali- 1.5 °C and 4.5 °C, with a preferred value of 2.5 °C (ref. 22): A brated with bomb-produced radiocarbon, is considerably useful index of the significance of a given CO₂ increase is the weaker than our vertical exchange calibrated by oceanic distri- factor In (Pco₂/278)/In (2) which, multiplied by 2.5.°C, will give bution of CFC-11. the equilibrium warming resulting from a given CO₂ increase The effect of ocean area is readily apparent from the results Thus, the atmospheric CO₂ concentration of 771 p.p.m. reached summarized in Table 1. An increase in the fertilized area from by 2090 in the business-as-usual scenario will give an equilibrium 9.7% to 16% leads to an almost linear increase in the effect of warming of 3.7 °C. The 107-p.p.m. additional oceanic uptake 900 Business as usual scenarios a 0 800 b Low lat Atmospheric CO₂ concentration (p.p.m.) 700 as usual CO₂ (p.p.m.) -50 600 Business High lat. 500 400 100 Constant High lat. emission 300 200 1900 1950 2000 2050 2100 150 Year 1900 1950 2000 2050 2100 Year 1,000 c 800 PCO₂ (p.p.m.) 600 FIG. 2 a, Future atmospheric CO2 concentration in the business-as-usual and constant-emission scenarios. The solid line is our prediction without iron fertilization, the dashed line shows what might occur with iron fertiliza- tion. b, The air-sea CO₂ difference of the iron-fertilized business-as-usual 400 scenario (dashed-lines) and the non-fertilized business-as-usual, scenario (solid lines). C, Dependence of the partial pressure of CO2 on CO₂ In high-latitude surface water (T=-0.22°C). With Increasing CO2 concentra- tions, the slope of the curve increases, and so does the change in Pco₂ for 200 a given change in ΣCO₂ 0 -100 -50 0 50 100 150 Perturbation 2CO₂ (umol kg ) 774 NATURE VOL 349 28 FEBRUARY 1991 LETTERS TO NATURE 900 estimate the CFC inventories. R. Fink developed our carbonate system algorithm and R. Slater helped with the phosphate data analysis. J, R. Toggweiler and J. Orr provided helpful comments on the manuscript. FJ. and U.S. acknowledge the hospitality of the Atmospheric and Oceanic Sciences Program during their extended visit to Princeton. This work was funded by subcontracts with Martin 800 Marietta Systems, Inc., under contract with the Carbon Dioxide Research Division, US Department of Energy: by a follow-up contract directly from the Carbon Dioxide Research Division of the Department of Energy: and by the National Science Foundation. Atmospheric PCO₂ (p.p.m.) 700 Business as usual Biases from natural sulphurization 600 in palaeoenvironmental recon- 500 struction based on hydrocarbon 400 biomarker distributions Constant Math E. L. Kohnen, Jaap S. Sinninghe Damsté 300 emission & Jan W. De Leeuw 200 Organic Geochemistry Unit, Faculty of Chemical Technology and Materials 1900 1950 2000 2050 2100 Science, Delft University of Technology, De Vries van Heystplantsoen 2, 2628 RZ Delft; The Netherlands Year FIG. 3 The effect of stopping iron fertilization after 50 years (dotted line) BIOMARKERS (chemical fossils) are sedimentary organic com- is depicted for the business-as-usual and constant-emission scenarios. The pounds whose basic skeletons suggest an unambiguous link with solid line is the unfertilized scenario; the dashed line is the scenario for known contemporary natural products, and were synthesized by continuous fertilization. biota present at the time of the deposition of the sediment. These compounds are commonly used to assess palaeoenvironmental resulting from iron fertilization gives an equilibrium warming conditions of deposition of Recent and ancient sediments¹⁻³ Satur- of 3.1.°C instead. On the other hand, the unfertilized constant- ated hydrocarbons are relatively easy to analyse and contain a lot emission scenario, which reaches 505 p.p.m. in 2090, gives a of geochemical information, and are therefore the most widely significantly smaller equilibrium warming of 2.2 °C, °C, which is used class of biomarkers in palaeoenvironmental reconstruction reduced to 1.4 °C with the 90 p.p.m. additional oceanic uptake Hydrocarbon biomarkers are biosynthesized as such or are derived resulting from iron fertilization. These numbers can be compared from functionalized biosynthetic lipids, such as alkenes, alcohols with the 0.9 °C equilibrium warming that can be expected from and acids, by diagenetically induced defunctionalization. Func- the present atmospheric CO₂ content of 355 p.p.m. It should be tionalized lipids may, however, also undergo an abiogenic reaction kept in mind that the transient warming is smaller than, and with hydrogen sulphide or polysulphides ('natural sulphurization") not necessarily proportional to the equilibrium warming, due during the early stages of diagenesis, and this may lead to selective primarily to the uptake of heat by the ocean. removal of specific hydrocarbon biomarker precursors. Here we The most important conclusion we draw from our calculations investigate the influence of natural sulphurization on hydrocarbon is that, although the effect of iron fertilization is large enough biomarker signatures in immature sediments from Italy and off to justify further study, the effect of a significant change in the Peru. We show that, if not taken properly into account, this process emissions of CO₂ is even larger. We emphasize the preliminary may lead to a severe bias in the interpretation of the geological nature of our calculations and the fact that all the assump- record. tions we have made have been biased so as to yield an upper Sedimentary functionalized lipids react with reduced sulphur limit. species (H₂S and HS) to form organic sulphur compounds (OSCs) and sulphur-bound lipid moieties in macromolecules (see ref. 6 for a review). The onset of sulphurization of organic Received 5 November 1990; accepted 5 February 1991. matter is controlled by the reactive iron (for example, ferri- 1. Booth, W. Washington Post A1 20 May (1990). hydrite and haematite) content of the sediment; because organic 2. Baum, R. Chem. Engrig News 68, 21-24 (1990). 3. Martin, J. H., Fitzwater, S. E. & Gordon, R. M. Global biogeochem. Cycles 4, 5-12 (1990). matter reacts with reduced sulphur more slowly than do sedi- 4. Knox. F. & McElroy, M. B. 1 geophys. Res. 84, 2503-2518 (1984). mented iron minerals and thus does so only after reactive iron 5. Siegenthaler, U.& Wenk, T. Nature 308, 624-626 (1984). oxides have first been converted to iron sulphides: In the region 6. Sarmiento, J. L & Toggweiler, J. R. Nature 308, 621-624 (1984). 7. Sarmiento, J. L. Toggweller, J. R. & Najjar, R. Phil. Trans R Soc. A325, 3-21 (1988). of upwelling in the Peru margin, sulphur incorporation into 8. Martin, J. H. & Fitzwater, S, E. Nature 331, 341-343 (1988). sedimentary organic matter starts in the top metre of the sedi- 9. Martin, J. H. & Gordon, R. M. Deep-Sea Res. 35, 177-196 (1988). 10. Martin J. H., Gordon, R. M., Fitzwater, S. & Broenkow, W. W. Deep-Sea Res. 36, 649-680 (1989). ment column', indicating that this process occurs in the early 11. Martin, J. H. Paleoceanography 5, 1-13 (1990). stages of diagenesis. Moreover, the identification of OSCs in a 12. de Baar, H. J. W. et al. Mar. Ecol. Prog. Ser. 65, 105-122 (1990). 13. Banse, K. Limnol. Oceanogr. 35, 772-775 (1990). Recent Black Sea sediment (age 3-6x 10³ yr) also points to an 14. Dugdale, R. C, & Wilkerson, F.P. Global biogeochem. Cycles 4, 13-20 (1990). early diagenetic sulphurization of organic matter⁸. Ten Haven 15. Anderson, G: & Morel, F. M. M. Limnol. Oceanogr. 27, 789-813 (1982). 16. Morel, F. M., & Hudson, R. J. in Chemical Processes in Lakes (ed. Stumm, W.) 251-270 (Wiley, et al.⁹ examined the extracts of almost 100 thermally immature New York, 1985): deep-sea sediments and found that approximately 70% of the 17. Peng. T.H. & Broecker, W.S. Nature 349, 227-229 (1991). samples contain OSCs. The widespread occurrence of OSCs, 18. Toggweiler, J.R. & Sarmiento, J.L. in The Carbon Cycle and Atmospheric CO₂: Natural variations Archean to Present Vol. 32; Geophysical Monograph Series (eds. Sundquist. E. T. & Broecker, and the fact that reduced inorganic sulphur species are present W. S.) 163-184 (American Geophysical Union, Washington, DC. 1985). in any anoxic organic-matter-containing Recent marine sedi- 19. Broecker, W. S. Peng, T.-H., Ostlund, G. & Stuiver, M. 1 geophys Res. 90, 6953-6970 (1985). 20. Weiss, R. F., Bullister, J. L, Warner, M. J., Van Woy, F. A. & Salameh, P. K. Ajax Expedition ment¹⁰, suggest that natural sulphurization of specific func- Chlorofluorcarbon Measurements (Scripps Institution of Oceanography Reference 90-6, La Jolla, tionalized lipids is a ubiquitous process. It seems inevitable that 1990): this selective removal of the precursors of hydrocarbon bio- 21. Broecker, W. S. Global biogeochem Cycles 4, 1-2 (1990). 22. Houghton, T., Jenkins, G.I. & Ephraums, J.J. (eds) Climate Change, The IPCC Scientific Assessment markers will alter the hydrocarbon biomarker distribution in (Cambridg, University Press, 1990). immature sediments, and thus the interpretation of the geologi- 23. Siegenthaler, U.& Deschger, H. Tellus 398, 140-154 (1987). 24. Peng. T. H. Takashi T. & Broecker, W. S. Tellus 399, 439-458 (1987). cal record as revealed by their apparent distribution. Hydro- carbon biomarkers are used in many (palaeo )environmental ACKNOWLEDGEMENTS. We appreciate the great help given to FJ. by M. Warner in setting up OUT CFC runs, and his generosity and that of R. Gammon in helping us to obtain the CFC date needed to reconstruction studies of immature sediments and of par- 775 NATURE VOL 349 28 FEBRUARY 1991 GLOBAL ENVIRONMENTAL CHANGE REPORT tion of net CO₂ emissions by 1-2% Methane Emissions from emissions "may account for half of per year starting now would be re- Lakes Increase as Air the observed trend of H2," and quired." This still amounts to at least Pressure Drops biomass burning, oxidation of non- a 20-50% cut in CO₂ over the next methane hydrocarbons, and various According to a report in the 25 Oc- 20-50 years, but the stepwise, year-to- anthropogenic activities could ac- tober issue of Nature (vol. 347, pp. year approach makes the target seem count for much of the rest. 718-719), the bubbles of methane less painful and uses a time frame that routinely escape from lakes may The 27 October issue of Science politicians can understand. become larger and more frequent News quotes F. Sherwood Rowland More Precise Temperature during periods of low air pressure. of the University of California, Irvine, as saying that methane oxidation Estimates Unlikely, Say According to the study's authors, should add "significantly more water Researchers Mark Mattson and Gene Likens of vapor to the stratosphere than would In what US climatologist Stephen the Institute of Ecosystem Studies, increasing levels of hydrogen, posing Schneider described as "the only orig- Millbrook, New York, USA, more an even greater threat to the ozone inal piece of climate science" that he than half the methane fluxes from a layer." Rowland, famous for his saw presented at the Second World Cli- New Hampshire lake during 64-day early work on CFCs and the ozone mate Conference, UK researchers Tom sampling period in 1989 occurred layer, has begun to focus much of his Wigley and Sarah Raper explained during a heavy rainfall on 5 August, attention on greenhouse gas emissions. that better estimates of the climate's and a tropical storm on 23 September. equilibrium response to an equivalent Mattson and Likens also cite carlier Iron Fertilization to doubling of CO₂ are unlikely to research that reported gas releases Remove CO2: What's the emerge in the next few decades. from soils during low-pressure Potential? If the current observed warming of events; if climate change eventually As GECR reported on 8 June (p. 6), 0.5°C resulted entirely from an en- increases rainfall in certain areas, and US scientist John Martin and col- hanced greenhouse effect, the re- increases the frequency of tropical leagues have proposed a way to searchers say, then a doubling of CO₂ storms, then this methane response reduce atmospheric CO₂ concentra- should result in a warming of only represents a potentially significant tions by enhancing plant productivity 1.25 - 1.55°C. Adding the very real positive warming feedback. in Antarctic waters. Although the factors of natural climate variability southern ocean is rich in nutrients, and data uncertainties widens the Molecular Hydrogen algae there are limited by a lack of range to 0.65 - 4.05'C. This range be- Increase May Pose Threat iron; adding iron to the ocean would comes even broader when the effects to Ozone Layer stimulate algal growth and take more of ocean circulation changes, solar The atmospheric concentration of CO₂ out of the atmosphere. variation, and sulfate aerosols are in- molecular hydrogen (H2) has in- But how much more? A new es- cluded. Sulfates, according to creased to a little over 500 parts per timate, by T.H. Peng and Wallace Wigley and Raper, may have offset billion since its estimated pre-in- Broccker of Oak Ridge National the greenhouse contribution to warm- dustrial level of around 200 ppb. Laboratory (Oak Ridge, Tennessee, ing by as much as one third, which And according to a report in the 25 USA) and the Lamont-Doherty means that SO₂ reductions currently October issue of Nature (vol. 347, pp. Geological Observatory (Palisades, underway in many Northern Hemi- 743-745), H2's concentration has in- New York, USA), respectively, con- sphere countries could result in a con- creased by about .5% per year bc- cludes that "after 100 years of totally siderably larger warming than might tween 1985 and 1989. Because H₂ successful fertilization, the atmos- otherwise be expected. oxidizes to form water vapor, and be- phere's CO₂ content would be cause an increase in water vapor in Wigley and Raper point out that no lowered by 30 parts per million [give the stratosphere may lead to an in- matter what global-mean tempera- or take 10 ppm]." In other words, ac- tures do over the next few decades, crease in polar stratospheric clouds, cording to Peng and Broecker, if CO2 the H₂ trend could be bad news for they won't reduce the range of uncer- were to reach 500 ppm in the year the ozone layer. tainty in estimates of the climate's 2090 under business as usual, a strict response to a CO₂ doubling. And, Why is molecular hydrogen increas- program of fertilization over the en- significantly, they claim that at least a ing? According to the study's tire century would lower the con- 2.5°C warming can be expected even authors, M.A.K. Khalil and R.A. Ras- centration to 455-485 ppm. Peng if the temperatures remain unchanged mussen of the Oregon Graduate In- told GECR that the relationship isn't for the next 15 years. stitute (Beaverton, Oregon, USA), linear - most of the effect occurs in the 1% per year increase in methane the carly stages of fertilization, and VOL. II, NO. 19 12 OCTOBER 1990 5 GLOBAL ENVIRONMENTAL CHANGE REPORT over a shorter time horizon (50 Study Finds High Levels enough to stabilize its global con- years), the atmospheric CO₂ con- of Methane in Eastern centration, "and eliminate any further centration would still be lowered by European Cities; Leaks incremental contribution of this gas 20-25 ppm. However, Peng cau- Blamed to the atmospheric trapping of in- tioned that the estimate is very op- frared radiation." F. Sherwood Rowland and colleagues timistic, and assumes the process of from the University of California at The IPCC's scientific report quotes a fertilization and CO₂ removal works Irvine report in the 4 October issue of figure of 15-20%; Rowland bases his "100% perfectly." More realistically, Nature (vol. 347, pp. 432-433) that calculation on the fact that methane CO₂ concentrations might be lowered recent methane levels in Berlin, has increased an average of 1% per by only 15 ppm over the next 50-100 Budapest, Krakow, and Prague were year during the 1980s, and that its at- years through an intensive fertiliza- "substantially higher than those mospheric lifetime is about 10 years. tion program. characteristic of most US cities." According to Rowland, "improved The two scientists will present the Rowland, who along with Mario controls on the tranmission of natural results of their study at the Fall 1990 Molina first brought the ozone-deplet- gas, particularly in eastern Europe meeting of the American Geophysi- ing potential of CFCs to light, says and Asia, offer a promising oppor- cal Union, to be held in San Francis- that leakage during natural gas trans- tunity to slow the global build-up of co from 3-7 December. mission is probably the culprit. methane..." See GECR's 22 June Rowland says a 10% reduction in issue, p. 6) for a related article. global methane emissions should be Industry Watch US to Facilitate fact that each industry is concerned prepared a strategy that calls for with its own specific issues will make 100,000 megawatts (MW) of wind Coordination of Industry's it hard to duplicate the broad generating capacity to be installed in Global Change Research programmatic scale that CEES has es- the EC by the year 2030. EWEA Efforts tablished for the Global Change Rc- produced the strategy on contract The Committee on Earth and Environ- search Program. GECR should note with the European Commission, and mental Sciences (CEES, formerly that PIGI's program would focus its recommendations are based on an CES), the government body that es- mainly on basic research, not on ef- appraisal of technological develop- tablished and oversees the US Global forts to mitigate global change. ments, realistic costs, a comparison Change Research Program, is trying with conventional power generation, to organize a similar program for US Henderson and colleagues are or- and reviews of markets, budgets, and industry research on global change. ganizing a workshop to be held 11-13 subsidies. February 1991 on industry's "current CEES has sct up an informal working and planned research that's relevant According to EWEA President David group, the Private Industry-Govern- to the Global Change Research Pro- Lindley, wind energy currently ment Interface (PIGI), to act as a gram." Interested parties should con- provides about 365 MW in Europe, communication point between tact Henderson at The Gcosat Com- with 250 MW of that being produced government and industry; the next mittee Inc., 601 Elm Street, Room in Denmark. Denmark's Energy step, says Dr. Frederick Henderson of 438C, Norman, OK 73019, USA. 2000 plan requires wind power to in- the Geosat Committee in Norman, Tel: +1-405-325-3329; Fax: +1-405- crease to 10% of the country's total Oklahoma, is to help industry or- 325-3327. energy production by the year 2000, ganize a research steering committee up from 2% today. The Netherlands that parallels CEES. Henderson ac- European Wind Energy has set a goal of 1,000 MW of wind knowledges that coordinating in- Pushes for Major energy capacity by the year 2000, dustrial research efforts may be quite Increases in Capacity and is currently soliciting bids for difficult, particularly if PIGI aims to Seeing an opportunity for growth in 250 MW worth of new turbines. Ger- be comprehensive. According to the face of the European many plans to add 250 MW of wind Geraldine Cox, vice president of the Communities' commitment to reduce energy by the year 2000, and is offer- Chemical Manufacturers Association CO2 emissions, the European Wind ing major subsidies (US $0.9 per and chair of the Global Climate Energy Association (EWEA) has kw/h to wind farmers); Italy wants Coalition's research committee, the 6 12 OCTOBER 1990 VOL. II, NO. 19 cations Sci. 288 A, 115 (1988). G. W. Viele, Eds. (Geological Society of America, Table 1. Schematic representation of t] 28. R. L. McMaster et al., Geology 8, 496 (1980). Boulder, CO, 1989), PP. 537-553. 1, field experimental treatments (4). All plots with 29. S. Mosher, Tectonics 2, 327 (1983). 39. W. R. Muehlberger and P. R. Tauvers, in ibid., pp. kangaroo rats removed (-) were compared 1 at con- 30. B. H. Reck and S. Mosher, J. Geol. 96, 677 (1988). 673-680. ssively 31. R. E. Zartman, and O. D. Hermes, Earth Planet. Sci. 40. J. Pindell and J. F. Dewey, Tectonics 1, 179 (1982). all other plots (+), except those with I Lett. 82, 305 (1987). 41. P. Bird, J. Geophys. Res. 83, 4975 (1978); Tectono- spectabilis removed. during 32. S. L. Dean, B. R. Kulander, J. M. Skinner, Geol. physics 50, 307 (1978). ysis is Soc. Am. Bull. 100, 299 (1988). 42. P. A. Ziegler, Tectonophysics 126, 303 (1986). Numb eozoic 33. J. Rodgers, The Tectonics of the Appalachians (Wiley- 43. D. H. Tarling, Proc. Ussher Soc. 4, 233 (1979); R. Treatment of Interscience, New York, 1970). Van der Voo, personal communication. nerica, plots 34. P. Geiser and T. Engelder, Geol. Soc. Am. Mem. 44. We benefited from field discussions with H. Maher 158, 161 (1983). and A. Dennis. A. Dennis, T. Engelder, W. R. 35. R. D. Dallmeyer, Am. J. Sci. 289, 812 (1989). Unmanipulated control 2 (+) Muchlberger, S. Schamel, W. E. Sharp, J. Shervais, 36. J. H. McBride and K. D. Nelson, Geol. Soc. Am. All rodents removed A. Snoke, R. Van der Voo, and H. Williams read 2 (-) Bull. 100, 436 (1988). All rodents and all ants early versions of the manuscript. This work has been 2 (- 158, 19 37. S. A. Graham et al., ibid. 86, 273 (1975); J. supported by the National Science Foundation removed Wickham el al., Geology 4, 173 (1976); K. D. grants EAR 8508123 and EAR 8803833 to D.T.S. Kangaroo rats removed 2 1. Nelson et al., Tectonics 1, 413 (1982); R. J. Lillie et Additional support has been provided through De- Kangaroo rats and 2 (- 77); R. al., Am. Assoc. Pet. Geol. Bull 67, 907 (1983). partment of Geological Sciences, University of Pogonomyrmex ants Duachita 38. W. A. Thomas and G. H. Mack, Geol. Soc. Am. South Carolina teaching assistantships and a Grant- removed W.A. Bull. 93, 6 (1982); G. H. Mack, W. A. Thomas, C. in-Aid of Research from the American Association ciety of A. Horsey, J. Sediment. Petrol. 53, 931 (1983); W. Dipodomys spectabilis removed 2 of Petroleum Geologists to P.E.S. 3. All ants removed A. Thomas, in The Appalachian-Ouachita Orogen in 2 (+ full. 88, the United States, R. D. Hatcher, Jr., W. A. Thomas, 26 June 1990; accepted 19 September 1990 Pogonomyrmex ants removed 2 (+ Seed addition 8 (+) DC. Am. A. Liu, ). 9, 809 dii, and D. merriami; or just the largest an ghanian Control of a Desert-Grassland Transition by a behaviorally dominant kangaroo rat, D Carolina Geolog- Keystone Rodent Guild spectabilis. An initial analysis of varianc blogical (ANOVA), with plots as the sample units to avoid "pseudoreplication" (6), showed tha JAMES H. BROWN AND EDWARD J. HESKE treatments in which all rodents or all kanga forton, T. Se- roo rats had been removed were similar to ological Twelve years after three species of kangaroo rats (Dipodomys spp.) were removed from each other in the plant variables analyze Excur- lumbia, plots of Chihuahuan Desert shrub habitat, density of tall perennial and annual grasses below. They differed significantly from a Geology had increased approximately threefold and rodent species typical of arid grassland had other treatments, except D. spectabilis re ibid., p. colonized. These were just the most recent and dramatic in a series of changes in plants moval. In addition, there were no detectable and animals caused by experimental exclusion of Dipodomys. In this ecosystem effects of ant removal or seed addition treat )uachita kangaroo rats are a keystone guild: through seed predation and soil disturbance they ments on these plant variables. Therefore W.A. have major effects on biological diversity and biogeochemical processes. the eight plots where all kangaroo rats wer ciety of absent were compared to the 14 plots wher- 19, 537 HE BIOLOGICAL DIVERSITY AND T of southeastern Arizona (4, 5). Twenty-four all kangaroo rats were present in the follow 7, 433 biogeochemical processes that char- plots, each 0.25 ha in area (50 by 50 m), ing analyses. The two plots from which jus pl. Soc. acterize an ecosystem depend on in- were fenced with fine wire mesh and as- D. spectabilis had been removed were exclud her, Jr., teractions of the organisms with each other signed at random to experimental manipu- ed from the analyses, because the vegetation United and with their abiotic environment. The lations. Treatments included removal of G. W. parameters were often intermediate between oulder, presence or absence of certain kinds of or- some or all rodent or ant species and addi- the two treatment classes and the limited ganisms, called "keystone species" (1), can tion of millet seeds (Table 1). Access of replication did not permit statistical resolu- 1319 dramatically alter the structure and dynamics different rodent species to appropriate plots tion. of ecological systems. In most ecosystems, was controlled by cutting holes of different Long-term removal of kangaroo rat vertebrate animals account for only a small sizes at ground level in the fences surround- caused a dramatic change in habitat, fron of the fraction of the biomass and energy flow (2), ing the plots. Populations of rodents, ants, desert shrubland to grassland (Fig. 1). Ef nont of but through predator-prey, competitive, birds, and plants were monitored systemat- fects of Dipodomys on vegetation were ana r., Ed. South and mutualistic interactions with other spe- ically at a grid of permanent sample sites lyzed by ANOVA, not only comparing plot: 1987), cies and by causing physical disturbance they within each plot. Between 4 and 16 Septem- where kangaroo rats had been removed tc can have disproportionately large effects on ber 1989, in order to better characterize plots where kangaroo rats were present, but Oliver, 3). habitat structure, species composition, and plant cover inside and immediately outside also comparing transects inside and imme- Progeny biogeochemical processes (2, 3). We show each plot, all plant species were counted at diately outside the plots where kangaroo rats la and that long-term experimental removal of a logical 10-cm intervals along eight 25-m transect had been removed (Fig. 2). Tall-statured al Sur- guild of three kangaroo rat species from a lines: four inside transects began 8 m from perennial and annual grasses colonized the desert ecosystem initiated changes that have the center of the plot and ran outward open spaces between the shrubs and in- 'g. 22, led to the conversion of the habitat from toward each corner; after a gap of 2 m on creased approximately threefold in the ab- ), 853 shrubland to grassland. either side of the fence, the other four sence of kangaroo rats. Much of this re- n, Jr., In 1977, experiments were begun on an transects continued for another 25 m out- sponse can be attributed to two species: the alluvial outwash plain with diverse desert side the fence. 3, 999 perennial Eragrostris lehmanniana, which in- shrub vegetation in the Chihuahuan Desert In ten plots, one of three different combi- creased more than 20-fold, and the annual 3). nations of desert rodent species was exclud- Aristida adscensionis, which increased ap- i, 735 Department of Biology, University of New Mexico, ed (Table 1): all rodents; three species of proximately threefold. There were comple- Am. J. Albuquerque, NM 87131. kangaroo rats, Dipodomys spectabilis, D. or- mentary changes in two species of short 250 21 DECEMBER 1990 REPORTS 1705 nnual grasses (Bouteloua aristidoides and B. same period, seed-eating birds decreased 30 Inside Dipodomys removed arbata); these were significantly less abun- their foraging on plots with rodents re- 1: Dipodomys present lant where kangaroo rats had been re- moved in comparison with plots where kan- noved. Tridens pulchella, a short-statured garoo rats were present (8). This was asso- 20 pecies and the only perennial grass that was ciated with an increase in herbaceous (both captured 10 tommon on the site when the experiments grass and forb) vegetation and a decrease in vere begun, was the only abundant species the amount of bare ground, conditions that 10 that did not show a significant response to made it more difficult for the birds to harvest cangaroo rat removal. The net result of the available seeds. Six to 12 years after the excluding either all rodents or all three Di- experiment was begun, and in response to the increase in grass cover documented 0 bodomys species for 12 years was a dramatic increase in the cover of tall grasses. above, six species of rodents that are charac- That kangaroo rats are a keystone guild in teristic of arid grasslands differentially colo- Cover (%) 30 Outside this ecosystem is further demonstrated by nized the grassy plots where kangaroo rats other changes that occurred on plots where had been removed (Fig. 3). Finally, we have Fig. rodei they were removed. During the first 10 years noted, but not quantified, greatly increased 20 funct after initiation of the experiments, large- accumulation of litter and longer persistence perer seeded winter annual plant species had in- of snow on the plots where kangaroo rats shade creased (Fig. 1), some by as much as several were absent. ized Two processes, seed predation and soil 10 thousand times, and small-seeded winter viven annuals decreased on plots with rodents and disturbance, appear to be responsible for the repre kangaroo rats removed (5, 7). During this changes in this desert ecosystem that were spp. both 0 ERIN PERG ARAD ANNG BOAR BOBA Vegetation type may Fig. 2. Effects of removing kangaroo rats (Di- lishn podomys) on densities of perennial and annual birds grasses. Plotted are mean values (±SE) of the distu percentage of cover for Eragrostris lehmanniana (ERLE), all other tall perennial grass species rats (PERG), Aristida adscensionis (ARAD), all other tence tall annual grasses (ANNG), and the short annual color grasses Bouteloua aristidoides (BOAR) and B. bar- speci bata (BOBA). (Top) Data from transects inside Al plots where kangaroo rats were either present or removed; asterisks indicate significant differences size-: between kangaroo rat removal and other plots. soil (Bottom) Data from transects immediately out- dant. side these plots. Asterisks indicate statistically significant differences between transects inside repo cated and outside plots where kangaroo rats were pres- ent or absent. (ANOVA, with plots as units of differ replication; *P V 0.05, ^ 0.01, V imen 0.001.) trans certai caused by removing kangaroo rats. Selective abser. foraging by kangaroo rats for large seeds is betw sufficient to explain the shorter-term re- when sponses to our experiments (5, 9). Large- grazil seeded winter annual plants increased in cause response to their release from rodent preda- shrut tion, and asymmetrical competition from effect these dominant large-seeded species then kang: caused the decrease in the small-seeded win- exclu: ter annuals. Increased availability of seeds prodi and reduced competition from kangaroo horse rats may account, at least in part, for the ed fro colonization of kangaroo rat removal plots no si; Fig. 1. Photographs across the fences surrounding plots by the granivorous Reithrodontomys rodents been from which kangaroo rats (Fig. 3). Physical disturbance appears to In were removed (left side of have played the major role in the longer- nism fence). (Top) Increase of the term responses to removal of kangaroo rats. ecosy. yellow-flowered, large-seed- ed annual, Lesquerella gordo- During their foraging, food caching, and single nii, 5 years after kangaroo rat burrowing activities, kangaroo rats make ically removal. (Bottom) Increase runways through the vegetation and move roo r. in rall-statured annual and large quantities of soil, creating many shal- behav perennial grasses 13 years af- ter kangaroo removal. low pits and little mounds. This disturbance lis) h: SCIENCE, VOL. 250 21 DE 1706 and distribution of other desert rodents (4, Brown, in Pattern and Process in Desert Ecosystems, W. 150 12), but it required the removal of all three G. Whitford, Ed. (Univ. of New Mexico Press, Albuquerque, 1986), p. 51. species to cause wholesale changes in vege- 3. J. A. Estes and J. F. Palmisano, Science 185, 1058 tation. On the other hand, the eight com- (1974); S. R. Carpenter, J. F. Kitchell, J. R. Hodg. Rodents captured son, BioScience 35, 634 (1985); J. C. Castilla and L. 100 mon and several rare species of desert ro- R. Duran, Oikos 45, 391 (1985); R. J. Naiman, C. dents that remained after kangaroo rats had A. Johnston, J. C. Kelley, BioScience 38, 753 been removed clearly did not play the same (1988); J. Pastor et al., ibid, P. 770. 4. J. H. Brown and J. C. Munger, Ecology 66, 1545 keystone role and were not able to prevent (1985). 50 the conversion of desert to grassland. 5. J. H. Brown et al., in Community Ecology, J. Dia- Twenty-five years after the concept of mond and T. J. Case, Eds. (Harper & Row, New York, 1986), p. 41. "keystone species" was first introduced, ex- 6. S. H. Hurlbert, Ecol. Monogr. 54, 187 (1984). 0 (DO KID) amples have been found in a number of 7. D. W. Davidson, D. A. Samson, R. S. Inouye, 0 20 40 60 taxonomic groups and habitat types (1, 3). Ecology 66, 486 (1985); D. A. Samson, T. E. Philippi, D. W. Davison, in preparation. Grass cover (%) It remains, however, to develop a general 8. D. B. Thompson et al., Ecology, in press. Fig. 3. Number captured per plot since 1983 of conceptual framework that will predict 9. J. H. Brown and D. W. Davidson, Science 196, 880 rodent species typical of grassland habitat as a which kinds of organisms play key roles in (1977); J. H. Brown et al., Annu. Rev. Ecol. Syst. 10, 201 (1979); R. S. Inouye, G. S. Byers, J. H. function of mean percentage of cover of all tall different kinds of ecosystems. Native species Brown, Ecology 61, 1344 (1980). perennial grasses plus Aristida adscensionis. Un- shaded symbols represent only the most special- are increasingly being eliminated from local 10. J. R. Hastings and R. M. Turner, The Changing Mile (Univ. of Arizona Press, Tucson, 1965); D. Sheri- ized grassland species, Sigmodon hispidus, S. ful- habitats and larger regions as a result of dan, Descrification of the United States (Council on viventer, and Baiomys taylori; shaded symbols human activities. It is critical to develop a Environmental Quality, U.S. Government Printing represent the above species plus Reithrodontomys theoretical basis for assessing the effects of Office, Washington, DC, 1981). spp. Least-squared regression lines are fitted, and 11. E. J. Heske and M. Campbell, Southwest. Nat., in these species on ecosystems so that, if extir- both are statistically significant (P < 0.0001). press; see also R. M. Chew, Am. Midl. Nat. 108, pation of keystone organisms cannot be 159 (1982); B. A. Roundy and G. L. Jordan, Southwest. Nat. 33, 425 (1988). avoided, their roles can be replaced by other 12. M. A. Bowers et al., Oecologia 72, 77 (1987). may facilitate decomposition of litter, estab- native or exotic species or by active ecosys- 13. We thank L. Valle, P. Mazzolini, and many other lishment of many annuals, and foraging of tem management. people for assistance with the fieldwork, S. Mistry al for help with the statistical analyses, C. Crawford, L. birds (8). Conversely, the reduction in soil Hawkins, M. Molles, and M. Taper for comments he disturbance following exclusion of kangaroo REFERENCES AND NOTES on the manuscript, and the National Science Foun- la es rats promoted the establishment and persis- dation (most recent Grant BSR-8718139) for sup- 1. R. T. Paine, Am. Nat. 100, 65 (1966); ibid. , port of the research program. er tence of tall grasses, and this in turn favored 103, 91 (1969). al colonization by specialized grassland rodent 2. R. M. Chew, Ohio J. Sci. 74, 359 (1974); J. H. 25 June 1990; accepted 3 October 1990 species. le Although kangaroo rats presumably cause or es size-selective seed predation and extensive s. soil disturbance wherever they are abun- & dant, the kind and magnitude of changes Characterization of "Peak E," a Novel Amino Acid ly reported here would probably not be dupli- de Associated with Eosinophilia-Myalgia Syndrome cated if kangaroo rats were removed from S- of different kinds of desert habitats. Our exper- imental site is near the zone of natural ARTHUR N. MAYENO,* FENG LIN, CHRISTOPHER S. FOOTE, transition from desert to grassland, so that DAVID A. LOEGERING, MATTHEW M. AMES, CRAIG W. HEDBERG, certain abiotic conditions or the presence or GERALD J. GLEICH e absence of keystone species can cause a shift is between alternative vegetation types. Else- Epidemiologic studies strongly associate eosinophilia-myalgia syndrome (EMS) with e- where in the southwestern United States ingestion of tryptophan containing a contaminant ("peak E"). Prior reports have e- grazing by domestic livestock is known to suggested that peak E is the di-tryptophan Nα-aminal of acetaldehyde. Spectral and in cause degradation of arid grassland to desert chemical studies now demonstrate that peak E is 1,l'-ethylidenebis[tryptophan]. This a- shrubland (10). At our site, however, the novel amino acid may be the etiological agent responsible for EMS, or it may be a m effects on vegetation of the exclusion of marker of a still unidentified causal agent. in kangaroo rats in combination with livestock Nα-aminal of acetaldehyde (4, 5). We present C exclusion were much greater than those is produced by the exclusion of cattle and A S OF AUGUST 1990, EMS WAS LINK- chemical and spectral data which show that ed to 27 deaths and over 1500 cases peak E is actually the isomeric l'-ethyli- Ю horses alone. Grazing livestock were exclud- (1). Epidemiological studies (2, 3) denebis[tryptophan] (1). ie ed from our entire 20-ha site since 1977, but have associated EMS with the ingestion of Peak E was isolated by high-performance ts no significant change in vegetation has yet L-tryptophan (Trp) produced by a single man- liquid chromatography (HPLC) (3). Fast ts been detected across the fenceline (11). ufacturer, suggesting that a contaminant is atom bombardment-mass spectrometry to In the present case the "keystone" orga- responsible for EMS. Recently we reported the (FAB-MS) revealed peaks at mass-to-charge r- nism whose removal caused large changes in discovery of a contaminant (peak E) in the Trp ratios (m/z) of 157, 231 (base peak), and S. ecosystem structure and dynamics was not a samples consumed by EMS patients that was 435 ([M + High-resolution FAB-MS id single species, but a guild of three taxonom- absent in the Trp consumed by asymptomatic gave a mass of 435.2041, consistent with ke ically related and ecologically similar kanga- controls (3). A significant association exists the molecular formula C₂₄H₂₇N₄O₄. ve roo rat species. Removal of the largest and between the presence of peak E and EMS- The proton nuclear magnetic resonance il- behaviorally dominant of these (D. spectabi- associated Trp (3, 4). Two groups have previ- (¹H NMR) spectrum of peak E in D₂O (6) ce lis) had significant effects on the abundance ously reported that peak E is the di-tryptophan was similar to that of Trp in terms of 50 21 DECEMBER 1990 REPORTS 1707 THE NTERPARLIAMENTARY CONFERENCE ON THE DROW GLOBAL ENVIRONMENT boxtles I HD. tel carf. ul Roina 1/25/91 whole theory of Fe seeding avore from 38 bottles 100. X will be a box re: this Stray X he released soon. in their glibal work I EARTH ON ICE John Martin's idea is simple and ingenious, and some say it's crazy. If you dump iron in the ocean, he says, you can draw carbon dioxide out of the atmosphere- and cool down an overheating planet. BY ROBERT KUNZIG OSS Landing on Monterey deep. Some people who heard about his M Bay is a knot of antique idea got enthusiastic, and he is now stores and a harbor tight pestered by people who have iron to get with boats that still fish rid of. Others, particularly environmen- the bay for albacore and talists, got mad. The National Research salmon. From a distance, Council has convened two workshops though, from the coast highway or from on Martin's idea. The American Society the artichoke fields of the Salinas Val- of Limnology and Oceanography just ley, what you see of Moss Landing are held a special symposium for the sec- the twin stacks of the Pacific Gas and ond time in its history. Martin has ap- Electric power plant. They dominate the peared on Good Morning America. The village like cathedral spires. And on a upshot of these learned discussions is hot summer day, when air conditioners that Martin's idea may work or it may in central California are pumping furi- not; but it is certainly worth investigat- ously, the PG&E stacks are spewing ing, and it is certainly no joke. 4,300 tons of carbon into the atmo- Somehow this bluff, shy but outspo- PHOTOGRAPH BY ANDY FREEBERG: MICROGRAPH BY DEE BREGER, LAMONT-DOHERTY GEOLOGICAL OBSERVATORY sphere. From the Moss Landing Marine ken man has managed to convince his Laboratories, which stand on the sand peers that it is worth talking about fertil- spit that protects the harbor from the lizing the Antarctic sea surface with a bay, John Martin, the director of the shipload of iron-that the single-cell labs, can gauge the region's demand for algae that float there might grow like electricity just by looking up at the crazy and in the process draw carbon stacks. Martin thinks he may have a dioxide out of the atmospheric green- way of getting some of that carbon back house. What accounts for Martin's per- out of the atmosphere. He says we might suasiveness? It is not charisma, though do it by dumping a few hundred thou- there is a disarming Yankee directness sand tons of iron into the ocean around about him. And it is not just that Antarctica. everyone these days seems to want to When Martin first made this sugges- talk about the greenhouse effect. The tion a few years ago, he was half joking. impact of Martin's idea really stems from "I got into this mess by shooting my its simplicity-and from the solid data mouth off," he says. But now he's in that back it up. John Martin at Moss Landing: Fertilize the ocean with iron, he says, and the phytoplankton-floating single-cell plants-will bloom. Inset: An electron micrograph shows a type of phytoplankton called a diatom; this one is relatively large-.001 inch long. More diatoms are on the following pages. DISCOVER APRIL 1991 55 would have blown everybody would scream at him, 'You out every window stupid old man!' He just completely "I'm not much of in Panama City. So infuriated a lot of people, and so they a that kind of slowed didn't want to believe him." a scientist. My them down. But it Martin and his assistants, though, de- was an interesting cided to heed Patterson's advice. They strength is project, and a tre- started taking off their shoes before en- mendous amount tering the laboratory so they wouldn't imagination-I of research came track in metal-rich dirt. They wrapped S out of it." each bottle of seawater in three plastic can imagine what's Martin's own re- bags. They kept the laboratory under S search involved tak- positive pressure, with filtered air com- going on in the ocean." ing samples of ing in and fans blowing out, so that no plankton-the sea's dust grains would waft in when the floating population door was opened. They even switched of tiny plants and to the brand of laboratory tissue that Dumping iron in the Antarctic, if it animals-from all over the Gulf of Pan- Patterson recommended as being partic- ever happens, would not be the weird- ama and measuring their metal content. ularly metal-free. "Our big goal in life," est project Martin has ever been in- He found that the water and the plank- says Martin, "was to get one damn num- volved in. In 1966, at age 31, he left the ton contained ample amounts of metals ber for lead that Patterson would believe- University of Rhode Island with a Ph.D. already, which suggested to the Atomic in plankton, in sea lions, in water, or in oceanography and took a job with the Energy Commission that a little radioac- whatever. So we started doing all this, Atomic Energy Commission. He was sent tive metal wouldn't do much harm. Mar- and it was amazing. All of a sudden all to Panama to work on Project Plow- tin got pretty good at these measure- these things started to fall into place.". share-the goal of which was to dig a ments, or so he thought at the time. But In Panama Martin had found that second, sea-level canal through the isth- in 1970, when he took a job at the many metals were naturally abundant mus by detonating a chain of 150 or so Hopkins Marine Station, a Stanford out- in seawater. Now he found that he had atom bombs. "What you do is, you set post on the Monterey peninsula, he was been wrong. When he cleaned up his the devices up across Panama, and you tired of plankton ("They're cruddy procedures and, eliminated sources of touch them off, and it's just like a wave things"), and he was tired of metals. He contamination, the water turned out to in the ground," says Martin. "You end wanted to switch to something new. be a lot purer. What's more, such metals up with a depression half a mile wide A colleague pointed out that his tim- as were in it were distributed in an and banks on either side. ing was all wrong. The early 1970s were orderly way. Before, when researchers "But when you blow off all these a time of swelling popular concern about took samples of the water column at underground shots, some radioactivity the effects of heavy metals in the envi- progressively greater depths, the con- is going to fall out on the sea surface. ronment. Caltech geochemist Clair Pat- centration of a metal would fluctuate And so they wanted to find out how terson, in particular, was warning about wildly and inexplicably. Now it followed much of this was going to be taken up the pervasiveness of lead pollution and a smooth profile, generally increasing by the food chain and eventually be- the dangers of lead poisoning. The toward the metal-rich mud on the bot- come a human health hazard. In the old Roman Empire had fallen, he said, not tom. Things made sense. days of atomic testing they used the to the Visigoths and the Huns, but to In the ensuing years Martin and his 'specific activity' approach, which said, lead poisoning. Prodded by his colleague group measured lots of different metals. the more of a stable element there is to at Hopkins, Martin read Patterson's pa- They measured manganese in seawater. dilute the radioactive element, the safer pers. They convinced him he should They measured mercury and cadmium you are. In other words, if you're a sea stay in the trace-metal business. When in sea lions, sea otters, and sewage. creature and you've got in the water an he moved to Moss Landing in 1972, he Those measurements were comparatively atom of radioactive zinc and a hundred set up a trace-metals lab. easy because there isn't much manga- atoms of stable zinc, the chances will be But Patterson proved to be a difficult nese, mercury, or cadmium floating a hundred to one that you'll take up the role model. "About the mid-seventies," around to contaminate a sample: Later stable element. So they wanted to know Martin recalls, "Patterson started telling Martin's group measured lead, which how much stable zinc was in the water, everybody, 'Your numbers are all bad. was harder; it was still being spewed and iron, and all these other things." You're contaminating your samples. out by cars running on leaded gasoline. Today Martin describes himself as an When you try to measure lead in water The biggest challenge of all, however, environmentalist, and he thinks nuclear or in plankton, everywhere you look or was iron. Martin had measured it in power plants are dangerous. But two touch is contaminated with lead that's Panama, but those numbers, he had decades ago the idea of nuclear excava- flowing out of exhaust pipes, and if you long since realized, were far too large. BY DEE LAMONT-DOHERTY OBSERVATORY tion didn't strike him as strange. "It could put on this special pair of glasses, Iron is highly insoluble in seawater; it didn't particularly bother me, I guess. you'd see lead all over the room.' And tends to glom on to organic particles Anyway, the bottom line was you have Patterson is enough of a strong individ- and sink to the bottom. So the amount PHOTOGRAPH GENE FELDMAN, a big mountain range in Panama, and if ual to go into a room full of scientists of iron in the water column had to be you knocked down the mountain, it and say, 'You're all full of crap!' Then very small. But nobody had been able 56 DISCOVER APRIL 1991 to say exactly how small, because con- them all the main nutrients-the ni- is a warmth to the man, although it tamination was almost impossible to trates and phosphates-they need to doesn't bubble. avoid. grow. Yet the nutrients go unused. The But Martin is also known for irritat- "Every time you get a speck of dirt, phytoplankton never bloom. That is, ing his peers. It is not just that he is that's five percent iron," says Martin. their population never explodes, as it competitive. He doesn't seem to feel he "And of course man uses iron for every- does, say, in the North Atlantic, where belongs to the club. "I'm not much of a thing. When you make a plastic bottle, the sea surface turns green each spring. scientist," he says. "I don't have a very it's usually extruded onto a stainless Why? In his talk that day Frost offered scientific mind-you know, math and steel form-there's this hot plastic flow- the conventional explanations: there is all that. My strength is imagination-I ing over stainless steel, and some of that not enough light in the Antarctic, he can imagine what's going on in the stainless steel is picked up in the plas- said; or it is too cold; or else the phyto- ocean. I guess I'm right-brained. Right- tic." Even the plastic bottles in the labo- brained people tend to get weeded out ratory, then, were not to be trusted as of science, and it's too bad. I knew a guy far as iron was concerned. at Rhode Island who was brilliant, but But after ten years of being he could never get his Ph.D. He just washed in nitric and hydro- wasn't able to concentrate long chloric acid-which is the enough on one thing. He procedure Martin's team would get bored." follows with all the im- Martin is impatient plements and even the with complex analytic walls in their all- arguments. He likes plastic lab-a filthy, simplicity. And he iron-infested bottle likes to needle peo- can get pretty ple. So after Frost clean. In a nut- finished explaining shell, that is the how Antarctic zoo- secret of how Mar- plankton could tin's colleague keep the phyto- Mike Gordon fi- plankton in check, nally succeeded, in Martin went up to the early 1980s, in him. "And more or measuring the first less in jest I said, reliable profile of iron 'Aw, baloney, it's just in seawater: he finally iron deficiency.' And he had a clean enough bot- didn't want to hear that. tle. By then Martin had re- It was too simple." tired from lab work. Reluc- Martin doesn't remember tantly, he had also quit going to what made him think of iron sea. Over the years the polio he deficiency. Other people had contracted in college in 1954 has made thought of it, too; the idea was in the him progressively less mobile, and now- air. Martin's wisecrack, though, was a adays he spends most of his time sitting kind of wishful boast-the wish being down, either in a wheelchair or behind Phytoplankton already bloom that the subject he had devoted his his desk. He has chronic pain. Gordon in the Antarctic, as this satellite career to might prove important in some and another colleague, Steve Fitzwater, image shows (red and yellow), broader context. Back at Moss Landing have taken over the hands-on work. It is but chiefly near the coast. he decided to try to deliver on his boast. slow, craftsmanlike work, and it requires He knew that phytoplankton, like dexterity. It is work that can be undone plankton are kept in check by tiny, every other living organism, need iron by a single errant dust grain. grazing animals, the zooplankton. to survive. Iron is an essential cog in the "So Mike showed me this iron pro- It was after Frost's talk that Martin cell's synthetic machinery. Phytoplank- file, casting pearls before swine," Martin first decided to shoot his mouth off. ton need iron to make chlorophyll, the recalls, "and I kind of ignored it." But molecule with which they capture sun- then in December 1986, at a conference PHOTOGRAPH COURTESY GENE FELDMAN, NASA/GSFC artin is well liked by those light. They need it to make nitrate re- in San Francisco, he heard Bruce Frost of the University of Washington give a M who know him well; his as- ductase, the enzyme with which they sistants are former students reduce nitrate to a form suitable for talk about an old problem in biological who have stayed with him for incorporation into proteins. Most funda- oceanography. In some regions of the years. The voice that sounds mental of all, they need it to make ocean, such as the Antarctic, the float- gravelly and short over the phone be- DNA, the stuff of their genes. So in ing single-cell plants, or phytoplankton, comes gravelly and friendly and even principle it was possible for a phyto- are notorious underachievers. Upwell- self-deprecating in person. The eyes twin- plankton population to be hobbled by a ing currents from the deep ocean bring kle, and now and then they smile. There lack of iron. DISCOVER APRIL 1991 57 no shortage of iron clean lab-whatever they do, they don't in the water. let it touch the deck, because it might "This is a real pain. But the winds pick up a few atoms of iron that it that blow over the would take back down on the next trip. It's on a rolling ship, Antarctic cross al- "This is a real pain in the ass," says most no land that Martin, who goes to sea vicariously. and maybe the deck isn't covered by ice. "This is hard. It's on a rolling ship, and So they drop on the maybe the deck is icy-it's terrible. But is icy-it's terrible. ocean almost no it guarantees a clean sample." dust and almost no With the clean samples they collected But it guarantees iron. In the Antarc- in the Gulf of Alaska, Martin's team did tic and in other a simple experiment. They measured a clean sample." places far from dry the amounts of iron and nitrate in the land, Martin rea- water and found that, according to the soned, phytoplank- best estimate of phytoplankton require- ton would have to ments, the plankton didn't have enough B Martin's problem, though, was that plank- make do with what little iron was al- iron to use all the nitrate. Then they ton get by on extremely small amounts ready in the water. And if Gordon's divided the samples into two groups. of iron. Just how small is not known measurements from the cruise. to Ha- One group they fertilized with iron; the precisely. But the best estimate is that waii were representative, that was very other they left alone. They watched the they use, at most, one atom of iron for little iron indeed-less than a millionth bottles for four and a half days. At the every 10,000 atoms of carbon, every of an ounce in a cubic yard of seawater. end of that period the bottles with the 1,500 atoms of nitrogen, and every 100 Even for a frugal plankter, that's just not added iron contained nearly nine times atoms of phosphorus. It didn't seem enough. as much chlorophyll-a measure of the likely, on the face of it, that there could The thing to do, then, was to see total mass of phytoplankton-than the be areas where the phytoplankton had whether Gordon's measurements held unfertilized bottles. The idea that phyto- plenty of the nutrients that they need up elsewhere in the ocean. In the sum- plankton may be limited by iron defi- plenty of but were unable to bloom for mer of 1987 Martin had a cruise sched- ciency had its first solid support. o lack of a few measly atoms of iron. uled to the Gulf of Alaska to make some This past year Martin's team went to p And yet Martin soon found evidence manganese measurements. But man- the Antarctic-"the big enchilada," as to back up his claim. It was buried ganese was old hat now. Martin's troops he puts it. The Southern Ocean is 8 among the debris on his chronically went after iron instead. They did the million square miles of nutrient-rich disorganized desk. Bob Duce, an atmo- same sort of experiments they've been water fed by vigorous upwelling cur- spheric chemist at the University of Rhode repeating in different places ever since. rents. Every year, during the six months Island and an expert on atmospheric On each of these cruises Gordon and of sunlight, phytoplankton bloom pro- C dust, had recently sent him a paper. In it Fitzwater bring along their own clean fusely in a narrow band along the Ant- Duce estimated that more than half the mini-lab-it's the size of a small house arctic coast. Those blooms support vast iron in the open ocean comes from dust trailer-and set it up on deck. They use herds of krill, shrimplike creatures that V settling out of the atmosphere, as op- special 30-liter sampling bottles that go are in turn the daily bread of seals, S posed to being upwelled from the deep through the sea surface closed (so the penguins, and whales. Away from the with the other nutrients. Then Martin inside won't get unearthed the iron measurements that coated with bilge Gordon had made on a cruise from water or fuel) and California to Hawaii. Gordon's numbers then snap open MICROGRAPHS BY DEE BREGER LAMONT-DOHERTY GEOLOGICAL OBSERVATORY. PHOTOGRAPH COURTESY JOHN MARTIN suggested that, if anything, Duce was automatically at 30 underestimating the importance of at- feet. The bottle is mospheric iron: there was so little iron lowered on Kevlar in the open ocean that about 95 percent line, because ordi- of the plankton's requirements must nary steel wire come from dust. might contaminate The circuit closed in Martin's mind: the water around Where there's no dust, there's no iron. the bottle. And Except for shallow coastal waters, where when the bottle the sea surface is close to the iron-rich comes back up, mud on the bottom, the distribution of now weighing more iron in the ocean is controlled by the than 100 pounds, geometry of winds and landmasses, par- Gordon and Fitzwa- ticularly arid landmasses. In the equato- ter either load it rial Atlantic, for example, winds sweep- onto a plastic cart ing off the Sahara carry dust as far west or manhandle it di- as Barbados. All along that path there is rectly over to the Far from the Antarctic coast diatoms are sparse 58 DISCOVER APRIL 1991 coast, however, the Southern Ocean is a back-of-the-enve- much less fertile. The major nutrients lope calculation. are there, the temperature and sunlight If 300,000 tons are similar, but the plankton don't bloom. of iron were broad- Gordon and Fitzwater did iron- cast in the ocean fertilization experiments in the Ross Sea, surrounding Ant- near the McMurdo base. They found arctica over a six- that adding iron to a bottle of seawater month growing sea- taken from near the shore didn't help son, he figured, it the phytoplankton much. That made would allow the sense; there was already plenty of iron phytoplankton to in the water. It came from the bottom convert all the avail- mud, and from atmospheric dust that able nutrients-in- had fallen on the glaciers over many cluding 2 billion years and then got dumped in the ocean. tons of carbon- But a few hundred miles offshore the into new organic story was different. When Gordon and matter. Those 2 bil- Fitzwater added iron to a seawater sam- lion tons would ple there, they got a fourfold increase in come from carbon the amount of organic matter in the dioxide that had dis- On the Antarctic cruise Martin's clean lab got a little chilly. bottle. "There was hardly any growth solved out of the without iron," says Martin. "And with it atmosphere into the surface waters. In with 280 parts per million in the nine- the stuff really took off." other words, they would come right out teenth century and 350 today. The same Now the evidence for Martin's iron- of our greenhouse. ice cores show that the amount of dust limitation hypothesis was strong. And Much of that carbon, Martin assumed, settling onto the Antarctic-and hence in the world of biological oceanography, would sink to the bottom of the ocean- the amount of up to 50 times this was big news-a new paradigm, in the form of plankton corpses or plank- greater during the Ice Age. The world one oceanographer called it. But a new ton parts or fecal pellets excreted by the was drier then, with vast deserts and paradigm about phytoplankton was things that eat phytoplankton. Oceanog- stronger winds that carried dust out to hardly enough to get Martin on televi- raphers know this carbon sinking goes sea. All that iron-rich dust, Martin sion. There was one more step to take. on all over the world; they call it the claims, may have led to massive plank- It was a logical step, and actually biological pump. They don't really know ton blooms in the Antarctic, which Martin had been referring to it all along how important it is. But they think that pumped carbon out of the atmosphere, in his papers. If adding iron to a bottle right now it may take as much as 3 which made the world colder. causes the phytoplankton in the bottle billion tons of carbon each year-half of If Martin is right, the amount of iron to bloom, then you ought to be able to all the carbon we pump into the atmo- in seawater-something he spent more fertilize the ocean itself. It wouldn't take sphere annually with our power plants than a decade learning how to measure very much iron to fertilize the whole and cars-and put it out of harm's way correctly-has a decisive impact on Southern Ocean, because phytoplank- in the deep ocean. All Martin suggests is Earth's climate. "You give me half a ton don't need very much. Martin did that it might be possible to speed up the tanker full of iron," Martin said this past pump in places May, shooting off his mouth again, this where, for lack of time to a Washington Post reporter, iron, it isn't run- "and I'll give you another ice age." ning up to speed. The pump may artin wasn't recommending an- have worked faster bles of ancient air M other ice age, of course, and before, he says. Bub- he doesn't truly think he could deliver one. He was just try- extracted from Ant- ing to get a point across. "I arctic ice cores re- don't waffle," he says. But for quite a veal that the car- while after that remark he almost bon dioxide level in wished he did. the atmosphere has Martin's idea has been attacked from fluctuated with the all angles. If there were huge blooms of advance and retreat phytoplankton in the Antarctic during of the ice sheets. the Ice Age, for example, some evidence During the Ice Age of them ought to be preserved in sedi- the carbon dioxide ments on the ocean floor. The dominant concentration was type of phytoplankton today are dia- only 200 parts per toms, which have hard shells of opal. but when iron is added to the water they multiply. million, compared This past year researchers at the Lamont- DISCOVER APRIL 1991 59 millionths of an Martin responds, somewhat testily, "I don't want ounce per square that he has in fact seen copepods hap- yard, and a steady pily grazing in his test bottles. But the supply would some- to go down in whole argument irritates him. "They how have to be take a very simple thing and make it maintained in the history as Martin's complex," he says of these critics. "And face of its. natural if you get into complexities, you can tendency to sink Mistake-the guy make all these arguments. If you keep it out of the surface simple, you just say, 'Hey, I got two waters. Martin who ruined the bottles, right, and in this one I'm going speaks vaguely of a to put a little bit of iron and in this one supertanker with a ecosystem." I'm not. And I'm going to sit here and huge fan on deck; watch it for a week and see what hap- iron dust coming pens.' And at the end of the week you up from the hold hold it up to the light and you can see on a conveyor belt all these glinting cells. And this one over Doherty Geological Observatory looked would be cast by the fan into the stiff here, the one without iron, has nothing for opal in sediment cores from the Antarctic winds. Alternatively, he says, in it. To my way of thinking you have a Southern Ocean. They found plenty in perhaps some sort of time-release pel- very dramatic effect." recent sediments, but in Ice Age sedi- lets might work. No one really knows. Martin would like to stop arguing. He ments they found almost none-just No one knows either what effect the would like to start preparing for what he the opposite of what Martin would pre- iron would have on the Antarctic envi- considers the ultimate test of his hy- dict. "This is one of the big flaws in my ronment. The ocean would not turn to pothesis: fertilizing a small square of hypothesis," says Martin. Perhaps, he green mush, but iron fertilization might ocean, perhaps 60 miles on each side. It says, the Ice Age phytoplankton were favor some species over others and would take a couple of years, as he sees mostly soft-bodied species that don't thereby alter the food web in unforesee- it, to thoroughly research the test area, fossilize, rather than diatoms. able ways. Martin thinks the simplest to map its currents and its weather Then there is the question of whether outcome is the most likely one: the iron patterns and its patterns of life. Only Martin's back-of-the-envelope calcula- would produce more diatoms and more then would the crop dusters or military tions are at all accurate-whether iron krill, which eat diatoms, and thus more cargo planes come in and start spraying fertilization would get rid of as much seals, penguins, and whales. ("When I iron. Meanwhile squads of oceanogra- atmospheric carbon dioxide as he get real facetious, I say I'm not inter- phers would be waiting on ships to your thinks. The Antarctic upwelling currents, ested in the CO₂ problem, I'm interested monitor the effects. it has been suggested, might keep most in repopulating the world's whale The experiment would not come of the plankton-fixed carbon from sink- stocks.") But it's also possible the iron cheap-it would certainly cost tens of ing; then the carbon would be quickly would yield blooms of organisms that millions of dollars, and perhaps more recycled into the atmosphere by bacte- krill don't eat and that would outcom- than a hundred million: The funding ria, which break down organic matter. If pete the diatoms. In that case the krill- agencies are wary. But in spite of all the that is true, then dumping iron in the based food web would collapse. flak Martin has taken, support for such Antarctic would have only a negligible Martin thinks all these concerns are a test seems to be growing among his impact on the greenhouse effect. People valid. They are not his area of expertise, fellow oceanographers. In the end his who specialize in making theoretical and he doesn't pretend to have the an- data are hard to ignore. models of the ocean are now arguing swers. He is not suggesting, he stresses Yet in the end there is also an argu- the point. Martin is not a modeler. He again and again, that we dump iron in ment against iron fertilization that no prefers to let them duke it out. the ocean now. He is only arguing for amount of data can answer. There is The modelers do agree on two things, focused research into the possibility. something fundamentally wrong, some however. One is that iron fertilization The one criticism that gets under his people would say, with the whole idea would at best be a palliative-never a skin comes from people who say he that we should attempt to "fix" nature cure-for the greenhouse effect. Even if hasn't really shown that phytoplankton on so grand a scale. Wrong because we it worked perfectly, it would remove are limited by iron-from people, in will surely foul it up; wrong because it less carbon from the atmosphere each other words, who question his data or is not our place. Does Martin not com- year than we are now putting in, not to his interpretation of the data. Perhaps, mit the sin of pride? mention what we have put in already. these people argue, adding iron to a "We're already involved in the big- The other point of agreement is that bottle of seawater doesn't really make gest experiment ever," Martin responds. iron fertilization would not be a one- the phytoplankton grow better. Perhaps "We're finding out what's going to hap- shot deal; it would have to be repeated, the bottle just didn't happen to include pen if we dump three billion tons of year after year, until we stop burning any of the large zooplankton, called CO₂ into the atmosphere every year. fossil fuels. Nor would it be just a mat- copepods, that graze on the plants and That is the biggest manipulation of the MICROGRAPH BY DEE GEOLOGICAL OBSERVATORY ter of pushing a few rusting cars over keep their population in check. With no environment ever. What I argue is we're the side of a ship. The iron would have grazers, it would look like the plants going to keep doing this. And we'd to be broadcast extremely thinly, a few had bloomed. better know about ways, if we have to, 60 DISCOVER APRIL 1991 Over the past 160,000 years, whenever the amount of iron-rich dust falling onto the Antarctic has gone Age of ice up, the amount of (thousands of years) carbon dioxide in the 20 40 60 80 100 120 140 160 atmosphere has gone down, and vice versa. The data come from 300 100 CO a 7,000-foot-deep Iron Antarctic ice core; the deeper the CO concentration (parts per million) 75 250 ice, the older it is. The iron peaks 50 Iron concentration (parts per billion) at about 18,000 and 160,000 years ago 200 correspond to ice- 25 age peaks. Martin thinks dust falling out of the atmosphere 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 caused Antarctic Depth of ice phytoplankton to (feet) bloom, taking up carbon dioxide and cooling the planet. to take CO₂ out of the atmosphere. you could go down there and add the Liebig, the nineteenth-century German "The Chinese are not burning their iron and not disturb the environment, chemist. Liebig discovered the law of coal right now, but they have massive then I'd say go ahead and do it." the minimum, which states that a plant reserves. As they industrialize they're will not grow unless it has its minimum going to want to use that coal. They're artin hasn't been spending requirement of each essential nutrient. not going to say, 'Gee, gang, we're going M much time at Moss Landing Liebig's law has always fascinated to use solar power so we don't put out lately. The earthquake in Oc- Martin. Now he thinks he has discov- greenhouse gases.' The average Chinese tober 1989 destroyed his lab- ered the first marine example of it, and peasant wants to have electricity and a oratories-pulled the rug right he would like very much to prove it refrigerator and maybe a motorbike- out from under the low stucco building, conclusively. The iron-fertilization idea, these people want to have the same damaging it so heavily it had to be he hopes, will help convince the public, crap we've got, and they're going to use condemned. Martin and most of the his peers, and the funding agencies that what they've got available. CO₂ is not other researchers have moved to Sali- the work is important. But Martin ad- going to go down next year, or level off nas, into temporary quarters provided mits that it makes him a little nervous. next year. It's going to go up by at least by San Jose State University. But Gor- "It's scary," he says. "I don't want to three billion more tons, and the next don is still at Moss Landing, doing iron go down in history as Martin's Mistake- year three billion more and probably analyses in the shadow of PG&E; he's as this guy who advocated adding iron, five. So by sixty years from now, you set up a clean lab in a trailer with and they did it, and it completely ruined may be going up ten or fifteen or twenty equipment salvaged from the ruins. "We the ecosystem. What I want is to do the billion tons a year. The purists say iron were very lucky," says Martin. "No one ocean experiment, and prove that iron PHOTOGRAPH BY BRIAN GRAPH DATA PROVIDED BY JOHN MARTIN fertilization is crap, and we shouldn't was hurt, and we didn't lose any of our is limiting-and then retire. even think about it-we should be con- equipment or samples. You lose your "When you work with trace elements, serving. That's very true. But what if we data and you're out of business." you'd like to think they're really impor- get into a situation where global warm- It's now three decades since Martin tant. But then you say, 'Who are you ing really does occur? first got interested in trace metals. That kidding-they're not important.' These "I'd see an ethical problem if we was back in graduate school, when he are the conversations you have with could go down and fertilize the Antarc- read about sheep in Australia whose yourself. And then, all of a sudden: tic and remove the CO2, but in the wool fell off because of vitamin B₁₂ 'Jesus! It looks like it is important. What process we would kill all the whales and deficiency. The deficiency was traced to the hell do I do now?" the penguins. Then I'd say, 'Don't do it. a lack of cobalt, and it was cured by Man has created this, he can stew in his adding tiny amounts of cobalt to each Senior editor Robert Kunzig wrote own juice. Leave the whales alone.' But acre of pasture. It was also back then "The Man Who Weighs Dinosaurs" on the other hand, if it turned out that that Martin learned about Justus von in the October issue. DISCOVER APRIL 1991 61 not global April 22, 1991 In my opinion, Discover Magazine 3 Park Avenue New York, N.Y. 10016 To the Editor: John Martin proposes iron fertilization of the Southern Ocean as at least a partial solution to the threat of global warming. ("Earth On Ice", April 1991) He's got it exactly wrong. Our manhandling of the environment created the problems we face; it in no -way- is part of the solution. It would be easy to attack Martin's suggestion on technical grounds: There is evidence that iron is not the limiting nutrient in the oceans; very recent research shows that, even if adding iron would increase plankton populations, the net amount of carbon sequestered would be modest; massive amounts of money and other resources would have to be mustered if the program were to be even marginally successful in achieving its objectives. hubris fundamental But my opposition to Martin's suggestion runs much deeper. The proposal is dangerous because it plays to wishful thinking that there is no need for change; and that, with a little creative engineering, we can out-smart nature once again. The truth, however, is that we cannot, without serious repercussions, continue our campaign of re- creating the environment. in the design of one civilization; You would think we would have learned that lesson by now. This is certainly not the first time we have undertaken such technological "fixes". Our determination to produce crops in areas not suited to produce them, for example, has poisoned the soil and, in some areas, created vast deserts of salt and pesticides. And consider our desire to adorn ourselves cheaply in silk by bringing gypsy moths into the country vast stands of diseased and dying trees are the legacy of that venture. The kudzu vine that chokes native plants and trees in Thinks the Southeast; the rabbits that overyran Australia; and the brown illegal trout that threatened the entire Yellowstone Lake ecosystem, are all testimony to our inability to gauge and control the effects of our grand eco-engineering schemes. And those were nothing compared to this proformal. It is time we stop kidding ourselves. The simple truth is that it is only by changing our behavior -- and not by further changing our on environment that we can effectively address the global can relation environmental problems we face. Study after study tells us we/make to make significant gains at minimal cost with existing technologies. The eart need for such action is clear. The ozone layer is being depleted at an alarming rate; we need to move quickly to end the use ozone destroying chemicals. We continue to experience record warmth, severe drought, interduction of Africarized has of the web of warning with to globalin and change to extensive forest fires, and intense storm activity; We we need cut our emissions of greenhouse gases in every sector of the economy. The very threads of life are being severed as rampant deforestation drives thousands of species to extinction; we need to preserve and protect sensitive ecosystems and manage our forests sustainably. So let's get on with it. I challenge our scientists to use their insights and creativity to devise ways for us to live in harmony with nature -- rather than speeding us further along the collision course that we have already begun to travel. are currently travelling on