Nature © Macmillan Publishers Ltd 1998

8

Powerful modes of computer modellingallow experimental strategies involving largedata sets to be devised and quickly refined.The Internet puts communities of scientistsin instant communication with one another.Synchrotron radiation sources with powersmany orders of magnitude greater thanwhen X-rays were discovered allow views ofunprecedented precision into materials.

Nevertheless, the problems of climatechange, flooding, pollution and energyshortage are all too familiar in many parts ofthe world, while strategies for dealing withthem often remain obscure. Research chal-lenges at the most fundamental level awaitthe efforts of geoscientists. For example,Miriam Kastner of the Scripps Institution ofOceanography in La Jolla, California, pointsout that we did not know of the existence ofsubsea vents until 1980. Since then we havediscovered that the entire volume of theoceans cycles through them once every 5 mil-lion years, and through the subduction zonesonce every 200 million years.

“As we learn more about the interactionof the solid Earth and the hydrologic cyclewe’ll understand more about earthquakes,and possibly will be able to predict fine grainchanges over time such as the Little Ice Age,”she says. She adds that already we have foundthat some classic models of climate and geo-chemistry are wrong and must be revised.But this can only be done if scientists fromdisparate disciplines cooperate.

Geoscientists also have a vital contribu-tion to make to human health in the area ofenvironmental toxins and pollutants. Thereis an urgent need for more knowledge of dustparticle size and chemistry, aerosol types,and the bioavailability of heavy metals, phos-phate, arsenic and other substances. Hereagain, scientists must communicate not justacross specialisms but across jurisdictions,because atmospheric pollutants are a globalphenomenon. That the geoscience and med-ical communities have historically been dis-connected from one another shows how“research is Balkanized,” says GordonBrown, of Stanford University’s SynchrotronRadiation Laboratory. He suggests that per-haps the toxicologist is the logical interfacethrough which this interaction will occur.

Charlie Alpers of the US Geological Sur-vey in Sacramento, California, says that clearcommunication to outsiders about researchproblems and findings is also crucial. “Weneed to communicate clearly with the public,and persuade them and their representativesthat what we are doing is worthwhile. Ulti-mately, public agencies and individual liveli-hoods depend on it.” He says he will empha-size the human welfare aspect of what hedoes in a forthcoming short course in sul-phate mineralogy, just as his agency accom-panies many of the reports it publishes withclear fact sheets to make the work compre-hensible to non-specialists, such as students.

Brown seconds this idea, saying that sci-entists ought to be able to make society careabout their work. “Who cares about the dis-tance between atoms?” he was recently askedby a political candidate in California. He gavean example of how the 0.4 Å differencebetween the radii of Cr6& and Cr3& quite pos-sibly allows the latter to mimic phosphate inlife systems, with potentially lethal results.He added that it is important from a publichealth perspective to accurately characterizethe bioavailability of different metals.

Although the interdisciplinary approachis mandatory for effective science and policy,it can take getting used to, admits Joe Smith,of the Department of Geophysical Sciencesat the University of Chicago. Fifteen yearsago the American Geophysical Union’s pub-lic affairs committee asked him to organize ascientific session on the ‘nuclear winter’issue. Once involved, he was appalled tolearn of the magnitude of the problem thatscience had helped to create, and it was at thistime that he began to alter his view of scienceand the responsibilities of the scientist.

Especially now, when the Earth sciencesare expanding so greatly in concepts andinformation density, he believes it is impor-tant for the Earth scientist to “think out theinterconnections between what he or shedoes and the various pieces of the tapestry ofhuman life”. Even though it may be easier towork on idealized problems, he has come torealize that members of the US NationalAcademy of Sciences like himself have a dutyto use their expertise for human welfare.

The cost of future research may be con-siderable by the historical standards of geo-science. For example, the current third gen-eration of synchrotron radiation sourcescosts about $15 million per cluster of experi-mental workstations, each of which shouldbe manned by two dozen scientists and engi-neers, says Smith. Available funds are too

scanty to serve the growing number of envi-ronmental geoscientists who require datafrom these instruments.

As relative latecomers to the sciencefunding competition, Earth scientists face astruggle to win adequate support. Smith ispreparing a presentation to the NationalAcademy of Sciences to raise $100 millionover five years to support staff at synchrotronradiation centres, with matching amounts tosupport research, teaching and public out-reach at universities and experimental sta-tions. He says this sum is a tiny fraction of themoney spent on nuclear weapons manage-ment, and is in the range of the cheapestrocket mission to an asteroid.Potter Wickware is a science writer in Oakland,California, USA.e-mail: wickware@worldnet.att.net

A risk propositionBrendan Horton

A geoscientist’s career can veer off academicand government tracks into involvement inindustry and even insurance. According toJohn Mutter, deputy director of ColumbiaUniversity’s Lamont–Doherty Earth Obser-vatory, there are concerns to be aware ofwhen too much of your funding comes froman industry–academic collaboration, espe-cially for researchers intent on landing a rareacademic post. But many scientists arediversifying their CVs to leave open employ-ment options outside the academic world.

“More and more people are building infunding from the private sector,” says Mut-ter, not only because it diversifies one’sfinancial support, but also because peoplesee the potential for a career in the privatesector. While Lamont’s partnerships havetraditionally been dominated by the oilindustry, “if we spoke two or three yearsfrom now, I’d [probably] be telling youabout valuing climate information”, saysMutter. With few exceptions, the industriesmost affected by climate have no researchculture to enable them to understand it.

The winds of change that blew aroundChris Barton, a research geologist with theUS Geological Survey since 1985, may nothave been hurricane force, but they werestrong enough to make him a G. K. GilbertFellow at IBM with Benoit Mandelbrot, the‘father of fractals’. He followed this up withanother fellowship at Lamont. These fel-lowships helped Barton’s research to evolvefrom studying the physical phenomena ofclimate catastrophes to studying the magni-tude of financial losses from these events,which in turn has affected the way reinsur-ance companies and the Federal EmergencyManagement Association evaluate risk.(Reinsurers insure major corporations andprimary insurance providers.)

careers and recruitment

494 NATURE | VOL 396 | 3 DECEMBER 1998 | www.nature.com

Further information on geoscience careers

GeoWeb Interactive: Jobs forGeoScience/Engineeringhttp://www.ggrweb.com/job.html

Earthworks http://ourworld.compuserve.com/homepages/eworks/

USGS Main homepage http://www.usgs.gov/

USGS Mineral Resourceshttp://minerals.er.usgs.gov/

NAS Geoscienceshttp://www.nsf.gov/home/geo/start.htm

AGI Careers in Geoscienceshttp://www.agiweb.org/career/carehome.html

Graduate School and the Job Marketof the 1990shttp://www.agu.org/eos_elec/97117e.html

The Lamont-Doherty Earth Observatoryhttp://www.ldeo.columbia.edu/

The Columbia Earth Institutehttp://www.earthinstitute.columbia.edu/

Alternative Careers in Science, edited byCynthia Robbins-Roth http://www.amazon.com

Nature © Macmillan Publishers Ltd 1998

8

According to Barton, the statistics of theinsurance industry have been well under-stood for almost 100 years. “It’s the law oflarge numbers”, which follows a gaussiancurve in terms of losses. He explains thatwith these natural events the mathematicsfollow a power law or fractal. Until recently,he adds, the use of large-number statisticsmeant the insurance industry was tremen-dously undercapitalized for the actual risk.

What has impressed Barton is that justas you can plot the strength of an event —be it the magnitude of an earthquake or theflood-discharge rate of a river — so you canplot dollar losses associated with each event(provided you have 50–100 years of data).“This behaviour was well understood fornatural hazards, but no one had ever triedto do this for the property losses.”

Hedge against catastropheGraciela Chichilnisky is a mathematicianand an economist at Columbia University.She is also director of the Program onInformation and Resources’ Risk Center atthe Columbia Earth Institute. Her interestin risk began in the early 1990s when shesaw how catastrophic events such as hurri-cane Andrew had shaken up the insuranceindustry. Knowing that hurricane frequencyis related to the El Niño Southern Oscilla-tion (ENSO) and with her knowledge ofchaotic, dynamical systems, Chichilniskybegan to assess the risks from natural geo-physical events, developing a theory of howthese risks could be classified. She designedfinancial instruments that could effectivelyhedge against catastrophic events, which bytheir nature cannot be predicted. One resultof her work is an instrument that she callsCatastrophe Bundles, a portion of which,the El Niño Index, is being commercializedby the Bermuda Stock Exchange in cooper-ation with Columbia University.

Chichilnisky’s Risk Center operates inthe grey area between publicly supportedacademic research and private interests, andallows the group’s intellectual property tobe commercialized. According to Chichil-nisky, this set-up should provide betterfinancial cover for the insurance industry,which may allow more companies to offerhurricane insurance at lower rates.

Some reinsurers, notably Swiss Re and

Munich Re, have built substantial in-houseresearch groups that include scientists hiredfrom academia. “I’m astonished to findmyself working for a reinsurance company,”says Jürg Trüb, head of the AtmosphericPerils Group at Swiss Re. “However, Iquickly learned that the insurance industryand, in particular, reinsurers need highlyspecialized geoscientists to analyse the risksof natural catastrophes.”

About one-third of his group’s work is inresearch and development. “Our work ismore ‘purpose-orientated’ and less in-depth research,” he says. The group adaptsand interprets the latest results in geo-science for the insurance industry. It mustexamine how weather forecasting andENSO could affect underwriting, identifynatural catastrophes that the company andits clients may be exposed to, and assesswhether the group thoroughly understandsand can manage these risks.

Beyond the academic sphereSince being involved in writing “Climatescience and insurance risk” (Nature 389,225–227; 1997), Anthony Michaels hasfound that much of his work is in givingcareer advice. As director of the WrigleyInstitute for Environmental Studies at theUniversity of Southern California, Michaelsheads a programme that will cross the uni-versity’s departmental boundaries, linkingscience, engineering, medicine and business

to the technical assessment of risk, returnand uncertainty in the insurance, bankingand other risk-related industries.

Michaels says they aim to train studentsin the degree programme “so that theycome out with something that has relevancebeyond the academic sphere. This is a fertilearea for climate and Earth scientists whohave some modelling background, goodmaths skills and an interest in a businessapplication of their background.”

Wrigley offers study programmes notonly in environment-related risks, such ashurricanes and earthquakes, but also in thepollution and health risks that affect health-and business-liability decisions on bothregional and global scales.

The Sloan Foundation has providedfunding to create two technical masters’degrees: one merges physics and businesswhereas the other is a masters’ degree inenvironmental risk. The National ScienceFoundation is also funding PhD pro-grammes in environmental science andengineering, one of which is in environ-mental risk assessment, says Michaels.

“The other part of the programme is theRisk Center that we’re putting together.”According to Michaels, this is still in theplanning stage. The centre is structured toenhance the interdisciplinary connectionswithin the university, so that faculty-levelprojects cross the spectrum of environmen-tal risk, and involve the College of Letters,Arts and Sciences, the Marshall School ofBusiness, the School of Engineering and theMedical School. These projects will look atnatural systems, the risks to human activityand how they affect business decisions.

Turning back the floodCynthia Rosenzweig, from the US spaceagency NASA’s Goddard Institute for SpaceStudies in New York, is involved in interdis-ciplinary programmes that have capturedprivate and public interest, as well as theirfunding. “After years of people paying lip-

careers and recruitment

NATURE | VOL 396 | 3 DECEMBER 1998 | www.nature.com 495

Further reading on climate risk, insurance and finance

A Calculus of Risk http://www.scientificamerican.com/1998/0598issue/0598stix.html

Christopher Barton at USGS http://coastal.er.usgs.gov/hurricane_forecast/barton2.html

The Risk Prediction Initiative, Bermuda Biological Station http://www.bbsr.edu/agcihome/rpi/rpihome.html

Columbia’s Program on Information Resources http://www.earthinstitute.columbia.edu/units_partnerships/pir/index.html

The USC Wrigley Institute for Environmental Studies http://wrigley.usc.edu/buss.shtml

US National Assessment http://www.nacc.usgcrp.gov/

Munich Reinsurance http://www.munichre.com/

Swiss Reinsurance Company http://www.swissre.com/

Since the 1960s,economic lossesfrom naturaldisasters on a globalscale have tripled,while insured losseshave risen fivefold(after Berz, NaturalHazards 5, 95–102;1992).

19651960 1970 1975 1980 1985 199030

25

20

15

10

5

0

Year

Bill

ions

of U.S

. dol

lars

(19

90 v

alue

)

Total economic lossesAverage economic losses per decadeTotal insured lossesAverage insured lossesper decade

CH

RIS

BA

RT

ON

, USG

S

Nature © Macmillan Publishers Ltd 1998

8

service to interdisciplinary work, it is reallyhappening now,” says Rosenzweig. She leadsthe Metropolitan East Coast Region for theUS National Assessment, which involvesteams of researchers and stakeholders fromsuch coastal zone sectors as infrastructure,water resources, human health and institu-tional decision-making. The NationalAssessment programme will study theregion’s vulnerability to climate change.The Metropolitan East Coast is crucial tothe assessment, says Rosenzweig, because“we are focusing on the role and response oflarge cities such as New York”.

Rosenzweig also heads an interdiscipli-nary science team involved in predicting theeffect of El Niño on agricultural and marinesystems as they relate to food supply. On theprivate side, she works with a consortium offour major agri-businesses, led by PioneerHy-Bred International, to evaluate howregions of interest around the world will beaffected by climate change.

Water and welfare:hydrology optionsPotter Wickware

Water supply and quality issues worldwidegive hydrologists a broad range of problemsand opportunities to occupy them in theyears ahead. Ismail Serageldin, vice-presi-dent for special programmes at the WorldBank and chairman of the World Commis-sion on Water for the 21st Century, says thatexisting technology can be applied forimmediate benefit in developed and devel-oping countries, for example in the reuse ofminimally treated water, harvesting rainwa-ter and preventing leaks. At the same time,researchers can work towards economicdesalinization methods, discovery of deepaquifers by remote sensing, engineered cropsfor maximum water utilization in aridregions, and other advanced technologies.

Some areas of need are going completelyunaddressed, he says. For example, moreneeds to be done to mitigate the effects ofmonsoon flooding and the associated prob-lems of wastewater flushing and water quali-ty. But little research is taking place anywherein the world in this important area. Basic sci-ence is also needed to generate adequatebaseline data to make accurate predictionsand decisions. “Most environmental statis-tics at present are unreliable or incomplete,and projections are also bedevilled by the dif-ficulty of modelling interactions among sec-tors,” he says.

Indeed, “why must we rely on water as themeans of sanitation, where water and humanwaste are dumped together, creating majorproblems and requiring treatments whichmost developing countries cannot afford?”

Serageldin asks. “Scientists should explorehow to cut the link between water and sanita-tion, and thereby dramatically reduce waterdemand.”

Hydrologists’ knowledge also plays anincreasingly important role in land use deci-sions, says William K. Michener, at theJoseph W. Jones Ecological Research Centerin Newton, Georgia. In the United States, atrend in favour of managed flooding andrestoration of wetlands can be seen in theClinton administration’s proposed CleanWater Initiative, which aims for a netincrease of 100,000 wetland acres per year.This will be paid for with $2.3 billion in newwater-related funding over five years.

The great Midwestern floods of 1993 and1995–96 taught the lesson that managedflooding is a more sensible strategy than theflood suppression attempts of the past, saysMichener. He points out that the licences ofmany older flood control dams are comingup for renewal by the US Department ofEnergy. Each licence review provides anopportunity to establish minimum waterflow levels and create flow regimes that moreclosely mimic natural ones. In some casesdams are dismantled altogether, leading tolarge-scale habitat restoration, as with the161-year-old Edwards Dam on Maine’s Ken-nebec River and similar proposals forstretches of Idaho’s Snake River. In 1993, 160licences affecting 262 dams on 105 riversexpired; half of these relicensing actions arestill under way. Some 550 more dams are duefor relicensing in the next 15 years.

Finally, hydrologists’ expertise is calledon for policy recommendations as effortsgain momentum to strengthen the CleanWater Act, reorganize the activities of the USArmy Corps of Engineers, and reframe thefederal flood insurance programme.

The after-effects of mining present a quitedifferent sort of problem and opportunityfor hydrologists. Kirk Nordstrom, a hydrolo-gist with the US Geological Survey in Boul-der, Colorado, describes the Iron MountainSuperfund site, where groundwater perco-lates through faults, saturating metal-sulphide formations and generating copiousacid drainage. The lack of buffering potentialin the surrounding matrix and the vigoroussulphide- and metal-oxidizing activity ofacid-adapted soil bacteria produce a witches’brew of concentrated heavy metals in a nega-tive pH solution upstream of fisheries, popu-lation centres and agricultural land in Cali-fornia’s upper Sacramento Valley. Althoughthe US Environmental Protection Agencyhas the situation in hand for the time being, apermanently engineered solution based onsound geological and hydrological knowl-edge has yet to be proposed.

Water supply and quality problems afflictmany parts of the world, but the same sets ofmultidisciplinary technical, social and polit-ical skills are needed everywhere to effective-

ly manage them. D. M. Smith, an environ-mental plant physiologist with the Instituteof Hydrology in Oxford, England, collabo-rates with scientists in Kenya to examinebelow groundwater interactions betweentrees and crops. Chiefly motivated by theproblem of land degradation and desertifi-cation, Smith says he tries to develop systemsof land use that balance exploitation ofresources, including water, with conserva-tion. This reduces to a problem of engineer-ing combinations of trees and crops that fit aparticular locality. But, after factoring in thesocial, cultural and economic issues, climatechange and high rates of population growth,the problems become complex. “As a conse-quence,” he says, “hydrologists working inthis field will be most effective when theywork together with, for example, agrono-mists, economists and anthropologists.”

He also stresses that the work must beparticipatory, with farmers and scientistsworking together. When he worked in Niger,West Africa, Smith says he would sit under ashady tree with groups of farmers to decideon experiments that could be done in theirfields. “The advantages of collaboration overprescription are plain,” he says. “Whensomething works, the farmers will have beeninstrumental in making it happen, will haveassessed the risks inherent in changing theirpractices and will therefore be more likely toadopt the measure permanently.”

Smith adds the qualification that,although the community-based approach iseffective for adaptive research, it is lessstraightforward for process-oriented re-search. For example, the investigation ofmicrometeorological control of soil evapora-tion is difficult with participatory methods.This makes it harder to obtain funding forthis type of research in developing countries.

Funding, of course, is critical to gettingthe job done at all. Just as water is maldistrib-uted in the world, so is money for water-related research. The United Kingdom cansupport research by a couple of hundred sci-entists at its Institute of Hydrology. But acountry such as Niger has scanty financialresources, although it has the scientists,trained locally and overseas. The solution, inSmith’s view, is to “twin” research institutesin the north and south. Scientists in develop-ing countries would be provided with fund-ing, training and access to state-of-the-arttechniques and instruments.

“I believe there is scope for twinningresearch institutes, rather as towns aretwinned across Europe — between Englandand Romania, for example. Such a schemewould be particularly apt for environmentalscience, because scientists from the north andsouth face similar issues. Ideally, twinningwould come with a long-term source of fund-ing for training and reciprocal visits to pro-mote collaboration, perhaps funded by theUnited Nations or World Bank,” he says.

careers and recruitment

496 NATURE | VOL 396 | 3 DECEMBER 1998 | www.nature.com